Determination of unbound-bound moisture interface of faecal sludges from different on-site sanitation systems.

Non-sewered sanitation (NSS) has been adopted by a significant population, especially in developing countries. For the effective operation of NSS, it is important to ensure optimal treatment of the collected faecal sludge (FS) in on-site sanitation systems. Solid-liquid separation is among the most important steps in the treatment of FS in NSS. The ability of the FS for solid-liquid separation has assumed greater significance in the design and operation of sludge treatment process. Although dewatering indices such as specific resistance to filtration (SRF) and capillary suction time (CST) have been extensively used to determine the limits of sludge dewaterability, these parameters do not provide information on the limits of unbound moisture. The unbound and bound moisture fractions of the FS samples were determined using different methods, and each method was evaluated for reliability. Different FS samples were collected from ventilated improved pit (VIP) latrines, urine diversion dehydrating toilets (UDDT) and septic tanks (ST). Batch settling tests (BST), centrifugation, thermogravimetric analysis (TGA), differential scanning calorimetry (DSC) and water activity (WA) measurements were performed on the samples. While the BST provided the settleability of FS at gravitational force, and centrifugation estimated the relative limit of solid-liquid separability upon application of a force, the boundary of unbound moisture or the interface of unbound and bound moisture fractions as determined by TGA, DSC and WA. Evidently there was a fine overlap of the unbound and bound moisture at the interface due to presence of some tightly held unbound moisture and loosely held bound moisture. The unbound-bound moisture interface, thus, can be experimentally determined to be between 51.07 and 60.35 % for VIP sludge, 51.48-64.38 % for UDDT sludge, 62.29-66.71 % for ST-wGW sludge and 59.11-60.37 % for ST-GW sludge. Although hypothesis testing revealed no statistically significant difference between the methods and sample types, it can be concluded that WA demonstrated the highest reliability in terms of accuracy, ease of measurement, rapidity and repeatability.