At the 2 nm generation, the semiconductor industry is no longer fighting only for smaller dimensions. It is fighting to control surfaces, interfaces, particles, molecules and chemical reactions with extraordinary precision.
For decades, semiconductor scaling was largely described as an engineering race: smaller transistors, tighter pitches, faster switching and greater computing density.
That description is now incomplete.
At advanced nodes, chemistry has moved much closer to the centre of the manufacturing problem. The industry is asking chemicals not simply to clean, etch, deposit or polish, but to perform those functions with extremely narrow process windows while leaving everything else untouched.
That changes the role of chemical purity.
A chemical can meet an analytical purity specification and still create a process problem. A trace metal, organic residue, particle, moisture level or degradation product may interact with a particular surface, precursor, resist or plasma environment in a way that only becomes visible later in the process.
This is why the transition toward 2 nm and beyond is increasingly becoming a chemical-control problem as much as a lithography problem.
ASML’s High-NA EUV platform illustrates the direction of travel. The company’s 0.55 NA EXE systems are designed for future leading-edge logic, while in 2026 Intel reported entering high-volume manufacturing for selected Intel 18A layers using High-NA EUV. At the same time, imec has been working on the resist, etch, metrology and process integration changes required to exploit the higher numerical aperture.
The important point is not simply that the optics are getting better.
Every improvement in resolution makes the chemistry less forgiving.
2 nm Is Not Simply a Smaller 3 nm
The phrase “2 nm” can give the wrong impression.
It is a technology-generation label, not a statement that every feature on a chip measures exactly 2 nm. What matters is that each successive generation demands tighter control of dimensions, materials, interfaces, electrical behaviour and variability.
That creates a difficult manufacturing equation.
A process step must remove one material without damaging another. A deposition must form a film with the required thickness and composition without introducing impurities. A cleaning chemistry must remove residue without changing the surface chemistry needed for the next step.
The process window gets narrower.
And when the process window narrows, chemistry becomes a control variable.
This is why advanced semiconductor manufacturing involves an interconnected sequence of cleaning, surface preparation, deposition, lithography, development, etching, stripping, CMP, metallization and inspection. A small chemical deviation early in the process can become a major electrical or yield problem much later.
The chemistry does not operate in isolation. It becomes part of the integration architecture.
Purity Is Becoming a Process Parameter
The old way of discussing chemical purity was relatively straightforward: identify the principal material, measure impurities and establish a specification.
Advanced semiconductor processing requires a more sophisticated view.
The relevant question is not only:
“How pure is the chemical?”
It is:
“Which contaminants are present, at what concentration, in what chemical form, and what will they do to this particular process?”
That distinction matters.
NIST’s work on semiconductor materials and measurement highlights the difficulty of detecting and characterising contaminants in advanced materials, including metals, particles and other near-surface impurities. As device structures become smaller and more complex, measuring the relevant contamination becomes increasingly difficult.
Entegris makes a similar point from the manufacturing side. Its contamination-control work focuses on particles, metals, organics and airborne molecular contamination across chemicals, gases, equipment and the fab environment. The company notes that semiconductor manufacturers continue tightening specifications as feature sizes decrease and 3D architectures become more prevalent.
This changes the definition of a high-performance chemical.
A supplier is no longer selling only a molecule.
It is increasingly selling chemical consistency under a defined process environment.
The Particle Problem Is Getting Smaller Than the Measurement Problem
Particles have always been a semiconductor concern. At advanced nodes, however, the challenge is becoming more fundamental.
A particle does not need to be large enough to be visible to the human eye to become a killer defect. At very small dimensions, the industry can reach a point where the defect of interest is difficult to detect directly.
That creates a difficult situation for process engineers: how do you control something you cannot reliably see?
This is why filtration, chemical delivery, packaging, storage and point-of-use purification are becoming integral parts of semiconductor process control rather than downstream housekeeping.
