Journal of Hazardous Materials 239–240 (2012) 348–354

Contents lists available at SciVerse ScienceDirect

Journal of Hazardous Materials

journal homepage: www.elsevier.com/locate/jhazmat

Effects of ozone and ozone/peroxide pretreatments on disinfection byproduct

formation during subsequent chlorination and chloramination
Xin Yang a,b,∗ , Jinfeng Peng b , Baiyang Chen c,∗∗ , Wanhong Guo b , Yongmei Liang b , Wei Liu b , Lu Liu b
a
Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, Guangzhou 510275, China
b
School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China
c
Harbin Institute of Technology Shenzhen Graduate School, Shenzhen 518055, China

h i g h l i g h t s

 O3 reduced THM and trihalogenated HAAs precursors in waters with SUVA > 2 L mg−1 m−1 .
 O3 and O3 /H2 O2 reduced HAN formation.
 O3 /H2 O2 generated more CH and HKs during chlorination than O3 .
 O3 or O3 /H2 O2 pretreatments combined with chloramination process is favored in controlling DBPs.

a r t i c l e i n f o a b s t r a c t

Article history: Ozone (O3 ) and ozone/hydrogen peroxide (O3 /H2 O2 ) can be used in water treatment facilities to remove
Received 7 July 2012 many organic micropollutants with taste, odor, and color implications. The effects of O3 and O3 /H2 O2
Received in revised form 8 August 2012 on the formation of disinfection byproducts (DBPs) in subsequent chlorination and chloramination pro-
Accepted 4 September 2012
cesses, however, are not well determined. In this study, we compared the yields of a series of regulated
Available online 11 September 2012
and emerging DBPs during sequenced O3 –Cl2 , O3 /H2 O2 –Cl2 , O3 –NH2 Cl, and O3 /H2 O2 –NH2 Cl oxidation
of 11 samples, each with different hydrophobicity, bromide concentration, soluble microbial products,
Keywords:
and humic substances. For most water, pretreatment with O3 and O3 /H2 O2 increased the formation of
Chlorination
Chloramination
chloral hydrate (CH), trichloronitromethane (TCNM) and haloketones (HKs) but lowered the yields of
Ozonation haloacetonitriles (HANs) during chlorination processes. Compared with O3 alone, O3 /H2 O2 in combina-
O3 /H2 O2 tion generated more CH and HKs during chlorination, and their extents of formation appeared to depend
Emerging disinfection byproducts on the O3 doses. In terms of chloramination, both O3 and O3 /H2 O2 reduced THM, HAA, and HAN formation
significantly without increasing CH, TCNM, or HKs. These results suggest that O3 or O3 /H2 O2 pretreat-
ments may provide some benefits for the chloramination process in controlling regulated and emerging
DBPs in waters without high bromide content.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction long-term guard for microbial control, chlorine (Cl2 ) or monochlo-

Ozone, a strong oxidant, has been used for disinfection, taste and
odor control, and color removal in some water treatment facilities
[1]. Ozonation can effectively remove a variety of organic microp-
ollutants [2]. Aided by H2 O2 , O3 can even remove certain resistant
micropollutants such as iodinated X-ray contrast agents [3]. How-
ever, because ozone reacts with many types of compounds, it is
rapidly consumed and no residual can be maintained. To provide a
∗ Corresponding author at: School of Environmental Science and Engineering, Sun
Yat-sen University, Guangzhou 510275, China. Tel.: +86 20 39332690.
∗∗ Corresponding author. Tel.: +86 13480727605.
E-mail addresses: yangx36@mail.sysu.edu.cn (X. Yang),
polar chen@hotmail.com (B. Chen).
ramine (NH2 Cl) is often used as a secondary disinfectant in distri-
bution systems. The impact of O3 and O3 /H2 O2 on the formation of
disinfection by-products (DBPs) during subsequent chlorination or
chloramination processes thus requires more attention.
Previous research has identified mixed effects of preozona-
tion for regulated DBPs. Although most studies reported that
preozonation can reduce the formation of trihalomethanes
(THMs) and haloacetic acids (HAAs) during chlorination or chlo-
ramination [4–6], some studies showed that the preozonation
process enhanced THM formation [7,8]. Similarly, the impacts
of O3 /H2 O2 oxidation on DBP formation vary widely depending
on species. It has been reported that, compared with O3 alone,
O3 /H2 O2 can remove 21% more of the HAA formation potential [9],
equally decrease or maintain THM formation potential [10], and
increase the formation of dissolved organic halogens. Formation

