In this paper, we developed a general cluster Monte Carlo algorithm where the transition probability is split into a selection and a flip part. The algorithm uses an adjustable parameter which is introduced to maintain a balance between these two parts. This algorithm is applied to the Blume-Emery-Griffiths model and permits us to generalize the mixed cluster algorithm (MCA) [Phys. Rev. B 54, 359 (1996)]. Using our general approach, we were able to explore regions of the parameter space which are not covered by the MCA. Furthermore, it can be used in the presence of a magnetic field.

• Received 13 June 2003

DOI:https://doi.org/10.1103/PhysRevB.68.064113