Geo-engineering is recognized as a promising technique to reduce the release of phosphorus (P) from sediments to water. To investigate the efficiency of lanthanum modified bentonite (LMB), in this study, a microcosm experiment with LMB capping of sediment cores, was performed up to 110 days. High resolution dialysis (HR-Peeper) and diffusive gradients in thin films (DGT) were employed to measure soluble reactive P (SRP) and labile P in overlying water-sediment profiles, respectively, at spatial resolutions from millimeter to submillimeter scale. Results showed that the concentrations of SRP and DGT-labile P decreased with treatment time. A top static layer, with low concentration and small variation (<30% RSD) of SRP and labile P, was formed on the 20th day after capping. The mean concentrations of SRP and DGT-labile P in this layer were stabilized at 5% of those before amendment on the 45th day in the 10 mm top layer, with the immobilization of P in the sediments achieving an ideal phase. Meanwhile, the upward diffusive gradients of SRP and DGT-labile P below the static layer were persistently relaxed, demonstrating a decrease in the release potential of P to the static layer. This was caused by the increasing difficulty of P release from labile solid P pool with time, which was reflected by the increases in distribution coefficient of P between the solid pool and pore water (K-d) and the adsorption rate constant (k(1)). It is suggested that the formation of the static layer is the key to efficiently retain the capping effect of LMB. (C) 2017 Published by Elsevier B.V.