For more than two decades, solid-state nanopores have hovered on the edge of commercial relevance. Researchers have repeatedly demonstrated their promise for single-molecule sensing: detecting DNA, proteins, and other biomolecules by monitoring ionic current as they pass through nanometer-scale holes in a membrane. But despite thousands of academic papers, commercialization has remained elusive. The bottleneck was never proof of principle; it was manufacturing. That barrier may now be breaking.

In a recent demonstration, imec showed the first wafer-scale fabrication of solid-state nanopores using extreme ultraviolet (EUV) lithography, producing pores of roughly 10-nm diameter across full 300-mm wafers with high uniformity. More importantly, the work reframes nanopores not as artisanal devices drilled one by one with electron beams or focused ion beams but as manufacturable semiconductor structures, built with the same tools and process discipline used for advanced CMOS.

“This is really about showing that nanopores can be made with the same repeatability mindset as chips,” Pol Van Dorpe, associate professor of quantum solid-state physics at KU Leuven and fellow at imec, told EE Times in an exclusive interview. “Once it’s possible to make thousands, or eventually millions, of pores in one go, a wider group of people can build assays, instruments, and eventually products.”

Why EUV changes the equation

Until now, most solid-state nanopores were fabricated using e-beam lithography, focused ion beam drilling, or transmission electron microscopy sculpting. These techniques can produce exquisitely small pores, but they are fundamentally incompatible with volume manufacturing.

EUV lithography brings a different value proposition. Rather than carving pores individually, EUV enables simultaneous patterning of dense nanoscale features with nanometer-level dimensional control, wafer after wafer. “In lithography terms, EUV is actually the most precise step in the whole flow,” Van Dorpe said. “The remaining variability mostly comes from etch, membrane formation, and release—not from the lithography itself.”

Imec’s approach combines EUV patterning with a spacer-based shrink process and a dual-mask strategy. A dense EUV mask defines an array of ~20-nm apertures, while a second, lower-cost deep-UV mask selectively “closes” unwanted openings. This hybrid approach solves a subtle but critical mismatch: EUV tools are optimized for dense patterns, while nanopore applications often require sparse, isolated pores. “This technique gives us flexibility,” Van Dorpe said. “You don’t need a new EUV mask every time you want a different pore layout. You keep one high-density mask and change the cheaper one.”

Technically, the process resembles a MEMS flow more than a transistor flow. Two wafers are processed in parallel: one containing the nanopore membrane, the other providing backside fluidic access. The nanopore itself is formed in a thin silicon nitride membrane, temporarily filled with a sacrificial plug to protect it during subsequent processing. The wafers are then bonded, thinned, and released to expose the final pore.

The result is a mechanically robust membrane with a well-defined nanopore and integrated fluidic access—fabricated entirely at wafer scale. According to imec, pores down to approximately 10 nm were achieved across full 300-mm wafers, with good within-wafer and wafer-to-wafer uniformity.

But Van Dorpe cautions that this is not the end of the journey. “We showed what the patterning capability can bring,” he said. “Now the optimization continues, especially in etch and release steps, so that you consistently hit very tight-sized windows.”

To prove that the pores were not merely “pretty TEM images,” imec performed DNA translocation experiments at scale. Double-stranded DNA passing through the pores produced clear current blockades with signal-to-noise ratios that correlated strongly with pore size across the wafer. “That correlation is key,” Van Dorpe said. “It shows we’re not just making pores; we’re making sensors.”

From sequencing dreams to proteomics reality

Much of the early excitement around nanopores focused on DNA sequencing. Biological nanopores have since captured that market, thanks to atomic-scale size control and enzymatic ratcheting. Solid-state nanopores, Van Dorpe argues, should not try to replicate that path. “For single-base DNA sequencing, solid-state pores are not there—and maybe never will be,” he said. “But there are many applications where you don’t need angstrom-level resolution.”

Proteomics is one of the most compelling. Unlike DNA, proteins cannot be amplified, and their concentrations span many orders of magnitude. Measuring them at scale requires massive parallelization—precisely where wafer-scale solid-state pores shine. “If you want unbiased protein analysis, you need to count enormous numbers of molecules,” Van Dorpe said. “That’s where arrays of solid-state nanopores become very powerful.”

Imec is deliberately positioning this work as a platform rather than a finished instrument. Today, the organization offers single-pore and small multi-pore chips, packaging, fluidics, and reference readout electronics. A multi-pore instrument capable of measuring up to 96 pores in parallel is expected by mid-2026.

What imec is not providing is the biology. “The chemistry, the sample prep, the assay design—that has to come from partners,” Van Dorpe said. “It’s a co-optimization problem. The pore and the assay have to be developed together.”

That openness is intentional. Rather than locking partners into long, custom development cycles, imec is standardizing interfaces—fluidic, electrical, and mechanical—so life-science tool developers can experiment quickly.

Biological nanopores already ship in commercial systems, but they come with limitations: fragile membranes, sensitivity to harsh chemistries, and limited CMOS integration. Solid-state pores offer robustness and manufacturability but less molecular specificity. Van Dorpe sees a future where the two converge. “Docking biological pores into solid-state platforms is very attractive,” he said. “It’s more complex, but it could combine the best of both worlds.”

Asked what would convince skeptics that solid-state nanopores have crossed from “emerging” to “manufacturing,” Van Dorpe was pragmatic. “A partner product,” he said. “Something in the market that clearly benefits from wafer-scale fabrication—not just incrementally but fundamentally.”

With EUV now in play, solid-state nanopores no longer look like lab curiosities; they look like semiconductor devices—waiting for the right applications to catch up.