QHD is our research framework. Crystal lattices vibrate in patterns, and some of those patterns are soft: they let the structure bend toward a new arrangement. We follow those harmonic distortions to find metastable materials that ordinary searches miss.
Quantum Harmonic Distortion is the idea that a material's own vibrations show where it can go next.
Every crystal vibrates in a fixed set of patterns, its harmonics, much like the notes of a string instrument. In a stable material every pattern springs back. In an unstable one, some patterns go soft: instead of springing back, the atoms slide along that pattern and the whole lattice bends into a new shape. QHD finds those soft harmonics, follows the distortions they open, and tests whether the new arrangement is a real, stable material.
The three words
The 3 T's · how QHD reads every distortion
Most computer searches for new materials start from structures that are already known and ask which ones are stable. That finds variations on the familiar. QHD starts from the other end: from arrangements that are unstable, and lets the instability show where the atoms want to go. The places a structure falls into are often ones no database lists.
That is how the J phase was found. Li₈Au had no entry in the Materials Project, and only an unanalysed cell in one high-throughput database. A lithium cage that could not hold its gold atom led us straight to it.
We placed a gold atom inside a tetrahedral cage of lithium. The cage was unstable. Its soft modes pulled it apart, and the lithium collapsed around the gold into a new close-packed structure, Li₈Au, with every gold atom held by 12 lithium neighbours. Every vibration of the new structure checked out as stable, and the result is published with a DOI.
It is a way of searching for new metastable materials, tested at every step by calculation and by predictions written down in advance. The broader ideas behind it, from the 2016–2018 journals, are treated as hypotheses until a test confirms them.
We placed eight lithium atoms in a perfect tetrahedral (diamond-type) cell and computed it from quantum mechanics with density functional theory. The perfect cage is a saddle point: two sets of its vibrations are unstable, the T1 and T2 modes (imaginary frequencies near 247 and 244 cm⁻¹). We followed them to see where the cage would go.
Each step lowers the energy: from 514 meV per atom above ordinary lithium, to 67 after the atoms follow their soft modes, to 30 once the cell itself is free to change shape (the latest step of a calculation still in progress).
Along the T1 mode the angles between atoms change in direct proportion to how far the mode is pushed: about 22 degrees per unit, with an almost perfect straight-line fit.
Bond lengths barely spread, growing only with the square of the amplitude. The cage shears rather than stretches: a harmonic distortion in the plainest sense.
