Detecting axion dynamics on the surface of magnetic topological insulators

Abstract

Axions, initially proposed to solve the strong CP problem, have recently gained attention in condensed matter physics, particularly in topological insulators. However, detecting axion dynamics has proven challenging, with no experimental confirmations to date. In this study, we identify the surface of magnetic topological insulators as an ideal platform for observing axion dynamics. The vanishing bulk gap at the surface allows for order $O(1)$ variations in the axion field, making the detection of axion-like phenomena more feasible. In contrast, these phenomena are strongly suppressed in the bulk due to the small magnetic exchange gap. We investigate two-photon decay as a signature of axion dynamics and calculate the branching ratio using a perturbative approach. Our findings reveal that the photon flux emitted from the surface is in-plane and orders of magnitude larger than that from the bulk, making it detectable with modern microwave technology. We also discuss potential material platforms for detecting axion two-photon decay and strategies to enhance the signal-to-noise ratio.

Figures (4)

• Figure 1: (a) Schematics of the experimental setup. The magnetic topological insulator is placed in a microwave cavity. Spins on the top surface point up in the red region and down in the blue region. When spin flip within the dashed line enclosed region, there is a finite probability that two in-plane photons will be emitted through the two-photon decay channel of axions. (b-c) A side view of the spin configuration before and after the domain wall movement, and the corresponding $\theta$ field. $\theta$ only changes a significant amount at the surface.
• Figure 2: The energy distribution of the emitted photons in the domain wall decay setup. $\omega$ and $\omega^\prime$ represent the energies of the two emitted photons, and $\Delta$ is the reduction in exchange energy during axion decay. The peak at $\omega+\omega^\prime=\Delta$ is very sharp. (b) is plotted along the $\omega^\prime=0$ line, i.e. the $x$-axis in (a).
• Figure 3: The bulk branching ratios of the photon channel in the domain wall shrinkage setup as functions of the axion size $a$. The spontaneous two-photon emission branching $\eta^B$ is independent of $a$. The stimulated emission branching $\tilde{\eta}^B$ is proportional to $a^4$. The branching of the stimulated emission is larger $\tilde{\eta}^B > \eta^B$ only when $a>a_c$.
• Figure S1: The bulk branching ratio $\eta^B$ as a function of $\log{w/a}$ in the bulk magnon decay setup. It can be concluded that the variation of $w/a$ amounts at most a $15\%$ change in $\eta^B$.