Description
Product Overview
The Thorlabs PF07-03-M01 is a Ø19.0 mm gold mirror with a dielectric overcoat, specified above 96% average reflectance from 800 nm to 20 µm on a 6.0 mm fused silica substrate. It is the broadband infrared workhorse of the group: not the highest number at any wavelength, but the one whose curve stays put across nearly two decades of spectrum.
The reason is the metal itself. Gold is a free-electron reflector with its interband absorption in the blue and violet, which is why it looks yellow and why it is useless below the visible. Above that edge it has nothing to absorb: no oxide growth, no absorption band in the near or mid infrared, and only a slow decline toward the far infrared. Aluminium, by comparison, carries interband structure near 800 nm that shows as a dip exactly where many laser diodes and Ti:sapphire systems sit, and silver, though brighter in the visible and near infrared, falls away more quickly at long wavelength. When a band is wide, or when the source has not been chosen yet, flatness of the curve is worth more than the peak value.
The overcoat is where the engineering compromise lives, and Thorlabs is unusually direct about it: the protective layer results in polarization changes that can be difficult to quantify and sometimes induces absorption in the mid to far infrared. In practice this means a protected gold mirror is not an unprotected gold mirror at 15 µm, and it is not the right optic for a polarisation-critical interferometer without measuring the effect first. What the overcoat buys in exchange is real: the film can be cleaned, it resists physical damage, and it does not have to be treated as a bare metal surface that will scratch if a tissue touches it.
The 19 mm format is the Polaris fixed-mount size, and for infrared work the argument for a fixed mount is stronger than usual. An 800 nm to 20 µm beam is invisible, so alignment cannot be done by eye on the spot; it is done on mechanical datums, and a mount whose seat repeats is what makes that work. The POLARIS-19S50/M holds a 6 mm thick, 19 mm optic on three axial contact points inside a Ø1 inch footprint, and the part number is laser engraved on the back of the mirror so an unmarked optic can be identified after it has been in a system for a year.
Key Features
- 96% average from 800 nm to 20 µm: a curve without absorption structure, which is what makes the part usable when the wavelength is not fixed in advance
- Highest nanosecond threshold of the group: 2 J/cm² at 1064 nm with 10 ns pulses at 10 Hz over Ø1.000 mm
- CW ratings for both common infrared lines: 500 W/cm at 1.070 µm (Ø0.089 mm) and 750 W/cm at 10.6 µm (Ø0.339 mm)
- Cleanable dielectric overcoat: the film can be wiped, which bare gold cannot, at the cost described above
- Polarization caveat stated by the manufacturer: the overcoat changes polarization in ways that are hard to quantify and can add mid to far infrared absorption
- Ø19.0 mm x 6.0 mm fused silica: λ/10 at 633 nm, 40-20 scratch-dig, parallelism within 3 arcmin, clear aperture 90% of diameter
Technical Specifications
| Model | PF07-03-M01 |
| Coating | Protected gold |
| Overcoat | Dielectric, protective |
| Wavelength range | 800 nm – 20 µm |
| Average reflectance | above 96% from 800 nm to 20 µm |
| Reflectance curves | published at 12 degree and 45 degree angle of incidence |
| Substrate | Fused silica |
| Diameter | Ø19.0 mm (0.75 in) |
| Diameter tolerance | +0.0 mm / -0.1 mm |
| Thickness | 6.0 mm (0.24 in) |
| Thickness tolerance | +/-0.2 mm |
| Clear aperture | 90% of diameter, about Ø17.1 mm |
| Front surface flatness | λ/10 peak to valley at 633 nm |
| Surface quality | 40-20 scratch-dig |
| Parallelism | at most 3 arcmin |
| Pulsed damage threshold | 2 J/cm² at 1.064 µm (10 ns, 10 Hz, Ø1.000 mm) |
| CW damage threshold | 500 W/cm at 1.070 µm (Ø0.089 mm); 750 W/cm at 10.6 µm (Ø0.339 mm) |
| Overcoat side effect | polarization change and possible mid to far infrared absorption |
| Cleanable | Yes |
| Source | Thorlabs official series specification table |
Applications
The mirror is chosen when the band is wide, the source is powerful, or the wavelength may change during the life of the build. The list below is what that looks like on a real bench.
