Deep Tech Brief #18

Welcome to this week's issue of Deep Tech Brief.

Every Saturday, we break down deep tech companies building genuinely hard things, in plain language. No jargon, no hype.

Three companies this week: silicon made from sand in Luxembourg, wheat that grows in salty soil from Sweden, and a ten-minute bedside blood test for stroke patients from Denmark.

Let’s dive in.

Silicon ends up in computer chips, aluminium alloys, chemicals and batteries, and the world depends heavily on one supplier for it. China accounted for almost 80% of global silicon production in 2025, according to the US Geological Survey. Kore's founder has warned that Europe would struggle to make phones, semiconductors and modern aluminium if those exports stopped.

There is a useful precedent here. Before 1886, aluminium was rare and expensive enough to be used mostly in jewellery. That year, Charles Hall in the US and Paul Héroult in France independently found a way to make it with electricity, by passing a current through molten salts. According to the Science History Institute, the price fell from about $12 a pound earlier in the decade to as little as 30 cents.

Kore Metals, based in Luxembourg, is applying the same family of technique, molten salt electrolysis, to silicon. Its process runs on renewable electricity and adapts the kind of equipment aluminium smelters already use. The company says it needs only sand, and not especially pure sand, and that raw materials would make up less than 5% of its production cost, compared with 35 to 40% in today's industry. It also says the process cuts emissions by up to five times.

The company has an unusual backstory. It is a spin-out of Maana Electric, a Luxembourg firm that in 2024 unveiled a working solar panel made from simulated Moon dust. When cheap Chinese solar panels made that business harder, the team turned its know-how to producing silicon itself.

Where they are: founded and led by chief executive Joost van Oorschot, Kore launched in October 2025 with what InnoEnergy described as a high seven-figure investment. By InnoEnergy's account, customers in aluminium, chemicals and batteries are already testing its silicon. Next comes a pilot plant in Luxembourg producing 500 to 1,000 tonnes a year, alongside a planned funding round of more than €10 million.

Worth watching, because silicon is a strategic material Europe barely makes, and Kore is borrowing a process that heavy industry has trusted for well over a century. The milestone ahead is the pilot plant, and then the economics of scale: by the founder's own estimate, a full industrial plant would need around 200 megawatts of steady electricity, so cheap clean power will decide where it gets built.

Learn more: koremetals.com

Salt is quietly taking farmland out of use. The UN Food and Agriculture Organization's first major global assessment of salt-affected soils in 50 years, published in December 2024, estimated that about 10% of the world's irrigated and rainfed cropland is affected. Rising seas, floods, droughts and poor irrigation all push more salt into the ground. In Bangladesh alone, around a million hectares are affected, according to OlsAro's chief executive.

Most crops, wheat included, struggle as salt builds up, and wheat matters enormously: together with maize and rice it is one of the three most important crops in the world.

OlsAro grew out of more than ten years of research at the University of Gothenburg and in London. The University of Gothenburg's description of the work shows how it was done. The team took a local Bangladeshi wheat variety and treated the seeds with a chemical that introduces random changes to their DNA, raising the genetic variation about a thousand-fold. They then screened around 2,000 lines and found 70 that tolerated very high salt levels. It is a classic plant breeding technique, not genetic modification, and OlsAro says its wheat counts as non-GM worldwide. What the company adds is AI that hunts through genetic data for the variations behind useful traits, so new varieties can be developed faster.

The results so far are promising. In saline conditions in Bangladesh, the company reports yields up to 41% higher than a moderately salt-tolerant variety. Field trials are running in Kenya, Pakistan, Egypt, India, Italy, Canada and Australia.

Where they are: led by chief executive Elén Faxö. A €5 million seed round this month was led by Industrifonden, with Nordic Foodtech VC co-leading and Future Food Fund, PINC, AgFunder and GU Ventures taking part, bringing total funding to about €7.5 million. A commercial launch in Bangladesh is planned for 2027 with Lal Teer Seed, and in India it is working with DCM Shriram, which OlsAro describes as the country's largest player in proprietary wheat seed.

Worth watching, because it has a launch partner lined up in its first market, and its non-GM route sidesteps the regulatory hurdles that gene-edited crops face in some markets. The milestone ahead is the 2027 launch, and then proof that farmers see the same yield gains season after season.

Learn more: olsaro.com

In a stroke, speed matters more than almost anything. A widely cited 2006 study in the journal Stroke estimated that a typical untreated stroke patient loses about 1.9 million brain cells every minute. Denmark alone sees around 12,000 strokes a year.

Some stroke patients are already on blood thinners such as apixaban. When one of them arrives in hospital, finding out quickly whether apixaban is in their system can be crucial before doctors decide on a treatment. MedicQuant, a spin-out from Aarhus University, is building a bedside test that gives that answer in about ten minutes, with what the company says is laboratory-level accuracy.

The technology comes from more than eleven years of research in DNA nanotechnology at the university's iNANO centre and chemistry department. The field treats DNA as a construction material rather than a genetic code, folding strands into tiny, precisely shaped structures that can be designed to recognise specific molecules. The company has not published the details of how its test works, but it is designed to sit at the bedside rather than in a central laboratory.

Where they are: founded in 2019 by Malthe Hansen-Bruhn and Line Debois Nielsen. A seed round of 22 million Danish kroner (about €2.9 million) earlier this year was led by Delphinus Venture Capital, with the Export and Investment Fund of Denmark and business angels taking part. This summer it secured a €2.5 million European Innovation Council Accelerator grant, with the option of up to €4 million more in equity from the EIC Fund. The money goes into product development, validation and preparing for clinical trials.

One to watch, carefully. It goes after a very specific, high-stakes moment in emergency care, and it is built on more than a decade of university research. The milestone ahead is clinical validation, showing that results at the bedside match the lab, followed by regulatory approval.

Learn more: medicquant.com

Know a deep tech company we should break down next? Hit reply and tell us. The best editions come from real tips.

See you next Saturday.