Thermoelectrics.
Waste heat, put to work.
Name the temperature you run at, the efficiency you need and the elements you can source.


- Waste-heat recovery
- Solid-state cooling
The material groups.
Each group brings its own structures and chemistry to the search.
Chalcogenides
Born with low thermal conductivity.
Skutterudites
Cages that trap heat and let current through.
Half-Heuslers
Tough at high temperature.
Oxide thermoelectrics
Stable where it is hot.
What you can aim for.
Set the properties that matter. Say what a candidate must meet and what it should push for, with cost and supply in view.
- Figure of merit (ZT)
- Power factor
- Low thermal conductivity
- Operating temperature
- Element cost
Starting points.
Start from a proven preset, or set your own targets.
Maximum ZT
The highest figure of merit the domain allows.
Lead-free
Thermoelectrics with no lead.
Waste heat
Power from industrial waste heat.
Four modes, in every domain.
Search from the target alone, improve a material you have, or watch the engine find sealed targets blind.
Open Discovery
You have a target, not a starting structure.
Prototype-Guided Optimization
You have a good material. Make it better.
Guided Rediscovery
Point it at a known material. Watch it arrive.
Blind Rediscovery Benchmark
Seal the answer. See if it comes back.
Every candidate, with its record.
Each candidate comes ranked against your targets, with its estimated properties and their uncertainty, the checks it passed and a structure file to export, ready for your lab.
Stop guessing which crystal to make. Ask for it.
Tell us the domain and what the material must do. License MS Lattice for your team, or let Norck take the discovery all the way to finished parts.

