Most material fights on an RFQ call are incomplete spectral requirements. Lock three lines first: wavelength (or laser line), transmit vs reflect, power density and environment. Index, hardness and cost are conditional. Harris’s SPIE handbook is still the right place to start; the table below is how we actually quote.
What we put on the print
| If the optic… | First substrate | Why not the other |
|---|---|---|
| UV–VIS–NIR, no thermal band | Fused silica / CaF2 | Ge is opaque; ZnSe is overkill and soft |
| 1.5–5 µm window, budget | Silicon | LWIR is closed; not a 10.6 µm lens |
| 8–12 µm camera | Ge or chalcogenide | ZnSe if you also need a HeNe; silica never |
| CO2 10.6 µm delivery | Laser-grade CVD ZnSe + AR | Ge runaway (Young 1971); ZnS usually higher α |
| High-power fold | Copper / metal HR | Do not force a transmitting window to do a mirror’s job |
| VIS alignment + LWIR | ZnSe or MS-ZnS | Ge has no visible path |
Two bands, two trains
Visible UI plus thermal sensing is two optical trains. One substrate across VIS + LWIR is either an impossible coating or compromised MTF. Split paths. Share the mechanics.
Send λ, AOI, clear aperture, figure at a named test wavelength, scratch-dig, coating, quantity and environment. Drawing attached. Spectral Explorer, then quote.
References. Harris, D.C., Materials for Infrared Windows and Domes, SPIE (1999). Young, P.A., Appl. Opt. 10, 638 (1971). Crystran ZnSe / Ge / Si / CaF2 / SiO2 data. Li, H.H., J. Phys. Chem. Ref. Data 13, 103 (1984) for ZnS/ZnSe/ZnTe indices.