Tailoring the resonant spin response of a stirred polariton condensate

TL;DR

This work addresses the limited spin coherence in exciton-polariton condensates by implementing a rotating, bichromatic optical trap that drives resonant spin precession and synchronizes trap rotation with the condensate's self-induced Larmor precession. The authors demonstrate an almost order-of-magnitude enhancement of the spin coherence time $T_2$ at resonance, with the resonance width tunable via the trap shape controlled by the intensity ratio $r$, and provide a comprehensive Adler-equation–based theory for mutual phase synchronization between the circular polarization components. Their model, incorporating pump-noise and nonlinear interactions, quantitatively reproduces the experimental resonance curves and $T_2$ variations, and is corroborated by full simulations of coupled polarization modes. The results reveal a robust route to optically control spin dynamics in polariton systems, with implications for spinoptronics, polariton-based qubits, and Floquet/time-crystal studies in driven quantum fluids.

Abstract

We report on the enhancement of the spin coherence time (T2) by almost an order-of-magnitude in exciton-polariton condensates through driven spin precession resonance. Using a rotating optical trap formed by a bichromatic laser excitation, we synchronize the trap stirring frequency with the condensate intrinsic Larmor precession, achieving an order of magnitude increase in spin coherence. By tuning the optical trap profile via excitation lasers intensity, we precisely control the resonance width. Here we present a theoretical model that explains our experimental findings in terms of the mutual synchronization of the condensate circular polarization components. Our findings underpin the potential of polariton condensates for spinoptronic devices and quantum technologies.

Figures (11)

• Figure 1: (a) Measured $g^{(2)}_{H,V}$ for the polariton condensate in a static annular optical trap at pump power $P= 2.7$$P_{th}$. (b) Measured $g^{(2)}_{H,V}$ for the condensate in the rotating optical trap ($r = 20 \%$) at $f=0.5$ GHz resonant to the self-induced Larmor precision frequency. The upper inset in panel (b) showcases the shape of the optical pump and induced energy splitting between the cross-polarized spin components of the condensate that is equal to the frequency of the external time-periodic drive. (c) The amplitude of the spin precession $\Delta g^{(2)}_{H,V}$ in the vicinity of the 15 ns time delay for (c) $r = 1 \%$ and (d) $r =20 \%$ demonstrating the resonant behavior. The upper insets in panels (c,d) schematically represent the shape of the rotating optical pattern for different $r$. The green point in (d) is an amplitude of the spin precession presented in (b).
• Figure 2: $g^{(2)}_{H,V}$ as a function of the stirring frequency $f$ and time delay of the HBT interferometer for (a) $r= 1\%$ and (b) $r= 20\%$. The purple dots in panels (a,b) represent the range of the $g^{(2)}_{H,V}$ retrieved within 2 ns vicinity of 15 ns time delay. The cross sections of panel (b) taken at $f = -0.5$ GHz and $f = 0.5$ GHz are depicted in (c) and (d), respectively. The light and dark blue arrows in (b) mark the positions of the cross-sections. Simulated auto-correlation functions $g^{(2)}_{HV}$ for (d) $f=0.5$ GHz and (f) $f=1.1$ GHz, the inter-polarization coupling strength is $\tilde{\sigma}=542$ MHz. (g) The experimental width ($1/\ T_2$) of the resonant curve as a function of $r$. The black dots are experimentally measured widths, the red dots - the widths of the synchronization ranges calculated from (\ref{['phase_equation']}). The black and red lines are to guide the eye.
• Figure 3: (a,b) The dependence of the average detuning $\Delta=\omega_{\uparrow}-\omega_{\downarrow} + < \dot \varphi >$ of the frequencies of $\uparrow$ and $\downarrow$ polarizations from the potential rotation velocity $\Omega$ for the ratio $r= 1$% ($\tilde{\sigma}=144$ MHz) and $20$% ($\tilde{\sigma}=542$ MHz), respectively. The synchronization range in (a,b) is shown with vertical dashed lines. The red lines in (a,b) depict stationary mutual phase $\varphi$ as a function of $\Omega$ within the synchronization range. Plots in panels (a,b) are obtained from the numerical simulations of Eq. \ref{['phase_equation']}. (c,d) The numerically calculated probability density of the mutual phase $\varphi$ as a function of $\Omega$ for the $r =20$ % ($\tilde{\sigma}=542$ MHz) and $1$% ($\tilde{\sigma}=144$ MHz), respectively. The experimentally measured mutual phases are shown with the white circles in panel (d).
• Figure S1: (a) The linewidth of the PL as a function of the pump power (in unit threshold). The data is accurately fitted by a linear fit $\gamma_{eff}=\gamma (1-\frac{P}{P_{th}})$. (b) The PL central frequencies (the energy of the polariton state) for different pump powers. The theoretical fit is $\omega_s=\omega_0+ \eta ( \frac{P}{P_{th}} -1)$ with $\eta=0.83$ where $\omega_0$ is the PL frequency at the threshold. The PL frequencies are shown as their detunings from $\omega_0$. (c) The experimentally measured frequencies of the condensate above the threshold and the theoretical curve $\omega_s=\omega_0 +\mu ( \frac{P}{P_{th}}-1 )$ for $\mu=0.1125$ meV versus the pump power. In (b) and (d) the reference frequency $\omega_0$ is the frequency of the condensate at the condensation threshold. In all panels, the open red circles are the experimental data and the black solid lines are the theoretical fits.
• Figure S2: The wash-board potential for equation (\ref{['phase_mut_phase_S']}). The bottom part of the figure illustrates the temporal evolution of the probability density starting from a delta function (schematically shown by thick blue line. )