Chemosphere 74 (2009) 854–859

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Wastewater recycling: Application of ozone based treatments to secondary effluents

Javier Rivas *, Olga Gimeno, Fernando Beltrán
Departamento de Ingeniería Química y Química Física, Facultad de Ciencias, Avenida de Elvas S/N, 06071 Badajoz, Spain

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

Article history: A secondary effluent from a municipal wastewater plant has been treated by means of several processes
Received 6 May 2008 based on the use of ozone. Application of O3 as the only oxidising reagent is capable of halving the initial
Received in revised form 30 July 2008 COD (35 ppm at a flowrate of 40 L h1) with a roughly 40% of mineralization degree. The system O3/H2O2
Accepted 31 July 2008
slightly enhances COD removal albeit it is capable of significantly increasing the mineralization level up
Available online 14 November 2008
to 70% (35 ppm of O3 at a flowrate of 40 L h1 in the presence of 0.01 M of H2O2). The combination of two
commercial activated carbons and ozone did not improve the results obtained in the single ozonation
Keywords:
process. However, the simultaneous application of ozone and UV radiation showed beneficial effects both
Ozonation
COD
in the COD removal rate and TOC conversion if compared to ozonation alone. The systems applied in this
Mineralization work can be acceptably modelled by empirical mathematical expressions based on the inlet ozone con-
Municipal wastewater centration, hydrogen peroxide concentration, COD, etc.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction
Water availability is a challenging problem facing our society in
some regions all over the world. This problem is aggravated when
the lack of rainfall extends over an unusually long period of time.
Drought usually results in water shortage that seriously interferes
with human activity. Water-supply reservoirs empty, wells dry up,
and crop damage ensues. Its seriousness depends on the degree of
the water shortage, size of area affected, and the duration and
warmth of the dry period.
Rapid population growth and increasing water withdrawal for
agricultural use culminate in a large population suffering from
water deficit. The irrigated agriculture accounts for approximately
70% of all freshwater withdrawals and more than 80% in the devel-
oping countries (FAO, 2008).
Once the regulatory limits are fulfilled, the normal procedure
after biological treatment of wastewater is the direct discharge into
natural water matrixes (rivers, oceans, etc.). Nevertheless, more
and more countries are focusing their efforts to reutilize secondary
effluents in those requirements not demanding a high standard of
water quality (drinking water or water for the food industry). Thus,
recycled water can be used in toilet flushing, garden watering,
street washing, fire protection, etc. Even if more stringent standards
of quality are adopted, reclaimed wastewater can be used for agri-
cultural purposes in crop production (The on Site Consortium,
2008). For this purpose, oxidising agents like ozone or other ad-
vanced oxidation processes based on its use are considered as
promising alternatives to conventional oxidation/disinfection tools
* Corresponding author. Tel./fax: +34 924 289385.
E-mail address: fjrivas@unex.es (J. Rivas).
like chlorine and its derivatives (Lee et al., 2008). Ozone is a power-
ful oxidant capable of reacting with a high number of organic and
inorganic molecules. The ozone versatility is based on the possibil-
ities of destroying contaminants through two different routes,
either by direct molecular attack and/or by decomposition into
HO radicals which, contrarily to ozone, actuates unselectively with
rate constants of the order 108–1010 M1 s1 (Beltrán, 2004).
The use of ozone in more sophisticated systems such as the
combination ozone/activated carbon (AC) (Faria et al., 2008),
ozone/hydrogen peroxide or ozone/UV radiation (Burns et al.,
2007) might improve the efficiency of the process by reinforcing
the second pathway of ozone performance, i.e., decomposition into
hydroxyl radicals. In the first case, adsorption of contaminants
onto the AC surface can also contribute to the global efficacy in
organic removal.
In this work, four ozone based systems (O3, O3 + H2O2, O3 + AC,
and O3 + UV) have been evaluated when applied to a real down-
stream effluent generated in a municipal wastewater plant.
2. Experimental
Biologically treated wastewater was collected from a municipal
wastewater plant in Móstoles (Spain) and immediately frozen after
characterization. The main properties of the effluent are displayed
in Table 1.
Experiments were conducted in a glass reactor equipped with
sampling port, temperature monitorization, ozone inlet and outlet,
and mechanical agitation. Temperature was kept controlled at
20 ± 0.1 °C by immersion of the reactor into a thermostatic bath.
When UV light was used, a low pressure mercury Heraeus
Hanau TNN 15/32 located in the middle of the reactor was used.

