← Latest papers
🔬 materials science

Multi-reference variational states resolve stripe selection and local dd-wave pairing in the Hubbard model

Using multi-reference neural variational auxiliary-field quantum Monte Carlo simulations on the two-dimensional Hubbard model, this study demonstrates that while half-filled antiphase stripe walls intrinsically host strong local dd-wave pairing, suppressing static stripe order alone does not yield long-range superconducting coherence across the lattice.

Original author: Honghui Shang, Jinlong Yang

Published 2026-11-13
📖 4 min read☕ Coffee break read

Original author: Honghui Shang, Jinlong Yang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the author. For technical accuracy, refer to the original paper. Read full disclaimer

High-temperature superconductivity remains one of the most enduring mysteries in modern physics. For decades, scientists have studied a class of copper-based materials called cuprates, which can conduct electricity without any resistance at temperatures far higher than previously thought possible. A central puzzle in understanding these materials is how electrons, which naturally repel each other, manage to pair up and flow freely. In the simplest models used to describe these materials, electrons move on a flat, two-dimensional grid. Within this grid, electrons sometimes organize themselves into parallel lines of charge and spin, known as stripes. For years, researchers have debated whether these stripes compete with superconductivity, blocking the flow of current, or if they actually help create it by providing a scaffold for electron pairs to form. The answer has been difficult to pin down because the energy differences between these competing states are incredibly small, and the behavior changes subtly depending on how many electrons are added or removed from the material.

To solve this, a team of researchers at the University of Science and Technology of China developed a new computational method to simulate the behavior of electrons on these grids with unprecedented precision. They focused on a specific condition where the material is doped with a precise amount of missing electrons, a state known to be critical for superconductivity. Using a powerful technique that combines artificial intelligence with quantum mechanics, they simulated a grid of up to one thousand sites, a size large enough to capture the complex interactions that smaller simulations miss. Their goal was to determine whether the stripes that form in the material are filled with electrons or only half-full, and to see if these stripes support the formation of electron pairs that could lead to superconductivity.

The researchers found that the answer depends heavily on the specific details of how electrons hop between atoms. When they adjusted the simulation to match the properties of real cuprate materials, the electrons rearranged themselves into half-filled stripes rather than the fully filled stripes predicted by simpler models. In these half-filled stripes, the electrons are more spread out, allowing them to move more freely across the boundaries between magnetic regions. This movement lowers the energy of the system, making the half-filled configuration the stable state. Crucially, the team discovered that within these half-filled stripes, electrons naturally form strong, short-range pairs. These pairs have a specific shape, known as d-wave pairing, which is the same shape required for high-temperature superconductivity. The formation of these pairs is an intrinsic property of the half-filled stripes themselves; they appear even when the stripes are locked in place and do not fluctuate.

However, the story does not end with the formation of these pairs. While the electrons pair up strongly within a single stripe, the researchers found that these pairs do not connect coherently across different stripes. The correlation between pairs on one stripe and pairs on a neighboring stripe fades away very quickly, disappearing within the distance of just one stripe spacing. This means that while the material is excellent at creating local pairs, it struggles to synchronize them across the entire grid to create a macroscopic superconducting state. The study suggests that the suppression of superconductivity in certain conditions is not because the pairs fail to form, but because they fail to link up over long distances.

The team also explored what happens when the stripes are allowed to move or fluctuate, a scenario that some theories suggested might help synchronize the pairs. They found that allowing the stripes to shift their position did not significantly change the strength of the local pairing, nor did it create a long-range connection between them. The local pairs remained robust, but the long-range order remained elusive. This result challenges the idea that the mere presence of fluctuating stripes is enough to explain the superconductivity seen in experiments. Instead, it points to a more complex picture where the local pairing is a natural consequence of the electronic structure, but the global superconducting state requires additional factors not captured in this single-layer model.

These findings provide a clearer microscopic view of the competition between stripes and superconductivity. The simulations show that the material naturally favors a state where electrons form strong local pairs on half-filled stripes, but this state alone does not guarantee a superconducting current flowing through the entire material. The inability of the pairs to communicate across the stripes suggests that other mechanisms, perhaps involving interactions between different layers of the material or more complex atomic structures, are necessary to unlock the full potential of high-temperature superconductivity. By resolving the nature of these local pairs and the limits of their connection, the study offers a solid foundation for future research to understand why some cuprates become excellent superconductors while others remain insulators.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →