Mechanism of decolorization and degradation of CI Direct Red 23 by ozonation combined with sonolysis.

Abstract

The decolorization and degradation of CI Direct Red 23, which is suspected to be carcinogenic, were investigated using ozonation combined with sonolysis. The results showed that the combination of ozonation and sonolysis was a highly effective way to remove color from waste water. The operational parameters, namely concentration of the dye, pH, ozone dose and ultrasonic density, were investigated during the process. The decolorization of the dye followed pseudo-first-order kinetics. Increasing the initial concentration of Direct Red 23 led to a decreasing rate constant. The optimum pH for the reaction was 8.0, and both lower and higher pH decreased the removal rate. The effect of the ozone dose on the dye decolorization was much greater than that of the sonolysis density. Intermediates such as naphthalene-2-sulfonic acid, 1-naphthol, urea and acetamide were detected by gas chromatography coupled with mass spectrometry in the absence of pH buffer, while nitrate and sulfate ions and formic, acetic and oxalic acids were detected by ion chromatography. A tentative degradation pathway was proposed without any further quantitative analyses. During the degradation, all nitrogen atoms and phenyl groups of Direct Red 23 were degraded into urea, nitrate ion, nitrogen and formic, acetic and oxalic acids, etc.

Full text

Page 1

Technical Note
Mechanism of decolorization and degradation of CI Direct Red 23
by ozonation combined with sonolysis
Shuang Song*, Haiping Ying, Zhiqiao He, Jianmeng Chen
College of Biological and Environmental Engineering, Zhejiang University of Technology,
Hangzhou 310032, People’s Republic of China
Received 15 December 2005; received in revised form 28 July 2006; accepted 30 July 2006
Available online 14 September 2006
Abstract
The decolorization and degradation of CI Direct Red 23, which is suspected to be carcinogenic, were investigated using ozonation
combined with sonolysis. The results showed that the combination of ozonation and sonolysis was a highly effective way to remove color
from waste water. The operational parameters, namely concentration of the dye, pH, ozone dose and ultrasonic density, were investi-
gated during the process. The decolorization of the dye followed pseudo-first-order kinetics. Increasing the initial concentration of Direct
Red 23 led to a decreasing rate constant. The optimum pH for the reaction was 8.0, and both lower and higher pH decreased the removal
rate. The effect of the ozone dose on the dye decolorization was much greater than that of the sonolysis density. Intermediates such as
naphthalene-2-sulfonic acid, 1-naphthol, urea and acetamide were detected by gas chromatography coupled with mass spectrometry in
the absence of pH buffer, while nitrate and sulfate ions and formic, acetic and oxalic acids were detected by ion chromatography. A ten-
tative degradation pathway was proposed without any further quantitative analyses. During the degradation, all nitrogen atoms and
phenyl groups of Direct Red 23 were degraded into urea, nitrate ion, nitrogen and formic, acetic and oxalic acids, etc.
? 2006 Elsevier Ltd. All rights reserved.
Keywords: Ozonation; Sonolysis; Azo dye; Degradation pathway; Pseudo-first-order kinetics
1. Introduction
There are many different classes of dyes, such as azo,
anthraquinone, metal complex, azo metal complex and
phythalocynanine, reflecting the chromophoric structure
of their constituent molecules. Of these dyestuffs, the azo
dyes are the most frequently used. They contain one or
more azo groups (–N@N–) bound to an aromatic group.
As they have been designed to resist chemical and photo-
chemical degradation, conventional chemical and biologi-
cal methods are not efficient means of degrading azo
dyes. On the other hand, anaerobic reductions often gener-
ate aryl amines, which are generally more toxic than the
parent compounds (Acuner and Dilek, 2004). It is known
that soluble azo dyes, when incorporated into the body,
are split into the corresponding aromatic amines by the
liver enzymes and intestinal microflora, which can cause
cancer in humans (Weisburger, 2002; Bhaskar et al.,
2003; Golka et al., 2004). Therefore, it is necessary to find
an effective method to remove colored dyes from effluents.
Ozone is a powerful oxidizing agent that can degrade a
wide variety of dyes in aqueous solution by destroying most
of the double bonds, such as C@C, C@N, and N@N
(Gogate and Pandit, 2004a; Martins et al., 2006). Ozone
can react with aqueous compounds either by direct O3
attack or indirect attack via hydroxyl radical (?OH) chains
resulting from the decomposition of O3, depending on the
operational pH (Chen et al., 2002; Zhao et al., 2004). How-
ever, there are limitations to the use of O3, such as the inten-
sity of energy required to generate ozone, its pH sensitivity
and its selectivity for organic substrates. The ozonation
process requires further research and development.
0045-6535/$ - see front matter ? 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2006.07.090
*Corresponding author. Tel.: +86 571 88320726; fax: +86 571
88320276.
E-mail address: ss@zjut.edu.cn (S. Song).
www.elsevier.com/locate/chemosphere
Chemosphere 66 (2007) 1782–1788