0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2008.07.090
 J. Rivas et al. / Chemosphere 74 (2009) 854–859 855

Table 1 further incubation at isotherm conditions depending on the type

Characterisation of secondary effluents of microbes to be tested.
COD 62 ± 4 Zn 47 ± 6
TOC 17 ± 1 Cu 6±2 3. Results and discussion
TIC 41 ± 4 Cd 2±2
BOD 41 ± 2 Cr 3±1
SS 77 ± 25 Fe 146 ± 113 3.1. Ozonation
DS 305 ± 44 Mn 30 ± 17
Al 93 ± 7 Ni <0.1 Some ozone reactions were carried out at increasing values of
As 5±2 Pb 9±2
ozone inlet concentration from 10 to 50 mg L1. Fig. 1 shows the
Sb 4±2 Se 5±4
results obtained. As seen from this figure, in terms of COD removal,
Units in mg L1 with the exception of metals in lg L1. ozone dose exerts a positive influence for values of ozone concen-
tration from 10 to 35 mg L1, however, a further increase up to
The main wavelength emission was at 254 nm with an intensity of 50 mg L1 did not appreciably affect the efficiency of the process.
4.5  106 Einstein L1 s1. Ozone was produced from an Erwin Dissolved ozone and ozone at the reactor outlet are immediately
Sander 301.7 laboratory ozone generator. detected after initiating each experiment. This suggests that com-
Activated carbons Darco 12-20 (Aldrich) and Hydraffin P110 pounds present in the secondary effluent are recalcitrant to ozone
(Lurgi) have the following characteristics according to the manu- attack (low rate constants) and, accordingly, slow regime reaction
facturers. Iodine index (mg g1): 600 and >950, SBET (m2 g1): develops. Mineralization of the organics present in the treated
650 and 1000, density (g cm3): 0.37 and 0.46, and moisture (%): wastewater is more difficult to achieve. With the exception of
<12 and <5. the run conducted at 10 mg L1, there are no differences between
Dissolved ozone was measured by following the decoloration of experiments. The maximum TOC decrease after 120 min of treat-
the 5,5,7 indigotrisulphonate. For that purpose a Thermo Spectron- ment barely reaches the 35–40% of the initial amount. Similar fig-
ic Hekios a spectrophotometer was used. Additionally, ozone in ures are reported by Zou and Zhu (2008) when also treating a
the gas phase was monitored by means of an Anseros Ozomat secondary effluent by the combination of ozone and light.
ozone analyser. Chemical oxygen demand (COD) was determined After 120 min of treatment, SS were reduced at an average per-
in a Dr. Lange spectrophotometer, the method based on the stan- centage of 80% (from 37 to 7 mg L1) in the case of experiments
dard dichromate reflux method. Biological oxygen demand (BOD) carried out with 20 and 35 mg L1 of ozone. When 50 mg L1 of
was measured following the respirometric method. For this pur- O3 was applied, this value decreased to slightly above 20% and even
pose, non-acclimated microorganisms from the municipal waste- lower values (7%) were obtained for the run completed with
water plant of the city of Badajoz were used. In order to assess 10 mg L1 of ozone. In parallel with SS, DS increased at the expense
the degree of mineralization, total organic (TOC) and inorganic car- of SS removal (i.e. for the run conducted with 20 mg L1 of ozone
bons (TIC) were determined by a Shimazdu TOC 5000A analyser by DS changed from 343 to 380 mg L1). It seems that ozone is capa-
directly injecting the aqueous solution. Suspended (SS) and dis- ble of converting some of the organics/inorganic molecules present
solved (DS) solids were obtained by means of the weight difference in particulates into soluble compounds.
after dryness of two aliquots of the effluent before and after filtra- Additionally, some metals are also reduced after ozone applica-
tion through 0.45 lm membranes. Metals were quantified by ICP tion. Hence, metal species (in parentheses elimination degree) like
after filtration. Microbial assays were conducted after filtration of Al (10%), As (55%), Sb (47%), Zn (70%), Cu (17%), Cd (100%), Cr
100 mL of wastewater onto membranes (0.45 lm pore size) and (74%), Fe (58%), Mn (26%), Pb (64%), or Se (34%) are eliminated when

