IBM has unveiled what it claims is the world’s first chip design with features smaller than 1 nanometre (nm), a milestone that could redefine the limits of semiconductor technology. The company’s “block of flats” architecture, which stacks transistors vertically rather than shrinking them horizontally, promises to extend the lifespan of Moore’s Law—the decades-old observation that transistor density doubles approximately every two years. However, with mass production still years away, the breakthrough remains a laboratory achievement rather than an immediate industry game-changer.
What Happened?
In a research paper published this week, IBM detailed a novel transistor design that pushes semiconductor fabrication beyond the 1nm threshold, a barrier long considered insurmountable due to quantum tunneling effects and other physical limitations. The company’s approach abandons traditional two-dimensional scaling in favor of a three-dimensional “high-rise” structure, where transistors are stacked vertically to increase density without expanding the chip’s footprint.
While IBM has not disclosed the exact dimensions of its prototype, it has suggested that the design could eventually enable features as small as 0.7nm. For context, the most advanced commercially available chips today—such as those from TSMC and Intel—are manufactured at 3nm and 2nm nodes, with 2nm production expected to begin in 2025. IBM’s breakthrough, if scalable, would represent a generational leap in computing power and energy efficiency.
Why It Matters
The semiconductor industry has spent decades chasing smaller transistor sizes to improve performance, reduce power consumption, and lower costs. However, as chips approach atomic scales, traditional scaling methods have hit fundamental physical barriers. Quantum tunneling—where electrons unpredictably pass through barriers—disrupts transistor functionality, while heat dissipation becomes increasingly difficult in ultra-dense designs.
IBM’s vertical stacking approach could circumvent these issues by allowing more transistors to fit within the same space without requiring further horizontal shrinkage. If successful, this could lead to:
– Faster processors with higher transistor counts, enabling more powerful artificial intelligence (AI), high-performance computing (HPC), and mobile devices.
– Lower energy consumption, critical for battery-powered devices and data centers, where power efficiency is a growing concern.
– Extended relevance of Moore’s Law, which has slowed in recent years as manufacturers struggle to maintain historical scaling trends.
However, the breakthrough also arrives at a time of geopolitical tension over semiconductor dominance. The U.S., China, and the European Union have all launched multi-billion-dollar initiatives to bolster domestic chip production, fearing reliance on foreign foundries. IBM, which sold its chip manufacturing division to GlobalFoundries in 2015, now operates as a research and licensing firm, partnering with foundries like Samsung and Intel to commercialize its innovations. This announcement could strengthen its position as a key player in next-generation chip technology, even if it no longer produces chips itself.
Background and Context
The semiconductor industry has long relied on shrinking transistor sizes to deliver performance gains. In 1965, Intel co-founder Gordon Moore predicted that the number of transistors on a chip would double every two years—a trend that held true for decades. However, as transistors approached the 10nm scale, manufacturers encountered physical limits that slowed progress.
Key milestones in recent years include:
– 2019: Samsung and TSMC began mass-producing 7nm chips, enabling significant performance improvements in smartphones and PCs.
– 2020: TSMC and Samsung introduced 5nm nodes, used in Apple’s A14 and M1 chips.
– 2022: TSMC and Intel announced plans for 3nm and 2nm production, with Intel targeting 2024 and TSMC aiming for 2025.
– 2023: IBM and other research labs began exploring sub-1nm designs, though none had demonstrated a viable path to mass production until now.
IBM’s vertical transistor architecture is not entirely new—companies like Intel and Samsung have experimented with 3D stacking in memory chips (e.g., High Bandwidth Memory, or HBM). However, applying this concept to logic chips (the “brains” of processors) presents unique challenges, including heat dissipation and manufacturing complexity.
Competing Claims and Uncertainty
While IBM’s announcement is a significant technical achievement, several questions remain unanswered:
1. Scalability and Yield
– The biggest hurdle for any new chip design is transitioning from lab prototypes to mass production. IBM has not provided details on yield rates—the percentage of functional chips per wafer—which are critical for commercial viability. Previous breakthroughs, such as IBM’s 2017 5nm test chip, took years to reach production.
