Fig. 4: The temperature dependence of \({m}_{{B}_{2g}}^{{B}_{2g}}\) for x = 0.068 (x ≈ x_c) cannot be described by a simple power law.

a \({m}_{{B}_{2g}}^{{B}_{2g}}\) for x = 0.068 (black line), the best Curie–Weiss fit \({m}_{{B}_{2g}}^{{B}_{2g}}\propto \frac{C}{(T-{{\Theta }})}+{m}_{0}\) (red line) and the associated residual (gray line). There is a clear temperature dependence in the residual indicating that Curie–Weiss does not fully describe the temperature evolution. b \({m}_{{B}_{2g}}^{{B}_{2g}}\) for a far underdoped sample x = 0.025 (black line), the best Curie–Weiss fit (red line) and the associated residual (gray line). The data and fit are taken from H.-H. Kuo et al.¹⁸. This sample has a structural transition at 98 K (dashed line). The magnitude of the residual is small compared to the residual shown in panel (a) indicating that Curie–Weiss is a reasonable approximation of the functional form. c Logarithmic plot of \(| {m}_{{B}_{2g}}^{{B}_{2g}}-{m}_{0}|\) vs. temperature for x = 0.068 (x ≈ x_c). No physically motivated value for m₀ linearizes the data, demonstrating that \({m}_{{B}_{2g}}^{{B}_{2g}}\) cannot be described by a power law over the whole temperature range.