In developmentPilot partners wanted
AuxeinAuxeinby Bucket Labs

Auxein: test the run before you run it.

Auxein is built to make every crystal run predictable.

A digital twin of your furnace, in development now at Bucket Labs.

Built from its geometry and thermal behaviour and corrected against your real growth data, so a recipe change can be tried in simulation before it costs a furnace run.

NVIDIAMember of NVIDIA Inception

We get there furnace by furnace, in the plant, with your engineers. Every hot zone we work on (the heaters, insulation and crucible at the core of the furnace) teaches the models something no off-the-shelf tool could know yet, and that is what the product is being built out of.

Built on what
already works.

A twin is only as good as the record it is corrected against. Our team has built and run these four in production, and each one is a layer Auxein stands on. Underneath them is the data foundation, including thermocouple drift detection, which keeps the temperature record the twin is corrected against honest.

  1. In-situ growth monitoringCatch problems mid-run
  2. Substrate defect metrologyTurn inspection into a measurement
  3. 3D boule reconstructionReplay the whole run
  4. Device yield & cost modelCatch expensive failures upstream
A seed crystal on a thin rod dipping into a crucible of glowing orange melt, where growth begins.

How a single crystal is grown

It starts as one continuous lattice. Grown from a melt, atom by atom, long before any wafer is cut from it. Three controls hold it in order. Temperature gradient, rotation and pull rate, held steady for days to months depending on the material. Those controls interact. Move one and the others move with it. The furnace is sealed, so the result of the choice arrives at the end of the run. We help you get it right in simulation first. Auxein gives your team a digital twin of your furnace to test process changes before you commit a run. You grow the crystal. We help you grow it better.

Physics first,
data second.

In crystal growth you cannot wait for data. A run takes days to months, depending on the material. The failure modes you most want to predict are rare by design, because the process has already been tuned to avoid them. A facility with twenty years of history may still have only a few hundred runs on record, and the useful labels are thinner than that.

Learning from observation alone does not work at that sample size. So we invert the order. We build the simulation first and use real runs to correct it.

  1. Build the twin from physics and geometry

    We start from the furnace itself. Hot zone geometry, heater configuration, insulation, crucible dimensions, material properties. That gives a thermal model of your furnace that produces sensible answers on day one, before it has seen a single one of your runs.

  2. Generate training data from the twin

    The twin can be run thousands of times across the parameter ranges you care about. Gradient, rotation, withdrawal rate, power profile. That generated set is large enough and varied enough to train a model, including regions of the parameter space you would never deliberately enter on real hardware.

  3. Correct against real runs

    Your furnace does not behave exactly like its model. As real runs accumulate, we use them to correct the twin and the model trained on it. Each run moves the simulation closer to your specific hardware. The design goal is that a handful of runs is useful here, because they correct a model that already works rather than teaching one from nothing.

  4. Test changes before committing a run

    Once the twin tracks your process, the aim is that you can ask it questions. What happens to the gradient if we change this heater ratio. Where does this profile put the interface shape. Auxein is designed to answer in hours, instead of committing months of furnace time to find out.

The simulation is what you see.
Most of the work is underneath.

Most growers aren't ready to adopt off-the-shelf simulation software, and they shouldn't have to be. Before a twin can predict anything, the data beneath it has to be trustworthy. That's the work we do with you, and every layer is useful on its own.

  1. Simulation and prediction (Auxein)

    What-if runs and predictions your team uses.

  2. In-situ monitoring

    See during a run whether it's on track, while there's still time to act.

  3. Predictive maintenance and defect trends

    Learn what normal looks like for each heater and temperature sensor, so drift is caught before it costs a run, and surface the process signatures that keep preceding failed runs.

  4. Data foundation

    Clean pipelines, data quality, validation and the infrastructure to keep it running.

That's why we start with discovery, not a demo.

How we work

Auxein aims to make that foundation more predictable before a single crystal is grown.

What's available today

What Auxein does not do

  • It does not control your furnace. We do not take a control action on your hardware. The output is information your team acts on.
  • It does not remove the need to run the furnace. Simulation narrows the set of experiments worth running. Real runs still decide the answer.
  • It does not arrive already knowing your process. The first version is built from physics and your own data. Accuracy against your specific hardware improves as your runs correct it.

The growth methods
we work in

Bridgman and vertical gradient freeze (VGF)

The melt freezes from the bottom up inside a crucible as the temperature profile slowly moves. Our primary area of competence, including high-pressure VGF for indium phosphide.

Gallium arsenide (GaAs)Indium phosphide (InP)Cadmium zinc telluride (CdZnTe)Compound semiconductors

Czochralski

A seed crystal is dipped into the melt and slowly pulled upward while rotating.

Silicon (Si)Germanium (Ge)Oxide crystals

Kyropoulos

The crystal grows down into the melt as the furnace slowly cools.

SapphireLarge optical crystals
“I truly believe a simulation for crystal growing would be a valuable tool.”
SVP of Manufacturing, crystal grower

Auxein is in development, and we are looking for pilot partners to build it with.

Become a pilot partnerHow a pilot works