Realization of Two-dimensional Discrete Time Crystals with Anisotropic Heisenberg Coupling

By combining IBM quantum processors with advanced tensor network methods, this study demonstrates the existence of a two-dimensional discrete time crystal in anisotropic Heisenberg systems, revealing a rich phase diagram that bridges the gap between simplified models and natural quantum interactions.

The Gist

Imagine you have a giant, complex dance floor filled with 144 dancers (the quantum bits, or "qubits"). In the world of physics, we usually expect that if you keep shaking this dance floor with a rhythmic beat, the dancers will eventually get tired, stop dancing in sync, and just move randomly. This random state is called "thermalization," and it's like the system forgetting its original choreography and turning into a hot, messy soup.

However, this paper describes a special kind of dance called a Discrete Time Crystal (DTC). In this state, the dancers refuse to forget their steps. Even though the music (the "drive") repeats every single beat, the dancers only change their formation every two beats. They are breaking the rhythm of the music to create their own, longer rhythm. This is a rare phenomenon where a system stays "out of equilibrium" and keeps its quantum memory alive, defying the usual laws that say everything should eventually settle down.

The New Twist: A 2D Dance Floor

Previous experiments with these time crystals were like watching a single line of dancers (one-dimensional). They were easy to simulate on regular computers, but they didn't look much like the complex, interconnected systems we see in nature.

This team took the experiment to a two-dimensional dance floor. They arranged the 144 dancers in a specific, honeycomb-like pattern on a real quantum computer (IBM's "ibm fez"). Instead of just simple "flip" interactions (like the old Ising model), they introduced a more complex, "Heisenberg" interaction. Think of this as allowing the dancers to not just flip forward and backward, but also spin and interact with their neighbors in multiple directions at once. This is much closer to how real magnetic materials work in nature.

The Experiment: Chaos vs. Order

The researchers wanted to see if this complex 2D dance could still hold its rhythm, or if the extra complexity would cause the dancers to immediately fall into chaos (thermalization).

They tested two different starting positions for the dancers:

1. The Checkerboard Start (Néel State): Imagine the dancers starting in a perfect alternating pattern (Up, Down, Up, Down).
2. The All-Up Start (Polarized State): Imagine every single dancer starting facing the same direction.

What they found:

• The Checkerboard Start: When they started with the alternating pattern, the dancers struggled to keep the rhythm. The "time crystal" rhythm faded away quickly. The system seemed to be fighting against the complexity of the 2D connections and the "spin-flip" interactions, eventually losing its memory of the start.
• The All-Up Start: Surprisingly, when they started with everyone facing the same way, the dancers held the rhythm incredibly well. Even with the complex interactions, they maintained a stable, repeating pattern that lasted much longer. The paper compares this to "quantum scars"—a fancy way of saying the system found a special, protected path through the chaos that allowed it to keep dancing in sync, almost like a ghost of a perfect memory that refuses to fade.

The Tools: Real Hardware and "Noise Cleaning"

Running this on a real quantum computer is tricky because these machines are noisy. It's like trying to record a symphony in a room where the wind is howling and people are shouting. The signals get distorted.

To solve this, the team used a clever trick. They ran the experiment on smaller groups of dancers (3x3 and 2x2 grids) to measure exactly how much the "noise" was messing things up. They then used this data to mathematically "clean" the results from the larger 144-dancer grid. It's like recording the wind noise in a small room and then using a computer to subtract that exact wind noise from the recording of the big hall, revealing the true music underneath.

They also used powerful classical computer simulations (using "tensor networks," which are like advanced maps of how the dancers are connected) to double-check that their cleaned-up data was actually showing a real time crystal and not just a glitch.

The Big Picture

The paper concludes that:

1. Time crystals can exist in 2D: Even with the complex, messy interactions found in nature (Heisenberg coupling), these systems can maintain a stable, rhythmic order.
2. It depends on how you start: The stability of this "time crystal" depends heavily on the initial state of the system. Some starting positions are fragile and break down, while others (like the fully aligned state) are surprisingly robust.
3. New Physics: This discovery shows that there are special "protected" states in driven quantum systems that can resist the usual tendency to turn into thermal chaos. This helps us understand how quantum systems can bridge the gap between perfect quantum order and the messy thermodynamics of the real world.

In short, the researchers successfully built a 2D quantum dance floor, proved that a specific type of complex interaction can still support a "time crystal" rhythm, and discovered that the secret to keeping the dance going lies in how you start the performance.

