Manufacturing Bits: June 10

Rapidly self-assembling thin films; the search for what comes after silicon zeroes in on three materials.


Self-Assembling Nano Films
Applying thin films with uniformity has always been an engineering challenge, but as feature sizes shrink the problem become even more pronounced. But a new approach developed by Lawrence Berkeley National Labs’ Materials Science Division could end up simplifying this process.

The new approach used chloroform as an annealing solvent to create self-assembling arrays of nanoparticles that can span an area of several square centimeters. Even better, this can be done in less than 60 seconds and is compatible with current manufacturing processes.

The initial study focused on “supramolecules,” based on copolymers combined with gold nanoparticles. By applying a solvent the particles self-assembled into hierarchically structured thin films, according to the lab, which is run by the U.S. Department of Energy.

Ting Xu, a polymer scientist at the lab, said the technique can be used to assemble nanoparticles across an entire silicon wafer. “You can think of it as a pancake batter that you can spread over a griddle, wait one minute, and you have a pancake ready to eat,” she said in a release. A paper on the subject can be found here.

Xu noted that within the supramolecules are microdomains, which range in size from a few nanometers to tens of nanometers. “To be compatible with nanomanufacturing processes, the self-assembly fabrication process must also be completed within a few minutes to minimize any degradation of nanoparticle properties caused by exposure to the processing environment,” Xu said.

AFM phase image shows a 50-nm nanocomposite thin film in lithographically patterned trenches that formed unidirectional nanoparticle arrays over macroscopic distances in just over a minute. The bright circular dots represent the 5 nm gold nanoparticles as illustrated by the schematic. Source: Lawrence Berkeley Lab.

After Silicon
There has been talk about post-silicon materials for more than a decade, but researchers at Sematech, Intel and Purdue University are zeroing in on molybdenum disulfide, gallium arsenide and germanium as likely candidates in three papers that will be presented at the VLSI Technology and Circuits symposium this week.

The great advantage of molybdenum disulfide is that it can be made extremely thin like graphene. The downside, and one that made it an unlikely candidate in the past, is high electrical resistance between metal contacts and the single atomic layers in this compound. By doping it with 1,2 dichloroethane—actually impregnating the molybdenum disulfide—the researchers showed a 10X reduction in contact resistance and a 100X reduction in contact resistivity. Researchers are looking at this material for flexible and transparent electronics.

A second paper examines GaAs, which already is complementary with CMOS manufacturing tedchniques. Researchers say the III-V material is now workable at the CMOS circuit level. A third paper looks at germanium for N-type transistors. In the past, it has been used almost exclusively for P-type transistors.

This graphic depicts the structure of an extremely thin semiconductor called molybdenum disulfide, which is particularly promising for future flexible and transparent electronic devices for displays, touch pads and other applications. The structure of molybdenum disulfide is a single-atomic layer of molybdenum sandwiched between single-atomic layers of sulfide and “doped” with a chemical compound called 1,2 dichloroethane (DCE). (Source: Purdue University/Lingming Yang)