Nanometer patterning with ice
Written by: Grant England AP225 Fall, 2011
Title: Nanometer Patterning with Ice
Authors: King, Gavin M., Gregor Schurmann, Daniel Branton, and Jene A. Golovchenko
This paper detailed the use of water ice as a photolithographic material akin to photoresist. By growing layers of ice on a substrate, and using e-beam or FIB to sublimate parts of the ice off, the authors were able to achieve very high (though not record-breaking) resolution for metal features deposited on the substrate.
Methods and Results
Figure 1 from the paper shows the apparatus used to deposit the thin ice layer as well as the metallization step. The paper detailed several methods used to form a gradient of ice removal for different parameters of the system. Once ice was selectively removed from parts of the substrate, the metallization via sputtering allowed for the creation of small metal features on the substrate that remained once the remaining ice was removed; however, the sample was also covered with pieces of metal from the ice-covered areas once all the ice is sublimated. This contamination could be circumvented by using an inverted chamber.
In addition to allowing metallization of small features, an interesting phenomenon was observed when a dose below the threshold for removing all of the ice was used on a silicon substrate. A reaction occurred between the water ice and the silicon to generate a thin layer of material (probably silicon oxide). The modification of the substrate structure is undesirable for the method as envisioned initially, but the creation of a thin layer could lead to interesting implications for SOI or p-n junction creation
This paper detailed an interesting method for the creation of small-scale metal structures on a substrate which, though not record-breaking, could allow for the creation of bleeding-edge technologies for which standard photolithography does not have the resolution. The side-effect of the patterning technique of using ice as a photoresist analog is that chemistry between the FIB ion and the substrate could potentially be exploited to generate extremely shallow and high-resolution implantation of a material different from that of the substrate for the creation of electronic components like tunnel junctions or gate insulators for nanoscale FETs.