Abstract
The scaling properties of the roughness of surfaces grown by two different processes randomly alternating in time are addressed. The duration of each application of the two primary processes is assumed to be independently drawn from given distribution functions. We analytically address processes in which the two primary processes are linear and extend the conclusions to nonlinear processes as well. The growth scaling exponent of the average roughness with the number of applications is found to be determined by the long time tail of the distribution functions. For processes in which both mean application times are finite, the scaling behaviour follows that of the corresponding cyclical process in which the uniform application time of each primary process is given by its mean. If the distribution functions decay with a small enough power law for the mean application times to diverge, the growth exponent is found to depend continuously on this power-law exponent. In contrast, the roughness exponent does not depend on the timing of the applications. The analytical results are supported by numerical simulations of various pairs of primary processes and with different distribution functions. Self-affine surfaces grown by two randomly alternating processes are common in nature (e.g., due to randomly changing weather conditions) and in man-made devices such as rechargeable batteries.