Energy Currency: Why Energy Is the Yardstick of the Age of Matter Transformation

Short answer: if we ever make materials by rearranging atoms instead of mining and smelting them, the scarce input will be energy. An economy whose real bottleneck is energy will measure things in energy. That makes energy a good unit for pricing things and a bad place to keep savings: you can price production in joules, but you cannot store them cheaply and without losses for years, and a joule is not worth the same everywhere and at every hour. This article does the arithmetic behind that sentence, from E = mc² in kilowatt-hours to the energy price of a titanium spoon, and checks a century of “energy money” proposals against it.

This is an engineering essay, not financial advice. Nothing here is a recommendation to buy, sell or hold anything.

A promissory note called “E = mc² cakewalk”

In the original list of industrial revolutions the sixth one had a cheerful name: “E = mc² cakewalk”. Transform atoms into destination materials, coal or iron into titanium spoons, houses and spacecraft, and, in one line, the condition: “for all of this we will need a much better energy source to easily deal with E = mc²”.

That line is a promissory note, and it has two halves. The first is about collecting energy, from a sensor living on microwatts to a floating platform delivering megawatts. This article is about the second. If energy becomes the main cost of making things, what happens to the thing we use to price them?

The question is not new. My 2023 article on energy currency already ended with the densest wallet imaginable: energy stored in matter itself, on the E = mc² principle. So let us start there, at the top rung of the ladder.

E = mc² in the units of a power bill

• 1 kg of mass, fully converted, is 8.99 × 10¹⁶ J ≈ 25 TWh, or 25 billion kilowatt-hours.
• The Energy Institute’s Statistical Review of World Energy 2026 puts the world’s total energy supply in 2025 at 600.3 EJ. Divided by c², all the energy humanity used in a year has a mass equivalent of about 6.7 tonnes.
• The other way round: one kilogram, fully converted, would run the whole world for roughly 80 minutes.

Classic energy harvesting and power generation span a ladder of fifteen orders of magnitude, from a quartz watch to a power plant. This is the rung above all of them. Mass is an absurdly dense energy store, and we use almost none of it.

How much of the mass we actually use

Burning carbon releases about one ten-billionth of the reactants’ mass as energy. Fission of U-235 releases about 0.09 %, deuterium-tritium fusion about 0.38 %. Only matter meeting antimatter releases 100 %, and antimatter has to be made first, at a huge loss, in quantities nobody could use. (These fractions are my own arithmetic from standard data: 393.5 kJ per mole of CO₂, about 200 MeV per fission, 17.6 MeV per D-T fusion.)

Fusion is no longer science fiction at the laboratory scale. On 7 April 2025 the National Ignition Facility fired 2.08 MJ of laser light onto a target and got 8.6 MJ of fusion energy back, a target gain above four; an LLNL article from April 2026 and a Physical Review Letters paper accepted in October 2026 still report it as the record. But read the denominator. The gain is measured against the laser energy that reaches the target. NIF’s lasers draw about 300 MJ of electricity to deliver a shot of about two megajoules, so the record shot returned roughly 3 % of what the facility took from the grid. A physics milestone, not a power plant. The sixth revolution lives or dies in that gap.

Three recipes for a titanium spoon

One line of the 2023 entry needs an engineer’s footnote, and it is a useful one: “from coal, iron atoms or even photons, we will create titanium spoons”. Each of those three starting points has a different energy price, and the price list already exists. It is the binding-energy curve.

From coal. Four carbon-12 atoms weigh more than one titanium-48 atom (two of the protons have to become neutrons on the way, and the energy balance does not care about the path). Net, about 48 MeV is released per titanium nucleus, around 27 GWh per kilogram of titanium. In other words, titanium made from carbon would be the ash of a fusion reactor. The catch is the Coulomb barrier between two carbon nuclei, far higher than for deuterium and tritium, which we can only just ignite.

