[Paper Review] Critical properties of the 2D Z(5) vector model

O. Borisenko, Gennaro Cortese+3 more|arXiv (Cornell University)|Jan 3, 2011

Complex Systems and Time Series Analysis3 references4 citations

TL;DR

This study investigates the critical properties of the two-dimensional Z(5) vector model using a newly developed Monte Carlo cluster algorithm for odd N, determining critical couplings β₁,c ≈ 1.0510(10) and β₂,c ≈ 1.1048(10), and estimating the critical exponent η ≈ 0.225 near the first transition and η ≈ 0.16 near the second, consistent with analytical predictions of η = 1/4 and η = 4/25, respectively, supporting the Berezinskii-Kosterlitz-Thouless phase transition scenario.

ABSTRACT

The two-dimensional Z(5) vector model is investigated through the determination of critical points and one critical index. To this purpose a new cluster algorithm has been developed valid for 2D Z(N) models with odd values of N. Results are compared with analytical predictions.

Motivation & Objective

• To determine the critical couplings β₁,c and β₂,c for the two-phase transitions in the 2D Z(5) vector model.
• To compute critical indices, particularly η, near the critical points to test universality and analytical predictions.
• To develop and validate a novel Monte Carlo cluster algorithm applicable to odd N, specifically N=5, where such algorithms were previously unavailable.
• To confirm the existence of a BKT massless phase between the high-temperature disordered and low-temperature ordered phases.

Proposed method

• A new cluster algorithm was designed for odd N, using spin-dependent bond probabilities based on sine-transformed spin differences to enable efficient updates.
• The algorithm employs a random reference spin n and defines bond activation probabilities depending on the sign of αᵢαⱼ = sin(2π/N(sᵢ−n))sin(2π/N(sⱼ−n)).
• Each cluster is flipped with probability 1/2 via the transformation sᵢ → mod(−sᵢ + 2n + N, N), preserving detailed balance.
• Finite-size scaling (FSS) was applied to susceptibility χᴹ and Binder cumulants Uᴹ, B₄ᴹᴿ, B₄ᵐₚₛ to locate critical points via crossing behavior.
• The effective critical exponent η_eff(R) was computed from spin-spin correlation functions Γ(R) to detect power-law decay in the BKT phase.
• Critical couplings were refined using both polynomial interpolation of cumulant crossings and χ²-based optimization with ν = 1/2.

Experimental results

Research questions

• RQ1 Does the 2D Z(5) model exhibit two distinct phase transitions consistent with the Berezinskii-Kosterlitz-Thouless scenario?
• RQ2 Are the critical exponents η = 1/4 at β₁,c and η = 4/25 at β₂,c numerically confirmed in the Z(5) model?
• RQ3 Can a new cluster algorithm for odd N be effectively implemented and shown to outperform standard heat-bath methods?
• RQ4 Is the critical coupling β₁,c reliably determined, or is finite-size scaling unreliable due to slow convergence?

Key findings

• The critical coupling for the first transition is estimated as β₁,c = 1.0510(10), with finite-size effects suggesting caution in extrapolation.
• The critical coupling for the second transition is β₂,c = 1.1048(10), consistent with infinite-volume extrapolation of pseudocritical points.
• The effective critical exponent η_eff approaches approximately 0.225 near β₁,c, close to the predicted 1/4 = 0.25.
• Near β₂,c, η_eff stabilizes at approximately 0.16, consistent with the analytical prediction of 4/25 = 0.16.
• The observed behavior of |M_L| and the ring-like distribution of complex magnetization confirm the existence of a BKT massless phase.
• The new cluster algorithm demonstrates strong performance advantages over standard heat-bath methods, enabling accurate simulations of odd-N Z(N) models.