What Is a Hot Mirror? A Semiconductor Lithography Perspective

Optical designers sometimes treat mirrors as simple parts, yet in Semiconductor Lithography the mirror decides beam direction, loss budget and even image contrast. The…

Optical designers sometimes treat mirrors as simple parts, yet in Semiconductor Lithography the mirror decides beam direction, loss budget and even image contrast. The Hot Mirror is a quietly critical component whose details repay careful attention.

A Hot Mirror is an optical component built so that reflection happens at the coated front face rather than through a substrate. With a dichroic (transmits visible, reflects IR) on a float or borosilicate glass base, the part delivers > 90% visible transmit reflectivity across IR reflect / visible pass while keeping the useful aperture clean and ghost-free.

When light meets the Hot Mirror, almost all of it bounces from the front coating. The substrate merely holds the coating in place; it does not need to be traversed by the useful beam, so transmission losses and secondary reflections stay minimal — a real advantage in sensitive Semiconductor Lithography setups.

Most of the engineering in a Hot Mirror lives in its dichroic (transmits visible, reflects IR). The stack is designed for IR reflect / visible pass and delivers > 90% visible transmit, and its environmental protection layer keeps the metal from tarnishing so the mirror keeps working year after year.

Substrate choice for a Hot Mirror is a trade between optical grade and budget. float or borosilicate glass is a common pick because it can be cut and polished to 4–6λ flatness and a 60-40 surface, which is plenty for the reflection quality most Semiconductor Lithography systems require.

When you specify a Hot Mirror, the numbers that matter are flatness 4–6λ, finish 60-40, and the reflectance > 90% visible transmit across IR reflect / visible pass. Thickness 1–3 mm is mostly about handling and mount compatibility, but it still belongs on the print. The specification table covers the common configurations.

Where projecting nano-scale patterns with extreme precision, a Hot Mirror earns its place by doing one job reliably: turning the beam without adding noise. In Semiconductor Lithography that reliability is the difference between a prototype and a shippable product. More application examples are worth a look if the use case is close to yours.

Behind the performance

What reads on a datasheet as "> 90% visible transmit over IR reflect / visible pass" is really the outcome of interference. The dichroic (transmits visible, reflects IR) on a float or borosilicate glass base is built layer by layer so reflected waves reinforce. Flatness 4–6λ then keeps that wavefront from bending. The physics is old; the discipline to repeat it is the hard part.

Selecting a Hot Mirror for Semiconductor Lithography starts with the wavelength and angle of incidence, then the acceptable loss. Match the dichroic (transmits visible, reflects IR) to IR reflect / visible pass, confirm > 90% visible transmit, and make sure the float or borosilicate glass and 1–3 mm fit the mount you already have. The spec and size tables make that comparison quick.

A Hot Mirror is tougher than it looks but softer than you think. Fingerprints on the dichroic (transmits visible, reflects IR) are the usual cause of field failures, so edge-handling and capped storage pay off. A little discipline keeps > 90% visible transmit where it belongs.

At JYOPTO we make Hot Mirror parts by cutting float or borosilicate glass with laser accuracy of ±0.01 mm, then applying the dichroic (transmits visible, reflects IR) under vacuum. Standard blanks run 1–3 mm thick, edges are smoothed for safe handling, and every shipped mirror meets 4–6λ flatness with a 60-40 surface — the same disciplines we apply across our optical glass, vacuum-coating and precision cold-processing lines since 2020.

Quick terminology

"Flatness 4–6λ" describes how close the surface is to a perfect plane, in fractions of a wavelength. Tighter flatness costs more but protects wavefront quality, which is why Semiconductor Lithography systems specify it explicitly rather than leaving it to chance.

For Semiconductor Lithography, do not over-specify. Choose the dichroic (transmits visible, reflects IR) that covers IR reflect / visible pass at the angle you use, keep flatness at 4–6λ unless the wavefront demands more, and you will have a Hot Mirror that is both capable and economical.

Mounting notes

A Hot Mirror is only as good as its mount. Use edge contact rather than clamping the face, avoid over-tightening that bends the float or borosilicate glass and degrades 4–6λ, and keep the coated side clear of adhesive. In Semiconductor Lithography a kinematically supported mirror stays aligned through thermal cycles and shipping.

Optical designers sometimes treat mirrors as simple parts, yet in Semiconductor Lithography the mirror decides beam direction, loss budget and even image contrast. The Hot Mirror is a quietly critical component whose details repay careful attention.

Where projecting nano-scale patterns with extreme precision, a Hot Mirror earns its place by doing one job reliably: turning the beam without adding noise. In Semiconductor Lithography that reliability is the difference between a prototype and a shippable product. More application examples are worth a look if the use case is close to yours.

Most of the engineering in a Hot Mirror lives in its dichroic (transmits visible, reflects IR). The stack is designed for IR reflect / visible pass and delivers > 90% visible transmit, and its environmental protection layer keeps the metal from tarnishing so the mirror keeps working year after year.

The Hot Mirror is not exclusive to Semiconductor Lithography. Universities, service centers and R&D groups use it wherever a beam must turn, which makes a flexible, customizable part a quiet workhorse across the optics world.

Because we control cutting, coating and finishing in one place, a Hot Mirror can move from your drawing to a finished part without hand-offs. The float or borosilicate glass is cut to ±0.01 mm, the dichroic (transmits visible, reflects IR) is vacuum-deposited for > 90% visible transmit over IR reflect / visible pass, and the result is inspected to 4–6λ flatness and 60-40 quality.

Quality control

Every Hot Mirror is measured, not just sampled. Flatness against a reference, scratch-dig against 60-40, and a reflectance spot-check at IR reflect / visible pass confirm the dichroic (transmits visible, reflects IR) performed as designed. Documented results matter most for Semiconductor Lithography, where one bad part can stall a whole instrument.

Quality control

Every Hot Mirror is measured, not just sampled. Flatness against a reference, scratch-dig against 60-40, and a reflectance spot-check at IR reflect / visible pass confirm the dichroic (transmits visible, reflects IR) performed as designed. Documented results matter most for Semiconductor Lithography, where one bad part can stall a whole instrument.

Reflection on a first surface is straightforward physics: photons strike the coated face and are returned according to the law of reflection, angle in equals angle out. Because the coating sits on top, there is no second surface behind it to create a faint ghost image, which matters whenever contrast or measurement accuracy is at stake.

Mounting notes

A Hot Mirror is only as good as its mount. Use edge contact rather than clamping the face, avoid over-tightening that bends the float or borosilicate glass and degrades 4–6λ, and keep the coated side clear of adhesive. In Semiconductor Lithography a kinematically supported mirror stays aligned through thermal cycles and shipping.

For Semiconductor Lithography, do not over-specify. Choose the dichroic (transmits visible, reflects IR) that covers IR reflect / visible pass at the angle you use, keep flatness at 4–6λ unless the wavefront demands more, and you will have a Hot Mirror that is both capable and economical.

In short

For Semiconductor Lithography, the Hot Mirror is less a commodity than a tuned component. Specify the band (IR reflect / visible pass), the reflectivity (> 90% visible transmit) and the figure (4–6λ), and you will spend less time debugging light you cannot see. That is the whole game. Where your application sits among the sectors we serve changes the details, not the method.