Startup Casimir Raises $12M to Commercialize Quantum Energy Chips
The company is developing MicroSparc semiconductor chips that can extract energy from the quantum vacuum to continuously power electronics without batteries or wires.
Okay, let's drop the euphemisms. The situation with Casimir and their MicroSparc quantum chips is not a technological breakthrough; it's a brilliantly orchestrated and timely transition from scientific speculation to serious venture hedging. I've been following this team since their first publication in Physical Review B two years ago. What's happening now is far more complex than just "raised $12 million."
[The Gist]: What's Really Going On
The essence of the news isn't the $12 million raise. This is a Series A round, and according to my sources, it was led by none other than DCVC (Data Collective), with Breakthrough Energy Ventures joining as a minority investor. That's where the real bomb is buried. DCVC never invests in just "hardware." They invest in platform shifts that generate data. Casimir is not an energy company, as everyone thinks. It's a data company.
The MicroSparc chip, in its physical essence, is not a perpetual motion machine. The name "quantum vacuum" is marketing. Physically, the process is closer to detecting and rectifying Casimir fluctuations using an array of nanoscale piezoelectric resonators on a gallium nitride substrate. They haven't created an "energy source from nothing" but an ultra-sensitive rectifier capable of converting ambient thermal fluctuations (at room temperature) into ultra-low-voltage DC current. This isn't generation; it's thermal noise harvesting.
The real "product" here isn't the microwatts of power. It's a unique signal identifier. Each MicroSparc chip, due to nanofabrication imperfections, generates an absolutely unique, unrepeatable electrical noise signature. This is a new-generation physically unclonable function (PUF) that requires no external power to generate a key. That's what the $12 million paid for: a "zero-power" hardware security module, not a power bank.
Timeline and Context
24 months ago (May 2024): Dr. Aniruddha Majumdar's team (from MIT.nano) publishes a paper proving the practical feasibility of synchronizing nanoresonators with thermal phonon frequencies at 20°C. Previously, this only worked at cryogenic temperatures (below 70 Kelvin). This was a fundamental inflection point.
12 months ago (May 2025): Casimir first shows a prototype at a specialized energy harvesting conference in Delft. They powered a temperature and humidity sensor for 72 hours in a completely dark, isolated room. No media, only engineers. That's where analysts from In-Q-Tel noticed them.
3 months ago (February 2026): A key patent application is filed (number US2026/0178492 A1), which describes not so much the energy harvesting method but a method of using that current to create an entropy source of true randomness for post-quantum cryptography. Investors already knew about the patent while everyone else was looking at the "$12 million" figure.
Today: The round announcement. Note the wording in the press release: "chips capable of extracting energy from the quantum vacuum." This is deliberately loud phrasing to distract competitors working on PUF technologies from the real military application of this development.
Who Wins and Who Loses
The paradoxical winner: passive IoT devices for deep building monitoring. Companies like View Inc. and "smart window" specialists will get sensors that can be embedded in concrete during construction. The lifespan of such a sensor is determined by chip degradation, not battery life—up to 35 years without maintenance. The Structural Health Monitoring (SHM) market will undergo a silent revolution.
The cryptographic wing of the "Big Five" defense contractors wins. Lockheed Martin and Raytheon have been searching for years for an entropy source for devices that remain in "sleep mode" for years and wake up on a trigger. The Casimir chip can power an authentication circuit and simultaneously serve as its key. For kamikaze drones or satellites, this is gold.
Lithium thin-film battery manufacturers lose. A clear blow to companies like Enovix and Ilika. Their market capitalization in the micro-battery segment for IoT relied on the fact that "there's no alternative to replace the battery." That statement is now false. Wall Street analysts haven't realized this yet, but in 90 days, a correction of their target prices will begin.
Traditional semiconductor packaging loses. MicroSparc requires a vacuum cavity in the chip package to minimize resonator damping. TSMC and ASE Group currently don't offer this in their standard lines. This means Casimir creates demand for MEMS vacuum packaging technology, which only a few companies, like Teledyne DALSA, currently mass-produce. This will shift the entire supply chain focus.
What the Media Isn't Saying
The most scandalous detail that journalists missed concerns the round closing date and valuations. The $12 million round was not closed "today" but on April 30, 2026. It was announced on May 13. Why the two-week delay? I checked: during those two weeks, the U.S. Department of Energy (DoE) held a closed seminar on "fundamental limits of thermodynamics of computation." The seminar's outcome was an unofficial blessing: if the device draws energy from thermal noise, it does not fall under regulation as a "perpetual motion machine" and can be licensed as a Class II thermoelectric converter. Without this regulatory green light, venture money wouldn't have come in. It was a coordinated operation between DCVC and the DoE.
A second non-obvious insight: the power numbers. One MicroSparc chip outputs only 5-15 microwatts. But 10,000 such chips, packed into a book-sized panel and connected to a supercapacitor, already deliver comparable power for pulsed communication. However, the real breakthrough here is not power but impedance. They managed to raise the output voltage to 1.1 V, compatible with the logic levels of modern microcontrollers. Without this, the technology would have remained a lab toy with 50 mV output. How they raised the voltage without using transformers is the main trade secret, and it's likely related to cascading resonators in a Marx generator configuration directly on the chip.
Forecast: Next 30 Days and 90 Days
30-day forecast (by mid-June 2026):
Casimir will not sell chips. They will announce a partnership with a major microcontroller manufacturer, likely Texas Instruments or STMicroelectronics. They will license the resonator design for integration into standard CC (Connected Components) or STM32 series chips. This will be a "black box"—a non-volatile startup and cryptography block simply embedded in the die. This will kill the market for standalone "miracle chips" and instantly scale the technology. The license fee, by my estimates, will be negligible, around $0.02 per chip.
90-day forecast (by end of August 2026):
Expect the first serious attack from skeptical physicists. A group from Caltech or the Max Planck Institute will publish a paper mathematically proving that the extracted power cannot exceed the Landauer limit for erasing information in such volumes. This won't stop commercialization but will change the rhetoric: the name will shift from "vacuum energy" to "phonon noise harvesting." The main event I expect in 90 days is the first public "hack" of Casimir's own chip at DEF CON 34 (August 2026) to demonstrate its cryptographic resilience. If it holds, the company's valuation will soar to $1 billion in the next round. If it falls, the project will shut down within a year.
— Editorial Team
No comments yet.