1
COMPARISON OF EFICIENCY OF VARIOUS PAXs IN WATER TREATMENT PROCESS
Dr. Mohamed Asheesh[1], B.Sc. Amel Salihagic[2], M.Sc, Eija Laine[3]
Abstract
Several water works in Finland are producing drinking water according to the requirements of EU directives. A lot of different chemicals are used in different units during the treatment processes. The most relevant chemicals used mainly for flotation and sedimentation are coagulant materials.
Oulu University of Applied Sciences has compared the efficiency of various PAX coagulants first through a jar test and later on in continuous steady flow conditions. The goal was to test these coagulants in similar conditions as those in water treatment plants.
Optimization of pH, dosing of coagulants and adjustment of rapid mixing time was studied using jar tests. Based on the jar test results, a model was developed and applied in the pilot plant. To evaluate the efficiency of various PAXs as coagulants in water treatment process,residual turbidity and aluminum were used as the main along with UV absorbance at 254 nm and turbidity were used as the main parameters along with iron and manganese.
Introduction
Coagulation is a process for increasing the tendency of small particles in an aqueoussuspension to attach to one another and to attach to surfaces such as the grains in afilter bed. It is also used to effect the removal of certain soluble materials by adsorptionor precipitation.The coagulation process typically includes promoting the interactionof particles to form larger aggregates. It is an essential component ofconventional water treatment systems. Although the removal of microbiological contaminantscontinues to be an important reason for using coagulation, a newer objective,the removal of natural organic material (NOM) to reduce the formation ofdisinfection by-products, is growing in importance.
The most common coagulants used in water treatment are aluminum and iron salts;ferric sulfate Fe2(SO4)3alum Al2(SO4)3∙14H2O and chlorides(1),due to its effectiveness in treating a wide range of water types and relatively low cost. Using the alum in coagulation involvesformation of an assortment of chemical species, called aluminum hydrolysis products,that cause coagulation.These species are formed during and after the time thealum is mixed with the water to be treated. Coagulants are sometimes formed (orpartially formed) prior to their addition to the rapid-mixing units. Examples includeactivated silica and synthetic organic polymers, and the more recently introducedprehydrolyzed metal salts, such as polyaluminum chloride (PACl), polyiron chloride(PICl) and polyaluminum sulfate (PAS)(2).PACl has advantage of being more effective at lower temperatures anda wider pH range than alum. PACl contains stable preformed aluminum species that are thought to be more effective at charge neutralization than alum due to a higher charge density. PACl preparation involves controlled neutralization of aluminum chloride (AlCl3(s)) with base. The degree of prehydrolysis is expressed as a neutralization ratio r (r=[OH-]/[Al]). High basicity PACLs have shown to improve filter effluent turbidity and improved NOM removal when compared to alum (3).
Various situation and research studies have identified that it is essential to optimize coagulation pH and the coagulants dose in order to achieve optimal NOM removal.
In this research study,the optimal conditions for coagulation process were investigated in two steps.The first step was jar testing - the most common and simple way of coagulation optimization. This method allows optimization of coagulation pH, dose of coagulants, rapid mixing time and other important parameters. In the second step optimal parameters obtained from the jar testswere transferredto the model of water treatment system (i. e. pilot plant).
The aims of this work were to compare effectiveness of various polyaluminum chloride (PAX) coagulants. Some of them are commercial coagulants like PAX-18 and some are novel PAX coagulants.The PAXs have similar basicity, ca. 1,3. PAX-a and PAX-b contain some additives which were expected to enhance the coagulation; PAX-a contains a Si-additive and PAX-b an organic additive. And also One of the aims is a commercial as well as a novel of PAX-b and PAX-c contain organic additive.
2. Optimization of the parameters effecting the coagulation by using jar tests
2.1 Jar test procedure
Jar testsare usually used as a small scale laboratory investigation and analysis tool to test and to improve the capability and function of different objects and elements.
