Comment on "effect of polydispersity on the ordering transition of adsorbed self-assembled rigid rods"

The critical behavior of self-assembled rigid rods on a square lattice was recently reinvestigated by Almarza et al. [Phys. Rev. E 82, 061117 (2010)]. Based on the Binder cumulants and the value of the critical exponent of the correlation length, the authors found that the isotropic-nematic phase transition occurring in the system is in the two-dimensional Ising universality class. This conclusion contrasts with that of a previous study [López et al., Phys. Rev. E 80, 040105(R) (2009)] which indicates that the transition at intermediate density belongs to the q=1 Potts universality class. Almarza et al. attributed the discrepancy to the use of the density as the control parameter by López et al. The present work shows that this suggestion is not sufficient, and that the discrepancy arises solely from the use of different statistical ensembles. Finally, the necessity of making corrections to the scaling functions in the canonical ensemble is discussed.