Semiconductor facilities that quote a particle class in the RFQ and receive a particle class back at acceptance often find the gap later — not in particle counts, but in yield data. Acids, ammonia, VOCs, or dopants originating from construction materials, process equipment, or personnel can reach sensitive surfaces without ever triggering a particle alarm, because the instruments watching the cleanroom are not designed to see them. The cost shows up as process instability or device-level defects after commissioning, at a point where tracing the contamination source requires testing that was never scoped. Separating what ISO 14644-1 governs from what ISO 14644-8 governs — and applying that separation at the specification stage — is the decision that keeps both requirements contractually visible and independently verifiable.

What ISO 14644-1 Covers For Particle Cleanliness

ISO 14644-1:2015 classifies cleanroom air cleanliness on a scale from ISO Class 1 through ISO Class 9 based on the maximum allowable concentration of airborne particles per cubic meter, across particle sizes from 0.1 µm to 5 µm. Those concentration thresholds define the particle cleanliness baseline for a space and form the basis for acceptance testing using light-scattering particle counters. What the standard does not address — explicitly and by design — is any form of chemical or molecular contamination. Particle counts and chemical concentrations are separate physical quantities governed by separate standards.

The measurement state specified for classification carries more procurement weight than it may appear to. ISO 14644-1 recognises three states: as-built, at-rest, and operational. Each produces a different particle count result for the same room, and each assigns different responsibility for controlling the conditions present during testing. A supplier can satisfy an as-built acceptance criterion without demonstrating that the room holds its class under process load. If the buyer’s scope does not specify the required measurement state, the acceptance result may be technically valid under the standard while providing limited assurance about real operating conditions. Specifying the state in the procurement document is not a formality; it defines who controls the test environment and what conditions must hold at the time of sign-off.

For procurement teams entering a semiconductor cleanroom project, the useful boundary to hold is this: ISO 14644-1 establishes what particle cleanliness means and how to measure it. It does not, and cannot, serve as a proxy for any form of molecular contamination control.

What ISO 14644-8 Adds For Chemical Concentration

ISO 14644-8:2022 addresses a separate category of air cleanliness: the concentration of airborne chemical substances under cleanroom operational conditions. Its scope covers acids, bases, volatile organic compounds, condensable hydrocarbons, and dopants — contaminants that are invisible to particle counters but capable of degrading surfaces, altering material properties, or interfering with deposition and etch processes in semiconductor manufacturing.

The standard establishes an air chemical cleanliness (ACC) descriptor format that allows buyers to specify chemical cleanliness precisely in procurement documents. That descriptor designates the chemical species or category of concern, the concentration limit, the sampling strategy, and a time-weighting factor where relevant. The measurable concentration range the standard works within spans from 100 g/m³ down to 10⁻¹² g/m³, which reflects the sensitivity required to detect chemically significant contamination levels in controlled environments. That range should be understood as a specification input — the appropriate limit for a given chemical species depends on the process sensitivity, not on a single threshold that applies uniformly to all applications.

The practical failure risk the standard addresses is specific. Chemical contamination reaching a cleanroom can originate from sources that a well-maintained particle filtration system does nothing to control: material outgassing from construction finishes, process equipment leaks, cleaning agents, or personnel. When a buyer’s acceptance package does not include an explicit ACC specification, those sources remain unaddressed in the supplier scope, and the contamination they produce has no contractual test against which it can be measured at handover. That absence may not surface until post-commissioning process data reveals a problem that no particle count log can explain.

Why AMC Cannot Be Treated As A Particle-Class Add-On

The assumption that a well-controlled particle environment implies acceptable molecular contamination is the most consistently damaging conflation in semiconductor cleanroom procurement. It passes incoming inspection while leaving airborne molecular contamination (AMC) entirely outside the acceptance package — not because the chemical contamination was controlled, but because no one tested for it.

The underlying reason the two cannot be substituted is mechanistic. Particle contamination follows a transport-and-deposition pathway that HEPA and ULPA filtration addresses by physically capturing particles at the filter media. Chemical contamination follows a different three-step pathway: generation at a source, transport through the airstream, and sorption onto a surface or into a material. Generation, transport, and sorption each require separate control measures, and none of those measures are built into a particle filtration system. A filter rated at 99.97% efficiency for particles ≥0.3 µm offers no defined performance against molecular species. The two systems are not redundant; they solve different problems with different media, different maintenance intervals, and different service-life assumptions.