0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jhazmat.2012.09.006
 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354 349

of emerging nitrogenous DBPs (N-DBPs) under sequenced Table 1

Raw water quality.
ozonation–chlorination (O3 –Cl2 ) and ozonation–chloramination
(O3 –NH2 Cl) processes has been studied infrequently. An increase Sample no. pH DOC UV254 SUVA Br (␮g/L)
in halonitromethanes (HNMs) with ozone pretreatment was (mg/L) (cm−1 ) (L mg−1 m−1 )
observed for selected waters [11,12]. Although haloacetonitrile 1 7.3 9.8 0.15 1.5 <15
(HAN) precursors were reduced at high ozone dosages, they were 2 7.3 20.6 0.31 1.5 <15
enhanced at low ozone dosages [12]. The formation of many other 3 7.4 13.9 0.22 1.6 <15
4 7.4 20.2 0.44 2.0 <15
DBPs during O3 /H2 O2 –Cl2 processes is not currently known.
5 7.3 10.2 0.15 1.5 <15
The reasons for these discrepancies thus require more research. 6 7.3 10.6 0.20 1.9 20
DBP formation is known to depend on many factors, including 7 7.4 5.0 0.12 2.4 20
precursor characteristics, reaction conditions (e.g., pH and tem- 8 7.3 6.9 0.11 1.6 20
9 7.3 5.6 0.20 3.6 25
perature), and oxidant type. On the one hand, both ozone and
10 7.6 6.4 0.09 1.4 25
H2 O2 induced hydroxyl radical are strong electrophilic agents that 11 7.6 5.4 0.12 2.2 1300
can break down nucleophilic centers on natural organic matter
(NOM), which leads to fewer precursors for some DBPs [13], but
on the other hand, ozone and hydroxyl radical may oxidize NOM to absorptivity of 2950 M−1 cm−1 [21]. A stock of free chlorine (HOCl)
produce greater amounts of methyl-ketone-like structures, which solution was prepared with 5% sodium hypochlorite (NaOCl) (from
are considered major precursors of THMs and total organic halide Sigma) diluted to 1100 mg/L as Cl2 . Monochloramine (NH2 Cl) solu-
(TOX) [14,15]. The two contrary mechanisms hence complicate the tions were prepared daily by reacting ammonium chloride and
effects of O3 and O3 /H2 O2 pretreatment of waters and make it dif- sodium hypochlorite at a weight ratio of 4 mg/L Cl2 to 1 mg/L
ficult to predict the DBP species [16]. N-NH4 + . The resulting solutions were standardized by diethyl-p-
NOM is precursor of DBPs. O3 preferentially reacts with unsat- phenylene diamine (DPD) ferrous titration.
urated bonds to oxygenated saturated functional groups in NOM,
such as aldehydic, ketonic and carboxylic groups [13,17,18]. The 2.2. Sample collection
addition of H2 O2 to O3 led to higher production rate of OH radicals.
However, the higher formation rate of OH is at the cost of a higher O3 The raw water samples were collected from 11 locations along
consumption. For example, the decomposition of O3 is accelerated the Pearl River in the Guangdong Province of China. The sampling
in the presence of H2 O2 at pH values higher than 5 [19]. In contrast sites are shown in Fig. S1 in the Supporting information. After col-
to O3 , OH radicals are believed to be non-selective oxidants, which lection, samples were shipped to the laboratory and kept at 4 ◦ C
react very fast with vast majority of organic compounds in water until use. All samples were filtered with a 0.45 ␮m membrane to
[13], among which organic molecules with unsaturated bonds react remove particles. Dissolved organic carbon (DOC) concentration,
faster with OH radicals than saturated C C bonds or C H bonds UV absorbance at 254 nm (UV254 ), and bromide ion concentration
[16]. Reactions of OH radicals with organic compounds can lead to were analyzed. Specific UV absorbance (SUVA) was calculated as
the formation of low-molecular mass acids, such as carboxylic acids UV254 divided by DOC. The water properties are summarized in
[20]. At equivalent O3 dosages, O3 /H2 O2 produced greater concen- Table 1.
trations of assimilable organic carbon (5–52%), aldehydes (31–47%),
and carboxylic acids (12–43%) compared to O3 alone [18]. There- 2.3. Experimental procedures
fore, pretreatment of water with O3 and O3 /H2 O2 may change NOM
properties, resulting in different DBP formation during subsequent Before reaction, all samples were diluted to 2 mg/L as DOC, and
chlorination or chloramination, which requires exploration. 10 mM phosphate buffer was added to pH 7.5. An O3 dose of 3 mg/L
This study is designed to investigate the effects of O3 and was applied to the samples, after which they were mixed for 5 min.
O3 /H2 O2 pretreatments on the formation of a series of DBPs during The samples for O3 /H2 O2 pretreatments were prepared by mixing
subsequent chlorination and chloramination processes. Six classes 1.5 mg/L H2 O2 with samples followed by O3 addition [10]. After
of carbonaceous DBPs and N-DBPs are selected that represent both reaction, a stoichiometric amount of sodium thiosulfate was added
regulated DBPs such as THMs and HAAs and emerging DBPs includ- to quench the residual O3 and H2 O2 . One aliquot of sample was
