Continuous phase transitions in diagrammatic Monte Carlo

In the diagrammatic Monte Carlo approach, properties of a strongly correlated system are represented by the series of Feynman diagrams in the thermodynamic limit. We demonstrate that the location of a continuous phase transition can be found from the series coefficients for diverging susceptibility in the normal state. We obtain the coefficients to high order such that they contain sufficient information to locate the transition by using the known critical behaviour, for a given universality class, to extract the location of the singularity in an accurate way. We demonstrate this by mapping the Néel transition of the 3d Hubbard model across a range of dopings and temperatures.