Page 2

The joint application of ultrasound (US) and other oxi-
dation techniques has yielded some interesting investiga-
tions that can improve degradation rates (Naffrechoux
et al., 2000; Ince and Tezcanlı ´, 2001; Gogate and Pandit,
2004b). In US/O3, the mass transfer and decomposition
processes of O3can be enhanced by US to improve the pro-
duction of free radicals (e.g.,?OH) to achieve better degra-
dation effects (Hart and Henglein, 1986; Weavers and
Hoffmann, 1998; Destaillats et al., 2000). This is a new area
of investigation, and studies are still being conducted on a
laboratory scale.
The role of ozonation with US and the related oxidation
process in treating several organic dyes has been examined
by different authors (Alaton et al., 2002; Neamtu et al.,
2004; Tezcanli-Gu ¨yer and Ince, 2004; Martins et al.,
2006). However, most of these studies paid little attention
to the pathway of degradation.
CI Direct Red 23 (4BS) was selected for study because it
is not quickly biodegraded and is widely used in the textile
industries. He et al. (2004) studied the degradation of
50 mg l?14BS solution in a 100 ml reactor with different
microorganisms and treatment processes. It was reported
that 4BS was completely eliminated in at least 6 h by mea-
suring the optical density at 500 nm using a spectropho-
tometer. When the concentration of 4BS was increased, a
longer reaction time was required. Zhu et al. (2000) used
photocatalytic technology during 4BS degradation but
did not investigate the degradation pathway.
The objectives of this study were to investigate the opti-
mum conditions for degradation of the azo dye 4BS and
the degradation mechanism itself when combining ozona-
tion and sonolysis.
2. Experimental
2.1. Reagents
4BS (MW = 813.73 g mol?1) was obtained from Zhe-
jiang Runtu Co., Ltd., in over 99% purity and used without
further purification. The solubility of 4BS in water is
30 g l?1at 80 ?C.
2.2. Apparatus
The experimental apparatus employed in this work is
showninFig.1.Thecylindricalpyrexglassreactor(diameter
100 mm, height 150 mm) was coupled with an ultrasonic
processor (YIY-UL04-B, Shanghai Yiyuan Ultrasonic
Equipment Co., Ltd.), an O3supply system (CHYF-3 A,
rated flow 50 mg min?1, Hangzhou Rongxin Electronic
Equipment Co., Ltd.) generated onsite by electricity using
drypureoxygenandanexhausttreatmentsystem.Theultra-
sonic processor (400 W output, variable power control) was
equipped with a 20 kHz converter (diameter 106 mm, height
225 mm) and a pure titanium probe (tip diameter 16 mm,
height 164 mm), the processing capability of which was
100–1000 ml. The glass reactor was sealed by a polyethylene
cap during operation. The probe was partially immersed in
the liquid and fixed at the center of the polyethylene cap.
The oxygen flow rate to the generator was monitored with
a rotameter incorporated into the ozone generator. In this
experiment, the ozone concentrations in the feed were kept
at 0.6–3.2 g h?1. Surplus ozone was passed into gas absorp-
tion bottles containing 2% KI solution. The experiments
were carried out in a thermostated bath.
Aqueous solutions (500 ml) containing 4BS at different
concentrations were used. 4BS was degraded with the
desired O3dose and/or US density. All experiments were
carried out at 25 ± 0.5 ?C. Samples withdrawn by a glass
syringe at various reaction times were passed through a
0.45 lm membrane filter and analyzed for solutes.
2.3. Analytical methods
The determination of 4BS in aqueous solution was car-
ried out on a UV-9200 ultraviolet–visible spectrophotome-
ter (Beijing Rayleigh Analytical Instrument Corp, Beijing,
China) equipped with a photoelectric detector at the maxi-
mum absorption wavelength of 500 nm.
Aromatic intermediates were detected by gas chroma-
tography coupled with mass spectrometry (GC/MS) (GC,
Varian cp3800 system; MS, Varian Saturn 2000 mass spec-
trometer). The GC was equipped with a WCOT fused silica
series column (30 m · 0.25 mm) and 0.25 lm film thick-
nesses and interfaced directly to the MS. The GC column
was operated for 2 min at a temperature of 80 ?C, which
was then increased to 250 ?C at the rate of 15 ?C min?1.
The other experimental conditions were EI impact ioniza-
tion 70 eV, helium as carrier gas, injection temperature
280 ?C, and source temperature 80 ?C.
Carboxylic acids and nitrate and sulfate ions were deter-
mined by an ion chromatograph (792Basic IC, Metrohm,
Switzerland) with a separating column (Metrosep A Supp
5–250, 250 mm · 4.0 mm), electrical conductivity detector
using NaHCO3 (1.0 mM)/Na2CO3 (3.2 mM)/methanol
(3%) eluent and a total flow rate of 0.6 ml min?1. The sam-
ples (20 ll) were injected manually through an injection
port.
1
2
3
4
5
6
7
off gas
oxygen
8
Fig. 1. Schematic of the experimental set-up. (1) Ozone generator; (2)
rotameter; (3) three-way valve; (4) to input ozone gas detection; (5)
thermostatic bath; (6) reactor; (7) ultrasonic generator; (8) absorption
bottle.
S. Song et al. / Chemosphere 66 (2007) 1782–1788
1783

