Urea-rubber matrices as slow-release fertilizers

The efficacy of a prototype slow-release fertilizer, urea-rubber matrix (URM) was assessed under flooded rice conditions. URM cuboids of size 0.5 × 1.0 × 0.4 cm were applied in comparison with prilled urea at levels of 50, 130 and 200 kg N ha^?1. URM was placed at the soil/solution interface in intimate association with rice seedlings whereas prilled urea was broadcast into the floodwater to simulate the normal application method of South East Asian farmers. URM cuboid sizes of 0.25 × 0.5 × 0.4 cm, 0.5 × 0.5 × 0.4 cm and 1.0 × 1.0 × 0.4 cm were similarly evaluated at a single rate of 130 kg N ha^?1; a broadcast URM treatment was also included. Different methods of prilled urea application including deep placement and split application were also studied at a single rate of 130 kg N ha^?1. It was found that the build-up of floodwater N (urea + NH ₄ ⁺ ) from URM during the 13 days following application was almost negligible irrespective of level or method of application. This was thought to result in low losses of N through ammonia volatilization as shown by higher rice grain yields in comparison with prilled urea at all levels of application. Deep-placed urea gave a comparable grain yield to that of broadcast URM. There was no significant difference in grain yield between URM applied by placement and by broadcast, suggesting that URM can be effectively applied either in intimate association with rice seedlings or by broadcasting to the rice field before, or after, planting. An attempt to predict the release of urea from URM was made using a diffusion-based simulation model. It was found that the model underestimated the actual release of urea from URM within the rhizosphere, probably due to the extensive penetration by rice roots of the URM cuboids.