Energy leaves the price of everything

The capability we assume

Assume the fuel was always already there.

Three funded programmes are building that device, and on the bench they look like three different machines. Work through the frames below and they resolve into one sentence — and then, in the last two frames, into the four ways they are still not the same thing.

Empty space is not empty. It is a structured medium carrying a real energy density, and not a small one: Eric Davis's comparison table puts the electromagnetic zero-point energy, integrated to the Planck scale, at about 10¹¹³ joules per cubic metre. Charles Chase, formerly of the Skunk Works revolutionary-technologies division and now chief technology officer of the UnLab, states it plainly: there is actually some structure in the vacuum.

That is not a fringe position. Dirac abandoned the empty vacuum for a filled sea, T. D. Lee coined the phrase vacuum engineering, and Frank Wilczek calls the same object the Grid. Nor is the field merely inferred: the Casimir force was confirmed to about five percent by Steve Lamoreaux in 1997 and to about one percent by Mohideen and Roy in 1998, in different laboratories. And in 2011 Wilson and colleagues published in Nature the result that ought to lead every discussion of this subject — a boundary modulated fast enough turns virtual photons into real, detected photons. Light out of nothing, peer-reviewed, reproduced since.

So assume a sealed device that draws on it. The governing rule is stated most precisely by the man building one. Chase: our physical theories say you cannot do anything with the vacuum — you cannot generate a current, you cannot generate a force — unless you break symmetry or you have a non-equilibrium situation. That is not an obstacle; it is the design specification, and every serious programme works inside it by engineering the asymmetry deliberately. Garret Moddel at Colorado Boulder uses an asymmetric optical and Casimir resonator: one cavity with more modes, another with fewer, a pressure difference between them, and a flow of electrons through a circuit. Harold White uses geometry, making the vacuum pressure non-uniform inside a Casimir power cell. Chase uses an asymmetric resonant tunnelling diode, where the fluctuations acquire a preferred direction. His own summary: all three are the same idea.

Assume fusion is the first door through. Drive the vacuum around the fuel out of balance and you raise the rate at which nuclei tunnel through the Coulomb barrier, which stops fusion being a confinement problem and makes it an electrostatic one. Lowering the barrier itself is the hope, not yet the claim. That is what a metal lattice already does through electron screening — NASA Glenn published exactly that in two 2020 papers in Physical Review C — and it is why Douglas Miller calls the architecture vacuum-catalysed fusion: harness the vacuum, then use fusion as the intermediary, because fusion is the most efficient energy generation available inside three-dimensional reality.

And assume the product is not a power station but a chip: a free energy microchip that goes in your phone, and your phone runs forever.

The physics is taught here: Chapter 2 and the zero-point field and Casimir course for the measured force, Chapter 6 for extraction and the funded programmes, Chapter 12 for lattice confinement fusion, and the Josephson junction course for the array engineering that turns one working element into a million.

First-order effects

Fuel leaves the price of power. Today an electricity bill is mostly fuel, plus the machinery to burn it and the wires to move it. Take fuel out and what is left is hardware and maintenance. The marginal cost of energy goes to zero — the economic content of infinite efficiency, and best said in exactly those cost terms: no fuel to buy, no fuel to lift, no fuel to ship.

The grid dissolves into the buildings. Central plants exist because combustion and turbines reward size. A device that works at the scale of a cabinet, a car or a handset rewards nothing of the kind. Generation moves to the room it serves, and the long lines survive for trade and balancing rather than delivery. Roughly eight percent of the world's electricity is lost between plant and socket; most of that loss has nowhere left to happen.

Power arrives where it never reached. Around seven hundred million people still live without electricity and more than two billion cook over solid fuel. Reaching them has always meant building the grid out to them first. A sealed unit with no supply chain behind it skips the grid, and the last mile stops being a mile.

Heat, cold, clean water and clean air. Space heating, process heat, refrigeration and desalination all become electricity problems the moment electricity is free. Seawater reverse osmosis takes three to four kilowatt-hours per cubic metre, and energy is most of what that water costs. Air pollution — linked to some seven million premature deaths a year — is mostly the smoke of burning things for energy, so the health dividend lands hardest in the poorest households.

Second-order effects

Energy-hungry chemistry stops being marginal. Aluminium takes around fourteen kilowatt-hours a kilogram to smelt, which is why smelters chase cheap hydro around the planet. Ammonia for fertiliser takes one to two percent of all world energy. Separating mixed waste back into clean feedstock is chemistry we understand and cannot afford. All of it becomes ordinary.

The world stops shipping energy. A large share of all sea and pipeline traffic exists to move coal, oil and gas from where it is to where it burns. That trade thins, the ports built for it change purpose, and the nations built on it face a transition better begun early — the peace page's subject.