From iron. Iron-56 sits at the top of the binding-energy curve, next to nickel-62, so splitting it into lighter nuclei is uphill. Turning it into titanium-48 and two helium nuclei costs at least 17 MeV per nucleus, about 9.5 GWh per kilogram of titanium. A 20-gram spoon: roughly 190 MWh, as a floor, before any accelerator inefficiency, which in practice multiplies the bill by orders of magnitude.

From light. This is the poetic one, and physics answers with a rule rather than a number. A photon above 1.022 MeV can turn into an electron and a positron; light really does become matter. But it always comes in matter-antimatter pairs, because baryon number (protons and neutrons count +1, their antiparticles −1) is conserved in every process ever observed. Light carries no baryons, so a kilogram of titanium made from light comes with a kilogram of anti-titanium and costs at least 50 TWh. Please do not drop it.

So the realistic reading of “matter transformation” is not creation from light. It is rearranging the protons and neutrons we already have: nuclear transmutation. It already works, at a scale that is a joke and a proof at once. In 2025 the ALICE collaboration at CERN reported that near-miss collisions of lead nuclei at the Large Hadron Collider knock three protons out of lead and leave gold behind: during Run 2 (2015–2018) about 86 billion gold nuclei, 29 picograms in total, across the four main experiments. Each nucleus lived for a fraction of a second before hitting the beam pipe. Medieval alchemists would have wanted to talk to the LHC’s electricity bill.

Notice what just happened. We compared three recipes for the same product and ranked them by their price in joules, without mentioning money once. That is exactly what a unit of account does.

A price list that already works

Before anyone moves an iron nucleus, an aluminium smelter already pays for metal mostly in electricity. The Hall–Héroult process takes on the order of ten-plus kilowatt-hours per kilogram of primary aluminium. At 15 kWh/kg, the middle of that band, the floor for “iron → titanium” (9.5 GWh/kg) sits more than half a million times higher, and that is before accelerator losses. The sixth revolution is not short of a price list. It is short of a source that can carry the bill.

What money does, and which job energy could do

Money has three classic jobs: unit of account (the yardstick for prices), medium of exchange (the thing that changes hands) and store of value (the thing you keep for later). A good monetary system does all three; most proposals for “energy money” quietly assume that because energy is good at the first, it must be good at the other two.

Some calibration first:

• Fiat money is government-issued currency not backed by a commodity. Foreign governments could still exchange U.S. dollars for gold until 15 August 1971, when President Nixon closed that “gold window”; from then on, in the words of the Federal Reserve’s own history, the international monetary system turned into a fiat one.
• A commodity standard, such as gold, ties the unit to a physical good. It limits how much money can be created, at the price of rigidity when the economy grows or shrinks.
• Cryptocurrencies are tokens whose ledger is kept by a distributed network instead of a central bank. They differ a lot from each other, and that difference turns out to be about energy.

A century of energy money

1921, Henry Ford. Ford wanted to take over the unfinished federal dam and nitrate works at Muscle Shoals on the Tennessee River and build a “Detroit of the South” around it. On 4 December 1921 The New York Times reported his monetary side plan; the subheadings say it best: “Plans to Show the World Natural Resources and Not Gold Should Be Behind Currency”, “Would Use Power as Money Unit” and, my favourite, “Would Keep Gold for the Arts”. Senator George Norris opposed handing the river to a private owner, Congress debated for years, and Ford withdrew his bid. The dam stayed in public hands and became part of the Tennessee Valley Authority under Roosevelt. The energy became a utility, not a currency.

1938, Technocracy Inc. The movement’s pamphlet The Energy Certificate describes an accounting unit issued to every person, and its properties are worth quoting exactly: “It is non-transferable and non-negotiable… It is non-cumulative, therefore cannot be saved; and it does not bear interest. It need not be spent, but loses its validity after a designated time period.” Whatever you think of the central planning around it, that is a unit designed on purpose for counting, not for saving. The same pamphlet sums up the economics in one phrase that has aged well: “the increased substitution of kilowatt-hours for man-hours”.

