Description
Product Overview
The Thorlabs PF1011 is a finished optical substrate with no coating on it. It is a Ø25.4 mm (Ø1 in) plano disc of Zerodur® glass-ceramic, 6.0 mm thick with a tolerance of plus or minus 0.2 mm, diameter tolerance +0.0 / -0.1 mm, polished to λ/10 peak to valley at 632.8 nm with 10-5 scratch-dig, parallel to within 3 arcmin, with a clear aperture of 90% of the diameter and a fine-ground rear face carrying a laser-engraved item number. Zerodur® is a registered trademark of Schott. There is no film on either face, and therefore no reflectance specification applies to this part: it is not a mirror until somebody coats it.
Buying a blank is a deliberate act, and it makes sense in two situations that have nothing to do with each other. The first is when the wavelength is not settled yet. A coating is chosen for a band, and a band that changes after the purchase order turns a coated optic into an expensive paperweight. A blank lets the optical design finish first: build the mount, measure the beam, decide whether the working line is 355 nm, 1064 nm or 10.6 µm, and only then send the substrate out for coating. The second situation is when the substrate quality is the point and the coating will be done by whoever is closest to the machine, in-house or by a local coating house that already handles the customer’s other optics.
The reason to pay for Zerodur rather than fused silica is thermal, and the mechanism is worth following step by step because it is the whole argument for the material. A mirror absorbs a small fraction of the power that hits it, even a good one, because no coating is lossless. That absorbed power raises the temperature of the substrate locally, in the region the beam actually covers, not uniformly across the disc. The substrate then expands, and because the heating is not uniform, the expansion is not uniform either: the surface develops a slight curvature where it was flat. A flat mirror turned into a very slightly curved mirror has a focal power, and a focal power in a fold mirror moves the focus downstream. On a marking system that shows up as a change in spot size through a shift. On an interferometer it shows up as tilt fringes that were not there at switch-on. On a multi-pass amplifier it shows up as a change in mode matching. Nobody changed anything, and the beam moved.
Zerodur breaks that chain at the second link. Its coefficient of thermal expansion is quoted as 0 ± 0.100 × 10⁻⁶/°C, against roughly 0.55 × 10⁻⁶/°C for fused silica, a difference of more than five times. The absorbed power and the local temperature rise are unchanged; what changes is how much geometry that temperature rise produces. With a near-zero expansion coefficient, the surface stays where it was, the focal power stays at zero, and the far-field spot stays put while the enclosure warms from a cold start to operating temperature. This is not a small refinement for a laboratory bench. It is the difference between a system that needs a thirty-minute warm-up before it can be trusted and one that can be measured from the first shot.
Two qualifications keep the claim honest. The near-zero figure is a property of the glass-ceramic over a limited temperature window around room temperature, not over unlimited temperature; Zerodur is engineered for thermal stability in service, not for use at cryogenic or elevated extremes, and a specification read without its temperature range is a specification misread. And a low expansion substrate does not rescue a bad mount. A mirror clamped by a steel ring into an aluminum block still sees the differential expansion of that assembly, and the substrate cannot help with a distortion that is imposed on it from outside. Zerodur removes one term from the error budget; it does not remove the others.
What arrives is a working surface, not a raw blank in the grinding sense. The polish is λ/10 at 632.8 nm with 10-5 scratch-dig, which is a better cosmetic grade than the coated mirrors in the same family are quoted at, and the reason is straightforward: the substrate is inspected before anything is deposited on it, and the coating step plus the handling that follows is what the looser grade on a finished mirror accounts for. In practice this means a PF1011 can be used uncoated where a good flat is wanted and no film is wanted, which is a narrower set of uses than people expect but a real one: a reference surface in a Fizeau interferometer is deliberately uncoated, because a weak Fresnel return from a clean polished surface is exactly what the measurement needs and a coated surface would return far too much.
The rear face is fine ground and engraved with the item number. That matters more on a blank than on a coated part, because a blank has no coating colour to identify it by: two discs of the same diameter, one Zerodur and one fused silica, look identical in a tray. The engraving is the only thing on the part that says what it is, and on a bench where both materials are in use that is not a trivial detail. Fused silica and Zerodur are both colourless, both polish the same way and both feel the same in the hand.