To determine optimal conditions for coagulation treatment, a series of jar tests was carried out to identify the optimal coagulation pH and dosage. A programmable jar testing apparatus (Kemira Kemi Flocculator) was used as follows: addition of coagulant to 1 l water samples followed by30 s rapid mixing at 400 rpm, 10min flocculation at 30rpm and 120min settling.
The coagulantdosage was measured by a calibrated micro-pipette. Settled water samples were collected at the level of 1/2 jar height measured from the bottom of the jar. Jar tests were carried out at room temperature, therefore the temperature of tested water was assumed to be between 18 and 19 ºC. The test water was a river water with a typical quality.
2.2 Adjusting coagulationpH
To determine the amount of acid-coagulant combinations needed to achieve a desired pH for each raw water sample, titration curves were developed for both the coagulants tested and sulfuric acid(4). Amount of sulfuric acid was determined from a titration curve as the volume of acid needed to decrease pH of raw water to the value for which adding of predetermined dose of coagulants decrease pH farther to the targeted pH value.
2.3 Optimization of coagulation pH and coagulant dosage
Determination of optimized coagulation conditions required evaluations of both the optimal coagulant dosage and pH. To determine the optimal pH for a selected coagulant dosage (50 ppm), jar tests were conducted using constant coagulant dose with varied pH of coagulation in range from 5,5 to 7. Sulfuric acid was used to reach the desired pH value before adding coagulant according to the titration curve mentioned above. The optimal pH of coagulation was identified as the highest pH at which there was maximum NOM removal for that specific coagulant dosage (5). Ultraviolet absorbance at 254 nm (2)was used as a measure of NOM removal.The ptimal pH regarding all the measured parameters was 6.0-6.2
According to the data shown in Figure 1, the optimal pH for coagulation with PAX-18 was about 6,2 because lower pH levels did not produce substantially greater removals of NOM. Other measured parameters including turbidity and iron concentration were also considered in determining the optimal pH.
To identify the optimal coagulant dosage at given pH, coagulant concentrations were varied in the range from 38 to 67 ppm in each jar while the optimal pH value was kept constant. According to the data shown in Figure 2, the optimal dose of PAX-18 was about 50 ppm.
Figure 1Optimizing coagulation pH by monitoring ultraviolet absorbance as measure of natural organic mater content at different pH, and influence of coagulation pH on turbidity and iron removal.
Figure 2 Optimization of coagulant dosage by monitoring ultraviolet absorbance as a measure of natural organic mater content at different doses of PAX-18, and influence of coagulant dose on turbidity and iron removal.
3. Jar tests results
3.1 Comparison of efficiency of various PAX coagulants by jar tests
Optimal conditions found for commercial coagulant PAX-18 (pH=6 and 50 ppm dosage) were applied in the jar tests using three different new coagulants (PAX-a, PAX-b and PAX-c) in order to find out the optimal dosage for these novel coagulants and to compare their efficiency.
For comparison of various coagulants, it is essential to maintain the same conditions during experiments. In our case the problem was variation in the raw water quality (alkalinity) between experiments. Raw water was taken from the Oulu River near the intake place used for the Oulu distribution system by thelocal water supply company. The quality parameters of raw water used in the jar tests varied as follows: pH = 9,64-9,55, alkalinity = 212-141 mg/l CaCO3, UVA254 = 0,259–0,325, and turbidity = 3,41-3,54 NTU except for test with PAX-c where turbidity was 11,24 NTU.
To evaluate the efficiency of different PAXs as coagulants in water treatment process, UV absorbance at 254 nm and turbidity were measured in settled water after the jar tests. According to the results presented in Figure 3, PAX-a, PAX-b and PAX-c showed similar efficiency of turbidity removal and therefore appeared to be better coagulants than PAX-18. However, it needs to be stressed once again that the turbidity of the raw water in the PAX-c test was much higher than in other tests. This might have been the reason for the best turbidity removal results for PAX-c. Regarding the efficiency ofUVA254removal as a measure of organic mater removal all tested coagulants showed similar performance for dosages in range 40-60 ppm.