Today, kilowatt-hour tokens. SolarCoin, running since 2013, rewards solar panel owners with one token (kSLR) per kilowatt-hour produced. The instructive part is the fine print: most rewards are calculated from the installation’s nameplate capacity with a standard estimate of about 1,314 kWh per kW per year; real monitoring data “may be used to refine” it. An energy-backed token, backed by an estimate of energy. That is not a scandal, it is the core engineering problem of energy money: someone has to measure the joule, and measurement costs money too.

Proof-of-work: the one place money really is backed by energy

The 2023 version of this article called mineable coins “mindless from energy perspective”. That is an honest summary of my irritation. It needs sharpening, not withdrawing.

Proof-of-work does not make energy disappear; nothing does. The electricity ends up as low-temperature heat, and what the network buys with that degraded energy is a security guarantee: rewriting the ledger would mean outspending the honest miners. In a strange way this is the one existing system where money really is backed by energy. Backwards: the energy is burned, not stored.

How much? The Cambridge Centre for Alternative Finance (Cambridge Digital Mining Industry Report, April 2025) estimated Bitcoin’s annual electricity consumption at 138 TWh, about 0.5 % of global electricity consumption, based on data from firms representing 48 % of mining activity. The same survey found 52.4 % sustainable sources in the mining energy mix (42.6 % renewables, 9.8 % nuclear), natural gas at 38.2 % and coal at 8.9 %.

Ethereum took the other road. Its Merge to proof-of-stake on 15 September 2022 cut its electricity use by more than 99.988 %, according to the Crypto Carbon Ratings Institute, to about 0.0026 TWh per year for the entire network. Security is bought with locked-up capital instead of burned kilowatt-hours, for some fifty thousand times less electricity.

My view, unchanged and unsolicited: for a monetary system, burning the electricity of a mid-sized country to order transactions is a heavy price when another design does a comparable job on a rounding error. That is a design judgement, not a physical law, and some mining does run on energy nobody else could use. Argue with the number, not with the thermodynamics.

Why 1 kWh is a bad constant

The 2023 article claimed that energy keeps an appropriate value in every place in the universe, and illustrated it with Mars: one kilowatt-hour buys a glass of water on Earth, while on Mars a glass of water could cost 1,000 kWh (or 10,000, two paragraphs later). The illustration is good. The conclusion runs the other way.

Cost floor. Melting a quarter-kilogram of Martian ice from −60 °C and warming it to 20 °C takes about 136 kJ, about 0.04 kWh (my estimate; the temperature is an assumption, Martian surface temperatures range from about +20 °C to −153 °C according to NASA). That is the thermodynamic floor. If the glass costs a thousand kilowatt-hours, nine hundred and ninety-nine of them are something else: the habitat, the drill, the logistics chain from Earth. For a multiplanetary species the real price of water is capital, not joules. Energy is a poor yardstick even in the example built to flatter it.

Time. Electricity is hard to store, so its price swings. Germany logged almost 575 hours of negative day-ahead prices in 2025, against 459 in 2024, with an average daily price spread of about €130/MWh (FfE analysis of EPEX Spot data). A kilowatt-hour at noon in June and at 7 pm in January are different goods.

Place. The same kilowatt-hour behind a congested grid is worth less than one next to the load. Form. A joule of electricity, of low-temperature heat and of diesel is the same physical quantity and three very different products.

A joule is conserved. A price per joule is not. That is also why I would drop the 2023 claim that inflation would be “naturally controlled” because humanity keeps getting better at producing energy. A peg fixes what the unit means; it does not by itself fix the price level, and rigid pegs have a history of their own (see: gold standard).

Two layers, only one of them hard

What survives from the original idea is its split into two kinds of energy payment, tidied up.

Layer 1: energy as a unit of account. Prices are quoted in kilowatt-hour equivalents and settled in ordinary money or tokens. Technically trivial: it needs conventions and metering, not new physics. It still inherits everything in the previous section.

Layer 2: direct settlement in energy. Energy itself changes hands: battery to battery, battery to grid, through hydrogen or other carriers. This is physics, not bookkeeping, and it pays every conversion loss, every percent of self-discharge and every kilometre of line on the way.

