Effects of Pre-ozonation on the Removal of Natural Organic Matter and Haloacetic Acids Precursors by Coagulation

Jelena Molnar1, Jasmina Agbaba1, Božo Dalmacija1, Aleksandra Tubić1, Malcolm Watson1, Dejan Krčmar1 and Ljiljana Rajić1


1 University of Novi Sad Faculty of Sciences, Trg Dositeja Obradovića 3, 21000 Novi Sad, Republic of Serbia; e-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it ; tel: +381214852725; fax: +38121454065




This study investigates the influence of oxidation pretreatment by ozone on the efficacy of coagulation for removing natural organic matter (NOM) from groundwater, which is used as a drinking water source for the Central Banat region (Northern Serbia). By using coagulation alone, the maximum removal of NOM and haloacetic acids (HAAs) precursors from water was achieved with 200 mg FeCl3/l (23% DOC and 44% HAAFP).When it comes to combined processes, changes in the NOM structure during pre-ozonation resulted in significant improvement of the coagulation efficacy in the NOM removal, compared to coagulation alone. For the total NOM removal, the most effective was a combination of a higher ozone dose of 3.0 mg O3/mg DOC with 200 mg FeCl3/l (46% DOC), while optimal HAA precursors reduction of 88% HAAFP was achieved using a lower ozone and coagulant dose of 1.0 mg O3/mg DOC and 150 mg FeCl3/l.

Keywords: natural organic matter, haloacetic acids, ozonation, coagulation




In the region of Central Banat (Republic of Serbia), as well as in the majority of communities in the region, groundwater is used as a water supply source by industry. One of the basic problems and the most common cause of unacceptable drinking water quality in this region is the high content of natural organic matter (NOM) in the groundwater.

Natural organic matter is a complex mixture of organic compounds present in all natural waters which consists of a range of different compounds, from largely aliphatic to highly coloured aromatics. The hydrophobic part is rich in aromatic carbon, having phenolic structures and conjugated double bonds, while hydrophilic NOM contains a higher proportion of aliphatic carbon and nitrogenous compounds, such as carbohydrates, proteins, sugars and amino acids. Hydrophobic acids constitute the major fraction of aquatic NOM, accounting for more than half of the dissolved organic carbon (DOC) in water (Swietlik et al., 2004). These hydrophobic acids may be described as humic substances classified as humic acids and fulvic acids (Matilainen et al., 2011).

The presence of NOM can cause serious problems to drinking water quality and its treatment processes. NOM often contributes to offensive taste and odours in potential drinking water sources and acts as a carrier for metals and various harmful organic chemicals. In addition, NOM is considered to be a major precursor for carcinogenic disinfection by-products (DBPs) such as trihalomethanes (THMs) and haloacetic acids (HAAs), which can occur during chlorination and contribute to bacterial regrowth and biofilm formation in drinking water distribution systems (Krasner et al., 2006; Matilainen et al., 2011; Joseph et al., 2012). Considering the highly complex and heterogeneous structure of natural organic matter, the number of compounds which can be formed during the disinfection process is large, although on average, approximately 46% of the organohalogenated compounds are the THMs and di- and trichloroacetic acid (WHO, 2011). Several studies reported that these compounds are related to the occurrence of cancer, growth retardation, spontaneous abortion, and congenital cardiac defects (Yang et al., 2000; Badawy et al., 2012).

Therefore, the U.S. Environmental Protection Agency (USEPA) regulates a maximum allowed concentration of 60 μg/L for the sum of five HAAs (monochloro- (MCAA), dichloro- (DCAA), trichloro- (TCAA), monobromo- (MBAA) and dibromoacetic acid (DBAA)) in drinking water, according to their current regulation guidelines (Farré et al. 2011). Furthermore, according to the Serbian Regulation (Official Gazette SRJ No. 42/98-4) and World Health Organization Guidelines (WHO, 2011), the maximum allowable concentrations for DCAA and TCAA in drinking water are regulated at 50 and 100 μg/L, respectively.

The increasingly stringent requirements for drinking water quality make it necessary to reduce the content of NOM (particularly humic and fulvic compounds) as the main DBPs precursors material. By applying conventional technologies, NOM can be removed by coagulation and flocculation, ion exchange, adsorption on activated carbon, and can also be removed by membrane filtration (Park and Yoon, 2007; Katsumata et al., 2008). Generally, the most common and economically feasible processes for NOM removal are considered to be coagulation and flocculation, with likely NOM removals of around 29-70% as DOC. Chemical coagulation is achieved by addition of inorganic coagulants, such as aluminium and iron salts. When added to water, aqueous Al(III) and Fe(III) salts are dissociated to their respective trivalent ions. Afterwards, they are hydrolyzed and form several soluble complexes possessing high positive charges, thus adsorbing onto the surface of the negative colloids (Matilainen et al., 2010; Joseph et al., 2012).