Figure 3 Comparison of different PAX coagulants in terms of the efficiency ofresidual UV absorbance at 254 nm as a measure of organic mater content (right) and residual turbidity
(left).
3.2 Comparison of floc settling rates
In order to compare different coagulants using the same raw water, settling rate was measured. The measurements were carried out using jar test procedure but this time samples were taken at the level of 2/3 jar height measured from the bottom of the jar at specified time intervals after stirring had ceased. Turbidity was measured immediately after sampling. The tests were performed with different doses of coagulants in the range from 30 to 60 ppm. For dosage of 30 ppm, settling was observed only for PAX-18. For the dosage of 50 ppm,the settling rates for PAX-a, PAX-b and PAX-c were similar (Figure 4). It was evident that the influence of the dose on the settling rate for PAX-18 is less explicit than for other tested PAXs (Figure 5). It should be stressed that the experiment with PAX-18 was performed two days earlier than experiments with the other coagulants and therefore alkalinity of raw water used in the tests with PAX-18 was 110 but for other coagulants alkalinity was 80 mg/l CaCO3.
Figure 4 Settling rate for 50 ppm concentration of various PAXs. The alkalinity of raw water used with PAX-a, PAX-b and PAX-c was 80 mg/l CaCO3, but for PAX-18 the raw water alkalinity was 110 mg/l CaCO3.
Figure 5 Settling rate for different doses of PAX-18 (left) and PAX-a (right). Similar profile was obtained for PAX-b and PAX-c as that of PAX-a.
4. Applying optimized parameters into continuously steady flow in pilot plant
4.1 Pilot plant description
The specifications of the small scale water treatment plant (Pilot plant):
The plant is able to treat 0,2 m3 /h surface water transported to the plant from the nearest lakes, river or sea, and 1m3 /h groundwater (from local well made inside the laboratory for the research and development purposes).
The plant consists of the following: preliminary treatment unit, chemical dosing and mixing unit, sedimentation and flotation unit, filtration, UV disinfection unit, andwater stabilization.Each unit is equipped with pH and conductivity meters which are connected with on-line monitoring system. The preliminary treatment unit, collector from sedimentation or flotation tank, and outlet from filter are equipped also with on-line turbidity meters.
The purpose of the plant is to support teaching, real demonstration during the teaching, to give students an opportunity to practice with a small scale water treatment plant and to develop some research and development projects with local and international partners.
The pilot plant experiments (Figure 6) were carried out in line with the results of jar tests.The flow of raw water was 6 l/min. The pH value of raw water was adjusted in the first unit by adding 0,3 M sulfuric acid or 0,3M sodium hydroxide according to the procedure described above and maintained by continuous additions of sulfuric acid or sodium hydroxide by peristaltic pump.
A dose of coagulants which had been optimized in the jar tests for each PAX was added to rapid-mixing chamber with the help of peristaltic pumps.
Coagulation and floc formation were taking place during 30 seconds in rapid mixing chamber and 30 minutes in slow mixing module. After that, there were 2 possible options for water treatment: settling in the sedimentation tank or flotation. Water from the sedimentation tank or alternatively flotation tank was lead to the collector tank, filtration module, UV disinfection and finally to the clean water tank.
1
Figure 6 Pilot plant
1
4.2 Pilot plant tests results
During the pilot plant experiment with PAX-18 as coagulant, pH value in slow mixing unit was 6,1 ± 0,15. The characteristics of raw water were as follows: UVA254=0,444, Turbidity=13 NTU, alkalinity=68 mg/l CaCO3 and pH = 7,95.
Samples were taken from the sedimentation and flotation tank outlet after twohours of sedimentation or flotation. The removal of organic matter and turbidity achieved was much lower in the pilot plant than in the jar tests.
For the dose of 50 ppm PAX-18, the following results were achieved; removal of UVA254 and turbidity after two hours of settling in the pilot plant were 42,3 % and 26,4 respectively, which is much lower than 76 % and 77,7 % in case of the jar tests. It needs to be noted here that there were significant differences in the characteristics of the raw water used in the pilot plant and jar tests.
