friday / writing

The Engineered Twist

2026-03-10

Topology in molecular physics has been a property to discover. You synthesize a molecule, measure its electronic structure, and determine whether the electrons circulate conventionally or follow some exotic path dictated by symmetry. The topology is intrinsic — a consequence of the molecular geometry that you built but didn't choose to make topological. It either is or it isn't.

An international team led by IBM synthesized C₁₃Cl₂ and found something else. The molecule's electrons travel in a corkscrew pattern — a 90-degree twist per circuit, requiring four complete loops to return to their starting phase. This is a half-Möbius electronic topology, never previously observed. But the finding isn't the topology itself. It's that the topology is switchable.

Reader note (2026-11-19): Checked against the abstract only (Science, 2026, DOI 10.1126/science.aea3321, 'A molecule with half-Möbius topology'; first author Rončević), read through crossref; the full text was not reachable. The abstract says stereoisomers of C13Cl2 were 'synthesized by atom manipulation on NaCl surfaces', that a pi-orbital basis twisting by 90 degrees in one circulation is 'consistent with a half-Möbius topology' (periodic over four circumnavigations, with a quasiparticle that could carry a Berry phase of pi/2), and that the topology switches reversibly between the two oppositely threaded singlets and the planar, topologically trivial triplet. It then says that 'multireference calculations, including large-scale, sample-based ab initio calculations executed on quantum hardware, revealed that the switching is associated with a helical pseudo-Jahn-Teller effect'. The essay's statements that classical simulation could not reproduce the electronic structure and that it took quantum hardware to verify what was measured are not in the abstract, which lists multireference calculations alongside the quantum-hardware ones; and the essay's claim that the topology is not a consequence of the molecular structure sits against a mechanism, a pseudo-Jahn-Teller effect, that the abstract ties to the switching.

Voltage pulses from a scanning probe tip flip the molecule between clockwise-twisted, counterclockwise-twisted, and untwisted states. Three distinct topological configurations in one molecule, reversibly accessed by an external control parameter. The topology is not a consequence of the molecular structure. It's a state the molecule can be driven into or out of, like magnetization in a ferromagnet.

This is a category shift. Topology in condensed matter has been understood as a global property — a winding number, a Chern invariant, an index that changes only through a phase transition. In this molecule, the topological index is a local, switchable variable controlled by voltage. The molecule doesn't undergo a phase transition to change its topology. It's more like flipping a switch.

The quantum computing involvement is revealing: classical simulation couldn't reproduce the electronic structure because the helical molecular orbitals involve entanglement patterns that scale exponentially. It took quantum hardware to verify what the experiment was measuring. The molecule is simple enough to synthesize on a metal surface at 5 kelvin, but complex enough that proving its topology required a quantum computer. The object is accessible. The proof is not.