Semiconductor materials innovation is becoming an increasingly important part of chip development as conventional scaling grows more complex. Transistor architecture, device dimensions and lithography still matter, but manufacturers are also looking to engineered materials to manage performance, power consumption and process variability. Atomera’s 2025 annual report describes this shift and positions its Mears Silicon Technology (MST) as a material-based approach intended to address several of these challenges.
The company develops and licenses semiconductor processes rather than designing or manufacturing integrated circuits. Its commercial model is therefore tied to integrating MST into customers’ fabrication flows, qualifying the resulting processes and, ultimately, enabling production under license. For engineers and operations leaders, the report highlights both the potential role of material changes in advanced manufacturing and the substantial evaluation work required before a technology reaches volume production.
Why semiconductor materials innovation matters
Moore’s Law historically advanced through smaller features and improvements in lithography. But as transistor scaling has become more difficult, the industry has increasingly combined geometry changes with new materials and device architectures. The report points to gate-all-around (GAA) transistors and dynamic random-access memory (DRAM) as areas where material engineering is relevant to next-generation designs.
These developments are occurring amid rising demand for semiconductor capability associated with artificial intelligence. Atomera identifies advanced logic, memory and efficient power delivery as areas connected to AI infrastructure needs. At the same time, smaller geometries and more complex structures increase the demands on process control and integration. Material solutions must therefore be evaluated not only for device performance but also for whether they can fit into manufacturing flows with acceptable disruption and cost.
How MST is intended to affect transistor performance
MST is described as a thin film of reengineered silicon deposited epitaxially. Atomera says the material can be used to shape doping profiles in semiconductor devices, including junctions, contacts and transistor channels. Its stated objectives include improving drive current, reducing leakage and limiting variability—factors that affect switching behavior, power use and device consistency.
The report illustrates two possible uses in GAA transistors. In one, MST is incorporated into the channel region with the aim of reducing gate leakage and off-state power. In another, it is used in the source/drain region to help control dopant diffusion into the channel, which the company says can improve control over transistor turn-on and operating variability. These are potential integration approaches, not universal outcomes: the effect depends on the device design and manufacturing process.
Atomera also describes MST as an additive technology intended to work with commonly used semiconductor manufacturing equipment and process steps. Because epitaxial deposition is already used in advanced structures, the company’s approach is to develop MST integration that can use established equipment and flows. Whether that translates into practical manufacturing benefits must be demonstrated for each application and customer process.
Applications extend beyond advanced logic
The report identifies several target markets for MST, reflecting the range of devices where materials can influence electrical behavior:
- Advanced CMOS: Logic and memory applications, including GAA transistor structures, where drive current, leakage and variability are important considerations.
- Power devices: MST-SP and MST-SPX are intended for different voltage ranges. Atomera says these options target characteristics such as on-resistance, breakdown voltage and reliability, with potential use in power management and higher-voltage applications.
- RF silicon-on-insulator: The company works with RF designers and manufacturers to evaluate MST on RFSOI wafers, including applications involving RF switches and low-noise amplifiers.
- Compound semiconductors: Atomera is developing MST as a possible buffer layer for gallium nitride on silicon. The aim is to address material-quality challenges associated with lattice mismatch; the report describes this work as under development and being tested with initial customers.
This range matters because the constraints differ by device category. A material change that is relevant to transistor leakage in logic may serve a different purpose in a power device or a compound semiconductor substrate. Application-specific evaluation is essential.
From simulation to semiconductor process integration
Introducing a new material into a fab is a staged engineering process. Atomera’s customer engagement model begins with technical planning and definition of evaluation parameters. Simulation tools, including its MSTcad software running with Synopsys Sentaurus TCAD, can help model potential effects before committing to wafer tests. The company says these models are calibrated against measured silicon results and can help identify applications for further integration work.
Simulation does not replace fabrication and qualification. The subsequent steps generally include depositing MST on customer wafers, building test devices through the customer’s process flow and reviewing results across test rounds. If a customer decides to proceed, an R&D license allows installation and internal qualification. The customer then carries out further process development and reliability testing. Commercial production and sales require a separate high-volume manufacturing license under Atomera’s described licensing structure.
This sequence reflects a key operational consideration: a promising material result is only one part of adoption. Process compatibility, repeatability, qualification requirements and the customer’s product objectives all influence whether a technology advances. The report also notes that high fab utilization and the cost of advanced wafers can make experimental runs challenging, increasing the role of simulation in selecting what to test.
Practical implications for chipmakers and integrators
For semiconductor manufacturers, the broader lesson is that material engineering is now closely connected to device architecture and manufacturing strategy. Evaluating a new material calls for a clear link between the targeted device limitation and measurable process or product objectives. Teams also need to consider where simulation can narrow the evaluation scope and where wafer-level testing remains necessary.
Atomera’s report presents MST as a candidate for several of these use cases, while emphasizing a commercialization pathway that depends on customer integration and qualification. Its stated benefits are company assessments, and successful adoption at production scale is not guaranteed. For technical decision-makers, the relevant questions are therefore specific: Does the material address a priority device constraint? Can it be integrated and qualified in the intended flow? And do the resulting characteristics support the product’s manufacturing and performance requirements?
Conclusion
As scaling becomes more dependent on complex architectures and engineered materials, semiconductor materials innovation is an important tool in the industry’s effort to balance performance, power and manufacturability. MST illustrates one approach: adding a material layer to influence transistor or substrate behavior while seeking to work within established fabrication practices. Its practical value will depend on application-specific evidence and successful process qualification. For fabs, designers and system integrators, the priority is to evaluate such technologies through a disciplined path from modeling and test wafers to reliable, commercially qualified production.