A technological investigation was carried out over a period of 2 years to evaluate surface water treatment technology. The study was performed in Poland, in three stages. From November 2011 to July 2012, for the first stage, flow tests with a capacity of 0.1–1.5 m3/h were performed simultaneously in three types of technical installations differing by coagulation modules. The outcome of the first stage was the choice of the technology for further investigation. The second stage was performed between September 2012 and March 2013 on a full-scale water treatment plant. Three large technical installations, operated in parallel, were analysed: coagulation with sludge flotation, micro-sand ballasted coagulation with sedimentation, coagulation with sedimentation and sludge recirculation. The capacity of the installations ranged from 10 to 40 m3/h. The third stage was also performed in a full-scale water treatment plant and was aimed at optimising the selected technology. This article presents the results of the second stage of the full-scale investigation. The critical treatment process, for the analysed water, was the coagulation in an acidic environment (6.5 < pH < 7.0) carried out in a system with rapid mixing, a flocculation chamber, preliminary separation of coagulation products, and removal of residual suspended solids through filtration.

INTRODUCTION

Surface waters are exposed to the impact of numerous pollutants. As a result of runoff pollution, including fertilisers used in agriculture, a continuous increase in surface water contamination has been observed. The resulting massive growth of aquatic plants, i.e. algal blooms, leads to secondary pollution of water. Due to an increasing concentration of water organisms and their productivity, the amount of organic substances in the water is high and the content is variable. Such substances, with various molecular weight and structure, can demonstrate different susceptibility to removal from water (Hem & Efraimsen 2001; Huber et al. 2011; Vasyukova et al. 2013). The main processes used to effectively reduce organic matter concentration during the course of surface water treatment include coagulation, sedimentation and filtration (Sohn et al. 2007; Lawler et al. 2013; Wolska 2014). A number of factors influence the effectiveness of coagulation and filtration for removal of organic substances from water. Detailed descriptions of such factors can be found in numerous publications (Pruss et al. 2009; Teixeira & Miguel 2011; Baghoth et al. 2011; Guminska & Klos 2012; Klos 2013). These reviews conclude that the correct way to choose the best surface water treatment technology prior to the design stage is through technological studies.

Intensification of removing organic contaminants from water during the coagulation of the volume is made possible by using optimal process parameters, including the proper dosage of coagulant, the optimum pH at which the process is performed, as well as the appropriate time and mixing intensity. The required total organic carbon (TOC) percentage removals increase with increasing raw water TOC concentration and decrease with increasing alkalinity. Furthermore, natural organic substances are effectively removed at a lower pH; thus – due to the greater cost water purification alkaline pH range – the required degree of removal of TOC decreases with the increase of its alkalinity.

The efficiency of the coagulation process has a significant impact on the final efficiency of the resulting water purification throughout the process system. According to many researchers, pH has a greater influence on the removal efficiency of organic matter than the type or dose of coagulant. The optimum pH at which the removal of organic compounds in the coagulation with aluminium sulphate is greatest is in the range from 5.0 to 6.0 (Qin et al. 2006; Volk et al. 2000).

On the basis of the conclusions drawn from the first stage of the research project involving laboratory tests and small flow pilot-scale studies (Pruss & Pruss 2013), a pilot-scale study was carried out with three process lines: coagulation with sludge flotation, coagulation and sedimentation with sludge recirculation, and micro-sand ballasted coagulation and sedimentation. The analysis of the results obtained served as the basis for selecting the optimum surface water pre-treatment technology. At present, the investment project, including design and construction of the water treatment plant, is in progress. The implementation of the water treatment technology process is forecast to be completed by 2017/2018.

METHODS

The study was carried out on a full-scale water treatment plant (Figures A1–A6, available online at http://www.iwaponline.com/wst/071/513.pdf). The purpose of these investigations was to select the more effective technology and propose a pre-treatment system for surface water characterised by low alkalinity, high temperature variability and seasonally by a very high content of organic substances resulting from algal blooms. Following a comparison of the surface water quality and the requirements set for the treated water (Table 1), total organic carbon, total iron, total aluminium, total suspended solids, and chemical oxygen demand (COD) KMnO4 were selected as the parameters of the water treatment effectiveness.

