Efficient mineralization of benzoic and phthalic acids in water by catalytic ozonation using a nickel oxide catalyst

Article abstract of DOI:10.1039/c5nj01385b

Organic acids (OAs) are usually used as refractory model compounds that require prolonged reaction time for their mineralization by different treatments. In this study, benzoic and phthalic acids in aqueous solution were significantly degraded and mineralized by catalytic ozonation in the presence of nickel oxide (NiO). For instance, only 42% mineralization was obtained by direct reaction with molecular ozone, while by catalytic ozonation 98% was achieved. These surprising results led to study in depth the role of NiO in the catalytic ozonation of both organic acids. With the aim of identifying the surface species responsible for the high activity of NiO, X-ray photoelectron spectroscopy (XPS) and high performance liquid chromatography (HPLC) were used as the main analytical tools. According to the results, the mineralization of both organic acids was directly influenced by the organic groups adsorbed onto the NiO surface. Therefore, NiO plays a relevant role as a catalyst forming complex surface compounds which are subsequently decomposed increasing the mineralization efficiency. In addition, a simplified kinetic study was conducted to characterize the effect of NiO concentration on the removal efficiency of both organic acids and the accumulation and degradation of the main byproduct (oxalic acid). These results were explained in terms of a combination of two reaction mechanisms such as conventional ozonation and indirect reaction based on OH radicals.

Full text of DOI:10.1039/c5nj01385b

NJC
View Article Online
View Journal
PAPER
Efficient mineralization of benzoic and phthalic
acids in water by catalytic ozonation using a
nickel oxide catalyst
Cite this:DOI: 10.1039/c5nj01385b
a
ab
a
b
Diana Magallanes, Julia L. Rodrıguez,* Tatiana Poznyak, Miguel A. Valenzuela,
´
c
d
Luis Lartundo and Isaac Chairez
Organic acids (OAs) are usually used as refractory model compounds that require prolonged reaction
time for their mineralization by different treatments. In this study, benzoic and phthalic acids in aqueous
solution were significantly degraded and mineralized by catalytic ozonation in the presence of nickel
oxide (NiO). For instance, only 42% mineralization was obtained by direct reaction with molecular
ozone, while by catalytic ozonation 98% was achieved. These surprising results led to study in depth the
role of NiO in the catalytic ozonation of both organic acids. With the aim of identifying the surface
species responsible for the high activity of NiO, X-ray photoelectron spectroscopy (XPS) and high
performance liquid chromatography (HPLC) were used as the main analytical tools. According to the
results, the mineralization of both organic acids was directly influenced by the organic groups adsorbed
onto the NiO surface. Therefore, NiO plays a relevant role as a catalyst forming complex surface
compounds which are subsequently decomposed increasing the mineralization efficiency. In addition, a
simplified kinetic study was conducted to characterize the effect of NiO concentration on the removal
efficiency of both organic acids and the accumulation and degradation of the main byproduct (oxalic
acid). These results were explained in terms of a combination of two reaction mechanisms such as
Received (in Nottingham, UK)
1st June 2015,
Accepted 23rd July 2015
DOI: 10.1039/c5nj01385b
ꢀ
www.rsc.org/njc
conventional ozonation and indirect reaction based on OH radicals.
degradation of organic pollutants and disinfection for several
decades however, it is very difficult for this route to obtain a partial
1
. Introduction
The increasing worldwide contamination of freshwater with mineralization of organic compounds. Accordingly, several ozone-
thousands of chemicals is a serious environmental problem based advanced oxidation processes have emerged, for example to
5
2 2 3 3
facing humanity in the present century. Industrial and municipal name a few, H O /O , UV/O and catalytic ozonation.
wastewater containing numerous toxic and recalcitrant compounds
Catalytic ozonation has been shown in many cases to be an
seeps into rivers, lakes, groundwater and coastal seas. Nowadays, effective process to eliminate organic refractory compounds in
1,2
6
,7
a challenge in wastewater and drinking water treatment is to wastewater. This kind of treatment was developed to over-
improve the existing processes and to design new ones, with the come the drawbacks of conventional ozonation such as the
purpose of removing a larger number of pollutants from water accumulation of recalcitrant biodegradables byproducts and
3
8
matrices. Therefore, various methods have been developed to partial mineralization. Commonly, activated carbons or metal
capture, degrade or mineralize organic pollutants in wastewater, oxides and supported metals have been the most studied
9
,10
such as ion-exchange, coagulation, flocculation, adsorption, catalysts in ozonation of toxic compounds in water.
