Swing-UP of Quantum Emitter Population (SUPER)

• SUPER is an off-resonant excitation protocol that employs two detuned pulses to achieve high-fidelity inversion of quantum emitters via coherent dressed-state dynamics.
• It leverages interference between pulse-driven dressed states to keep pump and emission spectrally separate, making it ideal for single-photon sources in semiconductor and cavity-QED systems.
• Precise control of pulse amplitude, detuning, and timing is critical, as small deviations can significantly reduce inversion efficiency and overall device performance.

Swing-UP of Quantum Emitter Population (SUPER) is an off-resonant coherent excitation protocol for quantum emitters in which two detuned optical fields, typically two red-detuned Gaussian pulses, drive population inversion without using a resonant pump. In its standard form, SUPER targets a two-level emitter and exploits interference between the two drives to create an effective modulation matched to a dressed-state splitting, allowing the system to be transferred from the ground state to the excited state with high fidelity while keeping the pump spectrum spectrally separated from the emission line. Within semiconductor quantum photonics, SUPER has been developed as an excitation strategy for on-demand single-photon sources, extended to cavity-QED settings, reinterpreted in fully quantized terms as a multi-photon process, and generalized to collective-state preparation in coupled-emitter systems (Bracht et al., 2021, Bracht et al., 2022, Heinisch et al., 2023, Vannucci et al., 2024, Kerber et al., 27 Aug 2025, Crowder et al., 31 Oct 2025).

1. Definition, scope, and historical development

SUPER was introduced as a coherent excitation scheme for a two-level quantum emitter using strongly detuned pulses that would not produce substantial occupation in an ordinary off-resonant Rabi picture. The original formulation considered both a frequency-modulated single-pulse realization and a practically relevant two-color version in which two pulses with different strong detunings of the same sign achieve the same swing-up effect (Bracht et al., 2021). In subsequent work, the two-color form became the standard implementation, particularly for semiconductor quantum dots, where both pulses are red-detuned by several meV and lie spectrally below the emission frequency, enabling straightforward pump rejection by spectral filtering rather than polarization filtering (Karli et al., 2022, Boos et al., 2022).

The central distinction between SUPER and conventional resonant $\pi$-pulse excitation is that no single field component is required to be resonant with the bare emitter transition. In resonant excitation, inversion is obtained by direct Rabi rotation on the bare transition. In SUPER, inversion arises from coherent dynamics in a dressed basis created by a strong off-resonant field and driven by a second off-resonant field tuned to the dressed-state splitting (Bracht et al., 2022). This also distinguishes SUPER from adiabatic rapid passage, which relies on chirped pulses and adiabatic following, and from phonon-assisted excitation, which uses dissipative relaxation rather than purely coherent population transfer (Bracht et al., 2021, Karli et al., 2022).

Experimentally, SUPER was demonstrated on a semiconductor quantum dot using two red-detuned phase-locked spectral components shaped from a broadband pulse. In that study, neither pulse alone produced appreciable excitation, whereas the two pulses together yielded coherent exciton preparation and emitted photons with properties comparable to resonant excitation (Karli et al., 2022). Later work investigated coherent dynamics and single-photon-source performance in greater detail, including extensive parameter scans and analysis of sample-limited indistinguishability (Boos et al., 2022). A systematic study on a GaAs quantum dot in a low-Q micropillar cavity reported near-unity population inversion of about $95\%$, single-photon purity $g^{(2)}=0.03$, and a shortened exciton decay time of about $200$ ps, while also finding polarized single-photon emission exceeding resonant two-photon biexciton-excitation saturation by a factor of $1.45$ when excitation and collection were aligned with the exciton dipole (Piccinini et al., 24 Oct 2025).

2. Dressed-state mechanism and semiclassical theory

In the standard semiclassical description, the emitter is treated as a two-level system with ground state $\ket{g}$, excited state $\ket{e}$, and transition frequency $\omega_0$. Ignoring phonons and spontaneous emission, the lab-frame system Hamiltonian is written as

$$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$

where $\sigma^\pm$ are raising and lowering operators and $95\%$0 is the total complex Rabi drive (Crowder et al., 31 Oct 2025).

For the two-pulse SUPER protocol,

$95\%$1

with carrier frequencies $95\%$2, Gaussian envelopes $95\%$3, relative delay $95\%$4, and relative phase $95\%$5 (Crowder et al., 31 Oct 2025). In the rotating frame of the first laser, and within the rotating-wave approximation, the SUPER Hamiltonian becomes

$95\%$6

where $95\%$7 (Crowder et al., 31 Oct 2025).