Table 1

Quality of raw water and requirements set for treated water

    Raw water
 
  Requirements for drinking water
 
Parameters and content Unit min max Requirements for treated water UE (Council Directive 98/83/EC) WHO (2011)  
pH – 7.28 7.89 6.5 < pH ≤ 8.5 6.5–9.5 a 
Temp °C 0.6 8.2 – – – 
Total hardness mval/L 2.31 3.09 ≤3.2 – b 
Total alkalinity mval/L 1.35 1.80 ≤2.0 – – 
Total suspended solids mg/L 2.8 39.0 ≤2.0 – – 
Total iron mgFe/L 0.508 1.360 ≤0.05 0.3 a 
Total aluminium mgAl/L 0.041 0.274 ≤0.1 0.2 0.1–0.2 
Chlorides mgCl/L 28 40 ≤90 250 250 
Sulphates mgSO4/L 50 75 ≤112 250 250 
TOC mgC/L 6.30 18.40 ≤4.0 No abnormal changes – 
COD KMnO4 mgO2/L 4.43 12.35 ≤7.5 5.0 – 
    Raw water
 
  Requirements for drinking water
 
Parameters and content Unit min max Requirements for treated water UE (Council Directive 98/83/EC) WHO (2011)  
pH – 7.28 7.89 6.5 < pH ≤ 8.5 6.5–9.5 a 
Temp °C 0.6 8.2 – – – 
Total hardness mval/L 2.31 3.09 ≤3.2 – b 
Total alkalinity mval/L 1.35 1.80 ≤2.0 – – 
Total suspended solids mg/L 2.8 39.0 ≤2.0 – – 
Total iron mgFe/L 0.508 1.360 ≤0.05 0.3 a 
Total aluminium mgAl/L 0.041 0.274 ≤0.1 0.2 0.1–0.2 
Chlorides mgCl/L 28 40 ≤90 250 250 
Sulphates mgSO4/L 50 75 ≤112 250 250 
TOC mgC/L 6.30 18.40 ≤4.0 No abnormal changes – 
COD KMnO4 mgO2/L 4.43 12.35 ≤7.5 5.0 – 

aNot of health concern at levels found in drinking water. An important operational water quality parameter.

bConsumers tolerate water hardness in excess of 500 mgCaCO3/L.

Raw water from the river at the maximum amount of 110 m³/h was delivered to the contact tank, where primary oxidation was carried out using chlorine dioxide added to the pipeline, prior to the static mixer. The dose of chlorine dioxide ranged from 0.8 to 1.25 mg/L. After preliminary oxidation, water was split into three independent technological lines where the following processes were carried out:

  • Coagulation + flotation and filtration through the anthracite and quartz sand bed (capacity of 10 m³/h). Coagulation products were removed together with air bubbles and transported to the water surface. Coagulant dose was 3.0–7.8 mgAl/L.

  • Coagulation + sedimentation with sludge recirculation and filtration through the anthracite and quartz sand bed (capacity of 20 m³/h). Coagulation products were removed by gravity flow in a multi-stream settler. Prior to its delivery to the sludge section, sludge was first re-circulated to the initial chamber. Coagulant dose was 2.4–5.0 mg Al/L.

  • Micro-sand ballasted coagulation + sedimentation and filtration through the anthracite and quartz sand bed (capacity of 40 m³/h). Removal of coagulation products was enhanced by micro-sand, which after sedimentation and hydrocyclone washing was returned to the process. Coagulant dose was 2.25–5.0 mg Al/L (Figure A6).

Aluminium sulphate was used as a coagulant, individually dosed into each of the process lines. Coagulation was carried out in an acidic environment (6.5 < pH < 7.0).

Subsequently, the water passed through a rapid filter with hydro-anthracite N and quartz sand bed. Filtration velocity was 5.0–7.5 m/h. The filter was filled with 300 mm of hydro-anthracite N (effective size de = 1.9 mm, uniformity coefficient < 1.5) and 700 mm of quartz sand (effective size de = 0.955 mm, uniformity coefficient = 1.28).

Incidentally, between 25 and 29 March 2013, 20 mg C/L powder activated carbon (PAC) Norit W 15 (Iodine number 1200, Molasses number 200, Methylene blue adsorption 22 g/100 g; total surface area (B.E.T method) 1,150 m2/g, particle size D50 = 15 μm) was added to the system before the filters. Hydraulic retention time for PAC before it entered the sand bed was about 10 min.