1 50
A C
40
0.8
o

CO g, ppm
CCOD/CCOD

30
0.6
3

20

0.4
10

1
0.2 40
B D
0.9
3
o

CO , ppm
CTOC/CTOC

0.8
2
3

0.7

1
0.6

0.5 0
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Time, min Time, min

Fig. 1. Ozonation of secondary effluents. Experimental conditions: pH = 7.8, T = 298 K, gas flowrate = 40 L h1, C O3 gðinletÞ : e, 10 mg L1; h, 20 mg L1; 4, 35 mg L1;
r, 50 mg L1. (A) COD; (B) TOC; (C) outlet ozone; (D) dissolved ozone.
856 J. Rivas et al. / Chemosphere 74 (2009) 854–859

35 mg L1 of ozone is fed to the reactor. The way metals are elimi-
nated from solution is unclear, likely these species are in solution
forming complexes with the organics initially present in the efflu-
ent. Once ozone does oxidise the complexing organics (i.e. humic
substances, fulvic acids, etc.) the remaining metals might undergo
hydrolysis and finally precipitate. Thus, initial ozonation pH was
approximately 8 showing minor variations after the treatment
process (less than 0.5 units in all cases), solubility products of the
analysed metals go from 2  1013 for manganese hydroxide or
7.2  1015 for cadmium hydroxide to values as low as 2.8  1039
for ferric hydroxide or 3  1034 for aluminium hydroxide (Kemij-
sko-Tehnološki Fakultet, 2008). In a similar way, evolution of carbon
dioxide at this pH due to organic matter mineralization can lead to
formation of insoluble carbonates (i.e. solubility products of carbon-
ates are from the order of 1.46  1010 for ZnCO3 to 7.4  1014 for
PbCO3). Alternatively although less probable, some metallic species
can directly be oxidised to forms of higher valence state whose
hydroxides are thereafter precipitated at the working pH.
Raw wastewater contained some colonies (4 colonies in
100 mL) of Feacal coliform, however, as expected, a test of pathogen
microorganisms after the ozonation revealed the absence of Sal-
monelle, Staphylococcus aureus, Feacal streptococcus, Pseudomonas
aeruginosa, F. coliform, and Clostridium sulphite reductors, regardless
of the ozone concentration used. Additionally, BOD5 values were
reduced in all cases more than 50% of the initial values, thus for
the experiment conducted with 35 mg L1 of ozone, BOD5 was low-
ered from 45 to 13 mg L1.
3.2. Ozone + hydrogen peroxide
The recalcitrant nature towards ozone of the substances present
in the secondary effluent suggests the utilisation of a technology
capable of generating more HO radicals to improve the oxidation
yield. The following reactions have to be considered (Rivas et al.,
2003):
O3 þ OH ! HO2 þ O2 ki ¼ 70 M1 s1
6
HO2 þ O3 ! HO þ  
O2 þ O2 kii ¼ 2:8  10 M 1
s 1
Fig. 2 displays the results obtained for a series of experiments
completed with different initial concentrations of hydrogen perox-
ide acting as a promoter. After comparison to single ozonation re-
sults, some significant facts are worth mentioning.
In terms of COD reduction, an initial dose of 103 M in H2O2 is
capable of accelerating the COD removal rate at the initial stages
of the reaction. Once hydrogen peroxide is consumed, the process
comes to a halt achieving a similar COD conversion after 120 min
of treatment if compared to the non-promoted experiment. Exper-
iments completed with initial H2O2 doses of 5  103 and 102 M
show an initial induction period proportional to the amount of pro-
moter added, nevertheless, after approximately 20 min from the
start, COD incessantly decreases to values close to 80% of conver-
sion in 120 min (70% achieved in the non-promoted experiment).
When the highest H2O2 concentration was used (0.05 M), the scav-
enging effect of the peroxide (reactions (3) and (4)) predominated
over its promoter action.
H2 O2 þ HO ! HO2 þ H2 O kTer1 ¼ 2:7  107 M1 s1 ð3Þ
9
HO2 þ HO ! 
HO2 þ OH 
kTer2 ¼ 8:3  10 M 1
s 1
ð4Þ
Dissolved ozone is detected only when hydrogen peroxide has
been consumed. For high dosages of H2O2 this species remains at
extremely low values in the frontier of the detection limit for the
analytical method employed.
The most relevant achievement when using hydrogen peroxide
is the mineralization improvement. Similarly to COD removal, a
dose of 103 M in H2O2 involves a high initial TOC elimination rate
but the improvement of the final TOC conversion after 120 min of
treatment is not achieved. An excess of the promoter impedes the
beneficial effect of the HO formed due to reactions (3) and (4).
However, H2O2 amounts in the range 5–10  103 M are capable
of increasing the final mineralization degree from below 40%
(non-promoted experiment) up to a value in the proximity of 70%.
Similarly to single ozonation, SS are quantitatively transformed
into DS. BOD values are minimized from 44 ppm to values in the
ð1Þ
range 6–7 ppm while metals also exhibit an important reduction
ð2Þ even higher than in the case of single ozone application (values

