Pages

February 13, 2009

More Durable and High Resolution Nanoimprint Lithography


Nano imprinting with Bulk Metallic Glass enables to directly replicate smallest features with high aspect ratio. Our current record is 13 nm and 50 for feature diamater and aspect ratio, respectively.

MIT Technology Review reports that researchers at Yale University have demonstrated that these nanoimprint molds can be created from more durable materials. This advance can broaden the commercially viability of nanoimprint lithography. They describe current work at 13 nanometer resolution using metallic glasses but are progressing to 1 to 2 nanometers using carbon nanotubes. Achieving 1 nanometer resolution would be one thousand times better than the 32 nanometer resolution lithography that Intel is commercializing now. Below details are presented from the website of the Yale research group.

In nanoimprint lithography, a mold made of a hard material such as metal or silicon is pressed into a softer material, often molten silicon itself or a polymer. The molds can then be reused. But both metals and silicon have limitations as mold materials. Silicon is brittle, and molds made of the material fail after about a hundred uses, says Schroers. Metal molds are more resilient but are grainy, and their features can't be any smaller than the grains in the metal itself--about 10 micrometers.

"In many ways, metallic glasses are an ideal material for nanoimprint lithography," says John Rogers, a professor of materials science and engineering at the University of Illinois at Urbana-Champaign, who was not involved in the work. "They're extremely strong, and they can be molded at extremely high resolution." Ordinary metals are crystalline. But bulk metallic glasses, which are created by cooling liquid metals very rapidly to prevent crystallization, lack such structure. Like silicon molds, they can be patterned very finely.

Schroers says that metallic-glass molds can be used millions of times to pattern materials, including polymers like those used to make DVDs. The Yale group has used the molds to create three-dimensional microparts such as gears and tweezers, as well as much finer structures. This week in the journal Nature, Schroers's group describes making molds with features as small as 13 nanometers.

"Theoretically, the size limit is the size of a single atom," says Schroers of the metallic-glass molds. Indeed, the Yale researchers hope to make molds that can form even finer structures by controlling the surface chemistry of the metallic glasses. But the main limitation on the molds is the structure of the metal and silicon templates used to make them. In the hope of further increasing the molds' resolution, Schroers is now developing templates made of nanostructures such as carbon nanotubes only one to two nanometers in diameter.








Jan Schroers lab site describes the nanoimprint work.

In February 2009, Nature will publish an article titled Nanomoulding using thermoplastic forming with bulk metallic glass. Golden Kumar , Hong Tang, and Jan Schroers developed a method to directly imprint features as small as 13 nm onto bulk metallic glass (BMG). We utilized thermoplastic forming of BMG and a favorable wetting behavior. This technique enables a reliable and economic process for nanoimprint lithography (NIL). Furthermore, as oppose to any other material/technology solution it provides a solution for both mold (template) and imprint material which makes it particularly interesting for high density data storage.



Wetting dominates the filling characteristics for mold diameters below 100 nm. For best control over the imprinting process a small < 100 MPa forming pressure requirement is ideal. Utilizing favorable wetting smallest features can be directly imprinted into BMGs as small as 13 nm with an aspect ratio of up to 50.



Schematic and experimental illustration of a processing technique based on the unique softening behaviour of BMG. First, BMG1 is embossed on a mold fabricated by conventional techniques. The mold can be any suitable material such as Ni, Si or alumina. The mold and BMG1 are separated, leaving a negative pattern of the mold imprinted on BMG1. Figure 4b shows an example of pattern transfer on Pt-BMG after embossing on the Ni mold. The patterned BMG1 can be used as a mold to imprint on a lower Tg metallic glass, BMG2. This step is demonstrated by using patterned Pt-BMG to imprint on Au-BMG. Alternatively, the crystallized BMG1 pattern can even be used as a mold for another amorphous sample of BMG1

The nature abstract on nanomoulding

Nanomoulding with amorphous metals

Nanoimprinting promises low-cost fabrication of micro- and nano-devices by embossing features from a hard mould onto thermoplastic materials, typically polymers with low glass transition temperature. The success and proliferation of such methods critically rely on the manufacturing of robust and durable master moulds. Silicon-based moulds are brittle and have limited longevity. Metal moulds are stronger than semiconductors, but patterning of metals on the nanometre scale is limited by their finite grain size. Amorphous metals (metallic glasses) exhibit superior mechanical properties and are intrinsically free from grain size limitations. Here we demonstrate direct nanopatterning of metallic glasses by hot embossing, generating feature sizes as small as 13 nm. After subsequently crystallizing the as-formed metallic glass mould, we show that another amorphous sample of the same alloy can be formed on the crystallized mould. In addition, metallic glass replicas can also be used as moulds for polymers or other metallic glasses with lower softening temperatures. Using this 'spawning' process, we can massively replicate patterned surfaces through direct moulding without using conventional lithography. We anticipate that our findings will catalyse the development of micro- and nanoscale metallic glass applications that capitalize on the outstanding mechanical properties, microstructural homogeneity and isotropy, and ease of thermoplastic forming exhibited by these materials


blog comments powered by Disqus