The physical interpretation is that the first pulse creates a large AC Stark splitting and a dressed-state doublet, while the second pulse introduces an oscillatory term at the beat frequency $95\%$8. SUPER is realized when that beat frequency matches the generalized Rabi splitting of the first drive. In the notation of the 2025 comparison study, the inversion condition is

$95\%$9

so that

$g^{(2)}=0.03$0

This gives resonant coupling between dressed states and yields a swing-up from the ground to the excited state (Crowder et al., 31 Oct 2025).

A dressed-state analysis made this interpretation explicit. Diagonalizing the emitter plus first pulse gives dressed states with splitting $g^{(2)}=0.03$1, and the second pulse drives transitions between them when its frequency difference from the first pulse satisfies $g^{(2)}=0.03$2. In the two-level case, this analysis yields analytic conditions for the second pulse detuning and pulse area required for inversion in the dressed basis (Bracht et al., 2022). Earlier work had already framed the swing-up effect as repeated off-resonant rotations of the Bloch vector around different axes, such that the net effect of alternating between strongly detuned configurations is a cumulative increase in excited-state population (Bracht et al., 2021).

The original proposal also studied a frequency-modulated version in which a single pulse has sinusoidally modulated detuning,

$g^{(2)}=0.03$3

with modulation frequency chosen close to the generalized Rabi frequency at pulse maximum. That work showed that full inversion can occur even when no spectral component is resonant with the transition, although for optical implementations the experimentally more relevant realization is the two-color protocol (Bracht et al., 2021).

3. Pulse structure, operating regime, and parameter dependence

SUPER is typically implemented with two overlapping Gaussian pulses of large area and substantial negative detuning. In the microscopic comparison of off-resonant excitation schemes for a quantum-dot single-photon source, the explicit SUPER parameters were: first pulse with $g^{(2)}=0.03$4, $g^{(2)}=0.03$5, and $g^{(2)}=0.03$6; second pulse with $g^{(2)}=0.03$7, $g^{(2)}=0.03$8, $g^{(2)}=0.03$9, delay $200$0, and phase $200$1 (Crowder et al., 31 Oct 2025). These are strongly detuned, high-amplitude pulses, and the total drive spectrum is many meV away from the qubit emission line.

For these parameters, the spectral overlap measure between the total driving spectrum and the qubit emission spectrum is

$200$2

meaning that the pulses are spectrally disjoint from the zero-phonon emission line (Crowder et al., 31 Oct 2025). This spectral disjointness is one of the protocol’s defining operational advantages in single-photon-source applications, because it removes the inherent $200$3 efficiency penalty associated with polarization filtering in resonant excitation schemes (Crowder et al., 31 Oct 2025).

The effective operating regime requires four conditions. First, both detunings must be much larger than the radiative linewidth, so that the pulses are far off-resonant. Second, the pulse amplitudes must be large enough that the generalized Rabi splitting is substantial. Third, the beat-frequency matching condition must hold at the time of maximum first-pulse amplitude. Fourth, the pulse delay must ensure temporal overlap near that maximum (Crowder et al., 31 Oct 2025). In practice, the comparison study notes that the parameters are not obtained by a complete analytic optimization but by scanning pulse parameters to identify combinations that achieve near-perfect inversion (Crowder et al., 31 Oct 2025).

Systematic scans confirm that SUPER operates over extended but structured regions of parameter space. For single emitters, large pulse areas, picosecond durations, and red detunings from a few meV to tens of meV were found to support high-fidelity inversion, with the exact optimum depending on pulse widths and timing (Bracht et al., 2021, Boos et al., 2022). In coupled emitters, similar scans over the two pulse areas showed broad regions where either the superradiant or the subradiant collective state is populated with high probability, indicating that the pulse parameters are not unique (Kerber et al., 27 Aug 2025).

At the same time, robustness is not uniform across all parameters. In the 2025 single-photon-source study, varying the second-pulse parameters around the optimum showed that coherence and indistinguishability remain excellent across a wide range, but efficiency is highly sensitive to the second-pulse amplitude and detuning. A $200$4 change in $200$5 reduces efficiency to about $200$6, a $200$7 change reduces it to about $200$8, and a detuning error of only about $200$9 meV can drive efficiency below $1.45$0, because the exact dressed-state matching condition is broken (Crowder et al., 31 Oct 2025). This establishes a characteristic trade-off: SUPER can deliver near-ideal performance under good calibration, but it demands tight control of pulse parameters.

4. Single-photon-source performance and comparison with competing schemes

Within single-photon-source applications, the main figures of merit are the efficiency $1.45$1, zero-delay second-order correlation $1.45$2, and indistinguishability $1.45$3. In the unified microscopic comparison of three off-resonant excitation schemes at $1.45$4 K, the following values were reported:

Scheme	$1.45$5	$1.45$6 / $1.45$7
Long dichromatic	0.4948	0.4619 / 0.5850
Short dichromatic	0.8519	0.0074 / 0.9889
NARP	0.9302	0.0036 / 0.9929
SUPER	0.9890	0.0012 / 0.9987

These values show that, among the off-resonant schemes considered, SUPER gives the highest efficiency, the lowest multiphoton probability, and the highest indistinguishability (Crowder et al., 31 Oct 2025).

