Efficient and reversible optical-to-spin conversion for solid-state quantum memories

This paper demonstrates a highly efficient and reversible optical-to-spin conversion process in $^{151}\textrm{Eu}^{3+}:\textrm{Y}_2\textrm{SiO}_5$ using an atomic frequency comb memory, achieving up to 96% efficiency for a 500 $\mu$s storage time by using a magnetic field to prevent Zeeman transition interference.

The Gist

The Quantum "Save Button": A Story of Light, Spin, and Perfect Recalls

Imagine you are a photographer trying to capture a single, fleeting moment—a lightning strike or a hummingbird’s wingbeat. In the world of quantum computing, that "moment" is a photon (a particle of light). These photons carry incredibly important information, but they are notoriously difficult to catch. They move at the speed of light, they are fragile, and if you don't catch them perfectly, the information is lost forever.

To build a "Quantum Internet," we need a Quantum Memory—a way to catch a photon, "save" its information, and then "play it back" later without changing a single detail.

This paper describes a breakthrough in making that "Save" and "Load" process almost perfect.

---

1. The Problem: The "Slippery" Photon

Think of a photon like a high-speed racing car. If you try to stop it with a standard brake (a traditional material), the car crashes, and the driver (the information) is lost.

Scientists use special crystals (in this case, a rare-earth-doped crystal called Europium) to act as a garage. The goal is to take the "speed" of the light and convert it into "spin"—a much slower, more stable form of energy. It’s like taking a speeding car and, instead of crashing it, instantly turning it into a parked car that stays exactly where you left it.

The catch? Most "garages" are leaky. When you try to convert the light to spin, some of the information leaks out. When you try to turn the spin back into light, even more is lost.

2. The Solution: The "Perfect Choreography"

The researchers, Chen and Afzelius, figured out a way to make this conversion reversible and highly efficient. They achieved an efficiency of 96%. In the quantum world, that is like being able to photocopy a document 100 times and having every single copy be identical to the original.

How did they do it? They used two clever tricks:

Trick A: The "Chirped" Pulse (The Smooth Landing)

Instead of hitting the atoms with a sudden, jarring burst of energy (which is like slamming on the brakes), they used "chirped" pulses.

Imagine a pilot landing a plane. If they just drop the nose and slam the wheels down, it’s a disaster. Instead, they gradually change the angle and speed of the descent. The researchers used pulses that "sweep" through different frequencies smoothly. This "adiabatic" approach allows the energy to slide from the light state to the spin state without any "bumps" that would cause information loss.

Trick B: The Magnetic "Organizer" (Clearing the Clutter)

In these crystals, the atoms are a bit messy. They have different "energy levels" that are very close to each other, like a crowded room where everyone is whispering at once. This "noise" makes it hard to hear the specific information you want to save.

The researchers applied a specific magnetic field to act like a strict librarian. The field forced the atoms into very distinct, organized rows. This "lifted the degeneracy" (cleared the clutter), ensuring that when they sent the control pulse, it hit exactly the right "shelf" and nothing else.

3. Why does this matter?

If we want to build a Quantum Repeater—which is essentially a signal booster for a quantum internet—we need to be able to store photons for a long time and retrieve them with near-perfect accuracy.

By proving that we can convert light to spin and back again with 96% efficiency, these scientists have provided a blueprint for a much more reliable quantum network. They’ve essentially moved us from a "blurry, low-quality recording" to a "high-definition, lossless save file" for the most delicate information in the universe.

Technical Summary

Technical Summary: Efficient and Reversible Optical-to-Spin Conversion for Solid-State Quantum Memories

Authors: Jingjing Chen and Mikael Afzelius (University of Geneva)
Subject: Quantum Information / Solid-State Physics

---

1. The Problem: Efficiency and Reversibility in Quantum Memories

Quantum repeaters and networks require quantum memories capable of long-duration storage and high efficiency. In Atomic Frequency Comb (AFC) memory schemes, the standard approach is to map short-lived optical coherence (photons) into long-lived spin coherence (atomic spins).