Energy and motion are the same purchase. This is the thesis's own point, not a footnote: when you extract zero-point energy from a medium you get energy, and you also get motion. The same asymmetry that lights a house pushes a vehicle — which is why the transport page is not a separate programme but the same device with a different job.

Heat becomes the number engineers design against. Every watt used ends up as heat somewhere, and the planet's balance does not care where it came from. At today's consumption that is nothing beside sunlight; at a hundred times today's it is a real design parameter, better built into the first unit than discovered at the thousandth.

Third-order effects and beyond

Energy poverty ends, and a shape of inequality goes with it. For most of history the gap between a rich place and a poor one has been readable in kilowatt-hours per person. That gradient flattens inside a generation, and what remains visible is knowledge, institutions, health and land.

Land comes back. Energy production is among the largest users of land on Earth: mines, fuel crops, reservoirs, corridors, refineries. Most of it becomes unnecessary. Whether that ground is rewilded, farmed or simply taken is a decision rather than a consequence — the largest ecological choice this technology hands anyone, and one to make with ecologists in the room from the start.

Ambition resets. When power is not the binding constraint, projects that were only ever dreams get costed seriously: atmospheric restoration, continental water management, industry moved off a living planet. Culture follows capability, usually about a decade late.

And the frontier under all of it is the field itself. If Planck's constant measures zero-point energy density, then changing the depth of that ocean changes the vacuum's permittivity and permeability — and therefore the local speed of light and the metric. Pull energy out of spacetime and you warp spacetime as you do it. Energy is where this begins, not where it ends.

A day in that world

She wakes before the light, the way bakers do, and the house is already warm. No click of a boiler, no smell of gas. The heat has been on all night because nobody thought about it.

In the yard the unit sits in its grey cabinet under the fig tree, about the size of a small refrigerator, humming so faintly she hears it only between birdcalls. Four years running, and the service card inside the door has one entry on it.

She fires the ovens — electric, all six, cheaper to install than the old flued ones. The bakery smells of steam and hot flour instead of diesel from the delivery yard. The village used to lose power for two hours most summer afternoons, and she used to bake around that. She has stopped thinking about it, which is the real change.

Her son runs the tap in the back kitchen. That water came up the hill from a small plant on the coast, pushed through membranes by the same kind of cabinet in a shed by the sea. Two summers ago the spring failed. This summer nobody noticed.

At nine the cold room clicks on and stays on: milk, butter, the yeast cultures in their jars, steady at four degrees through the hottest month. Her mother used to lose a batch a week.

She props the door open. The ridge above the village is uncrossed by any line — the poles came down two winters ago — and there is only the shape of the hill against the sky and someone's radio somewhere.

The bill comes once a year. It is for the box, not for the power. She pays it the way she pays for the roof.

Numbers that change

The first figure is not a cost. It is the size of the reservoir every other line on this page draws on, and it is the only one here that comes with its own measurement — so it gets its own axis rather than a row in the table.

Three densities, on one table

The section above gives three numbers in three different units, which is the one way to make sure nobody can compare them. Here they are in the same unit, on the same axis: how much energy a cubic metre of the universe holds as matter, as dark energy, and as the vacuum's zero-point field on the usual back-of-the-envelope estimate.

What to draw

The whole spanOnly the two that were measured

• Measured
• A calculation, not a measurement

Matter, ordinary and dark together

2.36E-10 J/m³ (-0.4)Computed from DESI's own flat-model fit: the matter share of the critical density that its Hubble constant fixes.

Dark energy

5.56E-10 J/m³The rest of a flat universe, from the same fit. The chapter rounds it to one nanojoule per cubic metre, and the round figure is the one to quote aloud.

The zero-point field, naive estimate

5.56E110 J/m³ (+120.0)Not a measurement, and not this site's claim. Add half a quantum for every mode out to a cutoff and you overshoot the measured value by the famous factor. It is drawn here as that multiple, which is the only form the literature ever states it in.

On this axis: On a scale wide enough to hold the calculated figure, the two measured densities are the same mark. That is the shape of the problem: everything anyone has ever weighed sits in the first pixel.

In the cosmological model used here, a negative-pressure component accounts for the accelerating expansion. The third bar is a calculation about what the vacuum should hold, and the distance between it and the second bar is the open problem this chapter is about.

• The expansion measurement these two rows are computed from
• Where the large factor is stated
• Where it is corrected

The same quantities, side by side

Every line above is one quantity read twice: what it is today, and what it becomes in the world this page assumes. Here they are in two columns, so the pairs can be compared instead of remembered.

