18 June 2026

 

 

For more than 50 years, the semiconductor industry has relied on a remarkably successful strategy: make transistors smaller, fit more of them onto a chip, and computing becomes faster, cheaper and more energy efficient.

That approach transformed everything from personal computers and smartphones to cloud computing and artificial intelligence. Yet the industry is increasingly confronting a challenge that engineers have anticipated for years. Silicon, the material that forms the foundation of virtually every modern computer chip, is approaching fundamental physical limits.

Researchers at imec, the Belgian nanoelectronics research centre widely regarded as one of the world's leading semiconductor R&D organisations, believe they have demonstrated one of the most promising routes beyond those limits.

Working alongside ASML and Taiwan Semiconductor Manufacturing Company (TSMC), imec recently announced the successful fabrication of advanced transistors based on 2D materials on industry-standard 300mm wafers. While commercial products remain years away, the achievement represents an important step towards manufacturing a new generation of semiconductor devices at industrial scale.

 

The research institute behind the breakthrough

 

Although ASML and TSMC are the most recognisable names in the announcement, the research itself originates from imec.

Founded in 1984 and headquartered in Leuven, Belgium, imec occupies a unique position within the global semiconductor ecosystem. Rather than manufacturing chips itself, the organisation serves as a neutral research hub where equipment suppliers, chipmakers, materials companies and academic researchers collaborate on technologies that may not reach commercial products for another decade.

Over the years, imec has contributed to numerous advances in transistor design, lithography and semiconductor manufacturing. As chip scaling becomes increasingly difficult, many of the industry's most important future technologies are first explored in imec's laboratories.

That is one reason the latest announcement has attracted attention far beyond the research community.

 

Why silicon is reaching its limits

 

To understand the significance of the breakthrough, it helps to understand what a transistor actually does.

Every computer chip contains billions of transistors. These microscopic devices function as electrical switches, constantly turning current on and off to process information, perform calculations and store data.

Every text message, video stream, AI query and internet search ultimately depends on vast numbers of these switches operating at extraordinary speeds.

For decades, semiconductor companies have improved performance by making transistors progressively smaller, a trend often referred to as Moore's Law. The problem is that silicon transistors are now approaching dimensions where controlling electrical current becomes increasingly difficult.

When transistors become extremely small, electrical current can leak through barriers that are supposed to block it. These leakage effects waste energy, generate heat and reduce efficiency.

The semiconductor industry has repeatedly postponed these limits through innovations such as FinFET and Gate-All-Around transistor architectures. However, many researchers believe entirely new materials will eventually be required if transistor scaling is to continue.

 

What are 2D materials and why are they different?

 

One of the most promising candidates is a class of materials known as 2D materials.

Unlike conventional semiconductor materials, which have thickness in three dimensions, 2D materials can be only a few atoms thick. One of the best-known examples is molybdenum disulphide (MoS₂), a semiconductor material that remains stable and functional even at extremely small dimensions.

Because these materials are so thin, they offer much stronger control over electrical current flowing through a transistor. That could help reduce leakage while allowing engineers to continue shrinking device dimensions.

Scientists have studied 2D materials for years, producing impressive results in laboratory experiments. The challenge has always been manufacturing.

Creating a small number of experimental devices in a research environment is very different from producing billions of transistors across industrial-scale wafers with acceptable yields and reliability.

That manufacturing challenge is where the latest breakthrough becomes important.

 

The breakthrough: manufacturing 2D-material transistors at scale

 

The collaboration between imec, ASML and TSMC demonstrated functioning 2D-material transistors on 300mm wafers, the same wafer size used throughout the world's most advanced semiconductor fabrication plants.

The researchers reported a transistor pitch of 50 nanometres for the demonstrated structures, representing one of the densest implementations achieved for this class of 2D-material devices.

Equally significant was the reported functionality rate of approximately 94% across the demonstrated test structures.

For emerging semiconductor technologies, such results matter enormously. A transistor architecture may perform well in a laboratory, but unless it can be manufactured consistently and reliably, it is unlikely to become commercially viable.

The results suggest that 2D materials are beginning to move from scientific research towards practical semiconductor manufacturing.

 

Why ASML matters

 

The announcement is particularly significant for the Dutch technology ecosystem because of ASML's involvement.

The Veldhoven-based company supplies the extreme ultraviolet (EUV) lithography systems used to manufacture the world's most advanced semiconductor chips. These machines are essential for producing cutting-edge processors used in smartphones, data centres and AI infrastructure.

Crucially, the demonstrated 2D-material transistors were fabricated using processes that are compatible with advanced semiconductor manufacturing techniques and existing lithography infrastructure.

That means future adoption would not necessarily require chipmakers to abandon existing production ecosystems and build entirely new fabrication facilities from scratch.

For an industry where individual factories can cost tens of billions of euros, compatibility with existing manufacturing infrastructure is a major advantage.

In other words, ASML's involvement suggests that the path from laboratory breakthrough to industrial production may be considerably smoother than it would be for a completely new manufacturing platform.

 

Why TSMC's involvement matters just as much

 

If ASML's participation demonstrates that 2D-material transistors can fit within advanced manufacturing flows, TSMC's involvement addresses a different question: does the world's leading chipmaker believe the technology is worth investigating?

TSMC manufactures many of the advanced chips used by Apple, NVIDIA, AMD, Qualcomm and countless other technology companies. More importantly, it has built its business on transforming promising semiconductor concepts into products that can be manufactured at enormous scale.

Commercialising a new transistor architecture is one of the most difficult challenges in the semiconductor industry. It requires years of process development, extensive reliability testing and substantial investment.

TSMC's participation does not guarantee that 2D-material transistors will eventually reach mass production. However, it does suggest that one of the industry's most influential manufacturers sees enough potential to explore how the technology could fit into future semiconductor roadmaps.

Together, imec, ASML and TSMC represent three critical parts of the semiconductor value chain: research, manufacturing equipment and large-scale production.

 

What could this mean for AI, smartphones and future computing?

 

As artificial intelligence models grow larger and more capable, demand for computing power continues to increase. At the same time, energy consumption is becoming one of the industry's biggest challenges.

More efficient transistors could help address both problems.

If future generations of 2D-material devices deliver on their promise, they could enable processors that consume less power while delivering greater performance. That could benefit everything from smartphones and laptops to AI accelerators and high-performance computing systems.

For consumers, the impact could eventually include smartphones with significantly longer battery life, more powerful on-device AI capabilities and computers that deliver greater performance without corresponding increases in energy consumption.

The technology could also help reduce power consumption in data centres, an increasingly important consideration as AI workloads continue to expand.

 

When will consumers see these chips?

 

Despite the significance of the announcement, consumers should not expect 2D-material processors to appear in commercial devices anytime soon.

Semiconductor innovation typically moves from research laboratories to pilot production, process optimisation and eventually high-volume manufacturing over many years.

Industry experts generally view 2D materials as a technology for the next decade rather than the next product cycle. If development continues successfully, early commercial applications could emerge during the 2030s, likely beginning with specialised high-performance computing applications before reaching mainstream consumer devices.

The latest results therefore represent an important milestone rather than an imminent commercial launch.

Silicon is unlikely to disappear. More likely, future chips will combine advanced silicon technologies with new materials that help overcome the scaling challenges facing today's transistors.

What the imec, ASML and TSMC collaboration demonstrates is that the search for those materials is no longer confined to academic laboratories.

For an industry built on the relentless shrinking of transistors, that may prove to be one of the most important semiconductor developments of the decade.