The structured contrast between what each standard governs reinforces why procurement documents must treat them as independent requirements.

The consequence of bundling the two into a single cleanliness claim is not just a compliance gap — it is a procurement visibility problem. When a combined claim is accepted at handover, the buyer has no documented basis for determining which chemical filter media were specified, whether any were specified at all, who owns the sampling responsibility during re-qualification, or when chemical filtration media requires replacement. Those decisions cannot be reconstructed cleanly from a particle acceptance record. If yield issues emerge later and an audit asks for evidence of AMC control, the absence of a separate chemical acceptance line is not a paperwork gap — it is the absence of any enforceable requirement.

How To Separate Acceptance Tests In The Supplier Scope

Particle and chemical acceptance tests are not just different in method — they are different in governing standard, test parameter, sampling approach, and the logic by which monitoring data can be used to modify re-qualification intervals.

For particle cleanliness, ISO 14644-2 sets a maximum 12-month interval for periodic particle count testing and allows continuous particle monitoring data to support extended re-qualification intervals where the facility has a defined monitoring program. For chemical cleanliness under ISO 14644-8, there is no equivalent fixed universal interval. The re-testing frequency is defined by the buyer’s ACC specification itself, based on the process sensitivity, the chemical sources present, and the stability of the control measures in place. Continuous particle monitoring cannot be substituted for chemical sampling under any reading of either standard — the measurement methods, instruments, and sampled quantities are not interchangeable.

The procurement consequence is direct: if chemical acceptance tests are not listed explicitly in the supplier scope as separate line items with their own governing standard, test parameter, and sampling protocol, they will not be performed at acceptance. A supplier quoting against a scope that references only ISO 14644-1 has no contractual obligation to demonstrate chemical cleanliness, regardless of what the buyer assumed was implied.

At the project stage where factory acceptance tests (FAT) and site acceptance tests (SAT) are being scoped, the question to ask is not whether chemical testing is generally appropriate for semiconductor cleanrooms — it usually is — but whether the specific chemical species, sampling locations, and concentration acceptance limits have been defined and assigned to a responsible party in the supplier scope document. If that definition has not happened before the RFQ is closed, the gap is typically priced out rather than resolved, because suppliers can return FFU counts and airflow figures quickly but cannot price chemical testing without knowing the contaminant categories, source assumptions, and sampling protocol the buyer intends to apply.

Specification Wording That Prevents Mixed Cleanliness Claims

A specification that references only ISO 14644-1 and a particle class number creates a single acceptance line. A supplier can satisfy it by delivering a space that meets particle count thresholds, and that satisfaction is complete under the contract whether or not any chemical cleanliness control was designed, installed, or tested. The vagueness is not a supplier failure — it is a specification gap that leaves no contractual basis for holding chemical cleanliness as an independent acceptance criterion.

The practical recommendation is to reference both standards separately and explicitly. ISO 14644-1 carries the particle class designation — ISO Class 7 or ISO Class 8, for example — with the measurement state specified. ISO 14644-8 carries a defined ACC descriptor that names the contaminant categories of concern, assigns concentration limits per species, defines the sampling strategy, and applies a time-weighting factor where the standard requires it. Those two references serve different functions and cannot substitute for each other. ISO 14644-8 further requires that the specification designate whether contaminants are specified at the individual species level, as a group, or by category — a distinction that matters because it determines what analytical method can satisfy the acceptance test and whether the result is verifiable against the stated limit.

A specification that omits the ACC descriptor leaves chemical cleanliness claims unverifiable at acceptance. Without a defined chemical species, a concentration limit, and a sampling protocol, any claim a supplier makes about chemical cleanliness during handover cannot be confirmed or disputed against a documented standard. That becomes a defensibility problem at re-qualification, at process audit, or when yield investigation requires tracing contamination history — precisely the situations where the absence of a baseline chemical acceptance record is most damaging. For projects where the process risk includes both particle-sensitive and chemically sensitive steps, building separate acceptance lines into the specification document before the RFQ stage is the point at which that risk is either managed or deferred.

For semiconductor cleanroom projects requiring particle-class compliance alongside chemical contamination control, understanding how filtration media selection and cleanroom module design interact with both acceptance regimes is a useful parallel reference. The Moduł do pomieszczeń czystych dla półprzewodników at youthfilter.com reflects the equipment scope typical of these environments. For a treatment of how filtration media types relate to different cleanliness requirements, Rodzaje filtrów powietrza do pomieszczeń czystych | Porównanie filtra wstępnego z ULPA covers the media-level distinctions that become relevant when particle and chemical filtration are specified separately.

