Most all atom-defined structures that have been built to date have been held together by extremely weak bonds and are therefore unstable at room temperature. Moreover, most systems of study have been fabricated on metal substrates that entirely mask any interesting electronic
properties of atomic (or molecular) assemblies. Patterned atomic dangling bonds on an
otherwise Hydrogen-terminated silicon surface are robust and unchanging to over 200 C and
offer useful electronic properties.
This area was pioneered largely by Joseph Lyding more than 2 decades ago. Many (many!) challenges stood in the way of tests and applications in this area. Over the last 20 years we have doggedly pursued and overcome innumerable fabrication and measurement issues related to
dangling bonds and as well have come to greatly advance understanding of the relations among
dangling bonds and of those with bulk silicon. Crucially, we recognize and utilize the gap-state
character of dangling bond structures to make 2D surface circuits that are isolated from the bulk – or if desired – connected with spatial and temporal control. Operation of binary atomic silicon logic gates and a binary wire have been demonstrated. These are derived from earlier, nonatomic,
field-controlled computing ideas developed by Lent, Snider, Porod and others at Notre
Dame. Our atomic building blocks retain fast, low power qualities while removing a cryogenic
operation requirement and while also providing unprecedented variance-free devices.
In this talk I will show new results including fabrication and multi-spm-probe characterization of nm-scale silicon wires that are intimately bonded to yet electrically isolated from normal bulk Si. Atom-defined wires and their quantized states will be shown. Also atomic lines that can be switched from covalent to ionic bonding will be shown. New boron in silicon results will
be shown. Atomic wire coupling – perhaps even a “read-out” of atomic dopant will be presented. Qubit, magnetometry, quantum metrology ideas flowing from this might be chatted about in our ad hoc sessions.
The many roles of multi-modal imaging, AI and modeling will be described.
Back to Workshop IV: Multi-Modal Imaging with Deep Learning and Modeling