Superconducting valence bond fluid in lightly doped 8-leg t - J cylinders

arXiv:2302.11633 · cond-mat.str-el · Submitted 2023-02-22 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Superconducting valence bond fluid in lightly doped 8-leg t - J cylinders".

Mira: Here's the gist of the paper. The gist The study reports that for lightly doped 8-leg t-J cylinders, ground-state correlations are nearly identical for both “doped quantum spin liquid” and “doped valence bond crystal” states, and upon increasing the cylinder width from 4 to 8, there is a significant strengthening of the quasi-long-range superconducting correlations and a dramatic suppression of any competing charge-density-wave order.

Kai: Model and Method The researchers employed density matrix renormalization group (DMRG) to study the ground state properties of hole-doped t-J models on the square lattice using a Hamiltonian that includes electron hopping terms and spin superexchange interactions between nearest neighbors and next-nearest neighbors <ref:2302.11633#pg5>.

Mira: The simulations were performed on cylinders with width W up to 8, with lengths L greater than W <ref:2302.11633#pg6>.

Kai: They considered doping concentrations of δ = 1/12 and 1/8, using parameters such as t1/J1 = 3 and specific ratios for t2/t1 <ref:2302.11633#pg6>.

Title and authors: Mira: Let's talk about the authors and title for this paper, "Superconducting valence bond fluid in lightly doped eight-leg t - J cylinders <ref:2302.11633#pg1,Superconducting valence bond fluid in lightly doped 8-leg t - J cylinders>."

Kai: The paper is by Hong-Chen Jiang, Steven A. Kivelson, and Dung-Hai Lee. They’re the ones doing this DMRG study on these specific t-J models.

Mira: The title itself points to a state they’re exploring: a superconducting valence bond fluid in lightly doped systems with eight legs for their cylinders.

Lev: From an experimental standpoint, I'm wondering what kind of fidelity we'd need to even begin measuring something like this correlated state in a real system.

Kai: That’s the first thing that comes to mind, Lev. We need a system where we can actually cool and measure these subtle correlations without destroying them during the process.

Mira: The paper is trying to establish if this superconducting behavior is plausible, even though they're dealing with uncertainties about the undoped system being either a quantum spin liquid or a valence bond crystal at these doping levels.

Lev: So, if it’s just a finite cylinder manifestation of something in two dimensions, how does that change the way we apply error correction?

Kai: It means we are looking at a finite size effect, but they are using this to build intuition for what happens in the infinite cylinder limit.

The paper's summary: Lev: So, let's look at the core of the paper. What did they actually report about these correlations?

Kai: They found that despite those uncertainties about the undoped system being a quantum spin liquid or a valence bond crystal, they believe their study provides plausible evidence that upon relatively light doping, the ground state becomes a nodeless d-wave superconductor.

Mira: That’s the big claim: the ground state is nodeless d-wave superconductivity. They support this by noting the isotropy of those SC correlations.

Lev: Isotropy is good for theory, but what about how strong is this correlation? Is it just a weak tendency or something more robust?

Kai: They looked at the power law decay exponent Ksc characterizing this order and found that for width four and six, it’s roughly comparable, but for width eight, it’s roughly a factor of two smaller.

Mira: That suggests that as the system gets wider, the power law decay of those superconducting correlations slows down a bit.

Lev: So what does this tell us about the 2D limit they are aiming for <ref:2302.11633#pg1>? Does it mean Ksc is approaching some specific value?

Kai: They suggest that this observation supports the speculation that the system has a nodeless d-wave superconducting ground state in the two-dimensional limit.

The paper's improvements: Mira: Now, let's talk about what they suggest as improvements or further implications for their picture.

Kai: They point to the lack of evidence of other orders in close competition with superconductivity on width eight cylinders as strong support for the SC being locally stable.

Lev: Stability is important, but stability doesn't mean it’s easy to achieve in practice. What about the charge density waves they mentioned earlier?

Mira: For width four and six, they saw a tendency to form charge stripes similar to earlier work at J2 equal point zero five, with Luttinger exponents Kc values of one point two for width four and one point four for width six.

Lev: And then the jump to width eight where they see that the charge oscillations are much weaker, with an exponent Kc greater than two.

Kai: That means the charge susceptibility doesn't diverge even at zero temperature on those wider systems.

Conclusion: Kai: So, wrapping up this paper on "Superconducting valence bond fluid in lightly doped eight-leg t - J cylinders," it seems they’ve shown that the SC state survives in the two-dimensional limit <ref:2302.11633#pg1,Superconducting valence bond fluid in lightly doped 8-leg t - J cylinders>.

Mira: They argue that hole hopping must generate significant superfluid stiffness without overly disrupting those local singlet correlations of the parent state, which is what supports this nodeless d-wave picture.

Lev: From an engineering view, it gives us a clue that doping can reach this SC state starting from what we might think is a weak valence bond crystal.

Kai: It’s showing that an essentially identical superconducting state can be reached by doping what we suspect is a weak VBC.

Mira: Overall, the main point of the paper is that this superconducting state survives in the two-dimensional limit, and it’s supported by the exponential decay of the equal-time single-particle Green function.

Lev: I think it sets a good benchmark for what we need to see in more complex systems when we try to model real hardware constraints.

Kai: Yeah, yeah. We’ll keep looking at these kinds of cylinder studies as we try to figure out how these materials actually work.

