A company that builds tools for condensed matter physics laboratories is now bringing these tools into wafer fabs.
| Twelve Years Starting from the Fudan University Laboratory
On September 28, 2026, the Shanghai Securities Regulatory Bureau disclosed that FERMI Technology (Shanghai) Co., Ltd. has initiated the IPO tutoring filing for A-shares, with Huatai United Securities as the tutoring institution.
Founded in Shanghai in 2014, the company is named after the physicist Enrico Fermi. Its core team consists of three Ph.D. graduates in condensed matter physics from Fudan University: Chairman Zhao Jiafeng, who is deeply engaged in the R&D of precision instruments; CEO Ou Hongwei, who is in charge of management; and CTO Xie Binping, who focuses on semiconductor equipment technology.
Over the past decade, it has done only one thing: providing domestically produced high-end instruments and equipment for the entire process of "R&D—preparation—measurement" of advanced electronic materials.
| Four Technological Foundations Supporting Domestically Produced High-End Instruments
Ultra-high vacuum is the physical starting point.
Molecular beam epitaxy (MBE), surface analysis, and thin film deposition must all be conducted in chambers with vacuum levels approaching outer space—even the residual impurity gas of a single molecular layer falling on the growing lattice will ruin the entire epitaxial layer. FERMI's ultra-high vacuum interconnection system comes in two forms: circular sample transfer and linear interconnection. They can be freely combined, allowing samples to flow between growth, processing, and characterization chambers without breaking the vacuum throughout the entire process. Tsinghua University and Southern University of Science and Technology (SUSTech) have purchased exactly this type of equipment.
The main line of cryogenic technology is helium-free electrical refrigeration.
Superconductivity, topological states, and quantum phase transitions all occur in the cryogenic temperature range. Traditional solutions rely on liquid helium immersion for cooling, whereas helium is a strategically scarce resource that China relies on almost entirely for imports. FERMI uses a Gifford-McMahon (GM) cryocooler to replace liquid helium as the cold source to achieve closed-loop operation: the sample stage of the Cryo-300 desktop vibration-isolated cryogenic optical cryostat can reach a minimum temperature of 3.6K, and multi-stage vibration isolation keeps vibrations below 200 nanometers; the 6-axis closed-cycle cryogenic sample holder can cool from room temperature to 6K in less than 4 hours, enabling six-degree-of-freedom adjustment of samples under ultra-high vacuum, and the entire system uses non-magnetic materials to adapt to strong magnetic field environments.
Thin film preparation is the main channel for transitioning from scientific research to industry.
Molecular beam epitaxy (MBE) heats elements into directional beams in an ultra-high vacuum chamber and deposits them layer by layer on a precisely temperature-controlled single-crystal substrate. Coupled with in-situ monitoring by RHEED (Reflection High-Energy Electron Diffraction), the thickness control accuracy can reach the single atomic layer level. It is almost the only means to grow quantum wells, superlattices, topological insulators, and oxide heterojunctions. Measuring the level of an MBE system often depends not on the chamber but on the core components: FERMI has independently developed high-capacity effusion cells, large-size substrate heaters, valved cracker sources (to control the beam switching and cracking efficiency of volatile elements such as As and P), and fully automated sample transfer robotic arms. These components have long been monopolized by overseas manufacturers; being able to develop them independently means the right to define the equipment is in its own hands. From research-oriented equipment like the MBE-300, it has advanced to the MBE1000-cluster production cluster system, meeting the basic threshold of SEMI S2 for entering mass production fabs.
Plasma technology is the fourth line.
The FERMI-RFPS-200 adopts a coil structure and can excite H₂, N₂, O₂, and rare gases. In MBE, it covers nitride growth, nitrogen atom implantation and doping, and high-energy hydrogen atom surface cleaning. The other branch is the ultraviolet (UV) light source, which started by serving scientific research spectroscopy and mass spectrometry users, and then extended downstream to lithography; at the 2nd Advanced Lithography Technology Seminar in June 2026, it exhibited its short-wave UV optical product line, including extreme ultraviolet (EUV) lithography light sources and optical systems.
| From Papers to Wafer Fabs: The Leap of Three Main Lines
Semiconductor thin films, namely wafer-scale MBE systems and ultra-high vacuum cleavage coating systems, are oriented towards the mass production of compound semiconductors and photonic chips, and have been delivered in batches to leading domestic semiconductor enterprises.