Technical Summary

Technical Summary: Realization of Two-dimensional Discrete Time Crystals with Anisotropic Heisenberg Coupling

Problem Statement
Discrete time crystals (DTCs) represent a non-equilibrium phase of matter characterized by the spontaneous breaking of discrete time-translation symmetry. While DTCs have been extensively studied in one-dimensional systems with Ising-like couplings, their existence in two-dimensional (2D) systems with interactions that more closely resemble natural physical phenomena remains an open question. Specifically, the stability of DTCs in 2D is theoretically challenged by the potential absence of many-body localization (MBL) in higher dimensions and the difficulty of simulating such complex quantum systems classically. Furthermore, previous experimental realizations have largely been restricted to simplified Ising models, leaving the behavior of DTCs under anisotropic Heisenberg interactions—common in systems ranging from single-molecule magnets to quantum dot architectures—unexplored.

Methodology
The authors combine state-of-the-art quantum hardware with advanced classical simulation techniques to investigate a 2D driven quantum system.

• Quantum Hardware Implementation: The experiment was conducted on IBM's 156-qubit Heron r2 processor (ibm_fez). The researchers utilized a 144-qubit subset arranged in a "decorated" (heavy) hexagonal lattice. The system evolution was governed by a Floquet unitary operator $U_F$, composed of three sub-cycle layers of disordered XXZ coupling gates ($U^{(k)}_{XXZ}$) and a periodic transverse kick ($U_X$). The coupling strength included uniform disorder to induce localization, while a tunable spin-flip coupling parameter ($\epsilon$) was introduced to study its effect on the phase.
• Classical Simulations: To validate hardware results and mitigate noise, the team employed tensor network methods. They utilized Matrix Product States (MPS) for 1D comparisons and two-dimensional tensor network states (2dTNS) with a topology matching the heavy hexagonal lattice to preserve interaction locality.
• Noise Mitigation: A renormalization technique was applied to recover noiseless values from hardware data. This involved extracting noise parameters from smaller subsystems (e.g., $3 \times 3$) and applying them to larger configurations ($3 \times 7$) to correct for depolarization and amplitude damping.
• Observables: The study analyzed time-correlation functions, spatial spin-spin correlations, Hamming distance distributions, and Quantum Fisher Information (QFI) to distinguish between ergodic, spin-glass, and time-crystalline phases.

Key Results

1. Existence of 2D DTCs: The study demonstrates the existence of a discrete time crystal phase in a 2D system governed by anisotropic Heisenberg interactions. The system exhibits persistent, period-doubled oscillations in spin dynamics, a hallmark of DTCs, despite the theoretical challenges of MBL in two dimensions.
2. Phase Diagram: A rich phase diagram was constructed, revealing transitions between three distinct regimes:
   • Spin-Glass Phase: Characterized by localized order.
   • Discrete Time Crystal (DTC) Phase: Characterized by stable subharmonic response and long-range spatial order.
   • Ergodic Phase: A featureless phase where thermalization occurs.
     The boundaries of these phases are tunable via the X-gate rotation angle ($\phi$) and the spin-flip coupling strength ($\epsilon$).
3. Initial State Dependence: A striking contrast was observed based on the initial state.
   • Néel State: Oscillations originating from the Néel state decayed rapidly under strong spin-flip coupling.
   • Fully Polarized State: In contrast, the fully polarized state demonstrated robust, stable subharmonic oscillations that persisted even for strong coupling ($\epsilon \sim 1$). This behavior resembles "quantum many-body scars" and "cat-scar DTCs," suggesting a mechanism where the system avoids thermalization through stable evolution trajectories that prevent eigenstate mixing.
4. Scaling and Stability: Analysis of order parameters across different system sizes ($2 \times 2$ vs. $3 \times 7$) indicated improved stability in larger systems, supporting the existence of a genuine phase transition rather than a finite-size artifact. QFI analysis confirmed slow entanglement growth in the localized and DTC phases, contrasting with the rapid growth observed in the ergodic phase.

Significance and Claims
The paper claims to present the first experimental investigation of a 2D discrete time crystal with anisotropic Heisenberg coupling. By successfully extending the study of Floquet matter beyond simplified Ising models, the work establishes that many-body localization mechanisms can provide sufficient protection against thermalization in 2D driven systems, allowing for stable subharmonic responses.

Crucially, the findings highlight the interplay between initialization, interaction anisotropy, and driving protocols in stabilizing the DTC phase. The observation of scar-like behavior in the polarized state offers a new perspective on weak ergodicity breaking, potentially connecting time crystals to quantum many-body scars and prethermal phases. The authors posit that these results lay the groundwork for exploring how driven systems bridge the gap between quantum coherence and emergent non-equilibrium thermodynamics, demonstrating that exotic phases of matter can be explored on current noisy quantum hardware through scalable noise renormalization methods.