Electrical conductivity of a monolayer produced by random sequential adsorption of linear $k$-mers onto a square lattice

Phys. Rev. E 94, 042112 (2016) The electrical conductivity of a monolayer produced by the random sequential adsorption (RSA) of linear $k$-mers onto a square lattice was studied by means of computer simulation. Overlapping with pre-deposited $k$-mers and detachment from the surface were forbidden. The RSA continued until the saturation jamming limit, $p_j$. The isotropic and anisotropic depositions for two different models: of an insulating substrate and conducting $k$-mers (C-model) and of a conducting substrate and insulating $k$-mers (I-model) were examined. The Frank-Lobb algorithm was applied to calculate the electrical conductivity in both the $x$ and $y$ directions for different lengths ($k=1$ -- $128 $) and concentrations ($p=0$ -- $p_j$) of the $k$-mers. The `intrinsic electrical conductivity' and concentration dependence of the relative electrical conductivity $\Sigma (p)$ ($\Sigma=\sigma/ \sigma_m$ for the C-model and $\Sigma=\sigma_m /\sigma$ for the I-model, where $\sigma_m$ is the electrical conductivity of substrate) in different directions were analyzed. At large values of $k$ the $\Sigma (p)$ curves became very similar and they almost coincided at $k=128$. Moreover, for both models the greater the length of the $k$-mers the smoother the functions $\Sigma_{xy}(p)$, $\Sigma_{x}(p)$ and $\Sigma_{y}(p)$. For the C-model, the other interesting findings are: for large values of $k$ ($k=64, 128$), the values of $\Sigma_{xy}$ and $\Sigma_{y}$ increase rapidly with the initial increase of $p$ from 0 to 0.1; for $k \geq 16$, all the $\Sigma_{xy}(p)$ and $\Sigma_{x}(p)$ curves intersect with each other at the same iso-conductivity points; for anisotropic deposition, the percolation concentrations are the same in the $x$ and $y$ directions, whereas, at the percolation point the greater the length of the $k$-mers the larger the anisotropy of the electrical conductivity, i.e., the ratio $\sigma_y/\sigma_x$ ($>1$).