Bloom Buster

How a pre-treatment pilot project by Veolia Water Solutions & Technologies (VWS) for the Fujairah 2 RO plant successfully tackled outbreaks of Harmful Algal Blooms

When Veolia Water Solutions & Technologies (VWS) was awarded the contract to construct the 128,000m³/day Reverse Osmosis (RO) component of the hybrid Fujairah F2 Independent Water and Power Project (IWPP) in the UAE, the contract, among other things, required VWS to establish the performances and reliability of the pre-treatment stage, a key reason being the increasing frequency and intensity of Red Tide or algal blooms along UAE’s Gulf of Oman coast where the Fujairah F2 plant is located.

Algal blooms spell bad news not only for the immediate aquatic and marine environment, but also for the operation of desalination plants, a critical issue given that desalination meets bulk of the Gulf region’s water requirements (In the case of UAE, it is 98%). The toxic variety, referred to as Harmful Algal Blooms (HAB), pose a major risk to humans if the toxins are not effectively removed from the seawater.

Seawater Reverse Osmosis (SWRO) tends to be more sensitive to source water conditions than thermal desalination. Algal blooms increase the organic load of the source water, which accelerates the biofouling of RO membranes and adversely impacts the performance and efficiency of the plant. To a large extent, the success of RO-based desalination depends on the effectiveness of the pre-treatment process, positioned between the source water and RO membranes and thus, plays a vital role in terms of assuring constant water quality prior to the RO stage. In fact, the past few years have seen growing cases of RO-based desalination plants along Arabian Gulf and Indian Ocean being shut down when their pre-treatment systems were overwhelmed by the extreme intensity and duration of algal blooms.

What sets the F2 apart is its advanced pre-treatment system consisting of a high speed Dissolved Air Floatation (DAF) process, followed by Dual Media Filtration (DMF) before the RO membrane. Despite never seen before HAB on the Gulf of Oman coast, the advance pre-treatment technique from VWS demonstrated its robustness to continuously produce pre-treated water with constant quality.

Pilot process

Often, the choice of a process used as pretreatment is supported by preliminary pilot tests. Therefore, a six month pilot plant trial was carried out at the F2 site from April to November 2008 to investigate the raw seawater quality, the performance of the pre-treatment under various conditions including algal blooms and optimisation of the chemical dosing.

The pilot plant from VWS comprised of its patented Veolia Spidflow high speed Dissolved Air flotation (DAF) clarification system coupled with DMF. An open seawater intake was installed in parallel of the future plant intake. The process flow diagram is given in Figure 1.

As seen in Figure 1, seawater is chemically conditioned with chlorination (using liquid sodium hypochlorite), pH adjustment (using sulphuric acid), coagulation (using commercial ferric chloride) and flocculation, followed by high speed floatation. Polymer injection option was provided between the coagulation and flocculation stages to obtain the required seawater quality in case there was a need for the same. The flow through the pilot plant averaged at 28 m³/hour. Recirculation rate was established between 5-11% of the influent flow. After floatation, only 1.16 m³/hour of the flow would pass through the DMF, which was made of sand and pumice. An option of injecting coagulant through a static mixer between floatation and filtration if needed too was provided.

The sea water parameters that were monitored during the pilot plant operation were pH, conductivity (mS/ cm), salinity, temperature (C), turbidity (NTU), TSS (mg/l) and Silt Density Index (SDI). To monitor and evaluate the process performances, the SDI of raw sea water was computed using the SDI 3 test while that of filtered sea water was computed using SDI 15 test.

During the pilot, the raw seawater reached quality levels never seen before. Hence, the pilot study distinguished between periods of normal quality sea water and degraded quality seawater. ‘Normal’ is characterised as relatively constant quality while ‘degraded’ represents extremely variable quality of seawater. Two periods of degraded quality seawater occurred during the pilot (with the second one extending beyond the trial period). A sampling and analysis campaign led by the United Nation University in Dubai traced the cause of the water degradation to Harmful Algae Blooms generated by the phytoplankton specie Cochlodinium polykrikoide. The initial signs appeared during August 2008, after which blooms occurred irregularly over a period of almost 10 months until the beginning of summer 2009. At the beginning of each bloom period, a drop in water temperature was recorded.