ing haloketones (HKs), chloral hydrate (CH), trichloronitromethane used for analysis of UV254 , DOC, and fluorescence. Another aliquot
(TCNM), and haloacetonitriles (HANs). Grab samples were obtained of sample was chlorinated or chloraminated for DBP formation
from 11 sites along China’s Pearl River, a waterbody typically dete- potential tests. In chlorination tests, excessive chlorine was applied
riorated during the past 10 years owing to the release of treated and to ensure a residual of at least 0.5 mg/L after incubation for one
untreated domestic wastewaters. Sample characteristics, including day. Chloramination tests were conducted in the same manner as
hydrophobicity, bromide concentration, protein-like products, and chlorination tests except that the preformed monochloramine was
soluble microbial substances, were examined. kept for a 3-day period. The effect of ozone doses on DBP forma-
tion was conducted by adding various doses of ozone (1, 2, 4 and
6 mg/L) to two water samples (No. 2 and 6). After ozonation for
2. Materials and methods 5 min, chlorination tests were conducted following the procedures
above. All samples were stored headspace-free in the dark at room
2.1. Reagents and solutions temperature (22 ± 1 ◦ C). After incubation, samples were quenched
by ascorbic acid and extracted immediately with methyl tert-butyl
Calibration standards, internal standards, and surrogate stan- ether (MTBE). The extracts were then subjected to DBP analysis by
dards for THMs, HAAs, HANs, HKs, CH, and TCNM analyses were gas chromatograph.
purchased from Supelco. H2 O2 solution was purchased from Sigma.
Gaseous ozone was generated with an ozone generator (CF-G- 2.4. Analytical methods
3-010 g, Guolin, Qingdao) and introduced immediately into the
ultrapure water using a diffuser placed at the bottom of the reac- DOC concentrations were measured using a total organic carbon
tor. The ozone concentration of the stock solution was determined analyzer (Shimadzu, 5000A) according to Standard Method 5310
using the direct UV absorbance method at 258 nm with a molar [21]. UV254 was measured by a UV–visible spectrophotometer
350 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354
(Shimadzu, Multispec-1501). Bromide was measured using an ion
chromatograph (Metrohm 882 compact IC plus) equipped with an
anionic column (MetrosepA supp 5) following Standard Method
4110 [21]. Fluorescence excitation–emission matrix (EEM) mea-
surements were conducted with a Hitachi F-4500 spectrometer,
which used a xenon excitation source with slits set to 10 nm for
both excitation and emission. The excitation wavelengths were set
from 200 to 400 nm in 2 nm increments, and the emission wave-
lengths were detected from 290 to 500 nm in 1 nm steps. Water
Raman peak (excitation at 350 nm) was checked before and after
sample analysis. No change in fluorescence intensity was observed.
Analyses of HAAs based on USEPA Method 552.2 were carried
out on a gas chromatograph (Agilent 7890) with an electron cap-
ture detector (ECD). Analyses of the other DBPs were based on
USEPA method 551.1. The column was an HP-5 fused silica capil-
lary column (30 m × 0.25 mm I.D. with 0.25 ␮m film thickness, J&W
Scientific). The minimum reporting limit (MRL) for all compounds
except HAAs was 0.2 ␮g/L. The MRL for HAAs ranged from 0.4 to
2 ␮g/L for individual species. The percentage recovery ranged from
94% to 109%. Experiments were carried out in duplicate. Relative
standard deviation of the two measurements was generally below
15%.
3. Results and discussion
3.1. Characterization of organic matter before and after O3 and
O3 /H2 O2
The Pearl River receives discharge of treated effluents from
four municipal wastewater treatment plants (Fig. S1). The samples
exhibited various DOC, SUVA, and bromide characteristics. Sample
No. 9 had the highest SUVA value (3.6 L mg−1 m−1 ), which indi-
cates NOM dominance, but most of the water samples exhibited
low SUVA values ranging from 1.4 to 2.4 L mg−1 m−1 , which indi-
cates that their NOM is less aromatic and probably impacted by
wastewater effluent [22,23]. As a result of tidal backflow action,
sample No. 11 collected from the downstream Pearl River contains
a high amount of bromide [24].
Typical fluorescence EEMs (sample Nos. 2, 3, and 9) are displayed
in Fig. 1. The fluorescence EEMs showed maxima at an emis-
sion wavelength of 340 nm and excitation wavelengths of 225 nm
and 290 nm, respectively, which are representative of the protein-
like substances or soluble microbial substances in wastewater
and biologically treated wastewater [25] (Fig. S2 in the Suppor-
ting information). The fluorescence EEMs of sample No. 9 exhibited
maxima at an emission wavelength of 425 nm and excitation wave-
lengths of 225 nm and 325 nm, which indicates the presence of
fulvic and humic-like substances. Overall, protein-like and solu-
ble microbial substances dominated (70–80%) the NOM of selected
