Exciton and biexciton preparation via coherent swing-up excitation in a GaAs quantum dot embedded in micropillar cavity

TL;DR

This work addresses coherent state preparation of exciton and biexciton states in a GaAs quantum dot to enable scalable quantum photonic devices. It systematically compares swing-up excitation (SUPER) to two-photon excitation (TPE) using a four‑level model that includes the biexciton state, demonstrating near‑unity exciton population ($P_X \approx 0.95$), high single-photon purity ($g^{(2)}(0) \approx 0.03$), and a fast emission time ($\sim200\,\mathrm{ps}$) under SUPER, with a polarization-enhanced brightness by a factor of $\approx 1.45$. The study furthermore shows the first experimental biexciton preparation via SUPER and uncovers a polarization-misaligned resonance enabling selective population of an orthogonal exciton dipole, revealing a new degree of coherent control. Altogether, SUPER emerges as a versatile, off‑resonant protocol for high‑fidelity, state‑selective quantum light sources, with strong implications for deterministic entangled-photon generation and scalable quantum photonics.

Abstract

Coherent control of quantum emitters is essential for scalable quantum photonic technologies. The recently proposed swing-up of quantum emitter (SUPER) scheme allows efficient and coherent preparation of single photons via off-resonant, red-detuned laser pulses, simplifying laser suppression and enhancing photon collection. We present a systematic study of SUPER excitation applied to a single GaAs quantum dot in a low-Q micropillar cavity. We perform a comparison of the key figures of merit against the well-established two-photon excitation (TPE). Despite requiring higher excitation powers, SUPER achieves near-unity population inversion of the exciton state ($\sim$95%) and high single-photon purity ($g^{(2)}=0.03$) comparable to that under TPE, while also exhibiting a shortened decay time ($\sim$200 ps) reducing the time jitter in the exciton population. A polarization-resolved analysis reveals that when both excitation and collection are aligned with one of the exciton dipoles, SUPER results in polarized single-photon emission, exceeding the resonant TPE saturation by a factor of 1.45. Under optimized excitation conditions, we also observe biexciton preparation via a distinct SUPER resonance, confirmed by the appearance of the biexciton emission line, constituting the first experimental demonstration of biexciton preparation using SUPER. These findings are in good agreement with a proposed four-level theoretical model that incorporates the biexciton state. We also report that a slight misalignment of laser polarization induces an additional SUPER resonance that selectively populates the orthogonal exciton dipole, without altering the nominal excitation polarization. This unexpected behavior reveals a new degree of freedom for coherent state preparation. Our findings establish the SUPER scheme as a versatile tool for state-selective exciton and biexciton control.

Figures (4)

• Figure 1: a Photoluminescence (PL) spectrum of the examined QD under CW above-band excitation at 650 nm. Inset: Scanning electron microscopy (SEM) of the micropillar (D = 1.58 µm) hosting the QD. b Illustration of the laser energies utilized for different excitation schemes in relation to the QD transitions and energy diagram schematics of the system. $\Delta_{\textrm{FSS}}\approx12$$\mu$eV. Detuning from the neutral exciton energy: $\Delta_{\textrm{TPE}}$=-2 meV, SUPER pulses: $\Delta_1$=-3.5 meV, $\Delta_2$=-7.5 meV. c Legend: Diagram of the employed convention for the excitation and detection angles in relation to the V,H-exciton states dipoles. Graph: Linear polarization anisotropy of the X PL under SUPER and TPE excitation for different orientations of the laser polarization. d Comparison of the X PL intensity under TPE and SUPER for different polarization configurations. Data was normalized by the maximum number of counts under TPE excitation for linear polarization of detection.
• Figure 2: Experimental (a,b) and simulated (c,d) excitation power scans of the X emission intensity under SUPER scheme for different detunings $\Delta_2$ for co-(a,c) and cross-(b,d) linear configuration of detection polarization, $\phi\approx 0$. Excitation parameters: $\Delta_1=-3.7$ meV, P$_1$=P$_2$. Simulation parameters: binding energy of biexciton E$_{b}=4$ meV, linewidth of the pulses FWHM=0.48 meV, P$_1$=P$_2$, $\Delta_1=-3.7$ meV.
• Figure 3: a Intensity of the neutral exciton (X) photoluminescence (PL) under SUPER excitation as a function of the detuning of the second excitation pulse and the detection angle of linear polarization. b Temporal decay of the X PL under SUPER excitation for different detunings of the second laser pulse, with the polarization of the laser aligned either along (upper panel) or between (lower panel) the dipole orientation of the exciton fine-structure-split states. c PL spectrum of the QD under SUPER excitation of the biexciton state (upper panel) and dependence of the X, XX intensity as a function of the detuning of the second pulse. (Data was normalized by the intensity observed at $\Delta_2=-6.1$ meV). d Intensity of the X PL under SUPER excitation while changing the detuning of the laser pulses. Dashed lines indicate the resonant conditions for exciting either the neutral- or the biexciton state. (P$_1$ is linearly increasing in the investigated range when tuning towards higher energies due to the limitation of the experimental setup.)
• Figure 4: a Polarization anisotropy scan of the neutral exciton (X) photoluminescence (PL) under SUPER excitation as a function of $\Delta_2$ with the laser polarization angle aligned at 90$^\circ$, 70$^\circ$ and 45$^\circ$ with respect to the V exciton state. b (upper) X PL decay of the V/H-states for $\Delta_2$ corresponding to either the F$_1$ or F$_3$ resonance fringe. b (lower) Quantum beats observed in the X PL signal while exciting at the F$_3$ resonance fringe. c Intensity of the X PL under SUPER excitation while changing the detunings of the laser pulses for cross-linear polarization configuration. Dashed line indicate the F$_3$ resonance fringe related to efficient population of the orthogonal X state. (Experimental conditions and colorscale same as in Fig. \ref{['fig:Fig3']}d). d Comparison of the X PL intensity under SUPER excitation for $\Delta_2$ corresponding to the F$_3$ resonance fringe and the observed quantum beat amplitude as a function of the polarization angle of the laser pulses. (QB amplitudes were measured for $\Delta_2$ tuned to the F$_1$ resonance fringe). e Linear polarization contrast ($\frac{I_{0^\circ}-I_{90^\circ }}{I_{0^\circ}+I_{90^\circ }}$) of X PL under SUPER for $\Delta_2$ corresponding to the F$_3$ excitation fringe.