Preparing 100-qubit symmetry-protected topological order on a digital quantum computer

Imagine you are trying to build a massive, intricate castle out of LEGO bricks. But there's a catch: you can only use a very specific, fragile type of glue, and if you try to build the castle too tall or too complex, the whole thing collapses under its own weight. This is the current state of quantum computers. They are powerful, but they are "noisy" and fragile; if you ask them to perform too many steps (a deep circuit) to create a complex state, the noise ruins the result before the job is done.

This paper is about a team of scientists who figured out how to build a 100-brick-high castle (a 100-qubit quantum system) that was previously thought to be too difficult to construct without it falling apart.

Here is the story of how they did it, broken down into simple concepts.

1. The Goal: Building a "Magic" Castle

The scientists wanted to create a specific type of quantum state called a Symmetry-Protected Topological (SPT) phase.

The Analogy: Think of a normal magnet. If you cut it in half, you get two smaller magnets. But an SPT phase is like a special kind of "magic" rope. If you cut the rope in the middle, the middle looks normal, but the ends of the rope suddenly start glowing with a special energy that the middle doesn't have.
Why it matters: These "glowing ends" (edge states) are incredibly stable. They don't get messed up easily by the environment. This makes them perfect for future quantum computers that need to store memory without losing data.

The challenge? To see these glowing ends, you need a rope that is 100 units long. Previous attempts could only manage ropes of about 20–40 units before the noise of the computer drowned out the signal.

2. The Problem: The "Too-Deep" Ladder

Usually, to build a quantum state, you have to run a sequence of instructions (a circuit).

The Old Way: Imagine trying to climb a ladder to reach the top of a 100-story building. If the ladder is too tall (too many steps), you get tired and fall off before you reach the top. In quantum terms, the "ladder" is the circuit depth. The longer the ladder, the more likely the computer makes a mistake.
The Limitation: To build a 100-unit SPT state the old way, the ladder would need to be thousands of steps high. Current quantum computers can't climb that high without falling.

3. The Solution: The "Smart Blueprint" (AQC)

The team used a clever trick called Approximate Quantum Compiling (AQC).

The Analogy: Instead of trying to build the castle brick-by-brick from scratch (which takes forever and is prone to errors), they first used a super-powerful classical computer (a traditional supercomputer) to design a perfect, compressed blueprint of the castle.
The Compression: They realized that even though the castle is 100 bricks long, the pattern of the bricks repeats in a simple way. They compressed this pattern into a very short, efficient set of instructions.
The Result: Instead of a ladder with 1,000 steps, they built a ladder with only 18 to 39 steps. It's like taking a shortcut through a tunnel instead of climbing the mountain.

They used a technique called Tensor Networks (think of it as a mathematical "folding" trick) to figure out exactly how to fold the instructions so they fit into a shallow ladder.

4. The Experiment: Testing the Magic Rope

They took their new, short-ladder instructions and ran them on a real quantum computer (an IBM machine named "Pittsburgh").

The Test: They checked three things to see if they successfully built the "magic rope":
1. The String Order: They checked if the "glowing ends" were connected to the middle in a hidden way. It's like checking if the two ends of a rope are tied together invisibly, even though the middle looks loose. They found the connection was strong, even across 20 units of the rope.
2. The Entanglement Spectrum: They looked at the "fingerprint" of the quantum state. In a perfect SPT phase, the fingerprint has a specific "double-decker" pattern (degeneracy). Their results showed this pattern clearly.
3. The Edge Modes: They looked at the very ends of the 100-unit chain. Just like the theory predicted, the ends had a special, stable magnetic property that the middle didn't have.

5. The Outcome: A New Era

The team successfully prepared a 100-qubit state with 98% to 99% accuracy.

Why this is a big deal: Before this, the best anyone had done was about 80 qubits, but with very simple patterns (low complexity). This team managed 100 qubits with complex, realistic patterns.
The Metaphor: If previous experiments were like building a small, simple sandcastle, this is like building a massive, detailed sandcastle that survived a storm.