samples, as shown in the Fig. S3 in the Supporting information.
In general, the UV254 values of the samples decreased by 31–77%
after O3 pretreatment and 16–64% after O3 /H2 O2 pretreatment.
As ozone is more selective for the removal of double bonds and
activated aromatic compounds than is hydroxyl radical [13], the
decreases in UV254 levels after O3 /H2 O2 pretreatment compared
with O3 alone suggest that the addition of H2 O2 to O3 may have
formed some hydroxyl radicals. O3 and O3 /H2 O2 oxidation showed
a minor DOC removal, which ranged from unchanged to 15% reduc-
tion. It is consistent with the literatures that reported the typical
reduction of DOC from 5% to 20% under typical ozone dose of
2–5 mg/L [26,27].
The fluorescence EEMs of selected samples (Nos. 2, 3, and 9)
after ozone pretreatment are displayed in Fig. 1. Although O3 /H2 O2
decreased the overall intensities of fluorophores to a larger extent
than did ozone, the EEM volume reductions in humic and fulvic
acid-like regions were constant regardless of O3 or O3 /H2 O2 use.
The normalized region-specific EEM volumes (˚i,n ), as calculated
by the method of Chen et al. [25], are shown in Fig. S3 in the Suppor-
ting information. On average, the EEM volumes decreased by 20%
to 40% after ozone pre-treatment. In region IV, which represents
the soluble microbial substances, O3 /H2 O2 removed approximately
10% more NOM than ozone alone. Soluble microbial products have
been characterized as high in polysaccharides, proteins, nucleic
acids, organic acids, and amino acids but low in aromatic compo-
nents [28], which appeared to favor reaction with hydroxyl radical
rather than ozone in this study.
3.2. Formation of DBPs
3.2.1. DBP formation by Cl2 and NH2 Cl
Overall, chloramination produced many fewer DBPs than
chlorination for the species examined in this study (Fig. 2). Tri-
haloacetic acids (THAAs) are the sum of trichloro-, bromodichloro-,
dibromochloro- and tribromoacetic acids (TCAA, BDCAA, DBCAA,
and TBAA), and dihaloacetic acids (DHAAs) are the sum of dichloro-
, bromochloro-, and dibromoacetic acids (DCAA, BCAA, and DBAA).
The median levels of total THMs, THAAs, DHAAs, and HANs were
0.22, 0.06, 0.045, and 0.021 ␮mol/mg C, respectively, after chlorina-
tion of raw samples. In comparison, the total THM, DHAA, and HAN
contents were 0.018, 0.02, and 0.0022 ␮mol/mg C, respectively,
after chloramination of samples. THAAs were not detected after
chloramination except in sample No. 11. The median concentra-
tions of CH, HKs, and TCNM were 0.06, 0.003, and 0.002 ␮mol/mg C,
respectively, after chlorination of the raw samples, and the yield of
TCNM was 0.001 ␮mol/mg C during chloramination. The concen-
trations of THMs, THAAs, and HANs were reduced by more than
80%, and the formation of DHAAs and HANs was reduced by 12–68%
when monochloramine was used. CH and 1,1,1-trichloroprapone
(1,1,1-TCP) were detected in some chlorinated samples but not in
chloraminated waters. No linear correlations were found between
the SUVA values and the formation potentials of THMs, DHAAs,
or THAAs (Fig. S4 in the Supporting information). Sample No. 11
exhibited the highest formation potential for THM, HAA, and HAN,
probably because of its higher bromide content than the other sam-
ples, as discussed below.
3.2.2. DBP formation with O3 –Cl2 and O3 –NH2 Cl
Compared with chlorination alone, THM formation in down-
stream samples (Nos. 8–11) was reduced by 1–33% after ozonation,
whereas THM formation in upstream samples (Nos. 1–7) was
increased by 1–45%. For chloramination, the THM reductions
ranged from 20% to 49% for all samples with ozone pretreat-
ment (Fig. S5). Both increases and decreases in THAA formation
were observed during chlorination with ozone pretreatment, but
no detectable THAAs were found if monochloramine was applied,
except in sample No. 11 (Fig. S6). The concentrations of DHAAs were
elevated in samples other than Nos. 10 and 11 during the O3 –Cl2
process. In contrast, a median reduction (22%) of DHAAs was found
for the sequenced O3 –NH2 Cl process (Fig. S7). HAN concentrations
were reduced in most samples for both the O3 –Cl2 and O3 –NH2 Cl
processes (Fig. S8). The only exception again was sample No. 11,
which contained 1300 ␮g/L of bromide in raw water. O3 –Cl2 treat-
ment increased TCNM formation by two to five times and O3 –NH2 Cl
increased it by one to two times (Fig. S9). Increases in the formation
of CH and 1,1,1-TCP were also observed during the O3 –Cl2 process
(Fig. S10).
In raw water samples, the water quality parameter SUVA was
an important parameter for determination of the removal rates
of some DBP precursors. During the O3 –Cl2 process, THM and
THAA formation were usually less in waters with high SUVA values
(>2 L mg−1 m−1 ) but were not altered or even enhanced in waters
 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354 351

