Advanced microcontact printing

Overview

Microcontact printing uses a soft elastomer stamp to bring molecules to a surface, where they adsorb and can act as resists for fabrication, modify the surface properties or provide binding sites for biomolecules. Our group has worked extensively on the application of microcontact printing, for example, as an alternative to photolithography in the fabrication of LC displays. Details of our work on printing self-assembled monolayers, its use for metal patterning as well as printing of metal layers can be found in the publications.

For the use of microcontact printing in the field of biological patterning, please refer to our work in Life sciences.

Figure 1 shows an optical micrograph of a lithograph by M. C. Escher, reduced 1000-fold, and replicated by printing and etching gold on silicon. The original has been scanned and transformed into a high-resolution SOI master by electron beam lithography (Extreme Lithography, Germany) with a pixel size of 200 nm. A stamp was molded off the master and used for printing alkanethiols onto a gold layer, followed by a selective etch to develop the pattern.

The quality and usability of microcontact printing depend critically on the transport of the molecules from the stamp onto the surface. We have investigated these processes in detail to find conditions that allow high-throughput printing, and we make use of the transport situation to print gradient surfaces.

Transport in microcontact printing

Conventional printing processes transfer ink from a surface to a substrate. Some inks used in microcontact printing, however, can diffuse into the bulk material of the stamp. The printing process is then a diffusion of molecules from the bulk onto the substrate. To produce high-quality prints, the diffusion has to be sufficiently fast, but at the same time limited enough not to cause surface diffusion of the ink that leads to bleeding.

We have shown that the diffusion of a common ink (hexadecanethiol) in the stamp elastomer (PDMS Sylgard 184) can be described using Fick's law with a diffusion constant of 5.9×10-7 cm2 s-1, and we have identified process windows that enable high-resolution prints. One such combination of process parameters is used in our high-speed microcontact printing process.

High-speed microcontact printing

Application of microcontact printing in an industrial process demands constant print quality throughout a sequence of multiple prints. This is a considerable challenge, given the varying dimensions and fill-factors of simultaneously printed patterns. Under standard printing conditions the alkanethiol ink is supplied from the bulk of the stamp, leading to an oversupply and pattern broadening in sparsely filled areas or insufficient protection in densely filled regions of the print.

Print quality can only be independent of the pattern fill factor when the alkanethiol ink for a single print is supplied solely from the patterned volume of the stamp (Figure 2B). Diffusion simulations reveal that high ink concentrations of 20 mM and printing times in the millisecond regime are required to reach this goal.

We built an instrument with a piezo actuator that is capable of printing in the millisecond time regime. Using this tool, we showed for the first time that alkanethiol monolayers with sufficient resist properties can be printed with a stamp-substrate contact time of only a few milliseconds (Figure 3).

Chemical gradients

Transport limitations in microcontact printing are, in general, a nuisance. It is possible to make use of them, however, for the printing of surface gradients. In thin stamps with varying thickness and low ink concentration, the transport of the ink is governed by the stamp geometry. Linear gradients can be printed with them: at the thin end of the stamp, very few molecules reach the surface, whereas a full monolayer is adsorbed at the thick end.

Figure 4 shows the surface chemistry evolution over the length of a gradient that was printed from a thin, wedge-shaped stamp. From one end to the other, the composition of the gradient changes from a pure hexadecanethiol monolayer (which was printed from the thick end of the stamp) to a pure perflurododecanethiol monolayer. The perfluorinated monolayer was grown in those parts of the surface not occupied by the printed gradient after the print, simply by immersing the entire surface into a solution of the perfluorinated thiol.