Industrial metrology holds the most strategic value: XPS (X-ray Photoelectron Spectroscopy) is based on the photoelectric effect. By measuring the kinetic energy of emitted photoelectrons, it deduces the elemental composition and chemical valence states within a few nanometers of the surface thickness. It is one of the few means that can directly read out "what chemical state the elements are in." In 2023, the first domestically produced industrial in-line XPS was introduced, compatible with non-destructive testing of 12-inch wafers. In December 2025, the first set of the Jingwei 100 series was delivered, forming complete coverage from the laboratory to the wafer production line.
UV light sources and plasma represent more long-term technological reserves.
| Published in "ACS Nano", Resonating with the Times
To judge whether a scientific research instrument is good, there is a more reliable indicator than parameters: whether it appears in the methods section of high-level papers. In May 2026, the Low-Dimensional Quantum Materials Team and the Topological Quantum Computing Team of the Beijing Academy of Quantum Information Sciences collaborated with the team from Renmin University of China and made progress in iron-based superconductivity research. The results were published in "ACS Nano" under the title "Reversible Control of Magnetic Order and Intrinsic Superconductivity in Strained FeTe Films via Precise Control of Stoichiometry," and the equipment used behind it was precisely from FERMI. Tsinghua University, Peking University, Fudan University, Nanjing University, University of Science and Technology of China, Institute of Physics of the Chinese Academy of Sciences, and even China Fusion Energy Corporation—from condensed matter physics to energy mega-science facilities—the customer list itself is an endorsement. Over the past decade, the company has accumulated nearly 200 authorized patents, been awarded the title of Specialized, Refined, Differential, and Innovative "Little Giant" by the Ministry of Industry and Information Technology (MIIT), and has R&D personnel accounting for more than 30%.
The proposals for the 15th Five-Year Plan have placed high-end instruments on a par with integrated circuits (ICs) and industrial mother machines as key areas requiring decisive breakthroughs through whole-chain promotion and research. On the market side, the global supply of XPS has long been concentrated in manufacturers such as Thermo Fisher Scientific, Japan's ULVAC-PHI, and Kratos under Shimadzu. The domestic substitution rate in China's high-end market was previously in the single digits, while demand is rapidly expanding—semiconductor interface analysis, solid-state battery electrode characterization, and catalyst research cannot bypass it; on the MBE side, France's Riber and America's Veeco are the leaders. Capital endorsement is equally clear: in the A+ round in 2021, Huawei Hubble Investment was introduced; in the C round in May 2026, more than ten national-level industrial capitals such as Furong Investment, Fudan Innovation, Industrial Mother Machine Fund, Yuanhe Puhua, and Shenzhen Capital Group were introduced.
| The Next Hurdle: From "Can Make" to "Can Mass Produce"
From the laboratory to the production line, beyond the technological leap, there is also an engineering leap.
The gap between a piece of equipment enabling a Principal Investigator (PI) to produce beautiful data and its ability to maintain stable yield and delivery consistency in a 24/7 wafer fab is a comprehensive reconstruction of the supply chain, manufacturing system, process methodology, and service network. When viewing the value of FERMI, what deserves more attention is that "vacuum + cryogenics + thin film + plasma" capability platform, rather than the short-term shipment of a certain model—what is most scarce right now is precisely this underlying tool capability that can span multiple tracks.
| Who Will Define the "Right to Measure"
Capital is now beginning to penetrate further upstream and into more "heavy" areas: scientific instruments, vacuum core components, and analysis and testing methodologies. These links have extremely high technical barriers and extremely low domestic substitution rates, but once established, they are long-term assets that are hard to shake. If the previous round of domestic substitution was about "seizing back the production lines," this round is more like fighting for the "right to measure and define standards."
The significance of FERMI reaching the doorstep of an IPO perhaps lies in proving that: the vacuum and cryogenic craftsmanship assembled by several Fudan physics Ph.Ds in the laboratory more than a decade ago can grow into the infrastructure supporting national strategic industries.
This article is written based on public information, including disclosures from the China Securities Regulatory Commission (CSRC) and the Shanghai Securities Regulatory Bureau, product information on the company's official website, government procurement and university bidding announcements, academic journals, and media reports. The views expressed in the article represent only the author's judgment and do not constitute investment advice.