It is assumed that the lower efficiency of coagulation in the pilot plant was caused by flocs strength. It was observed that flocs were breaking down during the transport from the slow mixing module to the sedimentation or flotation tank.
After the settling in the sedimentation tank, water was transported to the filtration module. The filtration through 35 cm quartz layer and 83 cm anthracite layer showed reduction in UVA254 from 0,262 to 0,070. In this case overall efficiency in UVA254 removal (after sedimentation and filtration) was 83,8 %.
Flocwas observer in the water collector unit after the sedimentation or flotation unit. This might have been caused by insufficient detention time in the sedimentation unit (2 hours) or flotation unit (2 hours), especially when the flocs were breaking down during the transport to the sedimentation or flotation unit. To increase the detention time, theflow after the slow mixing was split to two parallel lines (sedimentation and flotation). The dosage of coagulants was alsoincreased from 50 to 60 ppm.
Figure 7
Conclusions and Recommendations
During the testing of the above mentioned coagulants, several issues were noted:
The optimal conditionsdetermined when testing coagulants in jar tests might be quite different from those determined in the pilot plant conditions mimicking continuous steady flow of water treatment plants.
The results showed that the optimal conditions (eg. pH and dose) for the coagulant obtained in jar tests were useful as a starting point for determiningnew optimal conditions for steady flow in the pilot plant.
For different coagulants a new settling or flotation technologiesmight need to be developed to improve their efficiency.
Floc strength needs to be taken into account when developing coagulants so that the floc would withstand also the turbulent flow through the treatment plant.
To observe and compare the efficiency of the chemicals (coagulants) in flotation and sedimentation, the two units should be run as parallel treatment units under the same conditions.
Detentions time, type of flow and the floc transport are essential parameters that should be considered and controlled during the research process time.
The result of the research showed that the mentioned coagulants are probablyefficient coagulants and they could be applied to treat different types of raw water. To improve their efficiency suitable aid coagulants might need to introduced.The tested new coagulants gave clearly better sedimentation and residual Al in jar test …..
Acknowledgements
Financial support from the Finnish Funding Agency for Technology and Innovation (TEKES) is acknowledged, we wish to thank Land and Water Technology Foundation (maa- ja vesitekniikan tuki) for their financial support, also we would like to thank Oulu regional water supply and water works departments for their supports and advices. Great thanks to Water Supply and Sanitation Company Sarajevo B&H.
References:
(1 )American Water Works Association, Water Treatment,1986.
(2) Raymond D. Letterman. Water QualityAnd Treatment, Fifth Edition. American Water Works Association. 1999 McGraw-Hill, Inc.
(3) Kevin McCurdy, Kenneth Carlson, Dean Gregory. Floc morphology and cyclic shearing recovery: comparison of alum and polyaluminum chloride coagulants. Water Research 38 (2004) 486-494.
(4) Kimberly Bell-Ajy, Morteza Abbaszadegan, Eva Ibrahim, Debbie Verges, Mark LeChevallier. Conventional and optimized coagulation for NOM removal. American Water Works Association. Jurnal, Oct 2000, 92, 10.
(5) Jian-Jun Qin, Maung Htun Oo, Kiran A. Kekre, Frans Knops, Peter Miller. Imapcat of coagulation pH on enhanced removal of natural organic matter in treatment of reservoir water. Separation and purification technology 49 (2006) 295-298.
(6) Flotation, Kaj Jansson Kemira Chemical S.A, LUT , 2003
(7) Water Treatment, Mohamed Asheesh, Jouku Peltokangas, OAMK, 2006-8 teaching materials.
(8)Mark Hammer and Mark J. Hammer Jr., ISBN 0-13-097325-4 , Water Technology and Wastewater Technology, Fifth Edition, 2004
(10)American Water Works Association, Water Treatment plant Design,1971.
[1]Oulu University of Applied Sciences, Finland
[2] Water supply and sanitation Lt.d, Sarjevo, Bosnia Herzegovina
[3]Kemira Oyj, Finland