The View from the Laboratory
At Peking University, a team of researchers led by Professor Peng Hailin recently peered into a realm previously invisible—the microscopic dance of photoresist molecules in liquid developer. Using cryo-electron tomography (cryo-ET), they generated sub-5-nanometer resolution 3D images of the molecular interactions that determine how precisely circuit patterns are transferred onto silicon wafers. This breakthrough, published in Nature Communications, represents more than just a technical achievement—it’s a strategic move in the high-stakes global semiconductor competition.
For decades, the semiconductor industry has operated with a significant blind spot. The ‘development’ phase of chip manufacturing, where circuit patterns are etched onto silicon wafers, has been what industry insiders call a “black box.” Without the ability to observe molecular behavior in liquid environments, manufacturers relied on costly trial-and-error optimization—a major bottleneck at advanced nodes below 7 nanometers. Professor Peng’s team didn’t just open that black box; they installed a high-def camera, capturing molecular movements with such precision that researchers can now observe individual “tangles” that cause defects in finished chips.
The Geopolitical Chip Landscape
This scientific achievement unfolds against a backdrop of intense global competition and restructuring in the semiconductor industry. Nations and regions are racing to secure their positions in this strategically vital sector through massive investments and industrial policies.
The United States has mounted the most direct challenge to East Asian semiconductor dominance with its CHIPS Act, creating a $52bn incentive package to lure manufacturing back to American soil. This effort has yielded significant commitments from industry leaders like TSMC, which plans to invest $100bn in three new advanced logic semiconductor factories in the U.S.
Europe has taken a more targeted approach, approving €11bn for specialized facilities in Germany and Austria that focus on discrete power technologies and analog/mixed-signal integrated circuits crucial for automotive and industrial applications. The European Commission is already planning its next move with a potential “Chips Act 2.0” that would quadruple investment to maintain competitiveness.
Japan, once the undisputed leader in semiconductor manufacturing, is attempting a remarkable comeback through Rapidus—a public-private partnership that has begun trial production of 2-nanometer chips with technological support from IBM. This ambitious project aims to leapfrog from Japan’s current 40-nanometer capability directly to the 2-nanometer frontier, a technological gap that Rapidus’ CEO calls “Japan’s last chance to catch up with the world.’
The Technology Security Dilemma
Beneath these economic initiatives runs a strong undercurrent of national security concerns. The semiconductor industry finds itself at the center of a technological cold war, with export controls and supply chain restrictions becoming key instruments of foreign policy.
The United States has progressively tightened restrictions on advanced semiconductor technology exports to China, creating significant headwinds for Chinese chipmakers. The effects are tangible: Yangtze Memory Technologies, for instance, saw its equipment delivery times double from six to twelve months due to U.S.-Japan coordinated controls, forcing a 20% reduction in its 2026 production target .
This technological containment strategy has produced what industry analysts describe as a “limited decoupling”—a selective disentanglement of advanced semiconductor sectors while maintaining commercial relationships in less sensitive areas. The approach attempts to balance economic interdependence with security imperatives, but its implementation has drawn concern from industry leaders worldwide.
John Neuffer, SIA president and CEO, has emphasized that “American semiconductor companies cannot be absent from the Chinese market,” reflecting the industry’s precarious position between geopolitical tensions and commercial realities. NVIDIA CEO Jensen Huang offered an even blunter assessment, warning that continued restrictions could render the CHIPS Act a “complete failure” by depriving American companies of their largest market and essential revenue to fuel innovation.
China’s Strategic Pivot
Faced with these constraints, China has accelerated its drive toward semiconductor self-sufficiency, with the Peking University breakthrough representing a characteristic blend of fundamental research and applied industrial development. The approach exemplifies what Chinese officials term the “new whole-nation system”—a coordinated effort leveraging state funding, academic research, and industrial application.
The statistics reveal both progress and remaining challenges: China’s semiconductor market now accounts for 34% of global sales, reaching ¥1.35 trillion ($190 billion) in 2024, but its self-sufficiency rate for high-end chips remains below 20%, with core equipment and materials localization rates generally under 30%.
