In this work, the problem of adaptive stabilization by means of output feedback of an alternating activated sludge process is solved. The goal is to control the global nitrogen concentration at the outlet of the settler, with partial state measurement and unknown parameters. To solve this problem, an adaptive control law is designed by means of output feedback. The employed approach consists of the asymptotic reconstruction of a full state feedback stabilizing control law, without using a Lyapunov function and by applying a separation principle. Firstly, a full information feedback linearizing controller is considered and then, a dynamic output feedback (estimating the unmeasured state variables and unknown parameters) is synthesized, which asymptotically recovers the full information control law. The performance of the controlled system is showed through digital simulations.