Inverse Purcell Suppression of Decoherence in Majorana Qubits via Environmental Engineering

TL;DR

This work proposes an environmental-engineering strategy to suppress decoherence in Majorana-based qubits by implementing an inverse Purcell effect: shaping the bath density of states to be strongly suppressed at the qubit’s splitting frequency $ε$. With parity-conserving coupling $H_{int} = g \int dx\, Φ(x) (i γ_L γ_R) δ(x-x_0)$, the dephasing rate scales with the noise spectrum $S(ε)$, which in a typical 1D bath yields $Γ_φ \propto ρ(ε) T/ε$. By engineering the bath so that $ρ_{\text{engineered}}(ε) = ρ_{\text{free}}(ε) (ε/ω_c)^α$, the authors show that $Γ_{φ,\text{engineered}} ∝ ε^{α-1} ∝ e^{-(α-1)L/ξ}$, enabling exponential suppression with wire length for $α>1$ and a temperature-independent suppression factor $F_P = (ε/ω_c)^α$. The proposal leverages existing platforms such as high-impedance resonators, Josephson-junction arrays, and phononic crystals to realize the required spectral shaping at millikelvin temperatures, offering a practical route to enhanced coherence in topological qubits and potentially broader applications in quantum technologies.

Abstract

We propose a novel approach for optimizing topological quantum devices: instead of merely isolating qubits from environmental noise, we engineer the environment to actively suppress decoherence. For a Majorana qubit in a topological superconducting wire, the exponentially small energy splitting $ε\sim e^{-L/ξ}$ provides protection against local perturbations but renders it highly susceptible to pure dephasing from low-frequency environmental noise. We show that coupling via a parity-conserving operator ($iγ_Lγ_R$) to a bosonic environment yields a dephasing rate $Γ_φ\propto S(ε)$, where $S(ε)$ is the environmental noise power at the qubit splitting frequency. In the experimentally relevant regime where $k_B T \gtrsim \hbarε$ (with $T \sim 10-100$ mK), the noise power scales as $S(ε) \propto ρ(ε) k_B T/\hbarε$, leading to a dephasing rate $Γ_φ\propto ρ(ε) T/ε$. This exposes a fundamental challenge: the dephasing rate diverges as $1/ε$ for a standard environment, e.g., a 1D system with linear dispersion where $ρ(ε)$ is constant. We overcome this by designing environments with a suppressed density of states following $ρ_{\text{engineered}}(ε) = ρ_{\text{free}}(ε) (ε/ω_c)^α$. This creates an ``inverse Purcell effect'' that yields a temperature-independent suppression factor $F_P = (ε/ω_c)^α$. For $α> 1$, the engineered dephasing rate decreases exponentially with wire length, $Γ_{φ,\text{engineered}} \propto e^{-(α-1)L/ξ}$, meaning longer wires provide better coherence protection. This provides a quantitative design principle where environmental engineering transforms detrimental noise into a tool for coherence stabilization, while respecting fermion parity superselection rules.