The original’s most romantic line belongs here (spelling tidied): “the perfect situation will be when we can store a huge amount of energy in our pockets.” Be careful what you wish for. One kilowatt-hour is 3.6 MJ, the energy of about 0.86 kg of TNT. A 100 kWh car battery holds the energy of some 86 kg of TNT. And a measured Tesla 4680 cell stores 232.5 Wh/kg, so one kilowatt-hour still weighs about 4.3 kg in cells alone, before the pack around them. A pocket full of serious energy is either very heavy or very regulated, and the airline will have an opinion either way.

The thesis I would stand behind

1. In the sixth industrial revolution the cost of a material is no longer ore + mining + smelting. It is energy + recipe (know-how) + equipment, and the binding-energy curve sets the floor of the energy term for every recipe.
2. As recipes and equipment get copied and automated, the energy term dominates. Technocracy’s “kilowatt-hours for man-hours” is the same idea. This is a thesis, not a measurement, and the word “copied” is where it can fail.
3. An economy whose dominant real cost is energy will measure in energy. Energy accounting and kilowatt-hour units of account are worth designing long before anyone transmutes anything at scale; they are useful today for grids, factories and data centres.
4. A coin backed by stored energy is the weak part. Energy is expensive to store, impossible to hoard without losses, and not a constant price. Proof-of-work shows the only way money has managed to be “backed by energy” so far: by burning it.

Energy works as a unit for pricing things, not as a place to keep savings.

Summary

• 1 kg of mass ≈ 25 TWh. All the energy humanity used in 2025 (600 EJ) has a mass equivalent of about 6.7 tonnes; one fully converted kilogram would run the world for about 80 minutes.
• We use tiny fractions of that: 10⁻¹⁰ in chemistry, 0.09 % in fission, 0.38 % in D-T fusion. NIF’s record shot gave 8.6 MJ from 2.08 MJ of laser light, but only about 3 % of the electricity the facility draws.
• Matter transformation has a price list: titanium from carbon releases energy (but sits behind the Coulomb barrier), from iron it costs at least ~9.5 GWh/kg, from light at least 50 TWh/kg plus a kilogram of antimatter. The realistic route is transmutation; ALICE measured 29 pg of gold made from lead at the LHC.
• The price list already runs on aluminium: Hall–Héroult electrolysis is on the order of ten-plus kWh/kg. At 15 kWh/kg the iron-to-titanium floor is more than half a million times higher, before the accelerator.
• Energy money is a century old. Ford (1921) wanted power behind currency; Technocracy (1938) designed an energy certificate that “cannot be saved”. The useful lesson: pricing in energy and saving in energy are different proposals.
• Proof-of-work burns energy to buy security (Bitcoin ≈ 138 TWh/yr, CCAF 2025); proof-of-stake does a comparable job for about 0.0026 TWh/yr (Ethereum after the Merge).
• A kilowatt-hour is not a constant: form, time and place change its value, as 575 negative-price hours in Germany in 2025 show.

Every route to making things from atoms runs through energy. For the manufacturing side of the story, see the Fifth Industrial Revolution: “a tank of liquid, a wave source and energy”.

---

Rewritten and updated from “Energy Currency – Real tomorrow’s value” (energyharvester.eu, 2023).

Thank you for your attention, dear Reader.
Dominik Stępień, Eng.