Page 3

The pH values were measured via a pHs-25 (Rex Ana-
lytical Instrument Co. Ltd., Shanghai, China). The US
power dissipated into the reactor was adjusted and esti-
mated by calorimetry in order to ensure comparative US
conditions (Kimura et al., 1996). The ozone concentration
was determined by an iodometric method (IOA, 1987).
2.4. Procedures
Phosphate buffers were prepared previously in deionized
water by the reaction of calculated amounts of sodium
hydroxide solution and phosphoric acid solution to yield
an ionic strength of 0.0667 M. A 500 mg l?1stock dye solu-
tion, which was used in atest scheme and stored at 4 ?C, was
prepared by dissolving 500 mg of 4BS in 1000 ml of phos-
phate buffer. Dye solutions of various strengths were made
from the stock by proper dilutions with phosphate buffers.
In order to avoid interference from other substances and
to identify accurately the intermediates of the 4BS reaction
sequence and the pH changes during the US/O3process,
500 mg of 4BS were dissolved in 1000 ml of deionized
water, which was classified as a ‘‘non-buffered’’ dye solu-
tion. Samples of 10 ml were removed from the reactor at
30 and 80 min for the GC/MS analysis.
3. Results and discussion
3.1. Effect of initial dye concentration
The dye had a maximum absorption at 500 nm in aque-
ous solution in the UV–vis spectrum. During the decolor-
ization studies, the change in peak absorption of the dye
solution was monitored. The 4BS decolorization kinetics
obtained during US/O3fitted a first-order curve as illus-
trated in Fig. 2. Fig. 2a shows the decolorization of 4BS
at different initial concentrations at pH 8.0 with a US den-
sity of 176 W l?1and an ozone dose of 3.2 g h?1.
Pseudo-first-order rate constants at various initial con-
centrations of 4BS are given in Table 1. It is noticeable that
the rate constants decreased with increasing initial concen-
tration of 4BS. The rate constants for the degradation of
4BS are 3.7 and 0.9 min?1at the initial concentrations of
100 and 500 mg l?1, respectively. As the other operations
were not changed, the numbers of cavities and?OH radicals
generated in solution approached a steady state. Addition-
ally, competition between the dye metabolites and the sub-
strate for?OH became intense owing to the non-selective
properties of
removal rate of the apparent color could be decreased by
increasing the initial concentration of 4BS. Thus, reaction
mixtures containing more dilute initial solutions could be
expected to exhibit faster rates of decolorization.
?OH. Considering the above factors, the
3.2. Effect of pH
The pH of the solution is an important factor in deter-
mining the rate constant. The impact of pH (in the range
4.0–12.0) on the color decay of a 100 mg l?1solution of
4BS with a US density of 176 W l?1and an ozone dose
of 3.2 g h?1was studied by comparing the absorbance
decay rates. The results are shown in Fig. 2b and Table
1, where rate coefficients and regression statistics are pre-
sented. During the reaction, the pH of these solutions
changed by less than 0.2. As can been seen, the decay rate
coefficient at pH = 8.0 (rate constant was 3.7 min?1) was
much larger than the others. The rate constants obtained
at pH values of 4.0, 6.0, 10.0 and 12.0 were 2.2, 3.1, 2.7
and 2.4 min?1, respectively. This is due to competition
between several reactions. The various phosphate species
will react with the?OH at different rates depending upon
the particular species present. The reaction of
the various phosphate species would be expected to be
slowest in the pH range of 2.8–4.5 (Jiang et al., 2002). In
addition, the reaction of?OH with the dye might be easier
under alkaline than under acidic conditions. As a result,
the optimal operational pH was 8.0 during these experi-
ments.
?OH with
3.3. Effect of ozone dose and US density
Different doses of ozone were injected into the dye solu-
tions during sonolysis (176 W l?1) and the results are shown
in Fig. 2c and Table 1. After 1 min of reaction, the 4BS
removal observed was 64%, 78%, 89% and 98% with O3
doses of 0.6, 1.8, 2.7 and 3.2 g h?1, respectively. The
pseudo-first-order rate constant increased from 1.0 min?1
with an O3dose of 0.6 g h?1to 3.7 min?1with an O3dose
of 3.2 g h?1. This indicated that the dye degradation
depended upon the dissolution of ozone, which determines
the amount of?OH generation and affects the transfer of
ozone into the reaction solution. It also indicates that the
reaction is mass transfer limited. An increase in O3dose
could improve the mass transfer of O3 and cause an
increased O3concentration in the liquid phase and a higher
rate constant (Sevimli and Sarikaya, 2002). However, when
the O3dose and ozonation time were increased, the O3uti-
lization ratio decreased. Therefore, it was not necessary to
try to increase the concentration of O3to saturation level.
In order to understand the effect of US density on the
degradation rate, the reaction was conducted using
an ultrasonic horn (20 kHz) with different power settings
at pH 8.0 with a 4BS concentration of 100 mg l?1and an
ozone dose of 3.2 g h?1. The results are presented in
Fig. 2d and Table 1. When the US density ranged from
88 to 352 W l?1, approximately 98% of the 4BS was re-
moved after 1 min of US/O3treatment. Rate constants of
3.6, 3.7, 3.7 and 3.8 min?1were obtained at US densities
of 88, 176, 264 and 352 W l?1, respectively. In other words,
with an increase in US density, the change in the coefficient
of decay was negligible. According to Neppolian et al.
(2002), there are at least three factors affecting the charac-
teristics of US: power density, frequency and amplitude of
the system. The ultrasonicator used in this experiment was
kept at a constant frequency (20 kHz). A greater US
1784
S. Song et al. / Chemosphere 66 (2007) 1782–1788