Which quantities to show

All of themOnly the ones that go to zeroOnly the ones that change some other way

• Goes to zero
• Changes some other way

The same quantities, side by side

Quantity	Today	In the world this page assumes
Marginal cost of energyGoes to zero	Most of your bill, paid every month for a lifetime.	Zero. Nothing to buy, nothing to lift, nothing to ship.State it as a cost claim, which is what it is.
People with no electricity at allGoes to zero	Around seven hundred million, with more than two billion still cooking over solid fuel.	A sealed unit has no supply chain behind it, so the last mile stops being a mile.
Electricity lost between plant and socket	Roughly eight percent of the world's supply, on the wires.	Most of that loss has nowhere left to happen: generation moves into the room it serves.
Agreement between independent measurements of the Casimir force	About five percent in 1997 and about one percent in 1998, on different apparatus in different laboratories.	The same two numbers, read as the reason this is engineering rather than a proposal. Two instruments, one answer.
Money on the bench	Roughly twelve million dollars behind Harold White's Casimir power cell, backed by the Limitless Space Institute. No sheet in this library carries that figure; the awards it does carry are an NSF Phase I of $274,920 and a 2026 SpaceWERX Phase I.	The first line of a ledger, against a stated power-density target comparable to a solar cell — connectable in parallel and stackable in three dimensions.
Price of a first attempt	About half a million dollars for a first test wafer at a foundry, on top of the team and the design work.	Not the binding constraint. Foundry capacity is limited and, in this field, watched.A serious number for one person and a rounding error for a research budget, which is why several small companies are attempting it at once. The bottleneck is access to fabrication, not to physics.

On this ledger: 2 of 6 quantities go to zero. The other 4 change without disappearing, which is the half of this page that is easiest to misread.

What it would take

One complete ledger. A single device measured end to end over a closed cycle, delivering more than it consumes, with actuation and instrumentation counted in. Every serious programme names this milestone for itself. It does not have to be large. It has to be complete.

The Sandia structures. Chase's asymmetric-nanostructure cantilever deflected the hoped-for way and then proved contaminated by deposition stress in the film, so no conclusion was drawn. Sandia National Laboratories is fabricating the replacements — a named systematic and a named next measurement, which is exactly what a live experiment looks like.

A second laboratory saying the same thing. Replication turns a result into a field, and Chase's own survey is candid that nothing here has yet been independently replicated — which makes your independence the scarcest resource in the subject. Alongside it, an agreed measurement standard: every group counts differently, and a published protocol for what belongs in a vacuum-power ledger would let three teams compare results in an afternoon. Unglamorous, cheap, and probably the highest-leverage contribution available today.

Scaling from micro to kilo. A microwatt is a physics result; a kilowatt is a product. The road runs through arrays — many identical elements, phase-managed, fabricated together — the engineering in the Josephson junction course, where junctions are already made in their thousands on one chip.

A metered heat run on a loaded lattice. NASA Glenn put real nuclear reactions in deuterated metals into the peer-reviewed record in 2020; Clean Planet is working the same phenomenon with Tohoku University and industrial boilers, and Astral Systems is doing it in Britain. The open milestone is calorimetry: sustained excess heat, repeated elsewhere. A reactor that returns more than it draws is burning fuel — a separate sentence from anything about the vacuum, and the door that opens first.

Stewardship

Design the failure mode before the product. A device that fails cold, inert and locally is a household appliance; anything else is a hazard in a hundred million homes. Publish the safety architecture first, as a condition of the standard rather than an afterthought to it.

Keep the heat budget in the design. Waste heat is the real planetary ceiling in a world of free power. Measure it, publish it, set the limit early. It is far easier to build a civilisation around a known number than to retrofit one.

Make access the default, not the reward. The fastest way to rebuild scarcity is to lock this behind licensing tiers, and a licensed, government-issued chip is a plausible way for it to arrive. Open specifications, local manufacture and generous terms for low-income deployment put the first units where they do the most good: where the grid never reached.

Signals to watch

A complete net-positive ledger, published with its methods. One device, one closed cycle, every input counted. Watch Moddel, White and Chase, and the public funding lines already on record.

The Sandia cantilever comes back clean. New fabrication, the deposition-stress systematic removed, a deflection in the predicted direction. Then a second group. And when a buyer signs a purchase order for a unit rather than a study, watch the contracts, not the announcements.

Sustained excess heat on a deuterated lattice. The reactions are already published; calorimetry, repeated elsewhere, is the next line — and the one most likely to be crossed first.

Fusion arriving quietly. This does not end with a press conference. Disclosure is a process by which technologies are handed to the public — and fusion is first.