Dimer Adsorption on (100) Terraces With First- and Second-Neighbor Interactions$^{1}$

Adsorbed particles that binds to two nearest-neighbor sites, such as CO/Ni(100), are simulated as dimers covering infinitely long (100)-terraces of finite width $M$, with first- and second-neighbor interaction energies, $V$ and $W$. The set {\{}coverage, number of first-neighbors per site, number of second-neighbors per sites{\}}, and the entropy characterize the adsorption system. For attractive first-neighbors ($V>$0) there are two series of phase diagrams, for $M$ even and odd, which coincide in the infinite-$M$ limit, with one non-trivial phase {\{}$\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $, $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 4$} $, 0{\}} for $W/V \quad <$ -- $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $. For repulsive first-neighbors, the phase diagrams are obtained for $M \quad \le $ 7. In the infinite-$M$ limit, the non-trivial phases are, {\{}1/3, 0, 0{\}}, {\{}$\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $, $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 4$} $, 0{\}}, {\{}$\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $, 0, $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} ${\}}, and {\{}2/3, 1/3, 1{\}}. Past computations, which neglected second-neighbor interactions and considered V$<$0, found only two phases {\{}$\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} $, 0, $\raise.5ex\hbox{$\scriptstyle 1$}\kern-.1em/ \kern-.15em\lower.25ex\hbox{$\scriptstyle 2$} ${\}}, and {\{}2/3, 1/3, 1{\}}. Here, these results are recovered, and in the infinite-$M$ limit, Monte Carlo simulation and finite-size scaling are used to obtain the heat capacity and the critical temperature of the order-disorder transitions as a function of $W/V$. $^{1}$Work supported by NSF and the PSC (AP and DG), and by CONICET and the Fulbright Foundation(MP).