Samples for physicochemical analyses were collected from raw water, water after coagulation and water after filtration. The concentration of total suspended solids was determined only after coagulation. Other analysed parameters included TOC, dissolved organic carbon (DOC), COD KMnO4, alkalinity, pH and the content of aluminium and iron. In addition, to ensure a more meaningful interpretation of the results obtained, the organic matter fraction was determined during and after the algal bloom. In the autumn session of the study, when an algal bloom occurred, hydrobiological analyses of raw water were made. Phytoplankton biomass concentration and chlorophyll-a in plankton algae were also measured. In addition, phytoplankton taxa were identified. The physicochemical and hydrobiological analyses were carried out by accredited Polish laboratories, while the Liquid chromatography system coupled with organic carbon detectors (LC-OCD) analysis was performed by DOC-LABOR analytical company (http://www.doc-labor.de/) from Germany. The analytical methodology conformed with the standard method.

RESULTS AND DISCUSSION

The investigation was carried out in two seasons: in the autumn, from October to December 2012, and in the winter, from January to March 2013. The quality of water in the river changed significantly in October. In August, an algal bloom occurred which intensified over September and reached its peak during the first days of October when the biomass of the organisms equalled 20 mm3/L. The predominant taxa identified during the bloom included Cyanobacteria and diatoms (Figures A7–A9, available online at http://www.iwaponline.com/wst/071/513.pdf). Along with the plankton growth, the content of organic carbon in the water changed. The maximum TOC value (18.3 mg C/L) occurred in October, around 10 days after the peak algal bloom.

During the investigations, TOC occurred in raw water mainly in the dissolved form (90–98% as DOC). Depending on the season of the year, organic matter demonstrated a different degree of susceptibility to removal, as evidenced by the changing types and proportions of substances making up dissolved organic carbon observed over the same period.

Figure 1 presents organic matter fractions measured during and after the algal bloom. Humics were the predominant fraction of dissolved organic matter both during and after the algal bloom, accounting for 60–66% of organic content, respectively. The presence of biopolymers was undoubtedly related to the algal bloom and decomposition of dead phytoplankton. After the bloom occurred, dissolved organic matter also contained a small amount (1%) of organic acids.

Figure 1

Types of organic matter in raw water during the bloom and after the bloom.

Figure 1

Types of organic matter in raw water during the bloom and after the bloom.

Figure 2 presents selected quality indicators of raw water during the autumn and winter study sessions (Tables A1 and A2, respectively, available online at http://www.iwaponline.com/wst/071/513.pdf). The analysis shows that, during the study period, TOC concentrations changed continuously. The lowest values were observed at the end of January and beginning of February and the peak values occurred in mid-February. Changes in the TOC value of raw water had a clear impact on the concentrations of residual TOC after the coagulation and filtration process. Most of the time, TOC values, after coagulation, remained within the range from 4 to 5 mg C/L.

Figure 2

TOC in raw water and after coagulation, flotation or sedimentation and filtration required TOC value (TOC acceptable limit = 4.0 mg C/L).

Figure 2

TOC in raw water and after coagulation, flotation or sedimentation and filtration required TOC value (TOC acceptable limit = 4.0 mg C/L).

Between 15 and 19 November 2012 and 6 and 7 December 2012, an increase in the coagulant dose to 7 and 7.8 mg Al/L respectively, accompanied by a reduction in water pH to 6.5–6.7, resulted in a drop in TOC concentration to ≤4 mg C/L. A similar effect was achieved between 31 January and 1 February 2013 and 6 to 7 February 2013, after the initial dosing of ClO2 and after adding 5.2 mg Al/L at pH between 6.4 and 6.5. The result, however, could not be repeated between 13 and 15 February 2013 for a lower coagulant dose (4.5 mg Al/L) and the initial ClO2 dose increased to 1.2 mg/L and pH at 6.6.

Between 11 and 20 March 2013, pre-chlorination with 1.2 mg Cl2/L combined with an increased dose of coagulant (6 mg Al/L) and water alkalinisation, in order to maintain pH at the level of 6.4–6.5, helped to reduce the TOC value in the process of coagulation with sludge flotation of 4 mg C/L. In the two other process lines, the applied dose of 4.5 mg Al/L provided a decrease in TOC concentration to 4.4 mg C/L, on average.

