The experiment was carried out to evaluate the effects of processing (whole or ground) and including coffee hulls in elephant grass silage on the effluent production and composition. A randomized complete design in a 2 × 5 factorial design with two processes (whole or ground) and five coffee hull levels (0; 6; 12; 18 and 24 %) each with three replications. A split-plot design was used to assess the effluent composition, where the main plot consisted of a 2 × 3 factorial with two processes and three coffee hull levels (0, 6, 12%) and the split-plot consisted of the effluent collection days, with three replications. The total effluent production was 243.1 L/t silage for silage without hulls, 196.8; 93.2; 30.8; 3.1 and 149.1; 52.3; 30.8; 0.0 L/t for the silages with whole and ground hulls at levels of 6, 12, 18 and 24%, respectively. The hull processing and inclusion levels had a quadratic effect, according to the days, on the total solids and total nitrogen contents. A quadratic effect of the processing and days and the hull levels and days, respectively, on the composition in calcium and magnesium. The hull levels and the coffee hull processing had a linear effect on the phosphorus content in the effluent. There were effects of processing, hull levels and of the days on the contents of chemical and biochemical demands for oxygen. Coffee hulls were efficient in reducing effluent production in the elephantgrass silage, especially when ground, because they had more effect than when whole. The effluent composition is influenced by the day of collection and inclusion and processing of coffee hull levels.
absorbent additive; ensilage; environmental pollution; losses; processing