– Manufacturing at sub-1nm scales may require entirely new fabrication techniques, including extreme ultraviolet (EUV) lithography advancements or even next-generation tools like high-numerical-aperture (high-NA) EUV machines, which are still in development.
2. Performance and Power Efficiency
– IBM has not released benchmark data comparing its sub-1nm design to existing nodes. While vertical stacking could improve density, it may also introduce new challenges, such as increased resistance or signal interference between stacked layers.
– Heat dissipation remains a critical concern. More densely packed transistors generate more heat, which could limit performance gains unless new cooling solutions are developed.
3. Industry Adoption
– IBM no longer manufactures its own chips, meaning it must license its technology to foundries like Samsung, Intel, or TSMC. These companies have their own roadmaps and may be reluctant to adopt IBM’s design if it requires costly retooling of their fabrication plants.
– Intel, for example, is already investing heavily in its own 20A (2nm) and 18A (1.8nm) nodes, while TSMC is focused on 2nm and beyond. IBM’s breakthrough could face competition from these in-house efforts.
4. Geopolitical and Economic Factors
– The U.S. CHIPS and Science Act, which allocates $52 billion to boost domestic semiconductor production, could accelerate the adoption of IBM’s technology if it aligns with national priorities. However, China and the EU are also investing in their own chip ecosystems, creating a fragmented landscape.
– The cost of developing and deploying sub-1nm technology could be prohibitive, limiting its adoption to high-margin applications like AI accelerators and supercomputers before trickling down to consumer devices.
What to Watch Next
Several key developments will determine whether IBM’s breakthrough translates into real-world impact:
1. Manufacturing Partnerships
– IBM must secure commitments from foundries to license and produce its design. Watch for announcements from Samsung, Intel, or TSMC about potential collaborations.
– If IBM can demonstrate a clear path to commercialization, it could attract additional investment and partnerships.
2. Technical Milestones
– Future updates from IBM on yield rates, performance benchmarks, and power efficiency will be critical. The company has not provided a timeline for when its sub-1nm chips might enter production, but industry analysts suggest it could take until the end of the decade.
– Advances in lithography, such as ASML’s high-NA EUV machines, will play a crucial role in enabling sub-1nm manufacturing.
3. Competitor Responses
– Intel, TSMC, and Samsung are all pursuing their own sub-2nm and sub-1nm designs. Intel’s “RibbonFET” and TSMC’s “Nanosheet” transistors are among the competing architectures to watch.
– If IBM’s design proves superior, competitors may accelerate their own 3D stacking research.
4. Government and Industry Funding
– The U.S. and EU have earmarked billions for semiconductor research. If IBM’s technology aligns with national priorities, it could receive additional funding to expedite development.
– China’s semiconductor industry, which is heavily subsidized, may also attempt to replicate or improve upon IBM’s design, adding urgency to the race.
5. Market Applications
– Early adopters of sub-1nm chips are likely to be in high-performance computing, AI, and data centers, where performance gains justify higher costs.
– Consumer applications, such as smartphones and PCs, may take longer to benefit from the technology due to cost and power constraints.
Conclusion
IBM’s sub-1nm chip breakthrough is a landmark achievement in semiconductor research, offering a potential lifeline to Moore’s Law as traditional scaling methods falter. By stacking transistors vertically, the company has demonstrated a path to denser, more efficient chips that could power the next generation of computing. However, the road from laboratory to factory floor is long and fraught with challenges, from manufacturing hurdles to industry adoption.
For now, the announcement serves as a proof of concept—a glimpse into the future of chip design rather than an immediate revolution. The coming years will determine whether IBM’s innovation can overcome the technical and economic barriers to mass production. If successful, it could reshape the semiconductor landscape, reinforcing the company’s role as a pioneer in cutting-edge technology. If not, it may join the ranks of other promising but ultimately unrealized breakthroughs that litter the history of chip development.
One thing is certain: the race to sub-1nm is far from over, and the stakes—for industry, governments, and consumers—have never been higher.
Sources:
– BBC News World: [IBM hails new ‘block of flats’ design breakthrough for ultra tiny chips](https://www.bbc.co.uk/news/articles/cvg7vpyn5pxo)
Story synopsis gathered from: BBC News World — source
Corrections
If you believe this article contains an error, contact Herald Express with the source URL and supporting evidence.
