Science@Berkeley Lab nameplate Berkeley Lab logo
March 31, 2005
A Different Spin on Future Data Storage

The next generation of computers will be "instant-on," meaning they won't need to be booted up to move hard-drive data into memory. They'll also store data in a smaller space and access it faster, while consuming less power than today's machines — thanks in part to the development of magnetic random access memory chips, or MRAM. These MRAM chips will store data through the spin of electrons, giving them a distinct advantage over today's chips, which utilize electron charge.

MRAM computer chips use magnetization rather than electric charges to store bits of data; they can retain information even when electrical power is turned off. The world's first 16-megabit MRAM chip from Infineon Technologies and IBM research set records for high-density data storage in 2004.

Before the "spintronics" revolution can begin, however, a much better scientific understanding is needed of complex metal oxide materials that display the unique phenomenon called colossal magnetoresistance (CMR), including the materials known as manganites.

"With CMR manganites, the application of a magnetic field can cause the material's electrical resistance to change by as much as 1,000 percent," says Charles Fadley, a physicist affiliated with Berkeley Lab's Materials Sciences Division and a professor of physics at the University of California at Davis. "Today's best data storage devices are based on the GMR effect" — gigantic magnetoresistance — "in which the field application reduces the electrical resistance only 20 to 50 percent."

Fadley, working with his student Norman Mannella, now at Stanford University, used the exceptionally bright, soft x-ray beams (lower energy x-rays) and sophisticated experimental facilities of Berkeley Lab's Advanced Light Source (ALS) to shed new light on what could prove to be a crucial element of the CMR effect in manganites and other metal oxides. This is the surprising formation of a type of polaron, an electron that is somehow bound to a local distortion of the atoms in the crystal.

The distortion creates a sort of energy "well" that traps the electron, like a divot on a fairway that traps a golf ball. The type of polaron that Fadley and Mannella detected is called a Jahn-Teller polaron, after Hans Jahn and Edward Teller (the same Teller who later led development of the hydrogen bomb), who predicted such polaron distortions in a famous 1937 paper. Fadley and Mannella found that Jahn-Teller polaron formation took place in the CMR manganite only after it was heated past its Curie temperature, the point at which a material ceases to be magnetic.

"We've shown for the first time, using a combination of all of the primary ALS spectroscopies, that one of the most studied of the CMR manganites, a mixture of lanthanum, strontium, manganese, and oxygen, exhibits the formation of polarons above its Curie temperature" of about 350 degrees Kelvin, Fadley says. "The combination of ALS techniques we used showed much more directly than any previous measurements that one electron is localized to the magnetic manganese atoms, thus altering the electrical resistance of the entire material."

Manganites typically contain four or more constituents; manganese is often the only magnetic atom present. According to the findings of Fadley and Mannella, as the CMR manganites cool to below their Curie temperature, the Jahn-Teller polarons disappear, releasing the trapped electrons. The ability of the CMR manganites to conduct electricity is very different depending on whether or not a polaron is present.

Watching electrons move in a spintronic material; lanthanum strontium manganese oxide:

(Left panel) Two electrons in an inner shell of a manganese atom have slightly different energies, resulting from the interactions of their spins with the spin of the atom as a whole. In this graph, made by plotting the energy of electrons knocked out of the shell by photons from the ALS beamline, their energy separation at different temperatures (adjacent curves) appears as a trench-like feature. The separation increases as the temperature goes above the Curie point (blue line) and keeps increasing until no more separation is apparent (red line). When the temperature is decreased, the process reverses. The energy separation of these electrons is a direct measure of the atom's magnetic strength, which increases markedly above the Curie point due to the transfer of an electron to the manganese atom.

(Right panel) Meanwhile, the electrons in the innermost orbital shell of oxygen atoms in the compound simultaneously shift to a higher energy as temperature is raised, confirming the transfer of charge to the manganese atom above the Curie temperature. (Lanthanum and strontium in the compound show graphs similar to oxygen.) These results provide direct evidence of "polaron" formation: a distortion in the lattice of atoms surrounding the manganese atom, which traps an electron in its vicinity.

As Mannella explains, "Since the manganese atom and its surrounding oxygen atoms are much more massive than a bare electron, the polaron behaves as a negatively charged particle with a larger mass and lower mobility than an isolated electron."

Adds Fadley, "It seems clear that to understand CMR in manganites, you will have to take into account the effects of polarons. As for the temperature dependence of the polaron formation, it was much bigger than we ever imagined it would be."

Norman Mannella (center) and Charles Fadley at beamline 4.0.2 of the Advanced Light Source, where they used multiple spectroscopy techniques to detect the temperature-dependent formation of Jahn-Teller polarons in manganites, linked to colossal magnetoresistance (CMR) (Photo Roy Kaltschmidt)

The spectroscopic experiments carried out by Fadley and Mannella were performed using the multi-technique spectrometer/ diffractometer at ALS Beamline 4.0.2, which generates circularly polarized soft x-ray beams. A beam of light is circularly polarized when its electric-field component rotates around the direction in which the beam is traveling. The absorption of circularly polarized light by a magnetic material reveals much about the magnetic moments of its constituent atoms. Powered by one of the ALS's undulator magnets, beamline 4.0.2 is ideal for studying manganites and other materials of spintronic interest.

The CMR effect in manganites is an important subject of study for reasons other than its potential impact on high-density data storage devices. Some CMR materials conduct electricity via electrons with only one direction of spin (spin is a quantum-mechanical property, considered to be up or down), rather than equal numbers of electrons with either direction of spin, as in all of today's typical electronic devices. This means that CMR manganites can have nearly 100 percent spin polarization, making them excellent candidates for lightning-fast new logic devices, such as spin transistors and magnetic tunneling transistors.

"Beyond spintronic applications, our results could also have implications for the magnetic states of atoms under high pressure, as in the earth's core," Mannella says.

Geophysical studies have shown that iron-containing perovskites, metal-oxide materials with the same general crystal structure as the CMR manganites, can exhibit a marked reduction in the electron spin state of their iron atoms as they move through the earth's mantle toward conditions of extreme pressure at the earth's core. From an initial high spin state, the iron atoms drop to a low spin state as pressure increases. This leads to a gradual loss of magnetic moment, which has significant influence on the magnetic, thermoelastic and transport properties of the deep mantle. It could also affect the partitioning of iron between the upper and lower mantle, and between the lower mantle and core, with a possibility of iron enrichment in the deep mantle.

"The relationship between the perovskites in the mantle and pressure is based on the same effects that we're studying to observe changes in transition-metal spin [magnetic] states as a function of pressure and temperature," said Fadley. "Therefore, our results should be helpful to those who are developing a model of what the mantle really looks like."

Collaborating with Fadley and Mannella on this research were Berkeley Lab's Bongjin Mun, Corwin Booth, Stefano Marchesini, and See-Hun Yang. The research was funded by the U.S. Department of Energy, Office of Basic Energy Sciences (BES), Materials Sciences and Engineering Division.

Additional information