WaterSubject

As legislation with respect to drinking water tightens, there is a need to constantly improve on current water treatment processes and to evaluate the effectiveness of new technologies in producing drinking water that is free from pathogens and toxic chemicals. The removal of organic material from raw water is one area that has received much attention in recent years. Organic material can react with chlorine to form trihalomethanes (THMs), which can be harmful to health, and related organochlorine compounds that can cause taste complaints.

To comply with the THM legislation (100 ?g/L for total THMs at the customer’s tap), many surface water utilities have tried to remove organic material by increasing the coagulant dose, lowering coagulation pH or a combination of both – with the consequence that chemical and sludge disposal costs increase. Draycote water treatment works (WTW) is faced with the same issue.

Draycote WTW (Rugby, UK) is a conventional surface water treatment facility using dissolved air flotation (DAF), rapid gravity filtration, granular activated carbon and disinfection. The source water is low in colour, has low turbidity and is highly buffered, with alkalinity typically around 143 mg/L (as CaCO₃) and total hardness of about 290 mg/L (as CaCO₃). Total organic carbon (TOC) concentration is typically between 8.4 and 6.4 mg/L as C, of which 92% is dissolved organic carbon (DOC).

The Magnetic Ion EXchange (MIEX^®) DOC process, developed by Orica Watercare (Wigan, UK) has been reported as being successful in removing DOC from raw water, in particular, humic-rich water. This technology is based on the recognised premise that anionic resins have a high affinity for weakly acidic organic substances. The resin used in the MIEX^® DOC process is unique: the structure contains a magnetic compound that makes the resin beads behave like small magnets, which enables the resin to be separated from the treated water by settling at a high surface loading of the settler, and allows the resin to be used in a continuously, stirred tank reactor instead of a plug flow column system. In this way, the MIEX^® DOC process purportedly overcomes many of the problems associated with conventional ion exchange systems and makes this technology economically feasible for commercial, large-scale potable water treatment plants.

The results of jar test studies to evaluate the effectiveness of the MIEX^® resin for removing DOC from Draycote raw water were promising and justified further investigation of the MIEX^® process on a pilot plant scale at Draycote WTW.

The objectives of the study reported herein were as follows:

The MIEX^® pilot plant trial was conducted from 12 July to 1 December 2004 at Draycote WTW using a 30 m³/day continuously operating pilot plant provided by Orica Watercare. Raw water was fed to the MIEX^® pilot plant directly from the main plant’s supply line. The plant was run 24 h a day.

The parameters under which the pilot plant was operated at the start of the trial were selected using Orica Watercare’s previous pilot plant experience. The resin concentration in the contactor tanks was maintained at approximately 20 mL of settled resin per 1 L of raw water. A total contact time of 10 min was provided by the two contactors (5 min of retention time in each). The resin was recovered in the settler vessel and the treated water was sampled and fed to the DAF pilot plant. The MIEX^® plant was operated at a 4% resin regeneration rate (i.e. 96% recycled without regeneration). Resin regeneration was conducted with brine (120 g/L as NaCl) at 6-h intervals with the brine being reused five times before disposal. The waste stream, a highly concentrated DOC solution in brine, was collected in a 1000-L container and removed from the site by tanker. The MIEX^® DOC process is illustrated.

Over the trial period, four other operating regimes were appraised with a view to optimising plant performance, namely:

The DAF plant was provided by Purac (Kidderminster, UK) and operated at a 30 m³/day throughput. The coagulant used was ferric sulphate (Ferripol XL, E&A west, Grimsby, UK). HCl dosing facilities were also available.

Procedure for THM formation and measurement

Samples for measurement of chlorine demand were placed in 2.5-L amber glass bottles. Each sample was dosed with a 7-mL chlorine solution prepared by diluting 1-mL sodium hypochlorite solution to 100 mL with distilled water to achieve a chlorine residual of 1.5 mg/L. The chlorinated samples were then put into 100-mL amber glass bottles and placed in the dark in an incubator set at 9°C. Residual chlorine was measured after 2 days using a Hach pocket colorimeter (Hach Lange Ltd, Basingstoke, UK) in conjunction with ‘Palintest’ DPD No. 1 tablets (Palintest Ltd, Gateshead, UK) and sent to Severn Trent Laboratories (STL) for THM analysis in 40-mL amber glass vials containing sodium thiosulphate.

Sampling procedure

The raw water, MIEX^®-treated water and DAF-treated water were sampled several times a day. Temperature, pH and turbidity were analysed on-site immediately after the sample was taken. For off-site analysis (ultraviolet (UV)₂₅₄, DOC, colour and iron), samples were transferred into sample bottles and refrigerated before being taken to STL.