Rice chemists quench a liquid crystal and keep 70% of its alignment in the solid

The 70% figure is about orientation. Polarized optical microscopy tracks how well the molecular columns still point one way after the quench ^[8], and that is a different quantity from charge mobility or trap density in a finished device. The phys.org report does not describe a device measurement ^[22]. Even at the cooling rates that worked, roughly 30% of the liquid crystal's alignment was lost on the way to the solid ^[21].

If the front the team measured ran at about 100 micrometers per second, and that is roughly 10 million times the rate a commonly used model of conventional crystal growth predicts under similar conditions ^[7], then the model was expecting something near 0.01 nanometers per second ^[18]. At that rate, covering the same 100 micrometers takes about 10 million seconds, or roughly 116 days ^[19].

Kushal Bagchi, the corresponding author, attributes the speed to where the molecules start. "In the liquid-crystal phase, the molecules are already somewhat in the right position, so they don't have to move very much to get to the final crystal," he said ^[12]. In a martensitic transformation the molecules move together in a coordinated way instead of rearranging one at a time, and the product keeps some structural memory of the phase it came from ^[17].

Bagchi puts the claim against the textbook. "Martensitic transformations are usually thought of as exclusively solid-state processes," Bagchi said. "What we found is that a transition from something with fluidity into a solid can still exhibit the essential characteristics of a martensitic transformation" ^[10].

The evidence for the label is three signatures. The transition showed up only under fast cooling, the solid kept the structure of the liquid crystal it came from, and reheating largely reversed it, bringing back much of the earlier alignment ^[9]. "We had three qualitative indicators," Bagchi said. "Once we saw all three, we became confident that this was qualitatively similar to a martensitic transformation" ^[11].

Matteo Pasquali, a co-author and professor of chemical and biomolecular engineering at Rice, framed the manufacturing idea as a possibility. "It suggests that we may be able to use the structure that already exists in a liquid crystal as a kind of blueprint for building highly ordered crystalline materials," he said ^[13]. What has been shown is one compound, HAT6, aligned inside microscopic channels and cooled inside a narrow band: too slow and it crystallizes the ordinary way, too fast and it becomes a glass ^[4][6].

Thermal uniformity is what would limit a scaled-up version. At 100 micrometers per second, a front crosses a one-centimeter film in about 100 seconds ^[20]. The demanding part is holding that whole centimeter inside the cooling band for the duration, and the Rice work did it in channels a fraction of that size ^[4].