Neural network modeling of organics removal by activated carbon cloths

The adsorption of organic compounds onto an activated carbon cloth is studi ed in a dynamic reactor. An experimental design is carried out to investiga te the influence of operating conditions (initial concentration C-o, flow v elocity U-o, and bed thickness Z) and adsorbate's characteristics. A slow i ntraparticular diffusion is shown by flattened breakthrough curves, and ads orption capacities are high and range between 50 and 250 mg g(-1). The tran sfer zone Z(o), assessed by the Adams and Bohart equation, is low (3 mm). A ll experimental results are modeled by a neural network to take into accoun t the specific diffusion of cloths. Parameters related to the adsorbate-ads orbent affinity in a batch reactor are consequently introduced in the input layer of the neural network (intraparticular coefficient K-w and Freundlic h parameters K-f and Un), added to operating conditions whose influence was shown (C-o, U-o, and Z) and time t. The statistical quality of the neural network modeling is high (r(2) = 0.956). Furthermore, the Garson connection weight method allows the relative influence of input neurons to be determi ned. This analysis confirms the influence of parameters relative to adsorba nt-adsorbate affinity.