The microscopic explanation centers on phonon-induced dephasing and incoherent excitation. In the weak-phonon master-equation treatment, the time-dependent rates are

$1.45$8

$1.45$9

with phonon spectral density

$\ket{g}$0

where $\ket{g}$1 and $\ket{g}$2 (Crowder et al., 31 Oct 2025). For SUPER, the effective Rabi frequencies are so large that the phonon spectral density is exponentially suppressed at those frequencies, giving phonon rates on the order of $\ket{g}$3 GHz. In that regime, phonon-induced decoherence is essentially negligible (Crowder et al., 31 Oct 2025).

This contrasts sharply with symmetrically detuned dichromatic excitation. Earlier work showed that in an ideal two-level system, symmetric dichromatic pulses with no spectral overlap suffer from cancellation of accumulated pulse area, so finite inversion requires either phonon-assisted breaking of the cancellation or asymmetry between the two colors (Koong et al., 2020). In the later 2025 comparison, long dichromatic pulses were found to suffer phonon dephasing rates above $\ket{g}$4 GHz over long durations, leading to maximum excited-state population of only about $\ket{g}$5 and efficiency around $\ket{g}$6 (Crowder et al., 31 Oct 2025). SUPER avoids that trade-off because it moves the dynamics into a far-detuned, high-frequency regime where the phonon bath is effectively inert.

The same high single-photon quality was also emphasized in earlier work. The original proposal calculated, for a representative two-color SUPER pulse, single-photon purity $\ket{g}$7, indistinguishability $\ket{g}$8, and photon output $\ket{g}$9, values comparable to or slightly exceeding those of a resonant Gaussian $\ket{e}$0-pulse in the same model (Bracht et al., 2021). An experimental demonstration reported a corrected $\ket{e}$1, compared with $\ket{e}$2 under resonant excitation of the same quantum dot, supporting the conclusion that SUPER can yield single-photon purity comparable to resonant schemes while substantially easing pump rejection (Karli et al., 2022). A later study of coherent dynamics in a trion source measured corrected $\ket{e}$3 under swing-up excitation, with raw values nearly identical to resonant $\ket{e}$4-pulse excitation, while attributing the lower indistinguishability in that particular sample to environment-induced spectral broadening at the high powers required by SUPER rather than to the protocol itself (Boos et al., 2022).

5. Quantized interpretation, cavity-QED extensions, and collective-state preparation

A fully quantized treatment changed the conceptual interpretation of SUPER. Instead of treating the fields as classical drives, the two-color excitation was modeled as a two-mode Jaynes–Cummings interaction with pulsed couplings. In that picture, the total excitation number

$\ket{e}$5

is conserved, and the dominant SUPER process is not a one-photon absorption but a multi-photon redistribution between modes (Vannucci et al., 2024). Under optimal conditions, the simplest process is

$\ket{e}$6

meaning that the emitter is excited, mode 1 loses two photons, and mode 2 gains one photon (Vannucci et al., 2024). More generally,

$\ket{e}$7

so SUPER is a multi-mode, multi-photon coherent scattering process rather than a direct one-photon transition (Vannucci et al., 2024). A related few-photon analysis in an extended Jaynes–Cummings model derived the two-photon resonance condition

$\ket{e}$8

for the simplest few-photon channel, identifying the quantized analogue of the classical SUPER resonance (Richter et al., 2024). This suggests that the dressed-state interpretation and the multi-photon picture are complementary rather than contradictory.

SUPER has also been extended to cavity-QED systems. In a four-level quantum dot coupled to two orthogonal cavity modes, it was shown that the strong red-detuned pulses induce an AC-Stark shift that moves the exciton transition out of cavity resonance during the excitation window. This suppresses photon emission during the pulse, eliminates re-excitation, and yields almost ideal single-photon purity even in cavities approaching the strong-coupling regime (Heinisch et al., 2023). In the single-photon configuration, cavity population remained negligible until after the exciton was fully prepared, and the resulting emission probability was slightly below one but essentially flat versus cavity loss, while the purity remained essentially one across the tested range (Heinisch et al., 2023).

The same cavity-assisted decoupling mechanism underlies biexciton preparation and entangled-photon-pair generation. In a cavity tuned to half the biexciton energy, SUPER was shown to produce almost perfect biexciton initialization and concurrence values close to the ideal initially excited biexciton case; in a later analysis this was sharpened to concurrence up to $\ket{e}$9 and essentially temperature-independent values above $\omega_0$0 up to about $\omega_0$1 K, because the excitation and emission stages are dynamically decoupled by the Stark shift (Bracht et al., 2023).

