Filtration Characterization Method as Tool to Assess Membrane Bioreactor Sludge Filterability—The Delft Experience

Abstract

Prevention and removal of fouling is often the most energy intensive process in Membrane Bioreactors (MBRs), responsible for 40% to 50% of the total specific energy consumed in submerged MBRs. In the past decade, methods were developed to quantify and qualify fouling, aiming to support optimization in MBR operation. Therefore, there is a need for an evaluation of the lessons learned and how to proceed. In this article, five different methods for measuring MBR activated sludge filterability and critical flux are described, commented and evaluated. Both parameters characterize the fouling potential in full-scale MBRs. The article focuses on the Delft Filtration Characterization method (DFCm) as a convenient tool to characterize sludge properties, namely on data processing, accuracy, reproducibility, reliability, and applicability, defining the boundaries of the DFCm. Significant progress was made concerning fouling measurements in particular by using straight forward approaches focusing on the applicability of the obtained results. Nevertheless, a fouling measurement method is still to be defined which is capable of being unequivocal, concerning the fouling parameters definitions; practical and simple, in terms of set-up and operation; broad and useful, in terms of obtained results. A step forward would be the standardization of the aforementioned method to assess the sludge filtration quality.

Impact of suspended solids concentration on sludge filterability in full-scale membrane bioreactors

“…Table 3 shows the statistics of the cake layer parameters, namely the α R × c i product and s coefficient. As explained in our previous research the DFCm produces a hardly compressible cake layer (Lousada-Ferreira et al, 2014). In the DFCm the compressibility coefficient results are mainly between 0 and 0.3 (Lousada-Ferreira, 2011, Krzeminski, 2013, while theoretically it varies between 0 and 1, indicating no compression to total compression, respectively.…”

“…In the DFCm the compressibility coefficient results are mainly between 0 and 0.3 (Lousada-Ferreira, 2011, Krzeminski, 2013, while theoretically it varies between 0 and 1, indicating no compression to total compression, respectively. In the DFCm the cake layer mass and specific cake resistance are the main contributors of the total measured resistance (Lousada-Ferreira et al, 2014). The non-diluted samples' results in Table 3 show that "High MLSS" samples have an average α R × c i which is 5.5 times lower than "Low MLSS" samples.…”

“…The DFCm provides three types of results: ∆R 20 parameter, α R × c i product and s coefficient. The ∆R 20 is the cake layer resistance obtained after extracting 20 L of permeate per membrane area; the α R is the specific cake resistance which is multiplied by c i indicating the concentration of the cake layer particles; the s coefficient is the compressibility coefficient of the cake layer (Geilvoet, 2010, Lousada-Ferreira et al, 2014. The ∆R 20 parameter, contrary to the α R × c i product and s coefficient, does not have a direct physical meaning.…”

“…The filterability was measured according to the Delft Filtration Characterization method (DFCm), defined by Evenblij et al (2005) and further clarified in Lousada-Ferreira et al (2014). The DFCm measures filterability of activated sludge through a single transmembrane (TMP) filtration measurement at constant flux and cross-flow velocity, performed in a side-stream membrane installation with a defined operation and cleaning protocol .…”

“…The DFCm was applied in several MBR installations scattered around Europe (Moreau, 2010, LousadaFerreira, 2011, Krzeminski, 2013. The boundaries of the DFCm were clarified, namely on data processing, accuracy, reproducibility, reliability and applicability, leading to the conclusion that the DFCm was a convenient tool to research how filterability can be influenced by activated sludge characteristics such as MLSS concentrations (Lousada-Ferreira et al, 2014). In our previous work, the membrane tank sludge filterability and MLSS concentrations were measured at four full-scale MBRs scattered around Europe, during weekly campaigns at summer and winter seasons .…”