The activated sludge floc size distribution (FSD) is investigated by using different techniques and the results are compared against each other in order to gain insight into the FSD characteristics, as well as to detect the limitations of each measurement technique. The experimental set-up consisted of three devices coupled in series: a MastersizerS, an automated image analysis system (IMAN) and a CIS-100. The latter instrument has two measurement channels, based on time of transition (TOT), and image analysis (SHAPE) principles. In order to minimise the variability between successive measurements, the activated sludge samples collected from a pilot-scale SBR were flocculated until steady state was achieved. The results show that the MastersizerS and SHAPE devices yield similar volume weighted FSD. In contrast, the IMAN overestimated the floc size and TOT frequently showed a bimodal distribution. The number distributions from TOT and SHAPE were in agreement, while those generated by the MastersizerS were mainly located in the submicron range and those of IMAN corresponded to larger sizes. The experimental distributions show a good fit to the log-normal model. It is shown that the measurement principle is of utmost importance and results transformation may lead to data misinterpretation.

This content is only available as a PDF.