The most immediate pre-procurement check is whether the project specification currently contains two independent cleanliness references — one to ISO 14644-1 with a particle class and a measurement state, and one to ISO 14644-8 with a defined ACC descriptor — or whether it contains one particle class that is expected to cover both. If it is the latter, the chemical cleanliness requirement has no contractual definition and no verifiable acceptance criterion at handover.

Before closing the RFQ, the chemical contaminant categories relevant to the process must be identified, concentration limits must be assigned, and sampling responsibility must be designated in the supplier scope. That sequence cannot be completed after acceptance without reopening scope, cost, and schedule. The distinction between what ISO 14644-1 governs and what ISO 14644-8 governs is not an administrative detail — it is the specification boundary that determines which contamination risks are enforceable at handover and which are deferred to post-commissioning problem-solving.

Często zadawane pytania

Q: Does specifying ISO 14644-8 still make sense if the process uses only dry etching with no wet chemistry on site?
A: Yes, because the chemical contamination sources ISO 14644-8 addresses are not limited to process chemistry. Material outgassing from construction finishes, cleaning agents, and personnel are independent AMC sources that remain active regardless of whether wet chemistry is present. Dry etch processes can still be sensitive to trace acids, bases, or VOCs reaching the wafer surface, so the absence of on-site wet chemistry does not remove the need for a chemical cleanliness specification — it changes which contaminant categories are most relevant, not whether an ACC descriptor is warranted.

Q: Once the RFQ is closed with separate ISO 14644-1 and ISO 14644-8 acceptance lines, what is the first concrete step before FAT begins?
A: Assign sampling responsibility to a named party in the supplier scope document before any test protocol is drafted. ISO 14644-8 requires that the ACC descriptor define the sampling strategy, but it does not automatically designate who conducts chemical sampling at FAT or SAT. If that responsibility is unassigned when the acceptance test schedule is written, chemical sampling is routinely deferred or excluded by default, because it requires specialist instruments and analysis that differ from particle counter deployments. Confirming the responsible party — buyer, supplier, or third-party laboratory — while the scope is still open is the step that keeps the chemical acceptance line executable.

Q: At what point does quoting a single ISO particle class become genuinely acceptable rather than a specification gap?
A: Only when the process has no steps sensitive to molecular contamination and no materials present that outgas at concentrations relevant to the devices being manufactured. In practice, that condition is uncommon in semiconductor environments, but it can apply to back-end packaging or test areas where surface chemistry is not a yield driver and the contamination risk is purely particulate. The boundary test is process-specific: if yield or material integrity is affected by trace chemical species rather than by physical particles, a single particle class is insufficient. If the risk is exclusively particulate and the process documentation confirms this, a single ISO 14644-1 reference may be defensible.

Q: How does the choice between specifying contaminants at the individual species level versus by category affect what a supplier can realistically price and deliver?
A: Specifying individual species requires more targeted analytical methods and produces a verifiable result against a defined limit, but it demands that the buyer identify which specific chemicals pose process risk before the RFQ closes — something that requires process engineering input. Specifying by category is faster to write into a procurement document and easier for suppliers to engage with initially, but it produces broader acceptance criteria that may not be sensitive enough to catch the specific contaminant driving yield risk. The practical trade-off is that category-level specification keeps the RFQ moving but can result in a passed acceptance test that still leaves a process-relevant contaminant undetected. Where the process sensitivity is known, individual species specification is more defensible at audit and re-qualification.

Q: If a project is already past RFQ closure without an ACC descriptor, is there a practical way to recover chemical cleanliness traceability before handover?
A: Partial recovery is possible through a change order that adds a chemical baseline survey as a separate scope item before SAT, but it will not achieve the same contractual standing as a pre-RFQ acceptance criterion. A baseline survey can document chemical concentrations at operational state and create a reference for future re-qualification, but because no concentration limit was accepted by the supplier at contract stage, the survey result cannot trigger a contractual acceptance failure. The more realistic outcome is that the baseline survey becomes the starting point for an ongoing monitoring program rather than an acceptance gate. It limits future ambiguity without fully closing the gap created by the missing ACC descriptor at procurement.