Supercomputer muscle advances chip tech

IBM researchers are using computer-based simulations to drive new chip technologies. As reported in the scientific journal Physical Review Letters, a team of scientists at IBM’s Zurich Research Laboratory for the first time used advanced supercomputer-based models to examine a promising new material -- hafnium dioxide -- in silicon transistors.

The new material is key to IBM’s recently-announced “high-k metal gate” technology, the first major change to the transistor since the emergence of silicon semiconductors.

The Zurich scientists used an IBM Blue Gene supercomputer to determine why hafnium dioxide works so much better than other high-k materials considered by the industry. As a result, the researchers were able to gain a clear picture -- for the first time -- of the physics underlying the unique electrical behavior of hafnium dioxide when it mixes with silicon.

For this study, the IBM team simulated various material compositions using 50 different models of hafnium silicates, materials that form when silicon and hafnium oxides mix. These models contain up to 600 atoms and about 5,000 electrons, representing a realistic system. A single calculation of what’s called the dielectric constant was accomplished in only five days of computing time on the Blue Gene/L supercomputer in the Zurich lab. The complete simulation for all 50 models, about 250 days on Blue Gene, would normally take the most powerful laptop PC 700 years to calculate. This equals about 200 billion billion operations.