Entegris’ work on advanced chemical delivery specifically highlights the need to control contamination not only during chemical production, but also through packaging, transportation, distribution and final delivery to the fab. Its 2026 work on filtration also addresses nanoparticle behaviour, filter shedding and defectivity in advanced-node processes.
The lesson for chemical suppliers is important:
Purity can be lost after manufacturing.
A chemical that leaves the production plant within specification is not necessarily the same chemical that reaches the wafer in exactly the same condition.
The entire delivery chain becomes part of the chemistry.
Cleaning Is No Longer Just Cleaning
One of the clearest examples is wafer cleaning.
Traditional thinking treats cleaning as a relatively simple operation: remove particles, residues or unwanted material and prepare the surface for the next step.
At advanced nodes, cleaning is much closer to surface engineering.
The chemistry has to remove the target contamination while preserving:
- critical dimensions
- surface roughness
- film integrity
- underlying materials
- electrical properties
- interface chemistry
- adhesion characteristics
That is particularly difficult as structures become three-dimensional and multiple materials are exposed simultaneously.
Entegris’ 2026 Surface Preparation and Cleaning Conference work illustrates this shift. Its research includes advanced filtration for critical wet cleans, CFET defectivity reduction and wet-etch chemistry for molybdenum, silicon germanium and nitride-based applications.
The chemistry therefore needs selectivity, but selectivity alone is not enough.
It needs selectivity plus repeatability plus contamination control.
That combination is what makes advanced process chemistry difficult to formulate and manufacture.
EUV Has Turned Lithography Into a Chemical Problem
Extreme ultraviolet lithography is often presented primarily as an optics story.
It is also a materials and chemistry story.
EUV uses 13.5 nm wavelength radiation, and High-NA EUV increases numerical aperture from 0.33 to 0.55. That pushes resolution further, but it also changes the requirements placed on photoresists, underlayers, masks, development and downstream etching.
The resist has to produce the required chemical transformation when exposed to a very small number of photons while maintaining acceptable sensitivity, resolution and roughness.
That creates competing requirements.
You want the resist to respond strongly enough to exposure, but not so strongly that chemical diffusion destroys the intended pattern.
You want high resolution, but not at the cost of unacceptable line-edge roughness.
You want a thin film, but it still has to survive the subsequent process steps.
NIST’s 2026 work on EUV resist depth profiling highlights the importance of understanding nanoscale chemical changes within these increasingly thin resist structures.
So the lithography challenge is no longer simply:
Can we print the pattern?
It is:
Can we chemically create, preserve, transfer and measure the pattern without introducing unacceptable variability?
High-NA EUV Raises the Stakes Again
High-NA EUV is particularly interesting because it demonstrates how one technological improvement can create new chemical requirements elsewhere in the process.
Higher numerical aperture improves imaging capability. But the resist stack, process conditions, mask behaviour, etch transfer and metrology must all respond.
This is why imec’s High-NA work is not limited to the lithography tool itself. The organisation has identified resist chemistry, etching, metrology and process integration as important areas that must evolve alongside the optics.
This is a recurring pattern in semiconductor technology.
The bottleneck moves.
Improve lithography and resist becomes more critical.
Improve deposition and interface control becomes more critical.
Improve etch selectivity and contamination or surface damage may become the limiting factor.
Improve cleaning and the industry starts demanding better measurement of what remains.
There is no single chemistry that solves the scaling problem.
The entire process chain has to move together.
Etching: Removing Exactly the Right Material
Etching is another area where chemistry is being pushed toward extremely narrow operating windows.
The requirement is not simply to remove material quickly.
It is to remove the correct material at the correct rate while preserving the neighbouring layers and maintaining the intended profile.
As semiconductor architectures become more three-dimensional, the number of materials exposed within a single process increases. That makes selectivity increasingly important.
Entegris’ 2026 technical work, for example, focuses on wet-etch solutions for materials including molybdenum, silicon germanium and high-selectivity nitride applications.