Fig. 1. Fluorescence EEMs of sample No. 2 (a–c), sample No. 3 (d–f), and sample No. 9 (g–i). First column: raw water samples; second column: samples after O3 pretreatment;
third column: samples after O3 /H2 O2 pretreatment.

with low SUVA values (<2 L mg−1 m−1 ) (Fig. 3). On the basis of ear-
lier studies, it is generally agreed that preozonation of waters with
SUVA values greater than 2 L mg−1 m−1 can reduce THM and THAA
formation, but the effect of ozonation is less certain on waters
with SUVA values less than 2 L mg−1 m−1 [6]. This study hence is
in accord with previous research.
Preozonation enhanced DHAA formation in most of the chlori-
nated samples (Fig. S7 in the Supporting information). The trend of
DHAA precursor removal was opposite to that of THAA precursors.
A previous study demonstrated that during chlorination the
phenolic hydroxyl group is likely to form THAA, whereas carboxyl
and alcoholic hydroxyl groups are prone to form DHAA [29]. The
net increase in DHAA formation in this study thus may suggest
a conversion of these functional groups. Similarly, ozonation
is known to produce aldehydes and methyl ketone, which are
precursors of CH and 1,1,1-TCP, respectively [30]. In this study,
increased levels of 1,1,1-TCP and TCNM were observed in O3 –Cl2
and O3 –NH2 Cl treated waters. CH formation was not detected in
O3 –NH2 Cl waters, but it was enhanced during the O3 –Cl2 process.
3.2.3. DBP formation with O3 /H2 O2 –Cl2 and O3 /H2 O2 –NH2 Cl
During the chlorination process, the difference in THM for-
mation between O3 and O3 /H2 O2 pretreatments was within a
narrow variation (5%) for all samples except for sample No. 6. How-
ever, during chloramination, O3 /H2 O2 reduced THMs by 40–99%,
a much higher precursor inactivation than that by O3 (20–49%
reduction) (Fig. S5). Both increases and decreases in THAA, DHAA,
and TCNM formation were detected for the O3 /H2 O2 –Cl2 process
(Figs. S6, S7, and S9). Generally, O3 –Cl2 reduced HAN more than
did O3 /H2 O2 –Cl2 (15% more on average). In contrast, O3 /H2 O2
enhanced the formation of CH and 1,1,1-TCP by 2–10% and 3–27%,
respectively, for most of the samples.
352 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354

.4 .10
Chlorination (a)
outlier
90th percentile
.08
75th percentile
.3

( mol/mg C)

( mol/mg C)
Median

DBP yields (µmol/mg

DBP yields (µmol/mg

25th percentile
10th percentile .06

.2

.04

.1
.02

0.0 0.00
THM THAA DHAA CH HK HAN TCNM
.05 .010
Chloramination (b)

.04 .008
mol/mg C)

( mol/mg C)
DBP yields ((µmol/mg

DBP yields (µmol/mg

.03 .006

.02 .004

.01 .002

ND
0.00 0.000
THM THAA DHAA CH HK HAN TCNM

Fig. 2. The formation of DBPs from (a) chlorination and (b) chloramination without and with ozone or O3 /H2 O2 pretreatment (pH = 7.2, temperature = 22 ± 1 ◦ C, 1-day
reaction for Cl2 and 3-day reaction for NH2 Cl, number of samples = 11). First bar: without pretreatment; second bar: O3 pretreatment; third bar: O3 /H2 O2 pretreatment. ND,
not detected. (Note: THAAs were not detected in most samples except sample No. 11 during chloramination and are not displayed here.)

In this study, reduction of THM formation was found consis- 20% less HAA formation from chloramine than from free chlo-
tently during the chloramination processes, probably because of rine. HANs dropped from 0.018 ␮mol/mg C (median value) during
the lower oxidation ability of chloramines relative to chlorine the O3 /H2 O2 –Cl2 process to 0.0018 ␮mol/mg C (median value) dur-
(Fig. 2b). Whereas O3 /H2 O2 –Cl2 resulted in 0.18–0.31 ␮mol/mg C ing the O3 /H2 O2 –NH2 Cl process. Although concentrations of other
of THM, O3 /H2 O2 –NH2 Cl yielded less than 0.005 ␮mol/mg C emerging DBPs, such as TCNM, CH, and 1,1,1-TCP, were greatly
of THM. Formation of THAAs and DHAAs was greatly sup- enhanced by the O3 /H2 O2 –Cl2 process, they were undetectable
pressed by chloramination as well, resulting in approximately using O3 /H2 O2 –NH2 Cl. Therefore, the control of THM, HAA, and
many emerging DBPs may benefit from the combined use of ozone
and monochloramine.
80
Reduction in DBP formation after ozonation (%)

