An empirical study of an agglomeration network

Recently, researchers have reported many models mimicking real network evolution growth, among which some are based on network aggregation growth. However, until now, relatively few experiments have been reported. Accordingly, in this paper, photomicrographs of real materials (the agglomeration in the filtrate of slurry formed by a GaP-nanoparticle conglomerate dispersed in water) are analyzed within the framework of complex network theory. By datamapping from photomicrographs we generate undirected networks and as a definition of degree we adopt the number of pixel's nearest neighbors while adjacent pixels define a connection or an edge. We study the topological structure of these networks including degree distribution, clustering coefficient and average path length. In addition, we discuss the self-similarity and synchronizability of the networks. We find that the synchronizability of high-concentration agglomeration is better than that of low-concentration agglomeration; we also find that agglomeration networks possess good self-similar features.