Physicists at the National Institute of Standards and Technology and The Johns Hopkins University have found a surprise lurking at the heart of geometrically frustrated magnets, an odd class of materials that has tantalized condensed matter physicists for years with puzzling properties and hopes of new insights into the basic structure and organization of matter.

The surprise is connected to spin, a characteristic of electrons and the basis of magnetism in solids. When scientists cool a magnetic material sufficiently, electronic spins typically "freeze" into a periodic pattern throughout the material. Spins on neighboring atoms are usually either parallel or in opposition (antiparallel).

Collin Broholm teaching a recent class

Materials whose crystal structure is formed from square unit cells can readily achieve spin order, but in materials with a crystal structure that has triangular components, it can be geometrically impossible to achieve uniform global spin order, producing what physicists call a frustrated magnet.

Seung-Hun Lee, a Hopkins alumnus now working at NIST, and Collin Broholm, a professor of physics in the Krieger School of Arts and Sciences, learned how such a material relieves its frustration by bombarding a frustrated magnet, zinc-chromite, with neutrons. They found that the spins organize locally, rather than globally, with hexagonal groups of six atoms acting together to achieve local magnetic order without order on larger length scales.

"Our findings are directly related to a fundamental and methodological question--how to understand collective phenomena in strongly correlated many-body systems," said Lee, who is a NIST Center for Neutron Research staff physicist.

According to Lee, identification of this new, higher-order organizing pattern could be helpful to scientists working to advance their understanding of a variety of scientific topics, including how high-temperature superconductivity works, how proteins fold and how subatomic particles like quarks come together to form the components of atoms.

"While our interest is really in the fundamental properties of many-body systems, we're also excited about potential applications, which include new dielectrics for microwave applications and materials with a characteristic known as negative thermal expansion," Broholm says.

Such materials actually contract as their temperature increases. While they resemble magnetically frustrated materials, they are structurally rather than magnetically frustrated. Their atomic positions have trouble achieving an ordered configuration rather than their electronic spins.

"From an applications standpoint, negative thermal expansion is a very useful property," Broholm says. "For example, Lucent has been considering use of such materials to counteract thermal expansion in fiber optics."

Physicists have been interested in the concept of frustrated magnets for at least four decades. Broholm points out, however, that the phenomenon didn't get a name until 1979, when Jacques Villain, a French physicist who was a founding father of the field, wrote an influential article on the topic.

Two other types of frustration, structural frustration and ferroelectric frustration, are also known to exist. All involve "degenerate" ground states, where the normal rules of how interacting atoms develop order are blocked by the symmetry of a material's structure.

Broholm says interest in magnetically frustrated materials grew dramatically in the early 1990s after scientists discovered the first high-temperature superconductors. Many of these materials are created through a process called "doping" that involves introducing chemical impurities that yield mobile electrons and disrupt the ordered spin patterns of an insulating oxide antiferromagnet.

"The question emerged as to what's going on with the spins in high temperature superconductors--are they still fluctuating or are they frozen and, if so, in what pattern?" Broholm says. "That launched a search for materials that perhaps didn't have superconductivity but at very, very low temperatures had fluctuating spin states analogous to those in high-temperature superconductors. We were hoping they could help us sort out the possible novel states that one can have when the spins simply don't freeze."

Working at Bell Laboratories in the early 1990s, Broholm and colleagues Arthur P. Ramirez, Gabriel Aeppli and others were able to identify a material with the desired properties.

"We found that strontium-gallium-chromium-oxide is very pathological in that it has magnetic chromium atoms but no spin order at very low temperatures," Broholm recalls. "That material is actually very closely related to the material we've looked at here, zinc-chromite."

Broholm and Lee have been studying zinc-chromite from a variety of perspectives for several years. For their most recent paper, published in the Aug. 22 Nature, they used instrumentation at the NIST Center for Neutron Research to bombard the material with neutrons.

"It's all very quantum mechanical, so it's dangerous to make a classical analogy, but each neutron can be thought of as a little compass needle that we're sending into a system of large fluctuating compass needles," Broholm explains. "The neutron exchanges energy and momentum with the electronic spin system and, in doing so, is deflected and changes speed. By looking at the pattern of neutron scattering, one can indirectly glean unique information about electronic spin correlations--the relative orientation of fluctuating or static spins in the material."

After cooling the material to low temperatures, the scientists were able to show that the spins organized themselves locally rather than globally, with atoms from different facets of the crystals in zinc-chromite linking together to form hexagonal rings where all the spins are antiparallel to their two nearest neighbors.

The spins of each cluster of six atoms act in unison, producing what Lee and Broholm called a "spin director."

"While individual spins have strong interactions with nearest neighbors, spin directors are almost independent from each other, and therein lies the key to their apparent stability," Broholm says.

A parallel theoretical development by Princeton physicists Oleg Tchernyshyov, Shivaji Sondhi and Roderich Moessner provided an important clue for interpreting the data. Tchernyshyov has recently joined the Hopkins Department of Physics and Astronomy as an assistant professor.

Lee and Broholm are already looking at other frustrated magnets using the NIST neutron source.

"We have already found another material that has the same neutron scattering pattern as zinc-chromite," Lee says. "There also are some other geometrically frustrated magnets that exhibit different neutron scattering patterns, indicating that other types of spin clustering are realized in those systems. We would like to see if our approach can identify those different spin clusters."

Broholm notes that work is under way on a new neutron research instrument called Multi Axis Crystal Spectrometer at the NIST Center for neutron research. Jointly funded by Hopkins, NIST and the National Science Foundation, MACS will be optimal for determining nano-scale structure in strongly fluctuating systems.