The semiconductors of today—like the ones powering the device you’re probably reading on—contain billions of miniscule transistors printed onto thin wafers of silicon. As engineers design smaller, faster, and more energy-efficient electronic devices, they need even tinier transistors. Shrinking these components, however, is not as simple as making them smaller; it often requires new materials and lithography methods.
Now, researchers at the University of Chicago’s Pritzker School of Molecular Engineering (PME) have developed a “mix and match” design platform that points toward appropriate combinations of self-assembling block polymers to make semiconductors with the desired sizes and properties for technical applications. The new approach, described online in Nature Materials, creates patterns only a few nanometers in dimension—smaller than existing commercial transistors.
“We developed a platform where we can mix and match different components of these materials to give us the properties we want for specific applications,” said Paul Nealey, the Brady W. Dougan Professor of Molecular Engineering, who led the new research. “I think this will have a tremendous impact in this field as well as the broader area of polymer research.”
“We have been able to develop a flexible polymer platform that when combined with highly efficient chemistry gives us access to libraries of materials,” said co-senior author Stuart Rowan, the Barry L. MacLean Professor of Molecular Engineering. “This has allowed us to develop materials for a range of different applications quickly and efficiently.”
Speeding up semiconductor research
Traditional lithography—the process of printing transistors onto a semiconductor, or integrated circuit chip—uses a photographic process to etch complex circuit patterns onto silicon. However, as the desired size of the patterns decreases, this method doesn’t always work.
In recent years, researchers including Nealey and Rowan have developed self-assembling block copolymers as a new method for semiconductor lithography. These block copolymers spontaneously come together into nanoscale architectures that researchers can control through both the design of the polymers as well as external chemical cues.
However, there are nearly endless variations and combinations of these polymers that may be potentially useful in assembling semiconductors. Because of the way the properties of the polymers are intrinsically linked to the patterns they create, new polymers are required when researchers want to make new sizes of patterns on their semiconductors.
“We realized we didn’t need to just develop one material, but new, different materials for every resolution and application we might want to target,” said Nealey.
To figure out the best self-assembling materials for their applications, researchers have typically tested out one combination of block copolymers at a time. In the new paper, however, Nealey, Rowan and colleagues constructed a large material library with hundreds of block copolymers, all variations of the self-assembling copolymer known as thiol-functionalized polystyrene-block-poly (glycidyl methacrylate) (PS-b-PGMA). Then, they screened those new materials for their properties—including the size of the patterns they self-assembled into.
A tool for the future
Their large screen let the PME researchers build an atlas that correlated key parameters of different PS-b-PGMA copolymers with their chemistries. This enables researchers and chip manufacturers to plug in their desired dimensions and quickly develop desired materials, the team said.
“This idea of figuring out architectures and components to simultaneously optimize different properties is a vision that the field has been trying to achieve for a long time,” said Nealey.
Some of the results, Nealey added, surprised the group and underscored the importance of the unbiased screen—copolymers that they wouldn’t have otherwise tested ended up working well for certain architectures.
As the new platform is combined with machine learning approaches in the future, the researchers will likely be able to become even better at predicting the emergent properties of new copolymers.
“As the material dataset grows using this powerful, high-throughput platform, we can start to have a deeper understanding of the full capabilities and applications of these self-assembling block copolymers,” said Hongbo Feng, a former postdoctoral scholar in the Nealey and Rowan Labs, is a co-author of the paper, along with former graduate student Moshe Dolejsi and visiting professor Ning Zhu.
Citation: “High throughput platform to meet covarying materials manufacturing design criteria at multiple length scales for block copolymer nanolithography,” Feng et al, Nature Materials, November 10, 2022. DOI: 10.1038/s41563-022-01392-1
Funding: Semiconductor Research Corporation, National Science Foundation, Jiangsu Overseas Visiting Scholar Program