Key Features
- Zerodur® glass-ceramic substrate: coefficient of thermal expansion quoted as 0 ± 0.100 × 10⁻⁶/°C, more than five times lower than fused silica, which is the property the whole part exists for
- Uncoated and ready to be coated: choose the film after the wavelength is settled, whether that is UV-enhanced aluminum, protected aluminum, protected gold or protected silver, or a dielectric stack from a coating house
- λ/10 peak to valley at 632.8 nm: a flat that will not become the limiting aberration in an interferometer or a tight focus once it is coated
- 10-5 scratch-dig polish: the substrate grade, quoted before deposition; coated mirrors in the same family are quoted at 40-20 because the coating and handling come after this inspection
- 6.0 mm thick, plus or minus 0.2 mm: stiff enough for a three-point seat, thin enough for short working distances, and a defined thickness so a coating run can be planned against a known substrate
- Parallelism within 3 arcmin: a wedge this small keeps a rear-surface ghost close to the main beam, which matters when the part is used as a window or a reference flat
- Ø22.9 mm clear aperture: 90% of the diameter, leaving a handling annulus that a retaining ring can grip without touching the polished zone
- Fine-ground rear face with engraved item number: the only identification on an uncoated disc, and the reason a mixed tray of substrates can still be sorted a year later
Technical Specifications
| Model | PF1011 |
| Part type | Mirror blank, uncoated |
| Coating | None on either face; no reflectance specification applies |
| Substrate material | Zerodur® glass-ceramic (registered trademark of Schott) |
| Coefficient of thermal expansion | 0 ± 0.100 × 10⁻⁶/°C |
| Diameter | Ø25.4 mm (Ø1 in) |
| Diameter tolerance | +0.0 mm / -0.1 mm |
| Thickness | 6.0 mm (0.24 in) |
| Thickness tolerance | plus or minus 0.2 mm |
| Surface flatness | λ/10 peak to valley at 632.8 nm over the clear aperture |
| Surface quality | 10-5 scratch-dig |
| Parallelism | 3 arcmin or better |
| Clear aperture | 90% of diameter (Ø22.9 mm) |
| Rear surface | Fine ground, laser engraved with the item number |
| Edge finish | Ground, suitable for standard retaining rings |
| Mounting | Standard Ø1 in optic mounts, kinematic seats and retaining rings |
| Typical coating options | UV-enhanced aluminum, protected aluminum, protected gold, protected silver, dielectric stacks applied by a coating service |
| Source | Manufacturer published series specification table |
Applications
A blank is bought for the substrate, so the applications below are all cases where the substrate is the hard part and the film is the easy part.
- High-power carbon dioxide laser folds: at 10.6 µm the absorption of a coated mirror is small but the power is large, and the resulting temperature rise on a substrate with ordinary expansion is enough to walk the focus during a shift; a Zerodur substrate removes that term
- Ti:sapphire amplifier chains: a multi-pass amplifier with several folds accumulates wavefront error at every pass, and a substrate whose figure does not breathe with pump power keeps the chain aligned from cold
- Industrial cutting and welding heads: an enclosure that runs twenty degrees above ambient is the normal case, not the exception, and a low-expansion fold keeps the spot size constant from the first part of the shift to the last
- Reference flats for interferometry: used uncoated as a Fizeau reference surface, where a weak Fresnel return from a clean polish is the wanted signal and a film would be counterproductive
- In-house coating programmes: a laboratory that already runs its own deposition buys the substrate to its own figure and cosmetic standard rather than buying somebody else’s coating choice
- Designs where the wavelength is still moving: a prototype that starts in the visible and ends in the infrared does not have to throw away its mirrors when the source changes
- Replacement of a damaged coated part: when the mount and the design are fixed and only the surface is ruined, a blank plus a local coating run is often faster than waiting for a coated unit
Selection Guide
The first decision is whether to buy a blank at all. A blank is the right purchase when the answer to any of the three questions below is yes; otherwise a coated part is cheaper once the coating service and the handling risk are counted.
| Question | If yes | If no |
|---|---|---|
| Is the working wavelength still undecided? | Buy the blank, coat when the design settles | Buy the coated part in the band you need |
| Does the coating have to come from a specific supplier or process? | Buy the blank and send it to that supplier | Buy the coated part from stock |
| Will the part run at high power for hours, in a warm enclosure? | Buy the Zerodur blank and coat it for the band | A fused silica coated part is adequate |
| Is the surface wanted uncoated on purpose? | Buy the blank and use it as a reference flat or window | Continue down this table |
| Is delivery time the binding constraint? | Check stock on the coated variants first | The blank is the flexible option |
The second decision is the substrate material, and it is decided by absorbed power rather than by average power. A 5 W beam on a 10 mm spot deposits very little heat per unit area and a fused silica substrate will hold its figure; a 500 W beam on the same spot deposits a great deal and the same substrate will not. The useful way to think about it is temperature rise per unit of absorbed power density, and then to ask how much figure change that temperature rise is allowed to produce. Zerodur costs more and buys headroom in exactly that term.
Series Comparison
The PF1011 substrate is the base of a family: the same disc is offered with several films already deposited. The table shows what each version adds, so the choice between buying a blank and buying a finished part is visible in one place.
| Item | Coating | Band | Surface quality | Best use |
|---|---|---|---|---|
| PF1011 | None (blank) | Not applicable | 10-5 | Substrate first, film later; uncoated reference flat |
| PF1011-F01 | UV-enhanced aluminum with MgF₂ overcoat | 250 – 450 nm | 40-20 | Excimer and third-harmonic work |
| PF1011-G01 | Protected aluminum | 450 nm – 20 µm | 40-20 | General broadband folding, moisture tolerant |
| PF1011-M01 | Protected gold | 800 nm – 20 µm | 40-20 | Infrared systems, chemically inert film |
| PF1011-P01 | Protected silver | 450 nm – 20 µm | 40-20 | Preferred film for visible and near infrared beams |
One comparison is worth making explicitly because it is the question customers actually ask: why a Zerodur blank instead of a fused silica one at a fraction of the price. The answer is not reflectance, since neither surface carries a film. It is the expansion coefficient, five times apart, and it only becomes visible in a system that puts power on the optic for long enough for the substrate to warm. On a low-power alignment bench the two materials behave identically and the cheaper one is the correct purchase. On a kilowatt process head they do not, and the difference shows up as a focus that drifts through the shift.