chemical oxidation, ozone treatment, membrane filtration and However, until now, the mechanism of catalytic ozonation
photocatalysis. Ozonation has been a successful method for seems elusive and with a lot of contradictions. The literature
concerning catalytic ozonation has proposed several analytic
4
11
a
b
c
techniques in order to establish and/or explain its reaction
mechanism. Recently, our research group has proposed X-ray
Lab. de Ing, Qu ´ı mica Ambiental, ESIQIE-Instituto Polit ´e cnico Nacional,
Zacatenco 07738, M ´e xico, DF, Mexico. E-mail: ozliliana@yahoo.com.mx
Lab. Cat a´ lisis y Materiales, ESIQIE–Instituto Polit ´e cnico Nacional,
Zacatenco 07738, M ´e xico, DF, Mexico
5
photoelectron spectroscopy (XPS) as a basic tool to elucidate the
10
reaction mechanism in the catalytic ozonation. XPS can provide
Centro de Nanociencias y Micro y Nanotecnolog ´ı as-Instituto Polit ´e cnico Nacional,
Zacatenco 07738, M ´e xico, DF, Mexico
information related to the presence and nature of functional
groups onto the surface properties of the spent catalyst during
the organic compound degradation. Our previous studies showed
d
Departamento de Bioprocesos, UPIBI–Instituto Polit ´e cnico Nacional, Ticoman,
0
7340 M ´e xico, DF, Mexico
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
that 2,4-dichlorophenoxyacetic acid (2,4-D) elimination was 2. Experimental section
possible due to the surface complex formation of organic
compounds with active metal sites during catalytic ozonation.
XPS spectra revealed the presence of two important signals:
2
.1 Materials and reagents
1
0
The model solutions of both acids (BA or PA) were prepared
ꢁ
1
with 80 mg L of the organic compound (Sigma Aldrich, 99%)
and dissolved in distillate water. The initial pH was 3.2 and it
was not controlled during the ozonation. All chemicals used in the
experiments were analytical grade reagents. NiO (Sigma-Aldrich,
nickel-oxalate complex and NiCl , this information was used to
2
include these compounds in the reaction pathway.
Noble metals are efficient in the generation of surface
carboxylate complexes in ozonation but a low pH condition is
always necessary to achieve a higher degree of contaminant
99%) was used in the experiments as a catalyst.
1
2
degradation. Nickel oxide (NiO) has many applications such
1
3
14
as gas sensors, chemical catalysts, among others. NiO 2.2 Ozonation
significantly increased the mineralization degree during the
Ozone was generated from dried oxygen using an ozone generator
catalytic ozonation of 2,4-D at pH 3.1 and it also showed high
HTU500G (Azco Industries Limited, Canada) using the corona
discharge effect. The ozone concentration in the gas phase was
measured using an Ozone Analyzer BMT 946BT (BMT Messtechnik,
Berlin). The catalyst and ozone concentrations were kept at
1
0
stability. On the other hand, in most of the reported studies
on catalytic ozonation, the effect of the reaction parameters on
the degradation of different pollutants has been the main
1
5–17
research topic.
Therefore, there is a lack of investigation
ꢁ1
ꢁ1
ꢁ1
constant values, 0.1 g L and 29 ꢂ 1 mg L , respectively.
of the reaction mechanism associated with the chemical structure
of the contaminants, during the degradation and mineraliza-
tion steps.
In general, the catalytic ozonation is considered as a non-
selective process; however, the chemical structure of the initial
compound is a primordial factor that defines the interactions
with the catalyst during ozonation. Benzoic and phthalic acids
The ozone-oxygen mixture flow was 0.5 L min . All ozonation
experiments were carried out in a semi-batch type reactor (0.5 L)
at 21 1C. Details of analytic conditions and the ozonation
procedure can be consulted in ref. 10.
2.3 Analytical methods
were chosen as target compounds owing to the fact that their The concentration of BA and PA was determined by high
chemical structures are closely similar between them (the only performance liquid chromatography (HPLC) using a Perkin
difference being the number of carboxyl groups present in each Elmer Flexar instrument equipped with a DAD detector and a
molecule). Both compounds are toxic and exhibit low bio- ‘‘Prevail Organic Acid’’ 5m Grace column. In the case of PA, the
1
9,20
degradability.
Benzoic acid (BA) is found in wastewater mobile phase was acetonitrile and 25 mM KH PO4 buffer,
2
2
1,22
from the pharmaceutical industry.
3 4
In the case of phthalic previously adjusted at a pH of 2.6 with H PO (60 : 40 v/v)
ꢁ1
acid (PA), it has been used in the painting industries and the solution. The flow rate was set to 1 mL min . For BA, the
acylation–esterification in dyes. As a consequence, it has been mobile phase was acetonitrile : water (70 : 30 v/v), mixed using a
detected in different aqueous matrices, for instance: land- C18 (Grace) column. An external standard method and a
fill leachate water, surface water, sediments and even in KH
atmospheric aerosols.
