|
Su-Chan Park (The Catholic University of Korea) We study the critical behavior of the one-dimensional pair contact process with diffusion (PCPD) by extensive Monte Carlo simulations. We first investigate the critical decay of macroscopic quantities such as particle density and pair density. Developing a method to estimate the strength of a leading correction to scaling from the ratio of the density and the pair density, we estimate the exponent \chi of the leading correction to scaling as \chi = 0.37 \pm 0.01. With this \chi, we study the effective exponent of the critical density decay, to conclude that \delta = 0.185 \pm 0.010 which is different from the critical decay exponent of the directed percolation (DP), 0.1595. We then estimate other exponents from a finite size scaling (FSS) study and from microscopic absorbing phase transitions. Taking into account the special feature of the PCPD that the dynamics of isolated particles and pairs are different, we introduce three ensembles respectively called C-ensemble, D-ensemble, and G-ensemble and investigate critical behavior within each ensemble scheme. Our study on the FSS cannot give a conclusive value of the dynamic exponent z, while our study on microscopic absorbing phase transitions shows that there are two different sets of critical exponents; \eta = 0.285\pm 0.01, \delta' =0.145\pm 0.01, z= 1.639\pm 0.014 within the D-ensemble scheme, and \eta = -0.05 \pm 0.05, \delta' =0.45 \pm 0.05, z= 1.69 \pm 0.06 within the G-ensemble scheme. Although the C-ensemble yields another set of critical exponents, we argue that the C-ensemble contain misleading information regarding the critical behavior of the PCPD. These results are compared with a DP model with PCPD-like dynamics which gives known DP exponents \eta=0.313, \delta'=0.1595, and z = 1.5809 for all ensembles except the misleading C-ensemble. |