Built from roughly 3,700 aluminum bricks stacked in a New South Wales shed, a thermal store melts on the inside without changing shape and puts out superheated steam for 24 hours straight

Melting is the cheapest way to hold heat.

Warming a solid stores energy slowly and gives it back slowly, while melting it absorbs a large amount at one fixed temperature and hands the same amount back on the way down.

The trouble is what melts.

Aluminum particles are dispersed through a graphite matrix that stays solid, so the metal turns liquid while the brick keeps its outward shape.

A liquid needs a tank, a pump and somewhere for it to go when a seal fails, which is why molten salt plants are expensive buildings rather than expensive materials.

So the trick is to melt inside something.

Aluminum particles are dispersed through a graphite matrix that stays solid, so the metal turns liquid while the brick keeps its outward shape.

Why the block never needs a container

The two materials refuse to mix.

That is the whole design, a pairing chosen because neither will dissolve into the other at any temperature the system reaches.

The matrix does the holding.

Graphite conducts heat well and stays rigid throughout, so every melted particle is sealed inside its own pocket with no path to anywhere else.

That removes most of the plant.

No tank, no circulating pump, no pressure vessel and no spill, which turns a thermal store into something you can stack like a pallet load.

What is actually in the shed

The building sits near the coast north of Sydney.

Inside it is a stack of roughly 3,700 blocks, each about the size of a large house brick, in a structure a little under 40 feet long.

The rest of the box is small.

It runs about 10 feet wide and 13 feet tall, which is close to the footprint of a shipping container standing on its side.

The material is not exotic.

Aluminum and graphite are both commodity inputs bought by the ton, which is the argument for a store whose cost should fall with production volume rather than with chemistry.

The rating is modest.

The stack holds 5 megawatt hours and dispatches at 500 kilowatts, which is a demonstration scale rather than an industrial one.

The output is steam, not electricity

Heat exchangers take the energy back out.

Gas or fluid passes through the stack, picks up the heat as the particles solidify, and leaves as steam at a temperature the operator chooses.

The usable band is wide.

The system delivers steam anywhere from about 300 degrees Fahrenheit up past 1,000 degrees, and it held that continuously for a full day.

Steam is the point of the exercise.

Most heavy industry does not want electricity, it wants heat at a chosen temperature and pressure, which is demand a battery cannot serve without a second machine in between.

The comparisons come from the company.

It says the blocks store two to three times what conventional thermal storage manages and need 24 times less land than batteries of the same output, which are its own figures rather than an independent test.

The part that is missing from the brochure

This site has already had an incident.

In 2023 an overheating event at the same Tomago facility brought emergency services out, and the company described it as the sort of thing that happens at the front edge of a technology.

That context matters here.

A thermal store that had a thermal event is exactly the machine whose claim of holding its shape needs a demonstrated year behind it rather than a datasheet.

It also explains the pitch.

Every advantage the design claims is an absence, no liquid and no vessel, which is the answer to eight hours of storage arriving as a chemistry problem instead.

The demonstrator did run.

It was completed and operated through charge, storage and discharge cycles, and the demonstrator is the evidence the company is selling from.

The real test starts in Western Australia

The next one is 39 times larger.

A 195 megawatt hour store is planned for a mineral sands processing site at Kwinana, sized to deliver roughly 22 tons of steam an hour.

It is not built or funded yet.

The engineering design study only began in the middle of this year, with construction pencilled for 2027 and operation targeted for 2028.

Public money is carrying it.

A federal renewable agency is funding most of that study, and the company separately raised 11.9 million dollars in early 2026 to scale manufacturing.

Industrial heat is the actual market.

Around a fifth of global energy demand goes into process heat, and a store like this competes with a gas boiler rather than with steam turbines or a battery.

The order of events is the caution.

One demonstrator has run, the plant that matters is still a study, and the distance between those two states is where most storage companies stop.

Which leaves one honest sentence.

Roughly 3,700 aluminum bricks ran a full day of steam in a shed, and everything past that is still a drawing.