However, achieving near-unity reversibility—the ability to map the state from the optical domain to the spin domain and back again without loss—is extremely challenging. Two primary obstacles exist:

• Transition Dipole Moments: In crystals like $^{151}\text{Eu}^{3+}:\text{Y}_2\text{SiO}_5$, the transition dipole moments are very low, making it difficult to drive population inversion across the entire required bandwidth.
• Zeeman Degeneracy: At low magnetic fields, near-degenerate Zeeman transitions cause time-domain interference and complex modulations, preventing perfect mapping.
• Spatial/Spectral Inhomogeneity: Gaussian beam profiles and inhomogeneous broadening mean that simple $\pi$-pulses cannot efficiently address all atoms in the crystal.

2. Methodology

The researchers employed a combination of advanced pulse engineering, precise magnetic field control, and 2D numerical modeling.

• Material & Field Configuration: They used $^{151}\text{Eu}^{3+}:\text{Y}_2\text{SiO}_5$ under a moderate magnetic field of 231 mT. This specific field strength lifts the Zeeman degeneracy, ensuring that each addressed transition is a single, spectrally resolvable magnetic sublevel.
• Pulse Engineering (HSH Pulses): To overcome weak transition moments and wide bandwidths, they utilized Hyperbolic-Square-Hyperbolic (HSH) pulses. These are frequency-chirped, adiabatic pulses with smooth amplitude envelopes. They are designed to be "adiabatic," meaning they can achieve near-perfect population inversion across a broad bandwidth, provided the pulse is sufficiently long.
• Numerical Modeling: The authors developed a 2-dimensional Bloch-equation model. This model accounts for:
  • Spatial effects: The Gaussian intensity distribution and the overlap of input/control modes in both co-linear and crossed-beam ($\theta > 0$) configurations.
  • Spectral effects: The frequency chirp of the HSH pulses and the inhomogeneous broadening of both optical and spin transitions.
• Experimental Setup: They implemented a "minimum sequence" consisting of two optical and two spin population inversion pulses to achieve reversible mapping, including a spin-echo sequence to mitigate dephasing.

3. Key Contributions

• Optimization Framework: The development of a simulation tool that predicts how control mode waist, beam angle, and pulse duration affect efficiency.
• Lifting Degeneracy Strategy: Demonstrating that a moderate magnetic field (rather than near-zero field) is superior for achieving reversible, interference-free mapping.
• Adiabatic Control: Proving that frequency-chirped HSH pulses are the optimal solution for addressing the low-dipole-moment transitions in Europium-doped crystals.

4. Results

• Record Efficiency: The study achieved a total back-and-forth optical-to-spin conversion efficiency ($\eta_{\text{opt-spin}}$) of up to $(96 \pm 1)\%$.
• Robustness to Detuning: The efficiency remained high across the chirp bandwidth, confirming the adiabatic nature of the pulses.
• Angle Dependence: In crossed-beam configurations (used to reduce noise), the efficiency remained above 75% for angles up to $1^\circ$. The model successfully predicted that larger control beam waists are necessary to maintain overlap at higher angles.
• Storage Time: The high efficiency was maintained for storage times of 500 µs, with the model accurately predicting the decay due to spin dephasing.

5. Significance

This work provides a blueprint for high-performance solid-state quantum memories. By reaching near-unity conversion efficiency, the authors have removed a major bottleneck in the AFC protocol.

Implications include:

• Scalability: The methods are applicable to other non-Kramers systems, such as Praseodymium-doped crystals.
• Quantum Networking: The ability to store single-photon states with high signal-to-noise ratios at millisecond timescales is a critical requirement for implementing quantum repeaters.
• Future Directions: The paper suggests that combining these optimized pulses with cavity-enhanced or multi-pass configurations could push total memory efficiency ($\eta_{\text{tot}}$) significantly higher, moving closer to the requirements for practical quantum communication.