Beyond single emitters, SUPER has been generalized to the preparation of collective eigenstates in two dipole-coupled emitters at deep-subwavelength spacing. There, the relative optical phase between the fields on the two emitters selects the symmetry sector: $\omega_0$2 addresses the symmetric superradiant state and $\omega_0$3 the antisymmetric subradiant state. With suitable pulse parameters, final populations above $\omega_0$4 were obtained for the superradiant state and above $\omega_0$5 for the subradiant state, with corresponding cavity-emission antibunching values as low as $\omega_0$6 to $\omega_0$7 for the singly excited collective states (Kerber et al., 27 Aug 2025). This suggests that the swing-up principle extends naturally from single-emitter inversion to selective control of bright, dark, and hybrid collective excitations.

6. Misconceptions, limitations, and broader significance

A common misconception is that SUPER is simply resonant excitation in disguise, implemented through sidebands or weak spectral leakage. That interpretation applies only to certain frequency-modulated parameter regimes where a first sideband can become resonant. The original theoretical work explicitly identified a regime in which full inversion occurs although the first sideband remains detuned from the transition, demonstrating a genuinely off-resonant swing-up mechanism (Bracht et al., 2021). Similarly, the later two-pulse formulation and dressed-state analysis show that the essential resonance is in the dressed basis, not in the bare transition spectrum (Bracht et al., 2022, Crowder et al., 31 Oct 2025).

A second misconception is that all off-resonant excitation schemes are equivalent once pump–emission spectral overlap is eliminated. The 2025 comparison shows otherwise: long dichromatic excitation can lose up to about $\omega_0$8 of performance because widening the spectral separation requires larger instantaneous amplitudes that increase phonon-induced decoherence, whereas NARP and SUPER remain shielded from phonons by different mechanisms (Crowder et al., 31 Oct 2025). SUPER’s advantage is specifically tied to very large effective Rabi frequencies and far detunings that move the relevant dynamics beyond the phonon cutoff (Crowder et al., 31 Oct 2025).

The principal limitation of SUPER is not photon quality but control precision and, in some platforms, the high optical power required. In the single-photon-source comparison, efficiency was highly sensitive to small variations in second-pulse amplitude and detuning, even though coherence and indistinguishability remained robust (Crowder et al., 31 Oct 2025). In semiconductor samples, the required high powers can also perturb the local environment and reduce indistinguishability through spectral diffusion or heating, as observed in one trion experiment where the protocol itself retained nearly perfect purity but the measured Hong–Ou–Mandel visibility was limited by sample-specific noise processes (Boos et al., 2022). This suggests that the practical competitiveness of SUPER depends strongly on device architecture, pulse stability, and environmental engineering.

Broader applications are plausible but not universal. The multi-photon and multi-mode nature of the mechanism suggests relevance to few-photon frequency conversion and non-linear quantum photonics (Vannucci et al., 2024, Richter et al., 2024). The off-resonant, spectrally separated operation is attractive for solid-state emitters where resonant pumping is hindered by background scattering, and explicit analyses have already extended SUPER to WSe$\omega_0$9 emitters, where it improved exciton preparation fidelity from about $$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$0 under resonant excitation to $$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$1 for one emitter type and $$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$2 for another in a phonon-decoupling regime (Vannucci et al., 2024). In that material system, near-resonant phonon-assisted excitation still reached even higher preparation fidelity, up to $$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$3 or $$H_{\rm S} = \hbar \omega_0 \sigma^+ \sigma^- + \frac{\hbar \Omega(t)}{2}(\sigma^+ + \sigma^-),$$4 depending on the emitter, so SUPER is not uniformly optimal across all metrics and platforms (Vannucci et al., 2024). A plausible implication is that SUPER is best viewed not as a universal replacement for resonant or dissipative protocols, but as a distinct control primitive whose strengths are spectral separation, coherence, and compatibility with high-fidelity photon generation under suitable parameter control.

Taken together, the literature presents SUPER as a coherent off-resonant excitation framework that has evolved from a dressed-state inversion mechanism in a single two-level emitter into a broader family of control protocols spanning single-photon sources, cavity-QED architectures, quantized multi-photon scattering, and collective-state engineering. Its defining feature remains the same across these realizations: population inversion is achieved not by direct resonance with the bare transition, but by using interference between detuned fields to engineer the emitter’s dressed dynamics in a way that swings the population upward while preserving spectral separation between pump and emission (Bracht et al., 2021, Crowder et al., 31 Oct 2025).