A tunable solid-on-solid model of surface growth

TL;DR

This paper presents a Monte Carlo study of a tunable solid-on-solid surface growth model, revealing complex scaling behaviors and the necessity of multiple exponents to describe surface roughness.

Contribution

It introduces a parameterized model that captures diverse surface roughness regimes and demonstrates the need for more than two exponents for characterization.

Findings

Surface becomes spikier with negative or <0.

Surface width scales as or =0.

Different exponents indicate complex, possibly self-constraining surface behavior.

Abstract

We have performed a detailed Monte Carlo study of a diffusionless $(1+1)$-dimensional solid-on-solid model of particle deposition and evaporation that not only tunes the roughness of an equilibrium surface but also demonstrates the need for more than two exponents to characterize it. The tunable parameter, denoted by $\mu$, in this model is the dimensionless surface tension per unit length. For $\mu < 0$, the surface becomes increasingly spikier and its average width grows linearly with time; for $\mu = 0$, its width grows as $\sqrt{t}$. On the other hand, for positive $\mu$, the surface width shows the standard scaling behavior, $\la \sigma_m(t)\ra \sim M^{\alpha}f(t/M^{\alpha /\beta})$ where $M$ is the substrate size and $f(x) \to const (x^{\beta})$ for $x$ large (small). The roughness exponent, $\alpha = 1/2$ for $\mu \leq 2$, and $ = 3/5, 4/5 & \sim 1$ for $\mu = 5, 6 & 7$ respectively; the growth exponent, $\beta = 1/4$ for $\mu \leq 2$ and $= 1/2$ for $\mu > \sim 3.5$ respectively. These exponents are different from those of the height-difference correlation function,$\alpha ' = 1/2, \beta ' = 1/4$ and $z' = 2$, for higher values of $\mu$ suggesting thereby that the surface could be self-constraining.