Fig. 5: Finite-size scaling results for the long-time average of the infinite-temperature correlation function \(\overline{C(t)}\), averaged over a time window of Jt ∈ [50, 103], and then averaged over Ns ∈ [20, 1,024] disorder realizations depending on system size. | Nature Physics

Fig. 5: Finite-size scaling results for the long-time average of the infinite-temperature correlation function \(\overline{C(t)}\), averaged over a time window of Jt ∈ [50, 103], and then averaged over Ns ∈ [20, 1,024] disorder realizations depending on system size.

From: Unravelling quantum dynamics using flow equations

Fig. 5

a–d, Results for various system sizes, disorder types and strengths, and dimensionalities: a, 1D, random; b, 1D, QP; c, 2D, random; d, 2D, QP. Error bars indicate the variance over disorder realizations. The black squares indicate the results obtained from exact diagonalization, shown for small system sizes only. System sizes are L = 8, 10, 12, 16, 24, 36, 48, 64 and 100 in one dimension, and L2 = 9, 16, 25, 36, 49, 64 and 100 in two dimensions. e, The result of linearly extrapolating to L → ∞. Error bars indicate the uncertainty (variance) in the fits (shown as dashed lines in a–d) to extract the scaling behaviour. Lines are guides to the eye.

Back to article page