THM: this study

60 THM: from literature
THAA: this study .10
THAA: from literature CH .014
TCNM and 1,1,1-TCP yield (µmol/mg C)

40 TCNM
1,1,1-TCP
.08 .012
20
CH yield (µmol/mg C)

.010
0 .06
.008

-20
.04 .006

-40
.004
.02
-60
.002

-80
0.00 0.000
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0 1 2 3 4 5 6 7

SUVA (L mg-1m-1) Ozone dosage (mg/L)

Fig. 3. The relations between SUVA values and the percentage changes in THM and Fig. 4. The effect of ozone dosage on the formation of CH, TCNM, and 1,1,1-TCP
THAA formation after preozonation during subsequent chlorination. Data plotted during O3 –Cl2 oxidation (pH = 7.2, temperature = 22 ± 1 ◦ C, ozone to H2 O2 ratio is
include the results from this study and from the literature (Hua and Rechkow [6]; 1/0.4 mg mg−1 , 1-day reaction with Cl2 ). Black symbol: sample No. 2; white symbol:
Hu et al. [11]). sample No. 6.
 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354
Fig. 5. The effect of pretreatment on formation of DBPs during chlorination of (a) sample No. 7 and (b) sample No. 11 after O3 or O3 /H2 O2 pretreatment (pH = 7.2,
temperature = 22 ± 1 ◦ C, 1-day reaction with Cl2 ). First bar: without pretreatment; second bar: O3 pretreatment; third bar: O3 /H2 O2 pretreatment.
The effect of ozone dose was investigated in selected samples
(Nos. 2 and 6) by applying varying concentrations of O3 and H2 O2
while keeping constant ratios of O3 to H2 O2 . As shown in Fig. 4,
the increases in CH, TCNM, and 1,1,1-TCP concentrations were pos-
itively related to the increases in initial ozone dose. In contrast, the
formation of THMs and HANs did not vary significantly (less than
10%) with increasing ozone dose. Additionally, the formation of
DBPs varied widely in two tested waters. Sample No. 6 consistently
exhibited higher DBP formation than did sample No. 2 (Figs. S5–10),
which implies that the water characteristics also played an impor-
tant role in DBP speciation.
3.2.4. Effect of bromide on DBP formation
To obtain a clear understanding of the incorporation of bromide
on DBP formation, we employed the bromine incorporation fac-
tor (BIF) to indicate the degree of bromine substitution in DBPs.
The BIF is calculated for each DBP group using the following equa-
tion [31,32], and a higher value indicates a higher reactivity with
bromine than chlorine.
Number of bromide atom × DBP molar concentration
BIF =
Total DBP molar concentration
Here DBP refers to a certain class of DBP, e.g., THMs, HAAs, and
HANs. The BIFs for THMs, HAAs, and HANs during O3 /H2 O2 –Cl2 and
O3 –Cl2 treatments are displayed in Fig. S11 in the Supporting infor-
mation. BIF values in most of the samples were less than 0.5, except
for sample No. 11, which had BIF values of 1.4 for HAAs and 1.9 for
THMs. Bromine incorporation did not exhibit large variations dur-
ing chlorination among raw samples, samples pretreated with O3 ,
and samples pretreated with O3 /H2 O2 across many types of DBPs.
This suggests that the ozone and ozone/peroxide pretreatments
were able to alter the relative chlorine and bromine substitutions.
Clearly, high bromide concentration in water corresponds to
high BIF. In a typical sample without seawater influence (sample
No. 7), the dominant chlorinated species were chloroform, DCAA,
353
TCAA, DCAN, and TCNM, as shown in Fig. 5, and O3 and O3 /H2 O2 pre-
treatments reduced the formation of THMs and HANs but increased
the formation of 1,1,1-TCP, CH, and TCNM during subsequent
chlorination. But for sample No. 11, which contains 1300 ␮g/L of
bromide owing to the seawater backflow tidal effect, brominated
DBPs dominated the THMs and HAAs. Bromochloroacetonitrile
(BCAN) and dibromoacetonitrile (DBAN) were also produced in the
chlorinated sample No. 11. In addition, unlike other waters, the
total yields of THMs, HAAs, and HANs were much higher in sample
No. 11, and the formation of total THMs and HANs was enhanced
during O3 –NH2 Cl oxidation processes, as shown in Figs. S5 and S8.
The results thus suggest that the benefits of using O3 –NH2 Cl and
O3 /H2 O2 –NH2 Cl for DBP control might be compromised in high-
