For decades, the semiconductor industry has relied on one basic bargain: make silicon transistors smaller, and chips become faster, cheaper and more power-efficient. That bargain is getting harder to keep.
As AI models grow larger and data centres consume more electricity, chipmakers are running into a materials problem. Silicon, the foundation of modern computing, can only be shrunk so far before electrons begin to leak through transistor channels, wasting power and limiting performance gains.
Singapore-based Nexstrom believes the next step will not come from squeezing more out of silicon, but from replacing parts of the transistor with atomically thin semiconductor materials.
Also Read: NASA-linked, MIT-trained founders’ nSWX raises US$2M for AI chip packaging
The company has secured US$12 million in seed funding led by Xora Innovation, with participation from Foothill Ventures and SEEDS, an arm of SG Growth Capital. The round brings Nexstrom’s total capital raised to US$15 million, including US$3 million in non-dilutive funding.
Nexstrom said the capital will be used to commercialise its wafer-scale platform for two-dimensional (2D) semiconductor materials. Its near-term target is ambitious: to develop what it describes as the industry’s first 12-inch single-crystal 2D semiconductor wafer growth platform using production-ready manufacturing tools.
That matters because 12-inch, or 300 mm, wafers are the standard used by advanced chip foundries. Many promising semiconductor materials have performed well in laboratories but failed to make the jump to large, uniform wafers that can survive commercial manufacturing.
Why 2D materials matter
2D semiconductors are materials only a few atoms thick. One of the best-known examples is molybdenum disulphide, or MoS₂, a compound that has attracted research attention because it can help control electron flow at extremely small dimensions.
In simple terms, thinner channels give chipmakers better control over the movement of electrons. That could reduce leakage, lower power consumption and support further transistor scaling at a time when silicon is becoming harder to push.
This is especially relevant for AI and high-performance computing, where the economics of performance are increasingly tied to energy efficiency. Training and running large AI models already require vast amounts of computing power, and every incremental gain in chip efficiency can translate into meaningful savings for hyperscale data centres.
But the obstacle has never been scientific promise alone. The bigger question is whether 2D materials can be produced at the scale, consistency and cost that advanced foundries demand.
Also Read: Synopsys, A*STAR team up to tackle AI chip packaging challenges
“Silicon has fuelled decades of computing innovation, but the industry now needs a new materials platform to continue scaling performance,” said Dr Lance Li, Nexstrom’s co-founder and Chief Scientist. “For years, the challenge has not been demonstrating the promise of 2D materials, but manufacturing them at the scale and quality advanced foundries require.”
From research to foundry floor
Nexstrom was founded in 2024 and incubated through Xora Innovation’s venture-building model, which focuses on turning deeptech research into companies that can address industrial markets.
Its platform combines proprietary chemical vapour deposition hardware, process technology and wafer-scale 2D material growth. Chemical vapour deposition, or CVD, is a manufacturing process used to deposit thin films of material onto wafers.
Nexstrom’s approach is designed to fit into existing foundry workflows rather than requiring chipmakers to rebuild their manufacturing lines from scratch.
That compatibility will be important. In semiconductors, even promising materials face long adoption cycles because foundries are highly conservative environments. Any new process must deliver uniformity, repeatability and yield, while fitting into an industry already built around expensive equipment and tightly controlled production steps.
Nexstrom says its technology is aimed at continuous, single-crystal 2D material growth across 12-inch wafers. “Single-crystal” refers to material with a uniform atomic structure, which is important because defects and grain boundaries can affect electronic performance. In commercial chipmaking, uniformity across the wafer is just as important as performance in a single device.
The company is working with industry partners to validate its technology within existing chip manufacturing workflows. It did not name those partners.
A Singapore bet on upstream semiconductor technology
For Southeast Asia, Nexstrom’s financing lands at an interesting moment. The region has long been part of the global semiconductor supply chain, particularly in assembly, testing, packaging and equipment services. Singapore, Malaysia, Vietnam and the Philippines all play important roles in chip production, though most cutting-edge logic manufacturing remains concentrated in Taiwan, South Korea and the US.
Singapore has been trying to move further upstream, building on its base of wafer fabrication, precision engineering and research talent. The city-state already hosts operations from major semiconductor companies and has pushed deep tech as a strategic priority through public funding, university research and state-linked investment vehicles.
Nexstrom fits into that broader shift. Rather than building another application-layer AI company, it is attempting to address a bottleneck much closer to the physical foundations of computing. If successful, such technology would be relevant not only to AI chips, but also to data centres, advanced processors and future low-power electronics.
Still, the road from seed-stage materials company to semiconductor supplier is unusually long. Deep-tech hardware startups face lengthy validation cycles, high capital requirements and demanding customers. Unlike software companies, they cannot iterate through code alone; they must prove performance in physical systems, often over many years.
The competitive field
Nexstrom is entering a race that includes some of the world’s most sophisticated chipmakers, equipment companies and research institutes. Major players such as TSMC, Samsung, Intel and imec have explored 2D materials as possible candidates for future transistor channels, while universities and specialist materials companies are also working on graphene, MoS₂ and other post-silicon approaches.
Its challenge, therefore, is not merely to show a better material, but to build a platform that foundries can realistically adopt. That may be where a focused startup has room to compete: by solving one narrow but critical manufacturing bottleneck rather than trying to build an entire chip ecosystem around the technology.
Also Read: Southeast Asia’s chip-hub ambition is colliding with its chip-smuggling problem
Nexstrom’s technical foundation is led by Li, a recognised researcher in 2D materials and a Clarivate Highly Cited Researcher since 2018, a distinction covering roughly the top 0.1 per cent of researchers globally by citation influence. He has worked on single-crystal MoS₂ growth since 2012 and later led corporate research at TSMC on post-silicon electronics.
The company is also supported by advisors including Dr Sundar Ramamurthy, Dr Philip Wong, Dr Aaron Thean and Dr John Langan.
“The question is no longer whether 2D semiconductors matter. They are already on the technology roadmap of major semiconductor companies. The challenge is making them manufacturable,” said Wong, Board Advisor at Nexstrom and Inez Kerr Bell Professor at Stanford University.
That line captures both the opportunity and the risk. The semiconductor industry knows it needs new materials to keep scaling performance. But knowing what comes next is different from manufacturing it at commercial scale.
With its new funding, Nexstrom will expand platform development, deepen collaborations with foundries and grow its engineering and leadership teams. For Singapore’s deep-tech ecosystem, the company will be one to watch: not because 2D semiconductors are guaranteed to win, but because the next era of computing may depend on companies willing to work at the atomic edge of what silicon can no longer do.
The post Nexstrom lands US$12M to bring 2D semiconductors to 12-inch wafers appeared first on e27.