Zyvex Labs Research

UHV chamber

Zyvex Labs'
Ultra-High-Vacuum Scanning Tunneling Microscope

APMC PARTNERS & COLLABORATORS

ICSPI (Integrated Circuit Scanning Probe Instruments)
General Dynamics
National Institute of Standards & Technology ^1,
2
North Texas RCIC
University of Central Florida
University of Illinois
University of North Texas
University of Texas at Austin
University of Texas at Dallas ^1,
2,
Vought Aircraft

Institute of Materials Research & Engineering
Nanyang Technological University
University of New South Wales
Zyvex Asia

APMC approach to APM

This work has been supported by research contracts from DARPA (Feasibility of Atomically Precise Manufacturing, program ID HR0011-07-0049 and Atomically Precise Manufacturing, program ID N66001-08-C-2040) and the State of Texas' Emerging Technology Fund.

Our work is done in a custom-built UHV chamber, using a custom-built STM modeled on the Lyding design, running custom software we developed.

1) In an Ultra-High-Vacuum chamber, use a sharp tungsten tip...

A sharp STM tip - approximately 2nm radius of curvature on end.

The image on the left shows oxide and carbon contamination due to handling in air and TEM imaging after preparation.

The image on the right is a high-resolution TEM image showing the tungsten atomic lattice at the end of the tip.

We are investigating use of a single-atom tip to achieve even higher resolution.

2) ...on a hydrogen-passivated silicon surface...

Typical hydrogen passivated silicon <100> surface, showing 2x1 reconstruction and surface contamination. Each of the rows represents a dimer of silicon - a higher magnification image would show pairs of atoms crosswise to the major direction of the dimer.

Surface preparation is important, and much better surface quality is possible, but our priority right now is better precision and more reliable lithography. This surface shows numerous missing atoms and contaminants scattered about the surface, but these do not currently interfere significantly with our lithography development.

3) ...to pop off the hydrogen and perform atomic-precision lithography.

100 lines of silicon depassivation with better than +- 10% control over length.

Some imprecision in the lines is due to piezo creep and hysteresis during the several minutes this took to write. Work is underway to correct for some of these effects.

4) Then grow on the depassivated region via a self-limiting deposition process, and repeat.

APM cartoon

5) Make valuable products with atomic precision, including massively parallel versions of the lithography-based manufacturing hardware, which can then make greater quantities of atomically precise products less expensively, leading to still better manufacturing hardware... This results in a double-declining learning curve: as we get better at manufacturing products less expensively, the factory itself will also become better and less expensive. This contrasts with the semiconductor industry, where wafer fabs become geometrically more expensive with each new generation of technology.