Stanford Institute for Materials and Energy Sciences · Stanford University · University of California, Berkeley · Lawrence Berkeley National Laboratory

cond-mat.str-el

Submitted: 2023-02-22

Updated: 2023-02-22

Comments: 12 pages, 9 figures and 1 table

Journal ref: Phys. Rev. B 108, 054505 (2023)

DOI: 10.1103/PhysRevB.108.054505

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 81/100

The gist: The gist The study reports that for lightly doped 8-leg t-J cylinders, ground-state correlations are nearly identical for both “doped quantum spin liquid” and “doped valence bond crystal”

Key concepts

t-J Model: This is a mathematical model used to describe strongly correlated electron systems, specifically focusing on how electrons hop (kinetic energy) and interact via spin exchange. It's used here to simulate the behavior of materials like high-temperature superconductors.
DMRG: Density Matrix Renormalization Group is a powerful computational technique used to find the exact ground state properties of quantum many-body systems. It allows researchers to calculate how particles are arranged in their lowest energy configuration, which is crucial for understanding material behavior.
d-wave Superconductivity: This describes a specific type of superconducting state where the superconducting order parameter has a wave pattern similar to a d-wave function. The study suggests that this state is nodeless (no nodes), meaning superconductivity exists across the entire 2D material, which is important for its physical properties.

Summary

The gist The study reports that for lightly doped 8-leg t-J cylinders, ground-state correlations are nearly identical for both “doped quantum spin liquid” and “doped valence bond crystal” states, and upon increasing the cylinder width from 4 to 8, there is a significant strengthening of the quasi-long-range superconducting correlations and a dramatic suppression of any competing charge-density-wave order.

Model and Method

The researchers employed density matrix renormalization group (DMRG) to study the ground state properties of hole-doped t-J models on the square lattice using a Hamiltonian that includes electron hopping terms and spin superexchange interactions between nearest neighbors and next-nearest neighbors <ref:2302.11633#pg5>. The simulations were performed on cylinders with width W up to 8, with lengths L greater than W <ref:2302.11633#pg6>. They considered doping concentrations of δ = 1/12 and 1/8, using parameters such as t1/J1 = 3 and specific ratios for t2/t1 <ref:2302.11633#pg6>. The computational resources were substantial, requiring up to m = 60,000 states for W = 8 cylinders.

Superconducting Pair-Field Correlations

The ground state exhibits d-wave SC quasi-long-range order with a high degree of isotropy in the correlation function. The power law decay exponent Ksc characterizing this order is found to be roughly comparable for W = 4 and 6, but is roughly a factor of two smaller for W = 8, suggesting Ksc ≈ 0.6. This observation supports the speculation that the system has a nodeless d-wave superconducting ground-state in the 2D limit. Furthermore, for W = 8, there is no other apparent order in close competition with superconductivity.

Charge Density Wave Correlations

For W = 4 and 6 cylinders, the researchers observed a tendency to form charge stripes similar to earlier work at J2 = 0.5. The charge density oscillations are governed by a Luttinger exponent Kc, which is found to be Kc(W = 4) ≈ 1.2 and Kc(W = 6) ≈ 1.4. Crucially, for W = 8 cylinders, the charge oscillations are much weaker, and the extracted exponent Kc ≥ 2. This indicates that the charge susceptibility does not diverge even at T = 0 on wider systems.

Spin-Spin and Single-Particle Correlations

The magnetic properties are characterized by two correlation lengths, ξAF and ξF, where ξAF is associated with locally commensurate, Q = (π, π) correlations and ξF is associated with locally ferromagnetic correlations. The fact that ξAF is always less than W suggests that in the 2D limit it is finite. The equal-time single-particle Green function G(r) also shows exponential decay G(r) ∼ e −r/ξG, with extracted correlation lengths ξG decreasing with increasing W, suggesting that at this doping ξG is finite in two dimensions.

Dimer-Dimer Correlations and Conclusion

The dimer-dimer correlations fall with a decay exponent Kdimer ≈ Kc > Ksc. The overall results suggest that the SC state is a nodeless d-wave in the 2D limit, supported by the exponential decay of the equal-time single-particle Greens function. The physical picture suggests that hole hopping must generate significant superfluid stiffness without overly disrupting local singlet correlations of the parent state. This finding offers an additional clue in the search for new superconductors by showing that an essentially identical SC state can be reached by doping what is likely a weak VBC. The study concludes that the observed SC state survives in the 2D (W → ∞) limit.

Improvements for AI systems

The AI system can accurately predict whether a lightly doped t-J model in 2D will exhibit a nodeless d-wave superconducting ground state by analyzing cylinder width dependence, noting that Upon increasing the cylinder width from 4 to 8, we observed a significant strengthening of the quasi-long-range superconducting correlations.
The AI can distinguish between quantum spin liquid and valence bond crystal phases in the undoped system based on doping parameters, as it noted that for these values of J2/J1, earlier DMRG studies suggest that the undoped (δ = 0) system at J2/J1 = 0.5 is a quantum spin liquid (QSL), while at J2/J1 = 0.55 it is a valence bond crystal (VBC) in the 2D limit.
The AI can predict the expected decay exponent of superconducting correlations, Ksc, based on cylinder width, noting that "For large W, the exponent Ksc characterizing the power law decay of correlations is expected to decrease as Ksc ∝ 1/W; while we find roughly comparable values of this exponent for W = 4 and 6, it is roughly a factor of 2 smaller for W = 8."
The AI can identify the conditions under which charge density wave order is suppressed, recognizing that the dramatic suppression of any “competing” charge-density-wave order occurs upon increasing cylinder width.
The AI can assess the stability of superconductivity against competing orders, concluding that the SC is locally stable, and has the right properties for being a finite cylinder manifestation of a 2D SC because there is no other apparent order in close competition with superconductivity on W = 8 cylinders.
The AI can predict the divergence of superconducting susceptibility at zero temperature based on correlation lengths, inferring that the SC susceptibility should diverge as χsc ∼ T − (2−Ksc) as T → 0, and hence a smaller value of Ksc implies a stronger divergence.