The broader implication is significant for specialty chemical companies.
The winning chemistry may not be the most aggressive chemistry.
It may be the chemistry that gives the process engineer the best controllable window.
That is a very different value proposition.
CMP Shows Why Chemistry and Mechanics Cannot Be Separated
Chemical mechanical planarization is another useful example.
CMP combines chemical reactions with mechanical removal. The slurry, pad, filtration, cleaning chemistry and process conditions all influence the final surface.
A small change in slurry composition can affect removal rate. Particle behaviour can influence scratches. Cleaning efficiency can affect downstream contamination. Surface uniformity can influence subsequent lithography or deposition.
Entegris’ CMP portfolio reflects this integrated approach, combining slurries, pads, filtration, cleaning and process monitoring rather than treating each component as an isolated consumable.
DuPont is taking a similarly broad materials approach, with advanced-node offerings spanning CMP pads, EUV-related materials, mask cleaning and high-selectivity etchants. Its 2025 sustainability report specifically identifies new CMP, EUV and etch products for advanced semiconductor applications.
This is an important signal for the wider specialty-chemical industry.
The commercial opportunity is moving from individual chemicals toward integrated process performance.
The Interface May Become More Important Than the Bulk Material
At advanced nodes, the interface between two materials can be more critical than either material considered separately.
A film can have excellent bulk properties but still fail because the interface has poor adhesion, unwanted reactions, excessive roughness, trapped contamination or an unsuitable chemical state.
This becomes even more important as semiconductor architectures move toward increasingly complex 3D structures and advanced packaging.
Merck KGaA’s semiconductor-materials strategy illustrates how the materials ecosystem is broadening beyond conventional process chemicals. Its Electronics business includes semiconductor materials alongside metrology and inspection capabilities, reflecting the growing connection between materials performance and the ability to measure manufacturing processes.
That convergence is worth watching.
The next generation of materials suppliers will increasingly need to understand not only what a material is, but how its surface behaves, how it interacts with neighbouring materials and how that behaviour can be measured consistently at scale.
Metrology Is Becoming Part of the Chemistry
There is another uncomfortable reality in advanced semiconductor manufacturing:
You cannot control what you cannot measure.
As structures become smaller, measuring trace contaminants, thin films, interfaces and nanoscale chemical changes becomes harder.
That creates a feedback loop.
Better process control requires better metrology.
Better metrology exposes previously invisible variation.
That variation creates demand for better purification, better formulation, better delivery and better process chemistry.
NIST’s work in advanced semiconductor measurement reflects precisely this challenge, including efforts to characterise materials and contaminants at increasingly small scales.
This is why metrology should not be treated as an entirely separate equipment problem.
It increasingly defines what chemical suppliers can credibly guarantee.
The Supplier Opportunity: From Purity to Process Performance
This is where the semiconductor transition becomes particularly relevant for the chemical industry.
For a specialty chemical supplier, the opportunity is no longer simply to manufacture a higher-purity version of an existing molecule.
The opportunity is to understand the entire process environment in which that molecule will operate.
That could mean:
- controlling trace metals at extremely low concentrations
- managing particles and colloids
- preventing contamination during transport and dispense
- developing higher-selectivity wet and dry chemistries
- designing precursors for more controlled deposition
- developing resists and underlayers for advanced lithography
- engineering CMP slurries and pads as integrated systems
- developing materials that remain stable under increasingly aggressive process conditions
- linking chemical specifications directly to wafer-level defectivity
Entegris is already positioning around this broader model, combining contamination control, specialty materials, filtration, fluid management and process monitoring.
DuPont’s portfolio across CMP, EUV and etch points in the same direction.
For companies outside the established semiconductor-materials ecosystem, this creates a useful strategic question:
Are we selling a chemical, or are we solving a process problem?
The second proposition is harder to build, but it is also much more defensible.