Summary

The scientists used a smart compression algorithm to turn a massive, complex quantum problem into a short, manageable set of instructions. This allowed them to build a 100-unit quantum system on current hardware, proving that we can now study these exotic "magic" states at a scale that was previously impossible.

This opens the door to:

Better Quantum Memory: Using these stable "glowing ends" to store data.
Simulating New Materials: Understanding how real-world materials (like the nanographene mentioned in the paper) behave at the quantum level.
Future Dynamics: Watching how these systems react when you shake them up (non-equilibrium dynamics), which is something classical computers can't do anymore.

In short: They found a shortcut to build a quantum tower that was previously too tall to reach, and the view from the top is full of new physics.

Here is a detailed technical summary of the paper "Preparing 100-qubit symmetry-protected topological order on a digital quantum computer."

1. Problem Statement

Symmetry-Protected Topological (SPT) phases represent a class of quantum matter that extends the Landau paradigm, characterized by non-local order parameters and robust edge states rather than local symmetry breaking. While SPT phases have been realized in analog simulators (cold atoms, Rydberg systems, trapped ions) and specific material systems (nanographene), digital quantum computers offer unique advantages: programmability, arbitrary Hamiltonian preparation, and the ability to measure non-local observables like string order.

However, a significant bottleneck exists in preparing SPT ground states on digital hardware:

Circuit Depth: Traditional methods for preparing highly entangled ground states often require deep circuits, which exceed the coherence times of current Noisy Intermediate-Scale Quantum (NISQ) devices.
Fidelity vs. Scale: Previous demonstrations were limited to small system sizes or low entanglement (low bond dimension). Preparing large systems (e.g., 100 sites) with high fidelity and verifying their topological nature remains an open challenge.
Model Specificity: Many prior works focused on specific models (like the AKLT state) or required complex interactions. There is a need to demonstrate SPT preparation for standard spin-1/2 models with only two-body exchange interactions, which are more relevant to engineered spin materials.

2. Methodology

The authors propose a hybrid classical-quantum workflow to prepare and verify 100-site ground states of the Spin-1/2 Bond-Alternating Heisenberg Chain (BAHC) on IBM superconducting quantum processors.

A. The Physical Model

The target system is the 1D BAHC Hamiltonian:
$\hat{H} = \frac{1}{4} \sum_{i=0}^{N-2} J_i (\hat{X}_i \hat{X}_{i+1} + \hat{Y}_i \hat{Y}_{i+1} + \hat{Z}_i \hat{Z}_{i+1})$
where couplings alternate between $J_0$ (even sites) and $J_1$ (odd sites). The phase diagram includes:

Even-Haldane Phase: $J_0 > 0, J_0 > J_1$ (Trivial SPT).
Odd-Haldane Phase: $J_1 > 0, J_1 > J_0$ (Non-trivial SPT with edge modes).
Ferromagnetic Phase: $J_0, J_1 < 0$ .

The study focuses on four specific points in the phase diagram: $O_{1/2}$ (Odd-Haldane), and $E_{1/2}, E_{-1}, E_{-2}$ (Even-Haldane).

B. Classical Pre-processing (DMRG)

Ground State Calculation: The authors use the Density Matrix Renormalization Group (DMRG) algorithm (via the TeNPy library) to compute the Matrix Product State (MPS) representation of the ground states.
Compression: To reduce computational overhead for the next step, the MPSs are compressed to maintain at least 99.9% fidelity with the original state. This reduces the bond dimension ( $\chi$ ) from up to 40 down to 5–8.

C. Approximate Quantum Compilation (AQC)

Instead of using deep, exact circuit decomposition methods (like staircase circuits), the authors employ AQC-Tensor, a tensor-network-based approximate quantum compiling protocol.