Sources

• Energy Institute, Statistical Review of World Energy 2026 (75th edition, 30 June 2026): world total energy supply 600.31 EJ in 2025. https://www.energyinst.org/statistical-review ; summary of the same figures: IEEJ Special Bulletin, 16 July 2026, https://eneken.ieej.or.jp/data/13404.pdf
• Lawrence Livermore National Laboratory, “Target Breakthrough Enabled Fusion Record at NIF” (shot of 7 April 2025: 2.08 MJ laser, 8.6 MJ yield): https://lmf.llnl.gov/news/target-breakthrough-enabled-fusion-record-nif
• Clark et al., “Improved fusion yields at the National Ignition Facility via ramped-profile capsule doping”, Phys. Rev. Lett., accepted 16 October 2026 (record 8.6 MJ, target gain 4.1): https://doi.org/10.1103/2jhg-pclg
• AIP FYI, “National Ignition Facility Achieves Long-Sought Fusion Goal” (about 300 MJ of electricity per two-megajoule shot): https://www.aip.org/fyi/2022/national-ignition-facility-achieves-long-sought-fusion-goal
• ALICE collaboration, “ALICE detects the transformation of lead into gold at the LHC”, 8 May 2025; Physical Review C 111, 054906: https://alice.cern/alice-detects-the-transformation-of-lead-into-gold-at-the-lhc/
• NIST, Atomic Weights and Isotopic Compositions (atomic masses of C-12, He-4, Ti-48, Fe-56 used for the titanium recipes): https://physics.nist.gov/cgi-bin/Compositions/stand_alone.pl
• The New York Times, 4 December 1921, “Ford Hopes to Use Muscle Shoals as Step to End Wars”: https://www.nytimes.com/1921/12/04/archives/ford-hopes-to-use-muscle-shoals-as-step-to-end-wars-wants-the.html
• Encyclopedia of Alabama, “Wilson Dam and Reservoir” (Ford’s offer, Norris, withdrawal): https://encyclopediaofalabama.org/article/wilson-dam-and-reservoir/
• Technocracy Inc., The Energy Certificate (1938), full text: https://archive.org/details/theenergycertifi00unse
• Federal Reserve History, “Nixon Ends Convertibility of U.S. Dollars to Gold and Announces Wage/Price Controls”: https://www.federalreservehistory.org/essays/gold-convertibility-ends
• Cambridge Centre for Alternative Finance, Cambridge Digital Mining Industry Report (28 April 2025): https://www.jbs.cam.ac.uk/2025/cambridge-study-sustainable-energy-rising-in-bitcoin-mining/
• ethereum.org, “Ethereum’s energy expenditure” (CCRI estimate, Merge reduction > 99.988 %): https://ethereum.org/energy-consumption/ ; Merge date: https://ethereum.org/roadmap/merge/
• FfE, “German electricity prices on the EPEX Spot exchange in 2025”: https://www.ffe.de/en/publications/german-electricity-prices-on-the-epex-spot-exchange-in-2025/
• SMARD (Bundesnetzagentur), “The electricity market in 2025” (final count: negative day-ahead prices in 573 hours of 2025, 457 in 2024): https://www.smard.de/page/en/topic-article/217400/219038/record-high-for-solar-generation-in-each-quarter
• SolarCoin, Frequently Asked Questions (reward per kWh, nameplate-based estimate): https://solarcoin.org/frequently-asked-questions/
• Ank et al., “Lithium-Ion Cells in Automotive Applications: Tesla 4680 Cylindrical Cell Teardown and Characterization”, J. Electrochem. Soc. (2023), 232.5 Wh/kg at cell level: https://doi.org/10.1149/1945-7111/ad14d0
• NASA Science, “Mars: Facts” (surface temperature range): https://science.nasa.gov/mars/facts/
• International Aluminium Institute, Primary Aluminium Smelting Energy Intensity (definition: AC power of the smelter and DC power after rectification, MWh per tonne; the chart did not yield a single year as text): https://international-aluminium.org/statistics/primary-aluminium-smelting-energy-intensity/ ; IAI 2025 world average for electrolysis about 13,900 kWh per tonne, as cited by the Economic Chamber of Non-Ferrous Metals and Recycling: https://igmnir.org/en/knowledge/aluminium/
• Own calculations: mass-energy conversions, mass fractions of chemical, fission and fusion reactions, titanium recipe energies, Martian-ice melting energy, TNT equivalence (1 t TNT = 4.184 GJ by convention). Aluminium-to-titanium ratio: 9.5 GWh/kg / 15 kWh/kg ≈ 6.3×10⁵. The 13–16 kWh/kg band gives about 5.9×10⁵–7.3×10⁵. “More than half a million” holds across that band; with the IAI 2025 world average (13.9 kWh/kg) the ratio is about 6.8×10⁵. 15 kWh/kg is the middle of the Hall–Héroult range, not one transcribed IAI year.