Page 4

density means greater energy supplied so as to cause cavi-
tation bubbles at a higher temperature. Hence, the degra-
dation rate of 4BS should increase with power density.
However, when the US density increases, the dissipated
energy will increase. Meanwhile, the cavitation bubbles will
grow very large in the negative phase of the sound wave,
resulting in insufficient collapse of the cavitation bubbles
and acoustic screen; the degassing rates of O3 gas will
become higher, resulting in lower concentrations of O3or
radicals. Thereby, with an increasing US density, the incre-
ment of decolorization rate would be attenuated.
Fig. 3 shows the decolorization of the dye solution
(100 mg l?1) at pH 8.0 using US, O3and US/O3, respec-
tively. US and O3alone are less effective than the combina-
Table 1
Pseudo-first-order rate coefficients of various operation parameters during
degradation
Operation parametersPseudo-first-order
(k) (min?1)
Initial
concentration
(C0) (mg l?1)
pHOzone
dose
(g h?1)
US
density
(W l?1)
100
200
300
400
500
100
100
100
100
100
100
100
100
100
100
8.0
8.0
8.0
8.0
8.0
4.0
6.0
10.0
12.0
8.0
8.0
8.0
8.0
8.0
8.0
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
0.6
1.8
2.7
3.2
3.2
3.2
176
176
176
176
176
176
176
176
176
176
176
176
88
264
352
3.7
2.2
1.7
1.1
0.9
2.2
3.1
2.7
2.4
1.0
1.5
2.4
3.6
3.7
3.8
02468
0.0
0.2
0.4
0.6
0.8
1.0
01234
0.0
0.2
0.4
0.6
0.8
1.0
012345678
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.51.01.52.02.5
0.0
0.2
0.4
0.6
0.8
1.0
A/A0
Time (min)
100 mg l
200 mg l
300 mg l
400 mg l
500 mg l
-1
-1
-1
-1
-1
c
d
A/A0
Time (min)
pH = 4.0
pH = 6.0
pH = 8.0
pH = 10.0
pH = 12.0
A/A0
Time (min)
C(O3) = 0.6 g h-1
C(O3) = 1.8 g h
C(O3) = 2.7 g h-1
C(O3) = 3.2 g h
-1
-1
A/A0
Time (min)
P(US) = 88 W l
P(US) = 176 W l
P(US) = 264 W l
P(US) = 352 W l
-1
-1
-1
-1
Fig. 2. Effect of process variables on decolorization of dye: (a) effect of initial concentrations of dye: pH = 8.0; O3 dose = 3.2 g h?1; US
density = 176 W l?1, (b) effect of pHs: initial dye concentration = 100 mg l?1; O3dose = 3.2 g h?1; US density = 176 W l?1, (c) effect of O3dose: initial
dye concentration = 100 mg l?1; pH = 8.0; US density = 176 W l?1, (d) effect of US density: initial dye concentration = 100 mg l?1; pH = 8.0; O3
dose = 3.2 g h?1.
012345678
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
A/A0
Time (min)
US
O3
US + O3
Fig. 3. Comparison of US, O3and US/O3for color decay: initial dye
concentration = 100 mg l?1; pH = 8.0; O3dose = 3.2 g h?1; US density =
176 W l?1.
S. Song et al. / Chemosphere 66 (2007) 1782–1788
1785

