Skip to Content

Intel, IBM Overhaul Material for Next-Generation Microprocessor

Both companies modify transistor materials to make smaller, faster, more energy-efficient processors.
January 29, 2007

On Saturday, Intel and IBM separately announced a new type of transistor material that the companies say will lead to smaller chips, increased computing performance, and more energy-efficient computers. Both chip makers have plans to integrate the new material into the next manufacturing line of chips–known as the 45-nanometer generation–within a year. Ultimately, this advance should make computers better at playing video games, editing movies, and performing other processor-intense tasks.

An image of Intel’s static random access memory (SRAM) chip–containing more than one billion transistors–which is built from new materials, the company announced today. Separately, IBM and Intel announced new chip materials to help improve computing performance for next-generation microprocessors.

Intel, known for its personal-computer chips, has also announced that the new chips will appear in its next generation of “dual core” and “quad core” computers and servers, which are expected to be in production at the end of this year. IBM, well-known for its servers and high-performance computing systems, expects that products featuring its new chips will ship by early 2008.

Historically, the performance of microprocessors has doubled every two years, following a trend known as Moore’s Law. Performance has increased thanks to chip fabrication technology that has let transistors shrink with each generation, either allowing more transistors to be packed on a chip or allowing the chips to be made smaller. Currently, most computer and server microprocessors are made using what is known as a 65-nanometer process. The next generation of transistors will be made smaller using a 45-nanometer process. However, the companies’ new materials will provide this generation of chip with a performance boost that wouldn’t be possible otherwise.

One of the material changes will be in an important transistor component called the gate dielectric. This component helps control the flow of electrons, which turn the transistor on or off. For decades, the gate dielectric has been made of an insulating material called silicon dioxide. However, as transistors have shrunk, the layer of silicon dioxide has needed to become thinner. This poses a problem, however, because a thin layer of silicon dioxide lets electrical current leak through it, producing excess heat and resulting in poor device performance.

IBM and Intel found that a material called high-k produces the same benefits of silicon dioxide, but it can do so in a thicker layer. The material is based on the element hafnium, and both companies claim that it significantly reduces the amount of current leakage.

In addition to having to replace the gate dielectric, both companies were forced to rethink the type of material to use for the gate itself–the transistor component that ultimately turns a transistor on and off. Traditionally, gates have been made of polysilicon, a less structured form of crystalline silicon, which is used in the transistor. But in order to make the gate compatible with the new gate dielectric, Intel and IBM replaced polysilicon with a metal whose name neither company will disclose at this time.

Intel says that its new gate dielectric and metal gate allow transistors to be driven with 20 percent more current than before, which translates into a 20 percent increase in performance, says Mark Bohr, Intel senior fellow. “We think this is an important breakthrough that will really extend Moore’s Law.”

Bohr points out that high-k gate dielectric and metal-gate research are not new, and that research papers have been being published on advances in the field for years. However, he’s confident that Intel has found the best combination of materials that can keep its chip manufacturing on track into the next decade.

The key to implementing these new materials successfully is to make sure that they can seamlessly fit into a manufacturing line, says Bernard Meyerson, chief technologist at IBM. “You have to take a careful look at how this is implemented,” he says. IBM has found a way to add the new materials to the fabrication process without overhauling the entire process, he says, which could be expensive. Intel’s Bohr says that in terms of manufacturing the new chips at his company, “most of the process equipment is the same as in past generations.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.