Thin-Layer Magnetism Comes Alive

Freeman's interest in thin-layer magnetism began in the mid-1970s when he didn't believe experiments showing that the surfaces of bulk metals -- such as nickel, iron and cobalt -- have no magnetic moment: "We did computational investigations, and we found that at least theoretically, they weren't magnetically dead. They were more alive than the bulk." With iron, Freeman and collaborators discovered that magnetism at the surface actually increased 40 percent.

Freeman next modeled a single-atom layer, a "free monolayer," and discovered that the magnetic moment is even stronger than on a bulk metal surface. Experimentalists couldn't make free monolayers in the laboratory to test these predictions, so Freeman began simulating a metal monolayer on a surface of silver or gold, which because they have no unpaired electrons are magnetically neutral. He found that the magnetic moment dropped only slightly from the free monolayer value, less than 1 percent for iron.

Calculated electron spin density for the surface of iron (left) compared to the free monolayer (right). Dark blue indicates negative spin, and other colors are positive, increasing from light blue through pink. A larger area of positive spin in the free monolayer accounts for a magnetic moment of 3.18 bohrs, compared to 2.98 for the surface and 2.20 for bulk iron.

The increased magnetic moment that occurs in a monolayer, explains Freeman, happens because each atom has fewer atomic neighbors than in the bulk: "You have a reduced dimensionality. In a single layer, an iron atom, for instance, has four atoms surrounding it, whereas in the solid it has eight. The interactions with other atoms in the bulk tend to reduce magnetism."

In recent work on Pittsburgh's C90, Freeman and former post-doctoral fellow Ruqian Wu, now at the California State University-Northridge, determined that magnesium oxide is even more neutral than silver or gold. They also found that sandwiching layers of metal between neutral materials helps protect the metal and doesn't significantly reduce its magnetic properties. "You've covered it so it won't rust or oxidize," says Freeman, "so for practical purposes, multilayers are much better."

go back to the main page