1 1

0.8 0.8
2O2 H2O2o
o
CCOD/CCOD

0.6 0.6
/C

0.4 0.4
CH

0.2 0.2

1
0 0
3

2.5
0.8
2
CO , ppm
o
CTOC/CTOC

0.6 1.5
3

1
0.4
0.5

0.2 0
0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160
Time, min Time, min

Fig. 2. Oxidation of secondary effluents by the O3/H2O2 system. Experimental conditions: pH = 7.8, T = 298 K, gas flowrate = 40 L h1, C O3 gðinletÞ = 35 mg L1,
C H2 O2 : e, 5  102 M; r, 1  102 M; 4, 5  103 M; s, 1  103 M; N, 0.0 M.
 J. Rivas et al. / Chemosphere 74 (2009) 854–859 857

corresponding to the experiment with 0.01 M in hydrogen perox-
ide): Al (56%), As (98%), Sb (72%), Zn (99%), Cu (57%), Cd (100%),
Cr (100%), Fe (78%), Mn (49%), Pb (82%), or Se (96%).
3.3. Ozone + AC
Use of AC in the presence of ozone might involve the beneficial
effects of the adsorption, inherent to this material, and also the
promotion of ozone decomposition into HO. A first attempt was
therefore conducted by using Darco. The results obtained when
Darco AC was used were somehow disappointing (Fig. 3). Thus,
COD removal rate was similar to elimination achieved by the single
ozonation with the exception of the run conducted with 4 g L1
where a slight improvement in the initial rate was observed. Min-
eralization levels were even lower than those obtained in the sin-
gle ozonation (results not shown).
A second attempt was therefore carried out with a different AC
(Hydraffin). In this case, although COD depletion profiles were sim-
ilar to the single ozonation process, addition of Hydraffin showed a
better performance in terms of TOC conversions (Fig. 3). The
adsorption stage in the absence of ozone eliminated a roughly
25% of the initial TOC present. Increasing concentrations of ACs,
after an initial induction period, were capable of gradually enhanc-
ing the mineralization level. When the maximum amount of the
adsorbent was used (4 g L1), the lag period needed for free radi-
cals build up disappeared and an approximately 65% mineraliza-
tion level was achieved. The beneficial effect of AC can be
attributed to either the cumulative effects of adsorption + ozona-
tion and/or a potential catalytic action of carbon surface to decom-
pose ozone into hydroxyl radicals. Decomposition of ozone on
carbon surface leads to formation of hydroxyl radicals with a much
higher oxidation potential than molecular ozone (Alvárez et al.,
2006):
(
O2 þ HO
O3 þ ACðsÞ ! O3
HO2 ! HO þ  O2 þ O2
1
0.8
o
CCOD/CCOD
3.4. Ozone + UV light
The combination of ozone and ultraviolet radiation accumulates
the effects of two direct oxidation pathways (ozonation and pho-
