Novel surface treatment protects next-gen computer chips

A simple surface coating could make it easier to manufacture parts of transistors with less risk of damage.

Inside computer chips are billions of tiny transistors made from silicon. But the material is approaching its limits. In an effort to build smaller, more capable devices, researchers are exploring how they might build transistors with other materials alongside silicon, including a class of thin materials called transition metal dichalcogenides (TMD). One of the leading TMD candidates is a material called molybdenum disulfide. It is only three atoms thick: one layer of molybdenum sandwiched between two layers of sulfur.

To build transistors with silicon and TMD, manufacturers will likely need to remove atoms from only the top sulfur layer without damaging the layers underneath. The standard method for removing the top layer uses plasma, the state of matter that makes up the sun and the stars.

When TMD is exposed to a plasma under the right conditions, some of the plasma’s particles strike the surface of the TMD, knocking loose some of the atoms from the TMD. But it’s a fine line between hitting the TMD hard enough to remove the top layer of sulfur atoms and hitting it so hard that the molybdenum layer below is damaged. That tight margin makes it challenging to perfect the process so that all the atoms from the top layer are removed without any damage to the lower layers. Researchers from the Princeton Plasma Physics Laboratory used computer simulations to show that pre-treating molybdenum disulfide with oxygen or fluorine makes it easier to avoid impacting atoms below the top layer. Their research findings were published in the Journal of Physical Chemistry Letters.

The research team found that the energy needed to knock a sulfur atom loose falls from approximately 30 electron volts on an untreated surface to roughly 10 electron volts when the surface is coated with fluorine and to about 14 electron volts when coated with oxygen. That lower threshold matters because plasma ions don’t all carry exactly the same energy — there’s always some spread. On an untreated surface, the gap between removing sulfur and damaging the molybdenum layer below is narrow enough that some ions inevitably land in the danger zone. Pushing the sulfur-removal threshold down to 10 or 14 electron volts widens that gap, giving manufacturers a workable energy range where the top layer comes off cleanly and the rest of the material stays intact.

Instead of relying solely on brute physical force, the team used a chemical assist: When an incoming ion strikes an oxygen-coated surface, two oxygen atoms and a nearby sulfur atom combine into sulfur dioxide: a stable gas molecule that can drift away on its own. Fluorine works similarly, producing sulfur-fluorine compounds.

“We are not directly breaking the bonds,” said Yury Polyachenko, lead author of the study. “We are forming some intermediate products, such as sulfur dioxide. This intermediate product is much easier to break off.

“The next step is figuring out how much damage the process causes, not just whether it causes damage. After that, we want to see whether the same approach works for related materials — swapping molybdenum for tungsten, or sulfur for selenium — to find out how broadly this idea can be applied,” Polyachenko said.

Image credit: iStock.com/SweetBunFactory