Ansatz: A "brickwork" circuit ansatz is used, consisting of layers of nearest-neighbor two-qubit unitaries. This is chosen because the ground states of gapped 1D systems exhibit exponentially decaying correlations, which brickwork circuits capture efficiently.
Optimization: The parameters of the circuit ( $\vec{\theta}$ ) are classically optimized to maximize the fidelity $|\langle \psi_{target} | \hat{U}(\vec{\theta}) | 0 \rangle|^2$ between the circuit output and the compressed MPS target.
Initialization: The optimization is initialized with a state corresponding to the limit where one coupling is zero (a product of singlets), significantly improving convergence compared to random initialization.

D. Experimental Execution

Hardware: Circuits were executed on the IBM Pittsburgh quantum processor (127 qubits).
Error Mitigation:
- Pauli Twirling & TREX: Applied to mitigate readout and gate errors.
- Zero-Noise Extrapolation (ZNE): Used for string order measurements to extrapolate results to the zero-noise limit.
- Validation: "Identity circuits" were run to verify that the chosen noise factors and qubit layout were suitable for ZNE.

3. Key Contributions

Scale: Successful preparation of 100-site ground states, a significant increase from previous digital quantum simulations (e.g., 80 sites with low bond dimension).
Efficiency: Achievement of high-fidelity states using shallow circuits (18–39 CNOT depth), making them executable on current NISQ hardware.
Algorithmic Advantage: Demonstration that AQC with a brickwork ansatz outperforms traditional "exact ladder" or "approximate ladder" methods for these specific states, which would otherwise require prohibitively deep circuits (thousands of CNOTs).
Comprehensive Verification: Simultaneous observation of three distinct SPT signatures on hardware, providing robust verification beyond simple state fidelity.

4. Results

The team prepared and verified four ground states ( $O_{1/2}, E_{1/2}, E_{-1}, E_{-2}$ ) with the following results:

A. Circuit Performance

Fidelity: The compiled circuits achieved 97.9% – 99.0% fidelity relative to the uncompressed DMRG ground states.
Depth: CNOT depths ranged from 18 to 39, depending on the correlation length of the specific state (e.g., $E_{-2}$ with the longest correlation length required the deepest circuit).

B. Experimental Signatures of SPT Order

Non-Local String Order:
- Measured string order parameters ( $S_l$ ) for lengths up to $l=20$ .
- Result: In the Even-Haldane phase, the even string order remained non-zero at $l=20$ , while the odd string order decayed to zero (and vice versa for the Odd-Haldane phase). This confirms long-range SPT order.
Entanglement Spectrum Degeneracy:
- Measured the entanglement spectrum of reduced density matrices for cuts of length $l=1$ to $6$.
- Result: Observed the characteristic alternating even/odd degeneracy of eigenvalues. For example, in the Even-Haldane phase, cuts across $J_0$ bonds showed even degeneracy (two dominant eigenvalues), while $J_1$ cuts showed odd degeneracy. This is a robust fingerprint of symmetry protection.
Edge Modes (Odd-Haldane Phase only):
- Measured single-site magnetization near the chain boundaries.
- Result: Observed exponentially localized edge spins with a net magnetization of $\approx 1/2$ at the boundaries, decaying into the bulk. The extracted correlation length ( $\xi \approx 1.8$ ) matched theoretical predictions.

5. Significance and Future Outlook

Platform Validation: This work establishes current quantum computers as versatile platforms for large-scale studies of symmetry-protected quantum matter, moving beyond small, minimally entangled examples.
Foundation for Dynamics: The ability to prepare high-fidelity, large-scale SPT ground states is a critical prerequisite for studying non-equilibrium quench dynamics in regimes where classical simulation fails due to entanglement growth.
Quantum Information Applications: These states serve as resources for measurement-based quantum computation (MBQC) and topologically protected spin transport.
Limitations & Future Work: The method is currently best suited for 1D gapped systems with short-range correlations. The authors note that states with long-range entanglement or gapless Hamiltonians may require adaptive circuits or different ansätze. Future directions include probing excitation spectra, simulating defects, and extending these methods to 2D geometries.

In summary, the paper represents a major milestone in digital quantum simulation, successfully bridging the gap between theoretical SPT models and experimental realization on noisy hardware through the innovative use of tensor-network-based approximate compilation.