Two quotations for the same optical glass part in a hydrogen application can differ by a factor of three, and the reason is rarely the material. The specification is written around chemical resistance and pressure rather than around optical performance, and the processes that deliver those numbers are the expensive ones. This article breaks down where the money actually goes in glass for electrolysers and hydrogen instrumentation, and which tolerances are worth paying for.
Optical Glass Materials in Green Hydrogen
In green hydrogen, that translates into a specific set of requirements rather than a general preference for quality. Glass appears in two different roles: as a window into an electrolyser stack, where it lets an instrument or a camera observe the electrodes and the membrane without opening the vessel, and as the optical element in a sensor that detects hydrogen in air or measures its concentration in a process stream. The first role is dominated by pressure and chemistry, because a window in a cell sees hot alkaline or acidic electrolyte at pressure for years. The second role is dominated by the absorption feature being measured, because hydrogen absorbs in the near infrared and the optics are chosen around that line rather than around visible transmission.
Substrate and material selection
The substrate choice follows the chemistry of the cell. Alkaline electrolysers expose the window to a hot potassium hydroxide solution, which attacks ordinary optical glass, so borosilicate, sapphire or fused silica are the practical options, and sapphire is chosen where both the chemistry and the abrasion are severe. Proton exchange membrane cells run acidic, which rules out materials that leach alkali and pushes towards fused silica or quartz. Where the window is also a pressure boundary, the thickness is set by the mechanical design and the optical requirement becomes a constraint rather than a driver. For hydrogen sensors, the material is selected to transmit the measurement band without absorption features of its own, which in practice means fused silica or sapphire depending on how far into the infrared the instrument works.
Handling, cleaning and packaging
Glass for a hydrogen application is usually part of a sealed assembly, so the cleanliness that matters is the cleanliness inside the seal rather than on the outer surface. A particle trapped between the window and its seal becomes a leak path, which means the parts are handled in a controlled environment from the end of the coating process to the moment the assembly is closed, and the packaging has to preserve that state. Sapphire is hard and chemically inert but brittle, and it chips at the edge unless the chamfer is controlled, so edge protection in transit is a real requirement rather than a formality. Fused silica is more forgiving mechanically and less forgiving chemically, and it should not be stored against acidic or hygroscopic materials that can leave residues on the face.
The tolerances that actually matter
For an observation window, the tolerances that matter are mechanical and chemical: flatness enough to seal, thickness uniformity enough to keep the stress predictable, edge quality that will not initiate a crack under pressure cycling, and a surface that resists the electrolyte. Optical transmission is a secondary requirement in this role, which is the opposite of the usual optical specification. For a sensor optic, the tolerances invert: the transmission and the blocking at the measurement wavelength dominate, and the mechanical requirements are set by the housing. In both cases the specification should state the service environment explicitly, because a window qualified for 30 bar at 80 degrees in alkaline solution is a different part from one qualified for the same pressure in water.
What drives cost and lead time
Cost is driven by the pressure boundary rather than by the optical grade. A window that has to seal against a pressurised cell needs a controlled thickness, a defined edge profile and a surface finish that the seal can grip, and each of those adds a process step. The second driver is material: sapphire in a large diameter is expensive and slow, because the material is grown rather than cast, and a big window pushes both the material cost and the machining time. The third is qualification, because hydrogen equipment is usually certified as an assembly and the paperwork follows the part. Lead time is dominated by the material order for sapphire and by the coating cycle for coated parts, and neither shortens with order size in the way that a mature optical part would.
Requirements specific to Green Hydrogen
Hydrogen hardware adds constraints that a general optical specification does not carry. The service environment combines pressure, temperature and aggressive chemistry, and all three act at once on the same surface. Safety documentation follows the assembly, so the window supplier is usually asked for material certificates and dimensional records that can be folded into the customer's certification package. Where the gas is present, materials that are prone to hydrogen embrittlement are excluded, which is a mechanical rather than an optical criterion but it constrains the mounting design around the glass. Long service intervals matter more than peak performance, because opening a cell to replace a window is an outage rather than a maintenance task.
- Chemistry-led substrate choice: alkaline and acidic cells need different materials
- Pressure boundary tolerances that go beyond the optical specification
- Edge profile and surface finish qualified for repeated pressure cycling
- Material certificates that fold into the assembly certification package
Pressure equipment certification handled at the assembly level, material certificates for the glass batch, and a defined service environment covering chemistry, temperature and pressure cycling.
Selection data at a glance
| Material | Index (nd) | Typical use | Notes |
|---|---|---|---|
| Substrate | borosilicate, fused silica, quartz, sapphire | Chemistry dependent | |
| Alkaline resistance | sapphire and borosilicate suitable | KOH at temperature | |
| Acidic resistance | fused silica and quartz suitable | PEM cells | |
| Thickness | 3-20 mm for pressure windows | Mechanical driven | |
| Flatness | λ/2 or better on the sealing face | Seal dependent | |
| Edge profile | controlled chamfer, no chips | Pressure cycling | |
| Transmission | over 90 % at the measurement band | Secondary for windows | |
| Operating temperature | up to 90 °C continuous | Electrolyser dependent | |
| Documentation | material certificate and dimensional record | Per assembly |
Frequently asked questions
Why not use ordinary optical glass for an electrolyser window?
Because hot alkaline electrolyte attacks it. Ordinary optical glass contains alkali and alkaline earth oxides that are soluble in strong hydroxide solutions, so the surface etches and the part both weakens and clouds. Borosilicate, fused silica and sapphire all resist it, and the choice among them is settled by the mechanical duty rather than by the chemistry alone.
Does the pressure rating change the optical specification?
It changes which numbers matter. A pressure window is specified by thickness, flatness on the sealing face and edge quality before anything optical is considered, because those are what make the seal work. Transmission is still required, but it is usually easy to achieve once the mechanical requirements are met.
What should be documented for a hydrogen installation?
The glass batch and material certificate, the dimensional record for the finished part, and the surface finish on the sealing face. Certification is handled at the assembly level, so the supplier's data has to be in a form the customer can fold into that package rather than as a standalone optical report.
Need a quotation for optical glass materials?
Send your drawing, target specification or coating requirement and an optical engineer will reply within 24 hours with a price and a lead time.
Request a Quote Email an engineer