← DriftWest
Open research log · DriftWest Labs

Chasing the
Room-Temp Superconductor

Zero-resistance electricity at everyday temperatures is the last great prize in materials science. We can't build a lab on a home server — but we can teach a machine to read 21,000 known materials and point at where the treasure hides.

0.93
R² accuracy
±5 K
avg error
21,263
materials
108 s
to train
Watch first · 5 min

The treasure map, animated

Before the machine learning, here's the physics in plain language — what superconductivity is, why heat is the enemy, and the three unsolved gaps where a fresh idea could actually matter.

superconductors_ep1.mp4 · 1080p · 5m31sopen direct link →
The physics, in one breath

The glue is everything

Electrons normally repel and bump into the lattice, losing energy as resistance. In a superconductor they pair into Cooper pairs that glide with zero resistance. Heat is the enemy — it jiggles the atoms and rips the pairs apart. The whole quest is: keep the pairs glued at high temperature. What supplies the glue?

Glue A · phonons
250 K

Atoms vibrating in sync. Best result — but only under Earth-core pressure. Great glue, impossible plumbing.

Glue B · magnetic
135 K

Copper-oxide layers, works at normal pressure. Catch: 40 years on, nobody fully knows why.

Glue C · ???
?

An unknown mechanism at ambient pressure. This is the open door — and where an outsider hunch can matter.

What we built · runs on the home Optiplex

A model that predicts critical temperature

We trained a gradient-boosted model on the UCI Superconductivity dataset — 21,263 real materials, each described by 81 features derived from its chemistry. Given a formula it never saw, it predicts the temperature at which superconductivity kicks in.

R² = 0.93 on held-out materials, with 5-fold cross-validation at 0.928 ± 0.002 — the trustworthy number, no fluke. Average prediction lands within ~5 K of the real value. That matches the published state-of-the-art baseline (Hamidieh 2018) on turn one.

The model's confession

One lever rules them all

When we ask the model which features drive its predictions, the answer is startlingly lopsided. The spread in thermal conductivity among a material's atoms carries the bulk of the signal — and thermal conductivity is a phonon story. The machine is quietly shouting the physics: the glue is in the lattice vibrations.

Where the treasure is buried

Two findings that actually matter

1

The dominant lever is a phonon lever. Of all 81 features, the range of thermal conductivity alone accounts for ~70% of predictive power. It's a direct echo of the "phonon glue" from the map — the model rediscovered the physics from raw data.

2

Where it fails is the interesting part. Nearly every one of the model's biggest misses is a cuprate — the copper-oxide compounds nobody has a theory for. A feature-averaging model literally cannot see their secret. That's not a bug; it's the map pointing straight at the undiscovered physics.

The course ahead

Where we row next

Done

Baseline model

State-of-the-art Tc prediction trained and validated on this box.

3
Then

Crack the cuprates

Build a second model just on the families it fails on, and try to beat it.

4
The creative wedge

Feature injection

Invent a new descriptor — the "faking pressure with chemistry" hunch — and see if it moves the needle.

No snake oil

The brutal honesty

What this is — and isn't

  • A prediction is a lead, not a discovery. Someone still has to synthesize and measure a candidate to claim the prize.
  • The field is littered with mirages (LK-99, 2023). We stay skeptical by default.
  • But screening works — the near-room-temp hydrides were predicted on a computer first and made second. That door is open to us.