Digital programming of spin correlations in a fermionic lattice quantum simulator

This paper presents a hybrid analog-digital approach for a fermionic lattice quantum simulator that combines adiabatic preparation with digital collisional gates to engineer and measure target states with specific, long-range spin correlations, such as those found in Heisenberg chains.

The Gist

Imagine you are trying to build a complex, intricate sculpture out of tiny, invisible magnetic blocks. These blocks are atoms, and the way they stick together (or repel each other) creates patterns called "spin correlations." Scientists have long been able to let these blocks settle into natural patterns on their own, like sand dunes forming in the wind. However, they couldn't easily design specific, complex patterns from scratch, especially ones that require the blocks to "talk" to each other across long distances.

This paper describes a new "hybrid" method that combines two different ways of working with these atoms to build these specific patterns. Think of it as a two-step recipe: Analog Preparation (getting the raw materials ready) and Digital Programming (sculpting the final shape).

Step 1: The Analog Prep (The "Raw Dough")

First, the scientists take a cloud of atoms (specifically Potassium-40) and cool them down until they act like a single, unified quantum fluid. They trap these atoms in a grid of laser light, which acts like a series of tiny, one-dimensional tubes.

• The Goal: They want to create pairs of atoms that are perfectly linked, like dance partners holding hands. In physics, these are called "singlets."
• The Process: They use magnetic tricks to encourage atoms to pair up. However, the process isn't perfect; some spots have two pairs, some have one, and some have none.
• The Cleanup: To fix this, they use a "molecular shield." They turn the perfect pairs into molecules that are invisible to a specific color of light. Then, they blast the system with that light. The "lonely" atoms (the ones that didn't pair up) get hit by the light and kicked out of the system, while the perfect pairs remain safe.
• The Result: They are left with a clean line of "chained singlets." Imagine a row of couples holding hands: (Partner A - Partner B) - (Partner C - Partner D). This is their starting resource.

Step 2: The Digital Programming (The "Sculpting")

Now that they have their clean line of couples, they want to rearrange them to create a specific, complex pattern that nature wouldn't naturally form. This is where the "digital" part comes in.

• The Moving Walkway: The scientists use a technique called "topological pumping." Imagine a moving walkway at an airport that can slide atoms left or right without breaking their hand-holds. This allows them to move atoms to new positions without messing up their quantum connection.
• The Collision Gates: Once atoms are in the right spot, they let them "collide" in a controlled way. Think of this as a choreographed bump. When two atoms bump into each other, their internal magnetic spins swap or change in a precise way.
• The Programming: By moving atoms and making them bump in a specific sequence, they can "program" the system. They can take the initial pattern (A-B) - (C-D) and rearrange it into a new pattern where the connections are different, like (A-C) - (B-D), or even create long-range connections where the first atom is linked to the last one, skipping the middle ones.

The Proof: Checking the Work

How do they know they succeeded? They can't just look at the atoms with a microscope. Instead, they use a clever trick:

1. Rearrange: They move the atoms back to specific spots.
2. The Test: They apply a magnetic field that makes the atoms oscillate (wiggle) between being a "singlet" (holding hands) and a "triplet" (standing apart).
3. The Measurement: By watching how much they wiggle, they can calculate exactly how strongly the atoms were connected before the test.

They tested this by creating a pattern that mimics a "Heisenberg chain" (a famous model in physics). They showed that they could take their initial "chained" state and digitally transform it into a state that is over 99% identical to the perfect theoretical target.

Why This Matters

The paper claims this is a breakthrough because:

• Control: It moves beyond just waiting for atoms to settle naturally. It allows scientists to deterministically (reliably) create specific quantum states.
• Scalability: They proved this works on small chains of four atoms, but the method is designed to be scaled up to larger systems.
• Hybrid Power: It combines the best of both worlds: the stability of analog preparation (getting the raw materials ready) and the precision of digital gates (sculpting the final detail).

In short, the researchers built a machine that can take a messy pile of quantum particles, clean them up, and then use a digital "remote control" to arrange them into a specific, highly complex pattern that didn't exist before. This opens the door to studying materials and phenomena that are currently too complex for even the best supercomputers to simulate.

Technical Summary

Technical Summary: Digital Programming of Spin Correlations in a Fermionic Lattice Quantum Simulator

Problem Statement
The deterministic preparation of strongly correlated quantum states remains a central challenge in quantum simulation. While analog methods allow correlations to develop through Hamiltonian evolution from thermal ensembles or uncorrelated product states, they are limited by finite temperatures and closing energy gaps. Conversely, fully digital approaches offer programmable control but currently lack the necessary connectivity and coherence for fermionic systems on existing platforms. Consequently, the deterministic and programmable preparation of strongly correlated fermionic states, particularly those with specific long-range spin correlations, remains an open challenge.