tolysis), the formation of the system O3/H2O2 because of hydrogen
peroxide generation and the intrinsic photolysis of the peroxide to
produce two hydroxyl radicals:
O3 þ hmðsÞ ! H2 O2 þ 1=2O2 ð6Þ
H2 O2 þ hmðsÞ ! 2HO ð7Þ
Fig. 4 shows the results obtained in radiation experiments car-
ried out in the absence of ozone and hydrogen peroxide (photoly-
sis), in the absence of ozone and presence of the peroxide
(promoted photolysis), and finally in the presence of increasing
ozone concentrations and UV radiation. As observed, the photolytic
process is able to just remove the 15% of the initial COD. Addition
of hydrogen peroxide (0.01 M) increases the COD removal conver-
sion up to 65%. However, the best results are obtained when ozone
and UV radiation are combined. In this case, the initial COD deple-
tion is significantly enhanced achieving a roughly 95% COD elimi-
nation when 50 mg L1 of ozone was fed (final COD was roughly
3 mg L1). A similar trend was found in the case of TOC with a max-
imum mineralization level of 70% found when simultaneously
applying ozone and UV radiation.
3.5. Empirical kinetics
3.5.1. Ozone
The mechanism governing ozonation processes is extremely
complex (Rivas et al., 2006). Moreover, the complexity is exponen-
tially increased when a multicomponent matrix like wastewater
effluents is treated. The ignorance of compound nature and con-
centration in the secondary effluent, the presence of solids that
are further attacked by ozone (i.e. three phase system, i.e. gas–li-
ð5Þ quid–solid), etc. recommend the proposal of an empirical kinetic
law rather than considering a detailed mechanism of reactions.
This empirical model should be capable of modelling the
1
0.8

0.6
o
CCOD/CCOD

0.6
0.4
0.4
0.2
0.2
0
1
1
0
0.9

0.8
o

0.8
CTOC/CTOC

0.7
o
CTOC/CTOC

0.6 0.6
0.5

0.4 0.4

0.3
0 20 40 60 80 100 120 0.2
Time, min 0 20 40 60 80 100 120
Time, min
Fig. 3. Oxidation of secondary effluents by the O3/AC system. Experimental
conditions: pH = 7.8, T = 298 K, gas flowrate = 40 L h1, C O3 gðinletÞ = 35 mg L1. Top Fig. 4. Oxidation of secondary effluents by the O3/UV system. Experimental
figure: darco, bottom figure: hydraffin P110. CAC: e, 4 g L1; h, 1 g L1; r, 3 g L1; conditions: pH = 7.8, T = 298 K, gas flowrate = 40 L h1, C O3 gðinletÞ : e, 50 mg L1; 4,
4, 2 g L1; s, adsorption; N, 0.0 g L1. 35 mg L1; r, 10 mg L1; h, 0.0 mg L1 (C H2 O2 = 0.01 M); s, photolysis.
858 J. Rivas et al. / Chemosphere 74 (2009) 854–859

Table 2
COD removal model parameters in the treatment of secondary effluents by ozone based technologies