ChemBizR Expert POV: The Real Battle Is Not Purity Alone
A 30-Year Industry Perspective
From a ChemBizR perspective, I would be careful about describing this simply as a “purity race.”
Purity is the visible part of the story.
The deeper competition is process control.
In my experience working around specialty chemicals and technology-driven markets, the most valuable suppliers are rarely the companies that simply claim the highest specification. They are the companies that understand what happens after their product leaves the reactor.
That distinction is becoming critical in semiconductors.
A supplier can produce an exceptionally pure chemical, but if the packaging introduces particles, the filtration changes the chemistry, the material degrades during transport or the impurity profile behaves differently under the customer’s process conditions, the purity number alone has limited commercial value.
The industry is therefore moving toward a more demanding definition of quality:
molecular purity + particle control + delivery integrity + process compatibility + lot-to-lot consistency.
That has major implications for business development.
Chemical companies entering semiconductor materials should not begin with the question, “Which product can we sell?”
They should begin with:
Which process bottleneck can we remove?
That might be contamination. It might be etch selectivity. It might be precursor stability. It might be surface preparation. It might be residue removal. It might be a metrology problem.
The companies that identify those bottlenecks early will have a much stronger route into the semiconductor value chain than those simply offering another high-purity chemical.
This is also where ChemBizR sees an important opportunity for chemical companies in India, Europe and other emerging specialty-materials ecosystems. The entry barrier is high, but the opportunity is not limited to inventing an entirely new molecule. In many cases, differentiation can come from purification technology, formulation, analytical capability, contamination control, delivery systems, process development and application engineering.
The competitive question is shifting from:
“How pure can you make it?”
to:
“How reliably can you make the customer’s process perform?”
That is a much bigger opportunity.
What Chemical Companies Should Watch Next
For chemical and materials companies, several signals deserve close attention as 2 nm and sub-2 nm manufacturing develops.
1. Advanced lithography chemistry
High-NA EUV will continue to increase the importance of resist, underlayer, developer, mask-cleaning and process-control chemistry.
2. Selective etching
As more materials coexist in complex structures, selective removal without collateral damage becomes increasingly valuable.
3. Atomic-scale deposition
ALD and related processes increase the importance of precursor chemistry, thermal stability, surface reactions and delivery consistency.
4. Contamination control
Particles, metals, organics and molecular contamination will increasingly be managed across the entire chemical supply chain rather than only at the wafer.
5. Interface engineering
As structures become more complex, surface chemistry and interface behaviour will become increasingly important performance differentiators.
6. Integrated consumables
The strongest suppliers may increasingly sell combinations of chemistry, filtration, cleaning, polishing, monitoring and application support rather than isolated products.
7. Analytical capability
The ability to identify and control trace contaminants can become as commercially important as the ability to formulate the product itself.
The Bigger Shift
The semiconductor industry has spent decades learning how to make smaller structures.
The next phase is about learning how to make those structures chemically predictable.
That is a different challenge.
At 2 nm and beyond, a process engineer does not have much tolerance for chemistry that is “almost right.” A trace impurity, particle or surface reaction that would have been irrelevant several generations ago can become a source of defectivity, variability or reliability failure.
That is why semiconductor manufacturing is becoming a chemical-purity war, but even that description does not go far enough.
It is really a war for control at the molecular, surface and interface level.
For chemical companies, that distinction matters.
The winners will not necessarily be those with the longest catalogue of high-purity products. They will be the companies that can connect chemistry to process performance, prove consistency analytically, control contamination through the entire delivery chain and work closely enough with semiconductor manufacturers to solve problems that do not fit neatly into a product datasheet.
For ChemBizR readers, that is the strategic signal to watch.
Do not ask only which new semiconductor chemicals are coming. Ask which manufacturing problems are becoming chemically constrained.
Those problems are where the next generation of specialty-material opportunities is likely to emerge.
