What's useful
- Run records: process parameters for each run (temperatures, pressures, gas flows, times, tool settings).
- Measured results: metrology, yield, electrical test and characterisation data tied to those runs.
- Lab notebooks: experiment plans, observations and conclusions.
- Failed runs. Results that didn't work are as useful as the ones that did.
The key is the link: a recipe with its measured outcome. Parameters without results, or results without the recipe, are much less useful.
What qualifies
The listing is for semiconductor and materials companies of roughly 10 to 2,000 employees, which covers many UK compound semiconductor, materials and photonics firms as well as larger fabs.
What to check first
- Export controls. Semiconductor manufacturing technology is one of the most tightly controlled areas. UK export controls apply to process data and know-how, not just equipment, and US rules can apply to US-origin technology. Your export compliance lead should clear the scope before anything is shared.
- Customer and partner agreements. Foundry customers, joint development partners and funded programmes often own results or restrict their use.
- Trade secrets. Decide which recipes are core to your competitive position and keep them out of scope if needed.
- Grant terms. Publicly funded projects may carry conditions about how results are used.
What it's worth
Mercor lists potential company earnings of $100kâ$750k for semiconductor and materials companies, depending on what's in scope. That is Mercor's own figure, not a promise or guarantee from us.
Can we license older process generations only?
Yes. Excluding current processes is a common way to protect your competitive edge while still licensing a useful archive.
Our data sits across several tools. Is that a problem?
No, as long as runs can be matched to results, for example by lot or wafer ID.