Page 5

tion of US and O3. During US only treatment, the investi-
gated azo dyes were found to be fairly stable, with only 4%
of 4BS removed from solution in 8 min. Ozonation alone
can remove 16% and 97% of the dye in 1 and 8 min, respec-
tively. In the combined method, approximately 98% of 4BS
has been removed from solution in 1 min. Tezcanli-Gu ¨yer
and Ince (2004) also reported that the degradation rate
using O3, as well as US/O3, was much greater than that
using US alone during RB5 degradation. Both sonolysis
and ozonation are known to produce?OH radicals in aque-
ous solution, which may attack the 4BS molecule, thus
beginning the degradation. The poor effect of US alone
on the degradation may be attributed to the fact that US
treatment usually demands a long contact time (Ince and
Tezcanlı ´, 2001; Martins et al., 2006). In the combined
method, the US waves can improve the degradation effi-
ciency by enhancing the efficiency of O3dissolution and
the yield of free radicals via the mechanical action of US.
On the other hand, aeration of O3might increase the turbu-
lence of the aqueous solution, which would increase the
migration of some related substances from the collapsing
cavities into the bulk of the solution.
3.4. Degradation mechanism
In this part of the study, samples were taken from the
reactor periodically to analyze pH and nitrate and sulfate
ion concentrations. During the reaction, the pH changed
from 5.7 to 2.9. Organic intermediates generated during
the degradation of the oxidative dye detected by GC/MS
or IC are shown in Table 2. The reason that quantitative
analyses of these species was not done is because the reac-
tion system was very complex and a certain amount of
inorganic salts, such as Na2SO4, produced during the reac-
tion in water would affect the quantitative analysis of
organic moieties (Ohsawa et al., 2004).
The synergistic effect of combining ozonation with ultra-
sonic irradiation is observed only when free radical attack
is the controlling mechanism (Gogate and Pandit, 2004b).
Among the radicals,?OH is an important one and has the
highest oxidation potential of 2.8 V (Muthukumar and Sel-
vakumar, 2004). Hence, based on our results and assuming
that?OH radicals are the major reactive species, a substitu-
tion reaction mechanism was established for this system. A
tentative pathway for the degradation of 4BS during US/
O3was postulated and is presented in Fig. 4.
The partial charge shows that the?OH adds to the N(8),
C(14), C(22), C(27), and C(37) positions to form the state
S2. With the help of a hydrogen radical, N(8)–C(9),
C(14)–N(19),N(21)–C(22),
N(40) are potentially cleaved to form S3, D1, D2, D3, and
D4. Owing to the non-selective nature of radical reactions,
it is possible for the reaction of R1to yield D3and for the
reaction of R3to form D1.
For compound S3, azobenzol can be hydrolyzed in
aqueous solution to yield phenylium which is further oxi-
dized by
none. Then, the benzene ring is cleaved to form but-2-
enedioic acid and finally to form carboxylic acids such as
acetic acid and formic acid.
Under the influence of the?OH and aqueous solution,
state D1is then transformed into compound S5. The ben-
zene ring in 2-hydroxynaphthoquinone is then cleaved to
form state S11by further oxidation.
C(27)–N(32),andC(37)–
?OH to form hydroquinone and benzo-1,4-qui-
Table 2
Intermediates identified by GC/MS and IC
Symbol CompoundsStructural formulaAnalytical methods Samples time (min)
GC/MS IC30 80
D1
Naphthalene-2-sulfonic acid
NaO3S
pp
D2
Urea
NH2
O
NH2
ppp
D3
1-Naphthol
OH
pp
D4
Acetamide
NH2
O
CH3
pp
D5
D6
D7
Acetic acid
Oxalic acid
Formic acid
CH3COOH
HOOCCOOH
HCOOH
p
p
p
p
p
p
1786
S. Song et al. / Chemosphere 66 (2007) 1782–1788