diff --git a/examples/07_Dipole_Dipole.jl b/examples/07_Dipole_Dipole.jl index f56d46c9c..14020872f 100644 --- a/examples/07_Dipole_Dipole.jl +++ b/examples/07_Dipole_Dipole.jl @@ -46,7 +46,7 @@ minimize_energy!(sys_lr) plot_spins(sys_lr; ghost_radius=8, color=[:red, :blue, :yellow, :purple]) # Copy this configuration to the other two systems. Note that the original `sys` -# has a continuum of degenerate ground states. +# has a _continuum_ of degenerate ground states. sys.dipoles .= sys_lr.dipoles sys_lr_cut.dipoles .= sys_lr.dipoles; @@ -67,11 +67,12 @@ swt = SpinWaveTheory(sys_lr; measure) res2 = intensities_bands(swt, path) swt = SpinWaveTheory(sys_lr_cut; measure) -res3 = intensities_bands(swt, path) +res3 = intensities_bands(swt, path); # Create a panel corresponding to Fig. 2 of [Del Maestro and -# Gingras](https://arxiv.org/abs/cond-mat/0403494). That previous work -# underreported the energy scale by a factor of two, and requires slight +# Gingras](https://arxiv.org/abs/cond-mat/0403494). Dashed lines show the effect +# of truncating dipole-dipole interactions at 5 Å. The Del Maestro and Gingras +# paper underreported the energy scale by a factor of two, and requires slight # corrections to its third dispersion band. fig = Figure(size=(768, 300))