The Peking University advance represents a subtle but significant shift in China’s technological strategy—from playing catch-up in established technologies to pioneering new research pathways. By applying cryo-ET, a technique primarily associated with biological research, to semiconductor manufacturing, Chinese researchers have demonstrated the kind of interdisciplinary innovation that could accelerate progress across multiple chipmaking processes, including etching and wet cleaning.
Industry response has been remarkably swift. The technology has already been deployed in partnership with domestic photoresist manufacturers, with pilot production lines showing defect density reductions from 0.8 to 0.23 defects per square centimeter—bringing it to internationally competitive levels. Engineers report that development cycles for new photoresist formulations have been compressed from 18-24 months to just 6-8 months, dramatically accelerating the innovation timeline.
The Innovation Race Intensifies
The semiconductor industry’s global restructuring reflects a fundamental recognition that strategic autonomy in the digital age requires some degree of technological sovereignty. The Peking University breakthrough exemplifies several distinctive advantages in China’s approach: the ability to rapidly translate basic research into industrial applications (just eight months in this case), the willingness to repurpose techniques from seemingly unrelated fields, and the scale to deploy substantial resources against identified bottlenecks.
This coordinated strategy is showing concrete results beyond the laboratory. Domestic etching tools have increased their market share in mature processes (28nm and above) from 25% to 35% within a year, while companies like AMEC have seen their CCP etching machines adopted in 5-nanometer production lines at a 20% rate. Downstream, Chinese EV manufacturers including BYD and NIO have increased their adoption of domestic automotive-grade chips from 15% to 30% in a single year, creating a virtuous cycle of demand-driven innovation.
The Human Capital Dimension
The competition extends beyond financial investments and policy frameworks to encompass talent development—an area where China is pursuing a comprehensive strategy combining domestic education reform with international recruitment. The government has mandated that practical training account for 30% of semiconductor-related university curricula while establishing 100 industry-academia integration bases. Additionally, special programs offering tax incentives and research funding aim to attract global semiconductor experts, particularly from Chinese diaspora communities. This focus on human capital recognizes that technological leadership requires not just financial investment but sustained intellectual development.
Looking Ahead
The global semiconductor industry is evolving toward a more fragmented, regionally focused structure—a departure from the highly globalized model that dominated recent decades. This realignment reflects both geopolitical pressures and economic realities as countries recognize semiconductors’ critical role in everything from artificial intelligence and quantum computing to defense systems.
The Peking University breakthrough, while significant, represents just one front in this competition. As nations pursue technological sovereignty while maintaining unavoidable interdependencies, the industry appears headed toward what might be termed “technological spheres of influence”—partially separate innovation ecosystems with distinct characteristics and limitations.
For China, the path forward combines continued advancement in mature nodes (28nm and above) with targeted breakthroughs in specific advanced technologies where it can leverage unique research capabilities. The country’s enormous domestic market—projected to reach ¥3.5 trillion by 2030—provides a testing ground and absorption capacity unmatched by other regions.
Professor Peng’s “molecular navigation system” offers more than just a solution to a specific manufacturing challenge—it provides a metaphor for China’s broader technological strategy. As he noted in his research, “When we can see the dance of molecules, we can teach them to dance more beautifully.” In the high-stakes semiconductor race, the ability to see clearly—whether at the molecular level or the geopolitical scale—may prove the decisive advantage.
Links
Trendforce – Chip Breakthrough
Hqew – Global Semiconductor Industry Experiencing Strong Growth
Tippinsights – Proof 2D Chip Built For Space, Military Systems
Meet Taiwan – EU’s Planning “Chip Act 2.0” Policy
Semiconductor Industry Association – Global Semiconductor Sales Increase
Ta Kung Pao – Semiconductor Industry Accelerates Autonomy and enhances Resilience
Chinese Academy of Sciences – EU invests 1.1bn Euros in Two Semiconductor Plants
National Science Library – U.S. SIA Semiconductor Industry Report