Acrylic in a cleanroom: what passes and what does not
A cleanroom window is an interface: it has to seal, stay clear, and not shed. Most acrylic failures in a fab come from the edge, the gasket groove or the wrong sheet grade — not from the face.
A cleanroom window is an interface: it has to seal, stay clear, and not shed. Most acrylic failures in a fab come from the edge, the gasket groove or the wrong sheet grade — not from the face.
Four things come up on almost every enquiry in this sector. None of them is solved by choosing a thicker sheet.
A saw-cut edge sheds. Panels for clean areas are machined, then diamond-polished or flame-polished on the exposed edge, and packed double-bagged.
Sealing faces are held at ±0.1 mm and machined in one setup so the groove is coplanar with the frame face. Annealing after machining stops the groove moving after the seal is set.
Acrylic is not a low-outgassing material by cleanroom standards. Where the spec calls for it, we say so and quote polycarbonate or glass rather than imply compliance.
Standard acrylic insulates. Anti-static and ESD grades exist and are specified per panel, with surface resistivity stated on the packing list.
All of the below run on the same machines, in the same plant, with no operation subcontracted. Full detail sits on the Medical & laboratory line page.
Cleanroom work changes the priority order. In most sectors the first question is what the part has to survive; here it is what the part must not release. Particles, outgassing and static come before impact, and each one points at a different part of the specification.
A sawn or sanded acrylic edge sheds. A flame-polished or diamond-polished edge does not, because the surface is consolidated rather than abraded. That is the whole reason polished edges are mandatory rather than cosmetic on a cleanroom panel, and it is why we will not quote a sawn edge for one. Machined holes get the same treatment: a chamfer alone is not enough, the bore is polished or the hole is specified as a bonded insert instead.
Standard cast PMMA is a reasonable cleanroom material — it is amorphous, contains no plasticiser, and its outgassing is low. What changes the answer is cleaning chemistry. A panel wiped with IPA is fine; a panel wiped with acetone or an aggressive solvent degreaser will craze within weeks. Where the process uses solvent, we recommend polycarbonate or a coated sheet and will say so at quote stage rather than after the first failure.
This is the specification most often missed. Standard acrylic is an excellent insulator and will hold a charge of several kilovolts after a dry wipe — enough to destroy a gate oxide. For any panel in line of sight of a bare wafer or die, the sheet has to be static-dissipative. Surface grades sit at 10^6 to 10^9 ohms and decay with wiping over six to twelve months; volume-conductive grades last but cost roughly double. On a line cleaned every shift, we specify volume-conductive and treat the extra cost as maintenance avoided.
| Parameter | Standard PMMA | Cleanroom specification |
|---|---|---|
| Edge | Sawn, 1–2 mm tool mark | Flame or diamond polished, no shedding |
| Surface resistivity | > 10^14 ohms | 10^6–10^9 ohms (dissipative) or 10^4–10^6 (conductive) |
| Holes | Drilled, chamfered | Drilled, bore polished; bonded inserts where geometry allows |
| Packing | Carton, foam corners | Double-bagged, no cardboard in the inner pack |
| Cleaning tolerance | Soap and water | IPA acceptable; acetone and solvent degreasers not |
Cleanroom parts look expensive per square metre, and the reason is almost never the sheet. On a typical window or enclosure order: sheet 15–25 percent, machining 20–30 percent, polishing 20–30 percent, static grade or coating where required 10–20 percent, and clean packing 8–12 percent. Polishing is the line item buyers underestimate, because a polished bore on forty holes is more labour than cutting the panel.
There are three legitimate ways to reduce cost without touching the specification. First, reduce the hole count by using fewer, larger fixings — a panel with eight M6 holes is cheaper than one with twenty M4 holes and usually stiffer. Second, keep the polished area to the faces and edges that matter; a rear face that never sees a wafer does not need a diamond finish. Third, and most effective, standardise on one panel size across a tool set so the setup is amortised across the batch.
What we will not do is quote a sawn edge or a non-dissipative sheet as an equivalent. Both will pass a visual inspection and both will fail in service — one by shedding particles, the other by destroying product. When a buyer asks for the cheaper option, we quote it and state the consequence in writing, so the decision is a decision rather than an assumption.
| Item | What we need | Why it changes the price |
|---|---|---|
| Cleanliness class | The class the part sits in, and whether it is in line of sight of a wafer | Polishing and static grade are both driven by this |
| Static requirement | Dissipative (10^6–10^9 ohms) or conductive, and how often the part is wiped | Volume-conductive costs roughly double but does not decay |
| Cleaning chemistry | Which solvent is used, at what concentration | IPA is fine; acetone will craze within weeks |
| Edge and hole spec | Whether bores are polished or inserts are preferred | A polished bore on 40 holes is more labour than cutting the panel |
| Packing | Inner-pack restrictions (no cardboard), particle count on request | Doubles packing time; adds 3–5 days |
Where a part is still in design, tell us the wafer size and the working distance and we will propose the panel construction rather than quoting a drawing that will change twice.
The acceptance points below are recorded per batch, and the record travels with the packing list.
For a viewing panel or an equipment door it is used widely: it is light, it can be machined to a sealing face and it does not shatter into shards. It is not a low-outgassing material, so if your cleanroom standard requires an outgassing certificate for the panel itself, we will say that acrylic does not meet it and quote an alternative.
Yes, on a groove machined in the same setup as the sealing face, up to roughly 600 mm across. Above that, thermal movement during machining starts to matter, so the part is rough-machined, annealed and finished in a second setup. We will tell you which of the two routes your drawing needs.
Yes, as a specified sheet grade rather than a coating. The surface resistivity range is stated on the packing list so your ESD audit has a document to file. Cleaning protocol matters more than the grade: an alcohol wipe on a stressed panel will craze it regardless of the static specification.
Yes, and on a cleanroom part it is often required rather than optional. A drilled bore carries a tool mark that sheds, so we either polish the bore or specify a bonded insert instead of a through-hole. Polishing a bore is handwork and adds cost per hole, which is why the more economical answer is usually to reduce the hole count and use larger fixings.
Not as standard, because we are a fabrication shop rather than a certified cleanroom. What we do supply is the process record that supports a particle claim: polished edges, polished bores where specified, double-bagged packing with no cardboard in the inner pack. Where a buyer needs measured counts, we can arrange third-party testing at added cost and lead time.