Highly stable two-level current fluctuation in complex oxide heterostructures

Area of Science:

• Materials Science
• Condensed Matter Physics
• Quantum Computing

Background:

• Two-level systems in dielectric oxides are promising entropy sources for random number generators.
• Random telegraph noise (RTN) from these systems is crucial for generating random bit-strings for computing and cryptography.
• Classical oxide systems face instability issues due to defect migration and metastable states, hindering reliable RTN generation.

Purpose of the Study:

• To present a stable two-level quantum system for reliable random number generation.
• To investigate the use of complementary cation and anion point defects for enhanced system stability.
• To demonstrate the application of generated random bit-strings in stochastic machine learning algorithms.

Main Methods:

• Fabrication of a SrRuO3/LaAlO3/Nb-doped SrTiO3 heterostructure.
• Utilizing oxygen vacancies and antisite Ti defects as complementary point defects.
• Analyzing temporal electron localization and Coulomb interactions for two-level current fluctuations.
• Generating and testing random bit-strings for image super-resolution tasks.

Main Results:

• Demonstrated a stable two-level quantum system based on the heterostructure at room temperature.
• Observed highly stable RTN-like current signals attributed to defect interactions.
• Successfully generated random bit-strings from the stable fluctuations.
• Confirmed the applicability of these random bit-strings in stochastic machine learning for image super-resolution.

Conclusions:

• The SrRuO3/LaAlO3/Nb-doped SrTiO3 heterostructure provides a stable and reliable entropy source.
• Complementary interactions between cation and anion point defects are key to designing stable oxide-based electronic systems.
• This approach offers a pathway for developing robust hardware-based random number generators for advanced technologies.