Methodology
The authors present a hybrid analog-digital approach to engineer target states with specific spin-correlation patterns. The protocol proceeds in three distinct stages:

1. Analog State Preparation (Resource State):

   • The experiment utilizes a two-component degenerate Fermi gas ($^{40}$K) loaded into a three-dimensional dynamical optical superlattice, forming an ensemble of independent one-dimensional tubes.
   • Attractive interactions are applied during loading to form "doublons" (atoms occupying the same site with total spin $S=0$).
   • A purification step removes singly occupied sites ("singlons") by converting doublons into Feshbach molecules (which are transparent to resonant light) and selectively removing the singlons.
   • The doublons are adiabatically split into atomic singlets ($|\uparrow\downarrow\rangle - |\downarrow\uparrow\rangle)/\sqrt{2}$ by ramping a short-wavelength lattice. This yields a resource state of "chained singlets" ($|CS\rangle$), where approximately 82.2% of atoms are in singlet pairs, with over 12% forming four-particle chains.

2. Digital Programming (Gate Operations):

   • The system employs a combination of spatial reconfiguration and collisional gates.
   • Spatial Reconfiguration: Topological pumping (Thouless pumping) is used for bidirectional, state-independent atom shuttling. This allows atoms to be moved to distinct lattice sites without destroying existing entanglement.
   • Collisional Gates: Tunable partial swap gates are implemented via atomic collisions within the lattice. These gates are governed by a time-dependent Fermi-Hubbard Hamiltonian parametrized by dimer tunneling ($t$), energy offset ($\Delta$), and on-site interaction ($U$).
   • The gate operation is described by a unitary $\hat{U}(\alpha)$, where the gate angle $\alpha$ is continuously tunable via the Hubbard $U$ through a Feshbach resonance. The system operates in the direct-exchange regime ($|U| \lesssim t$), where the dynamical phase exhibits a first-order dependence on $U$, offering robustness against tunneling fluctuations compared to the superexchange regime.
   • By composing shuttling sequences with these tunable gates, the authors program a deterministic digital circuit that transforms local correlations into complex, non-local entanglement structures.

3. Measurement:

   • Spin pair-correlators $\langle C_{ij} \rangle = \langle S^z_i S^z_j \rangle 4/\hbar^2$ are measured using a readout circuit.
   • Atoms are rearranged via topological pumping to bring target pairs to neighboring sites.
   • A symmetry-selective detection procedure is employed: atoms in a spin-singlet configuration co-occupy the same energy level, while triplet configurations populate different bands. A conditional spin-flip operation acts exclusively on the singlet subspace, allowing the measurement of the singlet fraction via coherent singlet-triplet oscillations (STOs).

Key Results

• Non-Local Correlation Propagation: The authors demonstrated the ability to propagate correlations. Starting from a chained-singlet state ($|CS\rangle = |\text{singlet}_{12}\rangle \otimes |\text{singlet}_{34}\rangle$), they applied a swap gate to the inner atoms (2 and 3) to create a propagated state ($|P\rangle = |\text{singlet}_{13}\rangle \otimes |\text{singlet}_{24}\rangle$). Measurements confirmed the shift of the singlet signal from nearest-neighbor ($d=1$) to next-nearest-neighbor ($d=2$) correlations.
• Target State Preparation: The hybrid architecture was used to prepare a target state ($|T\rangle$) approximating the ground state of a four-fermion Heisenberg chain. By varying the gate exponent $\alpha$ in a four-gate quantum circuit, the experimental correlators matched theoretical predictions with no free parameters.
• Ground State Overlap: For the four-particle system, the initial chained singlet state had an overlap of $\sim 90\%$ with the Heisenberg ground state. The digital evolution improved this overlap to $>99\%$ with a specific gate sequence ($\alpha = 1/8$).
• Scalability and Robustness: Theoretical analysis for larger systems (longer Heisenberg chains) indicates that the analog-digital approach significantly improves ground state overlap compared to the resource state alone. The protocol remains robust against gate errors up to $10^{-2}$.

Significance and Claims
The paper claims to demonstrate a versatile framework for exploring strongly correlated many-body states characterized by non-local correlations. By separating bulk Hamiltonian evolution (analog) from gate-based digital refinement, the method enables the preparation of states beyond the ground or eigenstates of the underlying Hamiltonian.

The authors state that this approach paves the way for:

• The targeted preparation of strongly correlated states of matter.
• The measurement of higher-order correlations and non-local order parameters, including topological order and superconducting pairing correlations.
• Addressing connectivity and scalability challenges in fermionic quantum simulation, offering a pathway between specialized quantum simulation and universal fermionic quantum computation.

The work establishes a method to deterministically configure many-body states, transforming initial local correlations into complex, non-local entanglement structures using a hybrid analog-digital protocol in a fermionic optical lattice.