bromide waters.
4. Conclusions
This study evaluated the effects of O3 and O3 /H2 O2 pretreat-
ments on the formation of several regulated and emerging DBPs
during subsequent chlorination and chloramination processes. For
chlorination, O3 and O3 /H2 O2 pretreatment exhibited obvious ben-
efit for THM and HAA control for samples with SUVA values greater
than 2.0, but they simultaneously enhanced the formation of some
emerging DBPs, such as CH, 1,1,1-TCP, and TCNM. In contrast, O3
and O3 /H2 O2 generally were more advantageous for controlling
THM, DHAA, and HAN formation when used prior to chlorami-
nation, yet the use of O3 in the chloramination process should
be cautious for waters with high bromide concentrations. Higher
ozone dosages had minor effects on regulated DBPs but increased
concentrations of the emerging DBP species. Regardless of the
amounts of DBPs formed, O3 and O3 /H2 O2 pretreatment did not
alter the bromide ratios in many waters and for many types of
DBPs, which suggests that these pretreatments has not influenced
the subsequent chlorine or bromine substitutions.
354 X. Yang et al. / Journal of Hazardous Materials 239–240 (2012) 348–354
Acknowledgments
We thank the National Natural Science Foundation of China
(Project No. 51008316), China’s Fundamental Research Funds for
the Central Universities (Project No. 11lgzd15), and the Natu-
ral Science Foundation of Guangdong province, China (Project
No. 10451027501005074) to financially support this study. The
work is also supported by the Program for New Century Excel-
lent Talents in University (NCET-10-0860) and the Natural Scientific
Research Innovation Foundation in Harbin Institute of Technology
(HIT.NSFIR.2011127).
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.jhazmat.2012.09.006
References
[1] B. Langlais, D.A. Reckhow, D.R. Brink, Ozone in Water Treatment: Application
and Engineering, American Water Works Association Research Foundation and
Lewis Publisher, Chelsea, MI, 1991.
[2] P. Westerhoff, Y. Yoon, S. Snyder, E. Wert, Fate of endocrine-disruptor, pharma-
ceutical, and personal care product chemicals during simulated drinking water
treatment processes, Environ. Sci. Technol. 39 (2005) 6649–6663.
[3] J.E. Drewes, P. Fox, M. Jekel, Occurrence of iodinated X-ray contrast media in
domestic effluents and their fate during indirect potable reuse, J. Environ. Sci.
Health A: Tox. Hazard. Subst. Environ. Eng. 36 (2001) 1633–1645.
[4] D.A. Reckhow, P.C. Singer, The removal of organic halide precursors by preo-
zonation and alum coagulation, J. Am. Water Works Assoc. 76 (1984) 151–157.
[5] J.G. Jacangelo, N.L. Patania, K.M. Reagen, E.M. Aieta, S.W. Krasner, M.J. McGuire,
Ozonation: assessing its role in the formation and control of disinfection by-
products, J. Am. Water Works Assoc. 81 (1989) 74–84.
[6] G. Hua, D.A. Reckhow, Comparison of disinfection byproduct formation from
chlorine and alternative disinfectants, Water Res. 41 (2007) 1667–1678.
[7] B. Legube, J.P. Croue, D.A. Reckhow, M. Dore, Ozonation of organic halide pre-
cursors: effect of bicarbonate and bromide, in: Proceedings of the International
Conference – The Role of Ozone in the Water and Wastewater Treatment,
Selper, Ltd., London, 1985, pp. 73–86.
[8] J.D. Plummer, J.K. Edzwald, Effect of ozone on algae as precursors for tri-
halomethane and haloacetic acid production, Environ. Sci. Technol 35 (2001)
3661–3668.
[9] R. Lamsal, M.E. Walsh, G.A. Gagnon, Comparison of advanced oxidation pro-
cesses for the removal of natural organic matter, Water Res. 45 (2011)
3263–3269.
[10] R. Mosteo, N. Miguel, S. Martin-Muniesa, M.P. Ormad, J.L. Ovelleiro, Evaluation
of trihalomethane formation potential in function of oxidation processes used
during the drinking water production process, J. Hazard. Mater. 172 (2009)
661–666.
[11] J. Hu, H. Song, T. Karanfil, Comparative analysis of halonitromethane and
trihalomethane formation and speciation in drinking water: the effects of dis-
infectants, pH, bromide, and nitrite, Environ. Sci. Technol. 44 (2010) 794–799.
[12] P.C. Chiang, E.E. Chang, C.C. Chuang, C.H. Liang, C.P. Huang, Evaluating and elu-
