By Victoria Colling, Laura Zettler, Yanting Liu, Geoff Graham, Souleymane Ndiongue, Tricia Hamilton, Frederick Dubeau and Bill Jones
The Chippewas of Nawash Unceded First Nation community is located on the southwest coast of Georgian Bay, Ontario. The community’s current water treatment process includes coagulation, flocculation, pressure filtration, cartridge filtration, UV disinfection and chlorination. However, it is under a boil water advisory due to filtration and disinfection requirements not being met (Neegan Burnside, 2019). In conjunction with Ontario First Nations Technical Services Corporation (OFNTSC), the community will be upgrading its water treatment plant and distribution system.
It is expected that the upgraded distribution system will have a longer retention time, which may impact the formation of disinfection byproducts (DBPs).
In collaboration with the Chippewas of Nawash Unceded First Nation and OFNTSC, the Walkerton Clean Water Centre (WCWC) completed a pilot test to explore treatment options. Through pilot testing, WCWC works with clients to address their drinking water challenges. These projects can take place at the client’s location or at the WCWC Technology Demonstration Facility, which houses multiple pilot-scale treatment processes.
Using the community’s source water, WCWC investigated the formation of disinfection byproducts from two treatment trains: ultrafiltration, followed by granular activated carbon (GAC) filtration, and in-line coagulation, followed by ultrafiltration and GAC filtration.
DBPs, such as trihalomethanes (THMs) and haloacetic acids (HAAs) can form in chlorinated waters with organic precursors. The Guidelines for Canadian Drinking Water Quality provide a maximum acceptable concentration of 100 µg/L for THMs and 80 µg/L for HAAs as a quarterly running annual average (2017).
Overall, the purpose of this pilot test was to investigate if either in-line coagulation before ultrafiltration or GAC filtration after ultrafiltration are necessary processes to control DBPs, such as THMs and HAAs.
Raw water and water quality analysis
Raw water was collected at the Chippewas of Nawash Unceded First Nation water treatment plant and transported in a water hauling truck to WCWC in Walkerton, Ontario. The raw water was stored in a ground level, outdoor storage tank with a 40,000 litre capacity. It had low turbidity and low organics, as measured as dissolved organic carbon (DOC) and ultraviolet absorbance at 254 nm (UV254) as shown in Table 1.
During the pilot testing, samples were collected from the raw water and after each treatment process and analyzed at WCWC for turbidity, pH, apparent colour, true colour, UV254 absorbance, DOC and alkalinity.
Prior to pilot testing, the storage tanks were drained and cleaned, the ultrafiltration membrane was chemically cleaned and the GAC pressure filter was backwashed. Jar testing was completed to determine the optimal coagulant dosage for in-line coagulant assisted ultrafiltration.
Pilot Test One included ultrafiltration without in-line coagulation, followed by GAC pressure filtration. Pilot Test Two included in-line coagulation, followed by ultrafiltration and GAC pressure filtration. The pilot plants ran for a minimum of three hours.
The ultrafiltration system is a hollow-fibre vacuum driven membrane with a nominal pore size of 0.02 µm. During pilot testing, the effluent flow rate was approximately 9.3 GPM (35.2 L/min), the transmembrane pressure was -1.4 psi, and the permeate cycle was 30 – 60 minutes in between backpulses.
The GAC pressure filter consisted of virgin, coal-based carbon contained in a 0.75 ft3 pressure vessel. During the pilot testing, the flow was approximately 3 GPM (11.3 L/min).
For Pilot Test Two, polyaluminum chloride coagulant was used as coagulation before ultrafiltration. Based on the jar test results, the optimal coagulant dosage was 5 mg/L. Dosing the ultrafiltration pilot plant was variable, due to the fluctuations in flow rates of the feed water. However, the ultrafiltration was approximately dosed at 5 mg/L, based on the permeate flow rate.
Disinfection byproduct formation
A Simulated Distribution System method (Standard Methods 4710C) was completed to determine the potential formation of DBPs at each stage of the treatment processes. Grab samples were collected after a minimum of three hours of operation and transferred to a 250 mL chlorine-demand free, amber glass bottle. Samples were dosed at 2.0 mg/L and 3.5 mg/L of chlorine and stored at room temperature. After 2, 6 and 8 days of contact time, the chlorinated samples were tested for chlorine residual and sent to an accredited laboratory to measure THMs and HAAs.
Organics are low in the community’s source water, but they are precursors to the formation of DBPs. Time is another major factor in DBP formation and it is anticipated that this community will have long chlorine retention times in their distribution system. Therefore, it is important to assess the reduction of organics with the pilot testing treatment processes as the retention time in the community’s distribution system is not something that can be easily altered.
Figure 2 shows the DOC reduction at each treatment stage for each pilot test. As expected, the majority of the DOC was removed by the GAC filtration. When comparing the total DOC removal between the two pilot tests, it was found that the in-line pre-coagulation (Pilot Test Two) provided a total of 93% DOC reduction, whereas the same processes without coagulation (Pilot Test One) provided a total of 75% DOC reduction. Similar trends were found with monitoring UV254.
The formation of DBPs is not only dependent on the level of organic precursors and chlorine dosage, but also the reaction time. Although this source water has low organic precursors, the retention time is expected to be up to eight days in the distribution system. Therefore, THMs and HAAs were measured after 2, 6 and 8 days of contact time.
Overall, the formation of THMs and HAAs increased as the contact time increased from two days to eight days, as shown in Figures 3 and 4. The samples that were dosed with chlorine at 2 mg/L had slightly less THMs and HAAs, compared to samples that were dosed at 3.5 mg/L. Regardless of the contact time, GAC reduced the levels of THMs and HAAs formed, which was expected.
Ultrafiltration without coagulation (Pilot Test One) produced slightly higher levels of THMs and HAAs formed, compared to the processes with coagulation (Pilot Test Two). However, all DBP levels still remained below the maximum acceptable concentrations.
Overall, the GAC filtration provided a higher reduction in organic precursors and therefore a lower formation of THMs and HAAs than the coagulation followed by ultrafiltration. However, pre-coagulation may be considered to control biological fouling on the membrane.
The community, in conjunction with the engineering consulting firm, is in the process of designing a water treatment process and distribution system that is appropriate for this community.
Victoria Colling, Laura Zettler, Yanting (Helen) Liu, Geoff Graham and Souleymane Ndiongue are with the Walkerton Clean Water Centre. Tricia Hamilton is with Ontario First Nations Technical Services Corporation. Frederick Dubeau and Bill Jones are with the Chippewas of Nawash Unceded First Nation. This article appears in ES&E Magazine’s December 2019.