Congress Resources: Papers, posters and presentations

< Return to abstract list

Preliminary Assessment Of Glass-based Filtration Media For Drinking Water Treatment

Congress: 2015
Author(s): Jia-Qian Jiang (Glasgow, UK), Anna Cescon, Mark Haffy, Graeme Moore
Glasgow Caledonian University1, Scottish Water2

Keyword(s): Sub-theme 1: Water supply and demand,
AbstractJia-Qian Jiang1*, Anna Cescon1, Mark Haffy2, Graeme Moore2
1 School of Engineering and Built Environment, Glasgow Caledonian University, Glasgow G40BA, Scotland
2 Scottish Water, 6 Castle Drive, Carnegie Campus, Dunfermline, Fife, KY11 8GG, Scotland
* Corresponding Author: E-mail:; Tel: +44 1413318850
Filtration for water treatment can be improved with regards to getting high quality effluent, the use of emerging filtration media, optimisation of pre-treatment (e.g., coagulation), and reduction of process energy consumption (e.g., reducing the frequency of backwashes). Moreover, some materials such as pumice [1] and waste glasses [2] could be used as the filter media, since they are either easily available in the environment or can be recycled, which should represent a more sustainable approach to drive the Green agenda in achieving high water quality standards and responding to water pollution challenges.
Aims and Objectives:
This paper presents findings from research carried out to assess the performance of several emerging filtration media for drinking water treatment and their suitability to replace the conventional media -- while considering the influence of coagulation and pH adjustments on the filtration effluent quality.
The experiments were performed in a laboratory setting through a filtration unit formed by a coagulation/sedimentation tank and a filtration column. Model raw water was prepared through the addition of humic acid and kaolin particles. As well as sand and anthracite used as reference media, several media materials were selected for the filtration process: two glass based media (AFM and Enviro Glasmedia), one expanded clay, Filtralite, and one anthracite like media, Enverzit, were tested and considered in varying configurations, i.e. single and dual. All the materials selected comply with existing legislation for drinking water treatment [3]. The process was operated at a constant flow rate, maintained through a flowmeter located at the outlet of the column. The trials have been conducted in an intermittent mode: the process was operating for 4 hours each day, being then interrupted and reprised the following day. The end of the run was reached either after 40 cumulative hours of operations or when the head loss development had reached the maximum limit allowed, 55 cm.
Filtration was preceded by coagulation and sedimentation, performed alternatively with aluminium sulphate (alum) and polyaluminium chloride (PACl). The optimal dose and pH were determined through preliminary jar tests. Various parameters were considered to evaluate the effectiveness of the materials: turbidity, particle number concentration (ranging 0.9-140 µm), headloss development, visible absorbance (400 nm), UV absorbance (254 nm), DOC, residual aluminium/iron content, and suspended solids. Samples were collected and analysed from the model raw water, the supernatant after coagulation/sedimentation, the filtrate and two sampling points on the column.
Results and Discussion:
Figures 1-3 shows the turbidity, particle count and the head loss development of the filtration effluents. While 1 NTU is the turbidity limitation prescribed by the legislation [4], the water industry endeavours to maintain water turbidity below 0.1 NTU -- this parameter has become increasingly important in defining the performance of a filter medium. The particle count, especially when referred to particles smaller than 2 μm, represents a more precise method to test variations in the operational conditions when the turbidity values are low [5]. Head loss development influences the length of the run, and thus an indication of the energy consumption for the filter backwash.
Fig. 1 shows that regardless the media configurations and the type of coagulants used, turbidity values in the effluents were below 0.1 NTU for the majority of runs. The peaks visible for some of the configurations are likely due to the intermittent operations. The use of PACl increased the removal of turbidity for the study conditions. In relation to the residual number concentrations of 0.9 μm particles, below 1000 particles per litre was achieved for the effluent coagulated by the PACl (Fig. 2); among the configurations treated with alum, only AFM and Everzit reach comparable results.
The variation of head loss development is greatly affected by the coagulant used. For various filter configurations, the headloss development ranged 0-5 cm for the alum coagulated water and increased to between 10 and 25 cm for the PACl coagulated water. This could be explained that PACl promotes the formation of larger and more stable flocs, leading to a faster blockage of the filter media. The comparison of different media configurations suggests that glass media had lower head loss development, which represents a valid alternative to the conventional filtration media. Although the higher efficiency of PACl for the particle removal was observed in this study and is well documented in the literature [6], the adoption of a pre-polymerised coagulant should be evaluated case by case.
In terms of this preliminary study, alternative filter media could be employed with possible beneficial effects. The use of recycled glass media gives added values to materials that would be otherwise discarded, though the processing of these materials could be energy consuming. Further work is in planning where the most promising filter media and configurations will be assessed in full scale, against a range of more challenging water quality profiles -- typical across Scotland. At the same time, the environmental impact and cost effectiveness of emerging filter media will be evaluated, too.
The authors are grateful to the GCU Research Committee to offer Anna Cescon a studentship for this study. Thanks also go to the project sponsor, Scottish Water. The views in this abstract do not necessarily represent that from the company.
Figs 1-3 can not be uploaded. [1] Farizoglu B, Nuhoglu A, Yildiz E, Keskinler B. The performance of pumice as a filter bed material under rapid filtration conditions. Filtr. Sep. 2003, 40(3): 41-47.
[2] Water Development Services. Full Scale Operational Trials Involving The Use of Recycled Glass in Selected Markets (WRAP). 2005.
[3] DWI. List of approved products for use in Public Water Supply in the United Kingdom. 2013 (March).
[4] Scottish Statutory Instruments. The Water Supply ( Water Quality ) ( Scotland ) Regulations. 2001 (207).
[5] Hargesheimer E. E.; Mctigue N. E.; Mielke J. L.; Yee P.; Elford T. Tracking filter performance with particle counting. J. Am. Water Works Assoc. 1998, 90 (12): 32-41.
[6] Jiang J-Q. Development of coagulation theory and pre-polymerized coagulants for water treatment. Sep Purif Rev. 2001, 30(1): 127-141.
2011 IWRA - International Water Resources Association - - Admin