cidating the formation of nitrogen-contained disinfection by-products during
pre-ozonation and chlorination, Chemosphere 80 (2011) 327–333.
[13] U. Von Gunten, Ozonation of drinking water: Part I. Oxidation kinetics and
product formation, Water Res. 37 (2003) 1443–1467.
[14] J.C. Morris, B. Baum, Precursors and mechanisms of haloform formation in
the chlorination of water supplies, in: R.L. Jolley, H. Gorchev, H. Hamilton Jr.
(Eds.), Water Chlorination: Environmental Impact and Health Effect, Ann Arbor
Science, Michigan, 1978, pp. 29–48.
[15] S.D. Richardson, A.D. Thruston, T.V. Caughran, P.H. Chen, T.W. Collette, T.L.
Floyd, K.M. Schenck, B.W. Lykins, G.R. Sun, G. Majetich, Identification of new
ozone disinfection byproducts in drinking water, Environ. Sci. Technol. 33
(1999) 3368–3377.
[16] P. Westerhoff, G. Aiken, G. Amy, J. Debroux, Relationships between the struc-
ture of natural organic matter and its reactivity towards molecular ozone and
hydroxyl radicals, Water Res. 33 (1999) 2265–2276.
[17] F. Hammes, E. Salhi, O. Koster, H.P. Kaiser, T. Egli, U. Von Gunten, Mechanis-
tic and kinetic evaluation of organic disinfection by-product and assimilable
organic carbon (AOC) formation during the ozonation of drinking water, Water
Res. 40 (2006) 2275–2286.
[18] E.C. Wert, F.L. Rosario-Ortiz, D.D. Drury, S.A. Snyder, Formation of oxi-
dation byproducts from ozonation of wastewater, Water Res. 41 (2007)
1481–1490.
[19] J. Staehelin, J. Hoigne, Decomposition of ozone in water: rate of initiation
by hydroxide ions and hydrogen peroxide, Environ. Sci. Technol. 16 (1982)
676–681.
[20] J.V. Goldstone, M.J. Pullin, S. Bertilsson, B.M. Voelker, Reactions of hydroxyl
radical with humic substances: bleaching, mineralization and production
of bioavailable carbon substrates, Environ. Sci. Technol. 36 (2002) (2002)
364–372.
[21] APHA (American Public Health Association), American Water Works Associa-
tion, Water Environment Federation, Standard Methods for the Examination of
Water and Wastewater, 20th ed., American Public Health Association, Wash-
ington, DC, 1999.
[22] S.N. Nam, S.W. Krasner, G.L. Amy, Differentiating effluent organic matter (EfOM)
from natural organic matter (NOM): impact of EfOM on drinking water sources,
Adv. Environ. Monit. 3 (2008) 250–270.
[23] B. Chen, S.N. Nam, P. Westerhoff, S.W. Krasner, G. Amy, Fate of effluent
organic matter and DBP precursors in an effluent-dominated river: a case study
of wastewater impact on downstream water quality, Water Res. 43 (2009)
1755–1765.
[24] X.M. Wang, G.Y. Sheng, J.M. Fu, Y.S. Min, Influence of tidal action on the char-
acteristics of organic pollutants in Pearl River west channel and Liuxi River,
Geochimica 26 (1997) 54–59.
[25] W. Chen, P. Westerhoff, J.A. Leenheer, K. Booksh, Fluorescence
excitation–emission matrix regional integration to quantify spectra for
dissolved organic matter, Environ. Sci. Technol. 37 (2003) 5701–5710.
[26] Z.S. Can, M. Gurol, Formaldehyde formation during ozonation of drinking water,
Ozone Sci. Eng. 25 (2003) 41–51.
[27] L.Y. Lee, H.Y. Ng, S.L. Ong, J.Y. Hu, G. Tao, K. Kekre, B. Viswanath, W. Lay, H.
Seah, Ozone-biological activated carbon as a pretreatment process for reverse
osmosis brine, Water Res. 43 (2009) 3948–3955.
[28] D.J. Barker, D.C. Stuckey, A review of soluble microbial products (SMP) in
wastewater treatment systems, Water Res. 33 (1999) 3063–3082.
[29] A. Li, X. Zhao, H. Liu, J. Qu, Characteristic transformation of humic acid dur-
ing photoelectrocatalysis process and its subsequent disinfection byproduct
formation potential, Water Res. 45 (2011) 6131–6140.
[30] M.L. Trehy, R.A. Yost, C.J. Miles, Chlorination byproducts of amino acids in
natural waters, Environ. Sci. Technol. 20 (1986) 1117–1120.
[31] A. Obolensky, P.C. Singer, Halogen substitution patterns among disinfection
byproducts in the information collection rule database, Environ. Sci. Technol.
39 (2005) 2719–2730.
[32] X. Yang, W.H. Guo, Q.Q. Shen, Formation of disinfection byproducts from
chlor(am)ination of algal organic matter, J. Hazard. Mater. 197 (2011)
378–388.