Tuesday, May 13, 2008

Nanolithography

Subprojects and Devices:

Thermally and electrostaticlly actuated probe arrays
Thermally actuated probe array modeling
High density micro ink supply systems
Nanolithography Slideshow

Project Overview

The nanolithography project is a focused effort to develop tools that can modify surfaces with nanoscale resolution in a massively parallel fashion. Techniques involve the combination of directed "top-down" methods (direct probe control) and thermodynamically driven "bottom-up" methods such as self-assembly. Success is contingent on the ability to bridge the gap between macroscopic "real world" systems and the nanoscale dimensions that many modern technologies require.

The core process we use for surface modification is known as dip pen nanolithography (DPN). In this process, a modified atomic force microscope probe is coated with a surface binding chemical that is to be deposited. When the probe is placed in contact with the surface, a water meniscus condenses between them from the ambient environment and forms a liquid bridge. The bridge is then used as a diffusion pathway for chemical transport between the tip and surface.

This project is a collaboration with researchers in the Mirkin Group at Northwestern University who are developing a battery of compatible ink/surface chemistry systems. As a result of this research, direct chemical lithography may become a core enabling technology for applications ranging from semiconductor fabrication to protein synthesis.

Thermally and electrostaticlly actuated probe arrays

A critical component of the DPN system is the lithography probe. The only devices presently available for this task are are commercially available AFM probes. Although useful for demonstrating the concept, they are not optimal for converting DPN into a commercially viable process. Our goal is to develop high density probe arrays that allow many DPN patterns to be written simultaneously. Embedded actuators allow individual probes to be removed from the surface so different patterns can be created with different probes in the same run. Thermally and electrostaticlly actuated probes have been constructed to investigate the advantages and disadvantages of each method in various applications.

Images:

1) An array of thermally actuated DPN probes

2) An array of electrostaticlly actuated DPN probes

Related papers:

1) M. Zhang, D. Bullen, S-W. Chung, S. Hong, K. Ryu, Z. Fan, C. Mirkin, C. Liu, "A MEMS nanoplotter with high-density parallel dip-pen nanolithography probe arrays," Nanotechnology, Vol. 13, pp. 212-217, April 2002.

2) M. Zhang, D. Bullen, K. Ryu, C. Liu, S. Hong, S. Chung, C. Mirkin "Passive and Active Probes for Dip Pen Nanolithography," First IEEE Conference on Nanotechnology, October 28-30, Maui, HI, 2001.

Thermally and actuated probe array modeling

On the surface (ha!), DPN probes are similar to contact mode AFM probes and the probe arrays produced by other groups. In reality, the conditions necessary for performing parallel DPN create several issues that are not relevant in other scanning probe lithography techniques. In most of our systems, we have chosen thermal actuation and a fabrication method that does not require wafer bonding to ensure commercial viability. This yields devices that are simple to produce but have interrelated mechanical design issues related to probe geometry and materials. Different designs can result in radically different combinations of probe spring constant, actuator strength, operating temperature, substrate scratching risk, and so forth. These issues must be balanced carefully to produce working devices.

Do to the large number of conflicting constraints, an analytical design approach is being used to isolate optimal device layouts. Since, the optimal design depends heavily on the the application, and since there are no simple ways to correlate all the critical factors, a "shotgun" design approach is being used. In this method, large blocks of designs are evaluated based on their mechanical properties. The successful designs, along with the designs that fail certain criteria, are plotted against the independent variables of the simulation. This allows a rapid visual inspection of the design space and evaluation of various optimization schemes. The use of analytical analysis instead of FEA allows 10's to 100's of thousands of designs to be evaluated in a matter hours, including effects that are not easily modeled using FEA.

Images:

1) An example plot of successful thermal actuator designs for a given set of design constraints