- CO2 laser beam delivery: folding a 10.6 µm beam through a marking or cutting head, where the 750 W/cm CW rating at that wavelength is the number that matters
- Infrared thermal imaging and radiometry: relaying an uncooled or cooled detector field through a folded path without introducing a wavelength-dependent throughput that would have to be calibrated out
- High average power 1064 nm folding: amplifier chains and fibre delivery heads, where the 2 J/cm² nanosecond threshold and 500 W/cm CW rating leave margin that the other coatings do not
- Wideband infrared spectroscopy: a bench that runs from roughly 1 µm to 12 µm with grating or FTIR instruments and cannot afford a coating with an absorption feature in the middle of the range
- Multi-source laboratories: one fold shared by a 1064 nm laser, a 1.5 µm source and a thermal camera, where a coating optimised for one of them would compromise the others
Selection Guide
- Check the power before the reflectance: if the beam is high peak or high average power, this is the infrared coating to take. The mid-infrared enhanced gold reaches 98% but its pulsed threshold is twenty times lower, and its CW rating at 1.07 µm is twenty times lower
- Use the mid-infrared coating only above 2 µm: between 800 nm and 2 µm the protected gold is both more reflective and far more robust, so there is no reason to accept the lower threshold for a band the other coating already covers
- Do not plan on seeing the beam: at 1064 nm the spot is faint and beyond 1.5 µm it is invisible. Provide a flip-in alignment target or a visible tracer, and treat the mechanical datum as the reference
- Measure before trusting it in a polarisation-critical path: the overcoat changes polarisation in a way the manufacturer explicitly declines to quantify. For an interferometer that depends on extinction ratio, characterise the mirror or use an unprotected gold item
- Size for the beam, and then for the mount: 19 mm gives about 17 mm of clear aperture; a 10 mm beam is comfortable, a 16 mm beam is not, and a beam wider than the aperture will scatter off the coating edge into the detector
Series Comparison
The four infrared options are close enough in band that the deciding columns are the damage threshold and the shape of the curve, not the headline reflectance.
| Item | Coating | Wavelength range | Average reflectance | Pulsed threshold at 1064 nm | Notes |
| PF07-03-M01, this page | Protected gold | 800 nm – 20 µm | 96% | 2 J/cm² | Widest band with the highest threshold |
| PF07-03-M02 | MIR-enhanced gold | 2 – 20 µm | 98% | 0.1 J/cm² | Higher reflectance above 2 µm, one tenth the threshold |
| PF07-03-P01 | Protected silver | 450 nm – 2 µm; 2 – 20 µm | 97% / 95% | 1 J/cm² | Brighter in the visible, declines toward the far infrared |
| PF07-03-G01 | Protected aluminium | 450 nm – 2 µm; 2 – 20 µm | 90% / 95% | 0.3 J/cm² | Cheapest broadband option, absorption structure near 800 nm |
The same gold coating on smaller substrates covers compact layouts: Ø1/2 inch protected gold, Ø7.0 mm protected gold, and for low thermal expansion the Zerodur protected gold mirror.
FAQ
Why does this coating stop at 96% when the mid-infrared gold reaches 98%? Because they are optimised for different things. The mid-infrared enhanced coating is tuned to push reflectance above 2 µm and accepts a much lower damage threshold and unusable near-infrared performance to get there. The protected coating covers 800 nm to 20 µm at a flat 96% and keeps a 2 J/cm² pulsed threshold, which is the more useful combination for a general infrared bench.
Can I align the beam with a HeNe through this mirror? Not reliably. A 633 nm helium-neon beam is below the 800 nm band edge, where gold absorbs, so the return will be weak and the alignment misleading. Use a flip-in target, a visible tracer beam that shares the same path downstream, or the mechanical datum of the mount.
Does the overcoat affect polarization? Yes, and Thorlabs says so explicitly: the overcoat produces polarization changes that are difficult to quantify. For an application where extinction ratio or polarization state matters, either measure the mirror at the working wavelength and angle, or move to an unprotected gold coating.
Will it survive a 100 W CO2 beam? Compare the linear power density with the 750 W/cm rating at 10.6 µm, not the total power. A 100 W beam spread over 20 mm is 50 W/cm and is comfortable; the same power in a 1 mm spot is 1000 W/cm and exceeds the rating. Beam diameter at the mirror is the variable that decides it.
Is a protected gold mirror cleanable? Yes. That is the main practical advantage over unprotected gold, which scratches easily and cannot be wiped. Blow off loose contamination first, then use reagent-grade solvent on a lens tissue with a single-direction drag, and never scrub the surface dry.
Usage Precautions
- Assume the beam is invisible: use an infrared viewing card, a beam profiler or a fluorescent target to confirm the path. A 2 µm beam at full power will not produce a visible spot, and an unseen reflection off a fold is a genuine hazard
- Keep the optic dry and dust-free: gold itself does not oxidise, but contamination on the surface absorbs at 10.6 µm and becomes the seed of laser damage. A clean mirror at 750 W/cm is safe; a dusty one is not
- Do not over-tighten the retainer: the substrate is 6 mm fused silica and will bend under a heavy ring. A bent mirror in an infrared imaging path shows as a focus shift that is easy to blame on the detector
- Watch for polarization-sensitive layouts: if the bench includes a polarising beam splitter or a waveplate downstream, remember that the overcoat is not polarization neutral
- Store in a case, not in a drawer: the coating is hard enough to clean but not hard enough to survive contact with other optics. Loose storage is the most common cause of a scratched gold mirror
- Re-check throughput after service: a mirror that has been removed and refitted can come back with a contaminated face. Compare a power reading before and after, and clean before concluding that the laser has degraded
Disclaimer
Specifications on this page are compiled from Thorlabs official published material and are provided as a selection reference. The current official specification sheet governs; specifications and appearance may change without notice. Price and lead time are quoted by our sales team on request.
Contact Us
Changsha Soar Laser Equipment Co., Ltd. supplies Thorlabs infrared optics and can quote alternative coatings against a specific source, power density and band. Telephone +86 159 2722 0467, email 859818709@qq.com, or see www.soar-laser.cn.









Reviews
There are no reviews yet.