Between 25 and 29 March 2013, the coagulant dose used was 4.5 mg Al/L and powdered activated carbon (PAC) was added before rapid filters. As a result, in the process of coagulation, with sludge flotation, TOC was reduced to a level below 4 mg C/L.

An important parameter of the analysed processes, in addition to coagulation effectiveness, is efficient removal of coagulation products with water due to the filtration. It is assumed that water passing through a settling tank or separator that works effectively should not contain more than 2 mg/L of post-coagulation suspended solids. The results of relevant tests are presented in Figure 3. A review of the results shows that the process line, with sludge recirculation, offered the lowest degree of performance consistency. Due to frequent instances of sludge bulking, operation of the plant had to be stopped and the operating parameters had to be adjusted to the changing water quality and temperature.

Figure 3

Changes in the concentration of total suspended solids in the water after the full-scale investigation.

Figure 3

Changes in the concentration of total suspended solids in the water after the full-scale investigation.

After a micro-sand ballasted coagulation process, the concentration of total suspended solids often exceeded 2 mg/L. The process which was least sensitive to changes in water quality and temperature was that of coagulation with flotation (Table A3, available online at http://www.iwaponline.com/wst/071/513.pdf).

The studies showed that iron, aluminium compounds and COD KMnO4 were effectively reduced to values acceptable for treated water, i.e. total iron ≤ 0.05 mg Fe/L, total aluminium ≤ 0.1 mg Al/L and COD KMnO4 ≤ 7.5 mg O2/L, respectively, in the course of coagulation and filtration. Detailed analyses of these parameters are outside the scope of this article.

The results of tests and observations of the performance of the analysed coagulation systems served as the basis for selecting the optimum water treatment technology. The efficiency of the coagulation process depends on raw water quality, the coagulant used and operational factors, including mixing conditions, coagulation dose and pH.

The floc is removed from the treated water by subsequent solid–liquid separation processes such as sedimentation or flotation and rapid gravity filtration. Effective operation of the coagulation process depends on selection of the optimum coagulant dose and also the pH value. The wide range of the temperature of the water is also very important.

The results of tests and performance monitoring of the equipment used for coagulation with sludge flotation

A change in the water density and viscosity at low temperatures did not have an impact on the effective separation of suspended solids which formed flocs on the water surface, removed by a scraper. The system was operated with a capacity of 10–12.5 m3/h without any interruptions and, regardless of the throughput, the TSS concentration did not exceed 2 mg/L in 86% of water samples collected before the filter. Along with the sludge 0.35% of water was removed.

The results of tests and performance monitoring of the equipment used for coagulation and sedimentation with sludge recirculation

Since the resulting coagulation product flocs had similar density to the density of water and their hydraulic resistance was low, they were easily removed by water. Consequently, the operation was interrupted quite often and the plant worked with a capacity 20–25 m3/h lower than the nominal capacity of 40 m3/h.

The concentration of suspended solids in water samples collected before the filter exceeded the limit of 2 mg/L approximately 40% of the time. The concentration of suspended solids in separated water with sludge oscillated around 2 g/L. Along with the sludge, 1% of water was removed.

The results of tests and performance monitoring of the equipment used for micro-sand ballasted coagulation and sedimentation

Due to high density of sand agglomerates and coagulation products, water temperature did not have a major impact on the effective performance of the plant. The system worked with a nominal capacity of 40–50 m3/h. The concentration of suspended solids in water past the hydrocyclone ranged from 0.30 to 0.36 g/L. The concentration of suspended solids in water samples collected before the filter exceeded the limit of 2 mg/L approximately 85% of the time and continued to increase along with the capacity growth from 40 to 50 m3/h. Along with the sludge 6.0–7.5% of water was removed.

CONCLUSIONS

The raw water was characterised by varying content of dissolved organic matter, with incidental peaks of the undissolved fraction, mainly as a result of the summer algal bloom and, subsequently, due to the products of dead phytoplankton decomposition. Therefore, pre-treatment of the water required, first of all, removal of organic matter.

Historically, coagulation has been employed in water treatment to decrease turbidity and colour and remove pathogens. However, optimum conditions for turbidity or colour removal are not always the same as those for natural organic matter (NOM) removal. In the baseline coagulation, the coagulation conditions are optimised for turbidity removal, whereas optimised coagulation refers to dose and pH conditions optimised, especially for organic matter reduction.

