Summary
A Physical Review Letters paper reports a way to distinguish long-lived two-level systems from ordinary background noise in a superconducting qubit. The method uses patterns in quantum-jump measurements and changes in those patterns as the qubit frequency is swept.
An accepted Physical Review Letters paper reports a way to identify long-lived two-level systems (TLSs) interacting with a superconducting qubit. The authors use patterns in measured quantum jumps—transitions between qubit states—and find that those patterns can distinguish TLS-related behaviour from a conventional, memoryless background environment.
Why environmental memory matters
Models used to reason about fault-tolerant quantum computing often treat dissipation using the Born–Markov approximation. In this picture, a qubit loses energy to its surroundings without the environment retaining relevant information about the qubit’s past. The paper says this approximation can break down when a qubit couples to long-lived TLSs: these systems can become polarised during operation and retain memory of earlier qubit states.
That distinction matters for understanding qubit errors. If the environment remembers past states, its influence cannot be described simply as a constant, memoryless decay process. Identifying such behaviour helps researchers determine which physical sources contribute to the noise around a qubit.
Reading memory from quantum jumps
The researchers analysed quantum-jump dynamics arising naturally from thermal fluctuations. They report that the jump traces are non-Poissonian—a departure from the statistics expected for a simple constant-rate process—and contain information that can separate long-lived TLSs from a standard Markovian bath.
To distinguish the contributions, the team fitted Solomon equations to the measured dynamics. They also swept the qubit’s frequency. The resulting distinct peaks were each associated with a TLS that outlives the qubit, giving researchers a way to investigate the microscopic origin of the environmental defects.
The result is a method for probing a source of noise that can complicate assumptions used in modelling superconducting processors. It offers a route to characterising the qubit environment; the paper’s abstract describes that diagnostic advance rather than a change to quantum error-correction hardware or a demonstration of fault-tolerant operation.