Comparison of the behaviour of various geotextiles used in the filtration of clayey sludge: An experimental study

2019 
Abstract This paper presents the results of an experimental study of various geotextiles used to filter clayey sludge. The use of geotextiles to filter clayey sludge or suspensions of fine particles in water is more complex than that for filtering suspensions of granular soils. In practice, such applications generally use flocculants to postpone the formation of a low-permeability filter cake. The objective of the present study, which does not use flocculants, is to determine how geotextile characteristics affect the capacity of the geotextile to filter clayey sludge. Three key questions are addressed: (1) What are the main differences between vertical and horizontal filtration? (2) How do geotextile characteristics (nature, opening size, permeability, etc.) affect its capacity to filter clayey sludge (3) How do clayey sludge characteristics (i.e., grain size distribution and concentration)? and the type of flow (i.e., constant head or constant flow) affect the filtering capacity of geotextiles? To evaluate the capacity of a geotextile to filter clayey sludge, we propose three relevant criteria and analyse two filtration phases induced by different cake-formation processes (controlled by the geotextile and controlled by the filter cake). To determine the main differences between vertical and horizontal filtration, the settling of fines in the testing device and its influence on the results are analysed and discussed. This study shows that, for the various clayey sludge tested, the geotextiles (needle-punched nonwoven and thermally bonded nonwoven) with the smallest opening sizes (O 90  ≤ 60 μm) give the most promising results for filtering fines without the use of flocculants. Of these geotextiles, the thermally bonded nonwoven structure seems to offer the best filtration performance for the largest range of fines concentration in the sludge.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    19
    References
    13
    Citations
    NaN
    KQI
    []