O3 O3/H2O2 O3/UV
 1
kO3 ¼ 1:00  0:17  10 mg5
kO3 =H2 O2 ¼ 6:00  0:52  10 6
mg Om1
2 mg On
3 Lmþn1
min kUV = 2.5  103 min1
Om1
2 mg On
3 L
mþn1
min1

n = 0.79 ± 0.05 n = 0.95 ± 0.04 kO3 ¼ 1:00  0:17  105 mg Om1
2 mg On3 L
mþn1
min1
m = 2.11 ± 0.04 m = 2.00 ± 0.03 m = 2.11 ± 0.04
0 a n mnþa
kH2 O2 ¼ 6:10  0:17  1012 mg H2 O
2 mg O3 L n = 0.79 ± 0.05
00  0 0 0 0
kH2 O2 ¼ 4:30  0:15  1011 Lb mg H2 Ob
2 kO3 =UV ¼ 1:10  0:09  104 mg Om
2
1
mg On
3 Lm þn 1 min1
a = 0.680 ± 0.007 m0 = 1.92 ± 0.02
b = 6.3 ± 0.8 n0 = 0.63 ± 0.01
ðC CODexp  C CODcal Þ2 ¼ 159 (4 runs) ðC CODexp  C CODcal Þ2 ¼ 242 (4 runs) ðC CODexp  C CODcal Þ2 ¼ 206 (4 runs)

" #
kH2 O2 C aH2 O2
0
experimental trends found under the working conditions used. dC COD
Obviously the adjustable parameters should be contemplated as  ¼ kO3 =H2 O2 C m
COD C n
O3 gðinletÞ þ 00 ð10Þ
dt 1 þ kH O C bH O
mere modelling tools, i.e. they have no chemical significance and, 2 2 2 2

moreover, they depend on the initial guess used.
Recycling of wastewater is regulated (amongst other parame-
ters) by COD content. This parameter can be easily obtained with-
out the need of a specific and expensive apparatus as it is the case
for TOC analysis. Additionally, in oxidation processes even when no
appreciable TOC changes have been observed, low COD values indi-
cate the formation of small carboxylic acids (acetic, oxalic, maleic,
etc.) of non-toxic nature, i.e. contamination reduction is more cor-
related to COD depletion than to TOC elimination. Consequently,
COD was adopted as the parameter to indicate the wastewater
contamination level, and therefore, it was used in the modelling
process. As stated previously, the presence of dissolved ozone
immediately after the start of experiments suggests that the pro-
cess is chemically controlled, i.e. slow reactions develop in the
aqueous bulk and/or with the solid COD. A simple power kinetic
law was proposed of the form
dC COD
 ¼ kO3 C m n
COD C O3
dt
where kO3 , m, and n stand for the empirical rate constant for COD
depletion and empirical reaction orders regarding COD and ozone
concentrations, respectively. In order to simplify the mathematical
procedure, the concentration of ozone reacting with COD (either
dissolved or in solid phase) was considered to be proportional to
outlet ozone concentration (C O3 g ), so Eq. (8) was finally transformed
into
Again, since outlet ozone concentration immediately reaches the
steady-state conditions, ozone inlet concentration has been used in
lieu of the former. Once again, it should be pointed out the lack of
chemical significance of the adjustable parameters in Eq. (10).
At the sight of the curvature of H2O2 concentration profiles, it is
suggested that its decomposition must follow a kinetic rate with
order in the range interval 0–1. Accordingly a decomposition order
of 0.5 was initially considered:
dC H2 O2
 ¼ kPer C 0:5
H2 O2 ð11Þ
dt
By analytical integration:
qffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi k
Per
C H2 O2 ¼ C H2 O2  t ð12Þ
2
A plot of the square root of hydrogen peroxide concentration
versus time should lead to a straight line of slope kPer/2. When
H2O2 was expressed in mg L1 and time in min, kPer values ob-
ð8Þ
tained were 25.4, 28.0, 22.8, and 24.2  102 ppm0.5 min1 for
increasing values of initial H2O2 concentrations from 0.001 to
0.05 M.
Combining Eqs. (10) and (12), the adjustable parameters were
obtained to minimize the squared differences between experimen-
tal and calculated COD concentrations. Table 2 shows the fitting re-
sults. R2 values ranged in the interval 0.96–0.99. With minor
exceptions, errors below 10% are achieved.

