Enhanced Rydberg Blockade through RF-tuned Förster Resonance

Enhancing interactions between Rydberg atoms is a key challenge in contemporary quantum technologies. Stronger interactions enable faster Rydberg gates in digital processors and larger entangled states in analog simulation. Achieving the same interaction strength at lower principal quantum number addresses current constraints in available Rabi frequency and field sensitivity in large scale tweezer or cavity QED experiments. Here, we demonstrate a new technique using AC Stark shifts from a microwave drive to tune into a Förster resonance, thereby modifying the interaction scaling with distance from $1/R^6$ to $1/R^3$. We validate enhanced Rydberg interactions (in strength and range) by probing cavity Rydberg polariton blockade at $n=44$ in $^{87}$Rb, improving from $g^{(2)}(0) = 1.0 (1)$ in the Van-der-Waals regime to $g^{(2)}(0) = 0.38 (1)$ in the dipolar regime on the Förster resonance. Importantly, our technique allows minimal shifts of the original Rydberg state, suppressing detuning errors in gate protocols while maintaining quadratic insensitivity to DC electric fields.

💡 Research Summary

The authors present a novel method to dramatically enhance the interaction strength between Rydberg atoms by exploiting an RF‑driven AC Stark shift to bring a pair of Rydberg states into a Förster resonance. In conventional Rydberg experiments, two atoms prepared in the same Rydberg level interact only via a second‑order van‑der‑Waals (vdW) potential that scales as 1/R⁶ because the dipole‑dipole operator is odd under exchange and thus vanishes for identical states. When the energy defect ΔF between the target pair state |dd⟩ and an intermediate pair state |pf⟩ is tuned to zero, the two‑atom system hybridizes into symmetric and antisymmetric superpositions, yielding a resonant dipole‑dipole interaction that scales as 1/R³.

Previous work achieved ΔF=0 by applying static electric fields (DC Stark tuning). However, the DC approach shifts all Rydberg levels, making the system highly sensitive to background field fluctuations and introducing large detuning errors for the target state. The present work circumvents these limitations by applying a microwave (RF) field that is near‑resonant with a transition of the |p⟩ component of the |pf⟩ manifold but far detuned from the |d⟩ (target) and |f⟩ states. This selective AC Stark shift moves the |pf⟩ manifold while leaving |dd⟩ essentially unchanged, allowing precise control of ΔF without compromising the target transition frequency.

The experimental platform consists of a high‑finesse (F≈1200) optical cavity at 780 nm, loaded with ~1.5 × 10³ ultracold ⁸⁷Rb atoms. A two‑photon ladder scheme (780 nm probe + 480 nm control) creates electromagnetically induced transparency (EIT) and generates dark Rydberg polaritons—collective excitations that inherit the Rydberg component’s strong interactions. The target Rydberg state is the 44D₅/₂ (m_J=5/2) level. The chosen intermediate pair state is a superposition of (n+2)P₃/₂ and (n‑2)F₇/₂, denoted |pf⟩. For n=44 the natural Förster defect is ΔF≈65 MHz, far from resonance.

Using the ARC atomic‑physics library, the authors compute AC Stark shifts for the relevant states as a function of microwave frequency and polarization. They find that π‑polarized microwaves at 23.7 GHz produce a shift of the P‑state that is roughly 50 times larger than the shift of the D‑state, while avoiding vector (Zeeman‑like) shifts. This “magic” frequency yields a ratio of desired to undesired shifts that minimizes perturbation of the target transition, a crucial requirement for high‑fidelity quantum gates.

To locate the exact resonance experimentally, the authors monitor the mean‑field nonlinearity of the transmitted cavity probe light. By driving the cavity with a photon flux that yields on average three polaritons, they observe that as the microwave frequency approaches the Förster condition the blockade radius expands, suppressing simultaneous polariton excitation and reducing cavity transmission. Simultaneously, the EIT transmission spectrum exhibits a dip in transmission and a peak in linewidth at the resonance, providing a fast diagnostic for ΔF=0.

Once the resonance is identified, the authors perform a full two‑photon correlation measurement. In the vdW regime (no microwave), the second‑order correlation function at zero delay is g²(0)=1.0±0.1, indicating Poissonian statistics and negligible blockade. When the RF‑tuned Förster resonance is engaged, g²(0) drops to 0.38±0.01, a clear signature of strong photon blockade mediated by the enhanced dipole‑dipole interaction. This demonstrates that the interaction has switched from a short‑range 1/R⁶ to a longer‑range 1/R³ character, even at the relatively low principal quantum number n=44.

The authors highlight three key advantages of the RF‑AC‑Stark technique over DC Stark tuning: (1) Minimal shift of the target Rydberg level, reducing detuning errors in gate protocols; (2) Quadratic suppression of sensitivity to stray DC electric fields, because the static field can be nulled (E≈0) and only the microwave field contributes to the shift, leading to a residual dependence ∝(δE)²; (3) Rapid, bidirectional control of the interaction strength simply by adjusting microwave power or frequency, enabling sub‑microsecond switching suitable for dynamic quantum circuits.

By achieving strong dipolar interactions at lower n, the method alleviates two major experimental constraints: the need for high laser power (since transition dipole moments scale as n⁻³/²) and the heightened susceptibility to electric‑field noise (polarizability scales as n⁷). This opens a pathway toward faster Rydberg gates and larger entangled states in neutral‑atom arrays, tweezer‑based platforms, and cavity QED systems without the overhead of high‑n operation.

Future directions suggested include extending the scheme to multiple Förster channels simultaneously, applying it to two‑dimensional or three‑dimensional atom arrays for scalable quantum simulation, and engineering shaped microwave pulses to achieve time‑dependent interaction profiles. Such developments could further integrate RF‑tuned Förster resonances into error‑corrected quantum processors, analog quantum simulators of long‑range spin models, and hybrid light‑matter interfaces where controlled photon‑photon interactions are essential.