A Probabilistic Finite State Logic Machine Realized Experimentally on a Single Dopant Atom.

Area of Science:

• Quantum computing
• Nanoscale electronics
• Solid-state physics

Background:

• Conventional computing faces limitations at the nanoscale.
• Novel switching paradigms are needed for nanoscale logic processing.
• Single-atom devices offer unique computational possibilities.

Purpose of the Study:

• To design and experimentally realize a probabilistic finite state machine (PFSM) using a single phosphorus donor atom in silicon.
• To demonstrate a single-atom logic unit capable of simulating complex processes.
• To solve an optimization problem related to maximizing visitor dwell time in a simulated maze.

Main Methods:

• Utilizing a single phosphorus donor atom in a silicon matrix as the computational unit.
• Employing scanning tunneling spectroscopy (STS) for electrical addressing and probing.
• Leveraging the stochastic nature of electron tunneling for probabilistic operations.
• Analyzing macroscopic current output, not requiring single-electron sensitivity.

Main Results:

• Successful implementation of a PFSM in a single-atom device.
• Demonstration of simulating visitor flow through a four-room maze topology.
• Identification of optimal door opening rates to maximize time spent in a specific chamber.
• Validation of macroscopic current as a readable output for the computational process.

Conclusions:

• Single-atom devices can perform complex logic functions beyond conventional paradigms.
• Probabilistic computing can be realized at the atomic scale.
• This approach offers a pathway for developing new nanoscale computing architectures.