Spins in a ytterbium compound pick a shared axis 110 kelvin above its magnetic ordering point

YbMnBi2 had an odd habit. In a magnetic field, currents of electricity and heat moving through it are pushed sideways far more strongly than expected. They keep being pushed at temperatures where the compound's magnetic order has vanished ^[5]. Above its ordering temperature, an ordinary magnet's spins point in random directions ^[3]. Until now the behaviour had lacked a convincing explanation ^[5].

Earlier studies had suggested that some magnetic compounds host an in-between state, the spin nematic. In it, spins share a preferred direction without forming magnetic order, much as the rod-shaped molecules in a liquid crystal display do ^[4]. A state like that has no static magnetic pattern to detect. What it does have is uneven fluctuation: spins move more along some directions than others. That is why the experiment used neutrons, which carry spins of their own, with the experimenters setting the orientation of the beam's spins ^[7].

The unevenness appeared near 400 K as the crystal cooled from 450 K. Magnetic order arrives only around 290 K ^[8]. In Celsius, the nematic window runs from roughly 127 C down to 17 C ^[13], a band about 110 K wide ^[12].

The calcium sibling is the part of the design I find most persuasive ^[6]. It lacked the effect ^[9]. If manganese and bismuth were enough to produce the anisotropy, the calcium compound should have shown it too. The thing this doesn't tell you is which property of ytterbium matters, because the comparison isolates the atom and not its properties. The team's candidate is weight: heavy ytterbium ties the motion of electrons closely to their spins ^[10]. The paper's title calls the resulting twisted arrangement scalar spin chirality ^[11].

The neutron measurements are direct evidence for a fluctuating spin nematic, and phys.org's account calls the calcium result strong evidence for it ^[9]. The step from that state to the sideways currents is the team's proposed mechanism ^[10]. The phys.org account does not report a measurement or calculation that matches the size of the deflection to the measured fluctuations.

According to phys.org, the work could guide attempts to produce similar effects in other materials. It could also feed into spintronics, which carries information in spin instead of electric charge ^[14]. The account calls the effect especially promising for practical devices because it appears without magnetic order and at around room temperature ^[15]. I'd hold that view loosely. A chip needs the effect in a form it can build with, and these measurements were made on crystals of the compound ^[6].