When the bloom took place close to the seawater intake, high concentrations of the algae were sucked into the pre-treatment process. SDI values would shoot up and become difficult to measure due to the rapid clogging of the SDI membrane caused by the high concentrations of phytoplankton and the mucus produced by it. On many occasions, SDI 3 min could not be measured, and sometimes, even a SDI1 min was impossible to measure. The average SDI value increased from 20%/min during the normal quality periods to more than 27%/min during the blooms. Even parameters like TSS would swing from low (10 ppm) to extremely high values (30 – 35 ppm) during a very short period of time (under specific conditions of operation). Interestingly, Polynuclear Aromatic Hydrocarbons (PAH), oil and grease in the raw seawater levels were below the detection limit, which was the case during ‘normal’ periods.

As with seawater quality, the differentiation between normal and degraded seawater conditions was applied to plant operations too. During the degraded seawater conditions caused by algal blooms, the operation of the pilot plant had to be adapted to deal with varying seawater quality. Once the optimum operational parameters were identified, the same were strictly implemented during subsequent episodes too. The physical parameters of Spidflow and DMF remained unchanged, the floatation and filtration operational flows weren’t reduced, the recirculation rate was not increased; what was increased was the dosing of ferric chloride before floatation. The maximum dose was 11.1 mg/L as Fe. At such high doses, pH adjustment was not always necessary as ferric chloride was enough to lower the pH to 7 or less. A second coagulation point was started and tests carried out with an optimum dose of 1.0 mg/L as Fe.

After applying the adequate chemical dosages and new operating control mode, filtered water quality that was achieved under degraded conditions was well within the required criteria, with SDI 15 min below 4 %/min more than 75 percent of the time and below 5%/min all the time. For comparison, during normal seawater quality periods, SDI 15 min was below 5%/min all the time while SDI 15 min was below 4%/min more than 75% of the time. Table 1 shows filtered water quality data during the normal period, while Table 2 shows the same data for the degraded period after the adaptation phase. Both cases demonstrate the robustness of the process in continuously producing pretreated water with constant quality.

Figure 2 shows how filtered water quality improved during the process adaptation phase to reach the quality objectives.

Further, average filtration cycle duration during degraded conditions was lowered to 33.7 hours compared to 42 hours for normal seawater conditions, which is less than one per cent below the objective of 34 hours.

Another important operational parameter monitored was the floated sludge production. Spidflow is equipped with a mechanical scrapper which made it easy to control sludge. Under normal seawater conditions, average sludge production was 0.026% of the influent flow, with average concentration of three per cent. Under degraded seawater conditions, sludge production increased in line with the increase in suspended materials and coagulant dose. Average production increased to 0.086% of the influent flow, with an average concentration of 2.3%.

After finding the adequate operating parameters for both ‘normal’ and ‘degraded’ seawater quality periods, the process evaluation criteria were laid down as follows: – Average duration of filtration cycle of 40.5 hours (objective of 34 hours)

- SDI 15 of filtered water below 5 at all times (objective of 100%)

- SDI 15 of filtered water less than 4 on an average lower than two hours, resulting more than 90% of the actual SDI 15 tests lower than 4 (against an objective of 75%).

The pilot plant study provided valuable insights that were used in the deployment of Spidflow DAF clarification at F2 IWPP. Sanjay Sharma, Director – Sales, Design & Build, VWS said, “Given the inherent advantages of the Veolia Spidflow DAF for the removal of Red Algae, Oil and Colloids, the same can be incorporated either for augmenting existing SWRO plants as an add-on pre-treatment to the existing DMF, or as pre-treatment for new green field SWRO plants. During Red Algae events, the quality of water achieved after DAF is similar or even better as compared to sea water analysis during non-Red Algae condition enabling the SWRO to run smoothly and efficiently.”