In order to improve the efficacy of the coagulation process in NOM removal and disinfection by-products formation control, especially in water rich in NOM, before the coagulation process, an oxidative pretreatment with ozone can be applied. Ozone has the proven ability to decrease the concentration of DBP precursors and a number of microorganisms, and reacts with organic substances to increase their biodegradability (Kasprzyk-Horden et al., 2003; Chiang et al., 2009). Therefore, the effects of the pre-ozonation can be roughly classified as direct (direct removal of NOM, removal of colour, taste and odour) or indirect (increased removal of NOM by coagulation, flocculation and direct filtration processes) (Beltrán, 1995; Agbaba et al., 2003).

The main objective of this research was to investigate the effect of ozonation as a pretreatment process on the efficacy of coagulation for decreasing NOM content. In addition, the influence of combined pre-oxidation and coagulation were also investigated in terms of haloacetic acid precursors content in water rich in natural organic matter.


Materials and methods


Groundwater from Central Banat (Province of Vojvodina, Republic of Serbia) was used in the laboratory investigation. A mixture of water was taken from the two water-bearing layers at depths of 40-80 and 100-150 m, wells which are being used for the water supply in the town of Zrenjanin. The general characteristics of the raw groundwater are given by Molnar et al. (2012a). The groundwater investigated is rich in NOM (9.85±0.18 mg/L DOC, 0.493±0.10 cm-1 UV254, 309±15 μg/L HAAFP). The NOM is mostly of hydrophobic character (65% fulvic acid-FAF and 14% humic acid fraction-HAF), while the hydrophilic fraction makes up only 21% of DOC (Molnar et al., 2012b).

A standard solution mixture containing 2000 µg/mL of 6 HAAs in methyl tert-butyl ether was purchased from Supelco. Methyl tert-butyl ether used during the experiment was obtained from J.T. Baker and was Organic Residue Analysis grade. All other chemicals used were analytical grade and were used without further purification.

Experimental procedure

NOM content was measured in the raw groundwater as well as in the samples after each stage of the procedure: 1. coagulation and flocculation, 2. pre-ozonation 3. pre-ozonation, coagulation/flocculation. All experiments (DOC, UV absorbance at 254 nm, specific UV absorbance (SUVA) and HAA formation potential (HAAFP)) were performed in triplicate and are presented as average values. Standard deviation of measurement was less than 10%.

The efficiency of NOM removal by the processes of coagulation and flocculation at pH 7 was examined by the conventional jar test in 1-L beakers on a FC6S Velp scientifica apparatus at room temperature (23°C). Coagulation was performed with fast stirring (120 rpm) for 2 min, followed by flocculation with slow stirring (30 rpm) for 30 min. In the former process, 4% iron(III) chloride solution was used as a coagulant at doses of 50-200 mg FeCl3/L, and in the latter, anionic flocculant A-110 was applied at a dose of 0.5 mg/L.

After clarification and supernatant removal, the water sample was filtered through a cellulose nitrate membrane filter (0.45 µm) to remove the remaining colloids.

Ozonation was carried out at the natural pH of the water of 7.86 ± 0.20, in a 2 L glass column which was 85 mm in diameter. Ozone was generated electrochemically in an Argentox ozone generator with a 1 g/h capacity, and introduced to the water via a diffuser at the bottom of the column. The overall volume of water subjected to ozonation was 1.5 L, and samples were taken via a tap placed at the bottom of the column. The applied ozone doses were in the range of 0.4-3.0 mg O3/mg DOC. All chemicals used were of analytical grade.

Analytical methods

The changes in water NOM content were followed on the basis of DOC, UV254, SUVA and HAA formation potential.

Water samples were analyzed for DOC content on an Elementar LiquiTOCII, involving combustion at 850 °C. UV254 absorbance measurements were performed according to standard methods (APHA, 1998) on a spectrophotometer at a wavelength of 254 nm, with a 1-cm quartz cell, and the SUVA (m-1lmg-1) was calculated.

The concentration of ozone transferred to the water was calculated from the difference in the input and output ozone concentrations in the gas phase, which were measured under the standard conditions (273 K and 101.3 kPa) by iodometric titration (APHA 1998).

HAAFP in the raw water and in the samples after treatments were determined according to the standard method (APHA 1998). Formation potentials are analysed by simulating disinfection conditions in the laboratory. HAAs precursors content was measured using methods based on EPA Method 552. Analysis of HAAs was performed on a GC/μECD (Agilent 6890N) using a DB-608 column.The PQLs for the HAAs were five times the MDL values, and were 1.65, 0.63, 2.80, 0.20, 1.23 and 0.30 µg/L for monochloroacetic acid (MCAA), monobromoacetic acid (MBAA), dichlororacetic acid (DCAA), trichloroacetic acid (TCAA), bromochloroacetic acid (BCAA) and dibromoacetic acids (DBAA) respectively.