Hopping Hydrogen

---

Hopping Hydrogen

Hydrogen atoms are the kind that like to roam around, says a group of Carolina chemists.

The researchers, led by John Boland, associate professor of chemistry, were the first to catch the fickle atoms hopping around the surfaces of silicon wafers—the raw material for making computer chips.

Each wafer is built from layers of silicon or other semiconductor material. After a layer has been deposited, hydrogen blankets its surface, essentially capping the wafer and preventing the addition of more material. But if the temperature goes up, the hydrogen skips out, leaving behind "dangling bonds" on the surface-dwelling silicon atoms and allowing a new layer of material to form.

Knowing how hydrogen comes off a wafer's surface, then, is critical to understanding what happens when new layers form, Boland says. For example, it may explain why it's easier to add new layers to some surfaces than to others.

Boland's group—which included post-doctoral fellow Marc McEllistrem and graduate student Matthew Allgeier—heated wafers enough to let the hydrogen flit about the surface but not escape altogether.

Then the chemists watched the surface through a scanning tunneling microscope, which uses electrons instead of light so it can image individual atoms. Instead of following the hydrogen atoms themselves, the researchers tracked the more obvious bright spots created by the dangling bonds.

"It's tantalizing to see this chemistry in action," Boland says. "The movement is much more involved than we originally imagined."

Although other researchers had predicted hydrogen hopping and had collected other kinds of evidence for it, nobody had seen how the atoms interacted with a surface in such detail. And nobody, including Boland, had suspected that dangling bonds were as versatile as they now seem to be.

"We saw phenomena that were quite unexpected," Boland says.

For example, predictions said that an individual dangling bond would hop from one silicon dimer—a pair of silicon atoms—to its neighbor. Boland's group confirmed this at the relatively low temperatures of 640-660 degrees Fahrenheit.

They also saw that, at higher temperatures, the dangling bonds leaped further—even across a gap left by a missing dimer. In other cases, two dangling bonds that started life on the same dimer would unpair and hop away from each other. These lone dangling bonds often rejoined their partners later, showing that dangling bonds prefer to be paired.

Now the researchers are trying to find out why dangling bonds can scoot right by some defects but not others. This kind of information should bring the chemists closer to the bigger goal of understanding how hydrogen comes off a wafer's surface at even higher temperatures.

"It's exciting because the findings won't be limited to silicon," Boland says. "They'll apply to diamond and any other semiconductor surface. We don't know yet if what we learn will change the way people make these wafers, but it certainly has changed the way we think about them."

—Elizabeth Zubritsky