As demonstrated in the course of the full-scale investigation, the key treatment process was coagulation with aluminium sulphate conducted in an acidic water environment (pH = 6.5–6.9). The doses of the coagulant had to be determined in such a way as to make sure that after hydrolysis the pH value of water was not lower than 6.5. Most of the time, doses ranged from 4.5 to 6 mg Al/L and would guarantee TOC reduction in post-coagulation water to between 4 and 5 mg C/L.

The use of a PAC water purification system makes it possible to quickly react to changes taking place in the quality of water for a short time, so it is used mainly in cases where the water undergoes such changes or there is cyclic sudden increase in the degree of contamination. Reports in the literature show a very broad range of efficacy for removal of PAC dissolved organic matter present in the water. For example, the effectiveness of removal of organic substances at doses of PAC in an amount of 10 g/m3 was 20% (Sandrucci et al. 1995).

Technology research has shown that, to achieve TOC concentration below 4 mg C/L, PAC in an amount of 20 g/m3 needs to be added over a period of approximately 3–6 months, i.e. from the commencement of the algal bloom until a decrease in the share of phytoplankton decomposition products in the TOC load to be removed from the water.

In view of the study results and the observed system performance, it was concluded that the most advantageous coagulation process for the river water is coagulation with sludge flotation.

ACKNOWLEDGEMENTS

The author wishes to thank Professor Marian Błażejewski and the staff of AQUA S.A. companies from Poznan for the friendly atmosphere and effective collaboration during our investigation.

REFERENCES

REFERENCES
Baghoth
S. A.
Sharma
S. K.
Guitard
M.
Heim
V.
Croue
J. P.
2011
Removal of NOM-constituents as characterized by LC-OCD and F-EEM during drinking water treatment
.
Journal of Water Supply: Research and Technology-AQUA
60
(
7
),
412
424
.
Council Directive 98/83/EC on the quality of water intended for human consumption
.
Official Journal
L 330. 05.12.1998, 0032–0054
.
Guminska
J.
Klos
M.
2012
Analysis of post-coagulation flocs properties in aspect of coagulant choice
.
Environment Protection Engineering
38
(
1
),
102
114
.
Klos
M.
2013
Technological assumptions of coagulation and dissolved air troll system in water treatment process
.
Ochrona Srodowiska
35
(
3
),
39
43
.
Lawler
D. S.
Mikelonis
A. M.
Kim
I.
Lau
B. L. T.
Youn
S.
2013
Silver nano particle removal from drinking water: flocculation/sedimentation or filtration?
Water Science and Technology: Water Supply
13
(
5
),
1181
1187
.
Pruss
A.
Pruss
P.
2013
Organic pollutant removal from low alkalinity surface water
.
Ochrona Srodowiska
35
(
4
),
47
50
.
Pruss
A.
Maciołek
A.
Lasocka-Gomula
I.
2009
Effect of the biological activity of carbon filter beds on organic matter removal from water
.
Ochrona Srodowiska
31
(
4
),
31
34
.
Qin
J.
Oo
M. H.
Kekre
K. A.
Knops
F.
Miller
P.
2006
Impact of coagulation pH on enhanced removal of natural organic matter in treatment of reservoir water
.
Separation and Purification Technology
49
(
3
),
295
298
.
Sandrucci
P.
Merlo
G.
Genon
G.
Meucci
L.
1995
PAC Activity vs by-product precursors in water disinfection
.
Water Research
29
(
10
),
2299
2308
.
Sohn
J. G.
Amy
G.
Yoon
Y.
2007
Process-train profiles of NOM through a drinking water treatment plant
.
Journal of the American Water Works Association
99
(
6
),
145
153
.
Teixeira
M. R.
Miguel
L. M.
2011
The impact of natural matter seasonal variations in drinking water quality
.
Desalination Water Treatment
36
(
1–3
),
344
353
.
Vasyukova
E.
Proft
R.
Jousten
J.
Slavik
I.
Uhl
W.
2013
Removal of natural organic matter and trihalomethane formation potential in a full-scale drinking water treatment plant
.
Water Science and Technology: Water Supply
13
(
4
),
1099
1108
.
WHO
2011
Geneva Guidelines for Drinking-Water Quality
,
4th edn, World Health Organization, Geneva, Switzerland
.

Supplementary data