FAQ
Q: Does this part have a reflectance figure?
A: No. There is no coating on either face, so no reflectance specification exists for it and none should be inferred from the family it belongs to. An uncoated polished surface returns a weak Fresnel reflection, which is useful in a reference flat and useless as a mirror. If a reflectance figure is what you need, choose one of the coated versions in the comparison table.
Q: Why is the blank quoted at 10-5 when the coated mirrors are 40-20?
A: Because the substrate is inspected before deposition and the coated part is inspected after it. In practice the polish is good enough for 10-5, and what the looser figure on a finished mirror accounts for is the coating process and the handling that follows it. A customer who specifies 10-5 on a coated part is asking for something the deposition step cannot guarantee, which is why the catalogue does not offer it.
Q: Is Zerodur really zero expansion at any temperature?
A: No, and treating it that way is the commonest misunderstanding of the material. The near-zero coefficient applies over a limited window around room temperature, which is where optical systems live. Outside that window the material expands like other glass-ceramics. Zerodur is specified for thermal stability in service, not as an athermal material for arbitrary conditions.
Q: How does a warm mirror actually move the focus?
A: Through curvature, not through bulk expansion. The beam heats a patch of the surface, that patch expands more than the rest of the disc, and a flat surface becomes very slightly convex or concave. A curved fold mirror has focal power, and focal power moves the focus downstream. The shift is small in absolute terms and large compared with a diffraction-limited spot, which is why it is measured in process drift rather than seen by eye.
Q: Can I have it coated by someone else?
A: Yes, that is the normal use of a blank. Two practical notes: ask the coating house for its substrate handling procedure before shipping, because a 10-5 surface that arrives with a handling mark is no longer 10-5, and confirm that the deposition temperature is compatible with the substrate, since a coating process that runs hot enough can affect the figure of a finished polish.
Q: Can I use it uncoated as a window?
A: Yes, with the usual caution that an uncoated parallel plate produces two Fresnel reflections and a weak interference ripple across the transmitted beam. For a window in a low-coherence path that is harmless. For a coherent beam where ripple matters, a wedged or coated window is the better part, and this disc is not wedged.
Q: How do I tell it apart from a fused silica disc of the same size?
A: Read the engraving on the rear face. Both materials are colourless, both polish the same way, and both feel identical in the hand, so appearance is not a reliable test. This is the practical reason the item number is engraved on the back of a blank rather than printed on a bag that gets thrown away.
Usage Precautions
- Handle by the rim with gloves or finger cots. On an uncoated surface a scratch is permanent: there is no film to be re-deposited and no inexpensive rework, so the polish you damage is the polish you keep.
- Do not lay the polished face down on a bench, and do not stack two blanks face to face. Glass-ceramic is hard and brittle, and two polished surfaces in contact will chip at the edge under their own weight.
- Clean with solvent and a lens tissue in a single direction, after blowing off loose dust. There is no coating to attack with solvent, which makes an uncoated surface more forgiving chemically and less forgiving mechanically.
- Keep the part sealed until it is either mounted or shipped for coating. A polished surface that sits exposed collects dust that has to be removed later, and every removal is a chance to scratch.
- Protect the edges in transit to a coating house. Edge chips are the most common shipping damage on bare substrates and they are not repairable.
- Do not over-clamp. A 6 mm disc in a metal ring bends, and the λ/10 figure is measured before anybody tightens anything.
- Do not assume the figure survives mounting. Measure the assembled part if the application is wavefront-critical, because a mount can impose more error than the substrate specification allows.
- Record the item number before the part leaves your hands. Once it is at a coating house, the engraving on the back is the only link between the substrate you sent and the mirror you get back.
Disclaimer
Specifications on this page are compiled from the manufacturer published series tables and are offered for selection purposes only; the current official datasheet governs. This part carries no coating and therefore no optical reflectance performance is stated or implied. Coating services, price and delivery are quoted separately by our sales desk, and any third-party coating is outside the manufacturer’s specification.
Contact Us
Tell us the wavelength you expect to end up at, the power density on the optic, the beam diameter and whether the enclosure runs warm, and we will say whether a blank is the sensible purchase or whether a coated part already covers the case. If you intend to coat it elsewhere, tell us the deposition process and we will confirm that the substrate is suitable for it. Project and bulk quantities are quoted individually.








Reviews
There are no reviews yet.