Sustainable up-flow sand filter for domestic water supply
A decentralised system is among the solutions to providing treated water to isolated areas. Upflow sand filtration systems offer comparable treatment efficiency to downflow slow sand filters without clogging susceptibility. Upflow sand filter application as a point-of-use water treatment system was...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/39092/1/24%20PAGES.pdf https://eprints.ums.edu.my/id/eprint/39092/2/FULLTEXT.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A decentralised system is among the solutions to providing treated water to isolated areas. Upflow sand filtration systems offer comparable treatment efficiency to downflow slow sand filters without clogging susceptibility. Upflow sand filter application as a point-of-use water treatment system was studied. The current study aimed to identify the characteristics and suitability of local river sands as upflow sand filter media, evaluate the effects of sand grain size and filter bed height on operational pressure drops, and the efficiency of the upflow sand filter in removing total suspended solids (TSS), turbidity, total coliform, and ammonium (NH4+) at varied flow rates. The spatial and temporal profiles of microbial growth during acclimatisation and clogging effects throughout the filter operation were also assessed. Three river sand samples were studied regarding their physical characteristics and trace elements with a beach sand specimen as a comparison. The pressure drop of three sand sizes (0.15, 0.17, and 0.25 mm) and five bed heights (0.3–0.8 m) were determined. The results were contrasted against the Ergun, Kozeny-Carman, and Fair and Hatch equations. A pilot scale of the upflow sand filter was utilised to evaluate the removal efficiency of three operational velocities (0.072, 0.181 and 0.481 m h−1). An eight-week acclimatisation period was adopted to develop the microbial community within the filter bed, resulting in a total operational duration of 135 days. The river sand ranged from 0.075 to 1.7 mm in size and possessed a uniformity coefficient (Uc) of 1.8–2.7. Sands with a Uc of 3 and lower are preferable for filter applications. The samples also demonstrated high silica contents at 704.5, 604.24, and 405.41 g/kg, indicating purity. Conversely, beach sand contained the lowest silica (371.02 g/kg) and high chloride (30.15 g/kg) and calcium (537.17 g/kg) contents, which could leach out if employed as a filter media. Reduced grain size and increased bed height improved pressure drop. Similar maximum pressure drop velocities were recorded (0.835 m s−1 Vs), except at 40 and 80 cm bed heights at 0.747 and 0.878 m s−1 Vs, respectively. Subsequently, the filter bed was fluidised and stabilised following the maximum pressure drop. The prediction model exhibited that pressure drop could be projected with the Kozeny-Carman and Fair and Hatch equations pre- and post-fluidisation, while the minimum fluidisation velocity could be simulated via the Wen and Yu equation. The minimum fluidisation velocity was affected by grain size but was independent of the filter bed height. The turbidity and TSS removal ranged from 75 to 100% and 93.98% for NH4. Removal was mainly achieved deeper in the bed, where 90% turbidity and TSS removal occurred in the middle layer (30 cm deep), and over 86.73% NH4 was removed at 50 cm depth. The total coliform removal was beyond 89% but dropped at higher flow rates due to disturbances from the increased flow rate. The field emission-scanning electron microscopy (FE-SEM) images demonstrated that the microbial grew from the bottom and progressively moved upward in the filter bed. After eight weeks, substantial growth was observed at the bottom and middle layers and lesser on the top section. Dissolved oxygen concentration was also documented as the lowest (1.32 mg/L) in the middle layer, followed by the bottom layer (3.24 mg/L), indicating rapid microbial activities. Clogging was insignificant, with R2 of 0.0339, 0.4273, and 0.476. The current research demonstrated that upflow sand filters could be fabricated in isolated areas by employing local river sand. Moreover, for an average of 100 L per capita per day usage, the upflow sand filter would cater to a household of six. |
|---|