dC COD  
 ¼ kO3 C m n m n
COD C O3 g  kO3 C COD C O3 gðinletÞ ð9Þ 3.5.3. Ozone + UV radiation
dt This system was modelled by considering the cumulative ef-
It should be noticed that both inlet and outlet ozone concentra- fects of photolysis, ozonation, and HO generation enhancement
tions are similar after just 5 min of reaction, so inlet ozone concen- (considered proportional to ozone concentration):
tration can be used.
dC COD   m0 n0
By means of the finite difference method and the SOLVER opti- ¼ kUV C COD þ kO3 C m n
COD C O3 gðinletÞ þ kO3 =UV C COD C O3 gðinletÞ ð13Þ
dt
mization package included in EXCELÒ, the adjustable parameters

kO3 , m, and n were obtained. The values after the fitting process Table 2 shows the values of the adjustable parameters. R2 for
are displayed in Table 2. Errors are located inside the ±10% range UV/O3 runs was located in the range 0.91–0.99.
validating the mathematical expression used. R2 values ranged
from 0.967 to 0.998 for the worst and best fittings, respectively.
Acknowledgments

3.5.2. Ozone + hydrogen peroxide

Two opposite factors should be taken into account when con-
sidering this oxidation system. On one hand, the presence of
H2O2 may enhance HO generation by direct reaction of ozone
and the ionic form of the peroxide and, on the another hand, an ex-
cess of the promoter may lead to slowing down of the process due
to the predominant scavenging effect of hydrogen peroxide.
The empirical kinetic rate proposed should include the reaction
with ozone, the positive effect of adding H2O2, and also the nega-
tive effect of an excess of the latter, accordingly expression 10
was adopted:
This work was supported by the CICYT of Spain and FEDER
Funds (Project CSD2006/044). Dr. Gimeno (Ramón y Cajal Contract)
also thanks the Spanish Ministry of Science and Education.
References
Alvárez, P.M., García-Araya, J.F., Beltrán, F.J., Giráldez, I., Jaramillo, J., Gómez-
Serrano, V., 2006. The influence of various factors on aqueous ozone
decomposition by granular activated carbons and the development of a
mechanistic approach. Carbon 44, 3102–3112.
Beltrán, F.J., 2004. Ozone Reaction Kinetics for Water and Wastewater Systems.
Lewis Publishers, CRC Press Company, Boca Raton, FL.
 J. Rivas et al. / Chemosphere 74 (2009) 854–859
Burns, N., Hunter, G., Jackman, A., Hulsey, B., Coughenour, J., Walz, T., 2007. The
return of ozone and the hydroxyl radical to wastewater disinfection. Ozone-Sci.
Eng. 29, 303–306.
FAO, 2008. Water Quality and Environment Programme. <http://www.fao.org/
landandwater/aglw/waterquality/wastewater.stm>.
Faria, P.C., Órfa~o, J.J., Pereira, M.F., 2008. Activated carbon catalytic ozonation of
oxamic and oxalic acids. Appl. Catal. B-Environ. 79, 237–243.
Kemijsko-Tehnološki Fakultet, 2008. Solubility Product Constants. <http://
www.ktf-split.hr/periodni/en/abc/kpt.html>.
Lee, B.H., Song, W.C., Manna, B., Ha, J.K., 2008. Dissolved ozone flotation (DOF) – a
promising technology in municipal wastewater treatment. Desalination 225,
260–273.
859
Rivas, F.J., Beltrán, F.J., Gimeno, O., Carbajo, M., 2003. Homogeneous catalysed
decomposition of ozone in the presence of Co(II). Ozone-Sci. Eng. 25,
261–272.
Rivas, F.J., Beltrán, F.J., Gimeno, O., Carbajo, M., 2006. Fluorene oxidation by coupling
of ozone, radiation, and semiconductors: a mathematical approach to the
kinetics. Ind. Eng. Chem. Res. 45, 166–174.
The Consortium of Institutes for Decentralized Wastewater Treatment,
2008. Wastewater Reuse. <http://www.onsiteconsortium.org/files/wastereuse.
pdf>.
Zou, L., Zhu, B., 2008. The synergistic effect of ozonation and photocatalysis on color
removal from reused water. J. Photochem. Photobiol. A 196, 24–32.