2
PO
With respect to the degradation of used to identify the presence of products and byproducts during
benzoic and phthalic acids by catalytic ozonation, the studies BA and PA degradation. The mineralization degree of organic acids
are scarce. was monitored using a Torch instrument (Teledyne, Tekmar). All
4
buffer mobile phase, at a 210 nm wavelength, were
2
3,24
Some researchers have reported several details on the selectivity results are presented as the mean for triplicate determinations of
of the catalytic ozonation process over compounds with similar each sample.
18
chemical structure. T. Zhang et al. studied the selectivity of six
The pHPZC was determined using a Malvern Zetasizer at
carboxylic acids (acetic, citric, malonic, oxalic, pyruvic and succinic 25 1C and HCl titration was used to adjust the NaOH values for
acids) that have relatively low reactivity with molecular ozone. each sample. FTIR spectra were recorded using a LabRna
Their results clearly showed that the catalytic ozonation in the HR800 model with lens of Attenuated Total Reflection (ATR)
presence of CuO/CeO cannot degrade all carboxylic acids. Acetic, operating in the infrared middle range, with a diamond crystal
2
malonic and succinic acids were removed less than 20% in tip and a magnification of 36ꢃ in a reflection configuration.
comparison with the other acids. These authors employed IR XPS analyses were performed using a ThermoFisher Scientific
spectroscopy (FTIR) to characterize the surface interaction of six K-Alpha X-ray photoelectron spectrometer with a monochromatized
carboxylates that were found as part of a metal-carboxylate AlKa X-ray source (1487 eV). The base pressure of the system was
ꢁ9
complex.
10 mbar. Prior to XPS analysis, all samples were dried at 100 1C
The aim of the present work is to investigate the catalytic for 24 h. Subsequently, each of the specimen powders were
ozonation of toxic organic acids (BA and PA) over the NiO dispersed and embedded in a 5 ꢃ 5 mm indium foil, after these
catalyst by using FTIR and XPS as characterization techniques. were fixed with Cu double side tape to the sample holder. The
In particular, we are interested in studying the C1s and O1s position of the C1s peak at 284.5 eV was monitored on each
regions, after catalytic ozonation of the organic acids, which sample to ensure that no binding energy shift due to charging
provide useful information about the surface interactions on had occurred. Narrow scans were collected at 60 eV analyzer pass
the NiO catalyst.
energy and a 400 mm spot size.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
2.4 Kinetic model
The model of the organic acid (OA) decomposition was proposed
to obey the so-called pseudo-monomolecular dynamics. This
approach does not take into account some aspects of the reaction
nature. However, the complexity of parallel-series reactions and
the presence of the catalyst in the reactor limited the application
of more complicated and realistic model where all the byproduct
variations (accumulation and degradation) are considered.
The modelling strategy considers two different types of
reactions: the first one was the decomposition of the initial OA
while the second one describes the accumulation of the main
byproduct (oxalic acid or OXA) generated as a consequence of
decomposition of the initial OA. The first type of reaction was
modeled using a decaying exponential function defined as:
ꢁ
kt
OA(t) = OA(0)e
(1)
where OA(t) is the concentration of OA measured during the
reaction, OA(0) is the initial concentration of the same organic
acid and k is the pseudomonomolecular reaction rate constant.
For the second reaction, the model selected to characterize the
accumulation of byproducts was:
OXA(t) = OXA(t*)(1ꢁeꢁ
f
k t
)
(2)
where OXA(t) is the concentration of the oxalic acid measured
during the reaction. The term refers to OXA(t*) that is the
concentration of the same byproduct at the time t* that coincides
with the time where the maximum level of this organic was detected.
The parameter k is the corresponding pseudomonomolecular
f
accumulation reaction rate constant. Finally, the third type of
reaction was modeled using the same decaying exponential
function defined as:
ꢁ (t*ꢁt)
k
d
OXA(t) = OXA(t*)e
(3)
Fig. 1 (A) Removal of BA and PA as a function of ozonation time and
where OXA(t) is the concentration of the byproduct to be
decomposed and OXA(t*) is the maximum concentration of
(
B) TOC decay during conventional and catalytic ozonation with NiO.
d
oxalic acid. The parameter k is the pseudomonomolecular
decomposition reaction rate constant corresponding to this
selected byproduct.
ozone and in the presence of the catalyst. By the first route
Fig. 1A), it is clearly noticed that the degradation rate of
benzoic acid was higher than that obtained for the PA, since
0% of BA degradation was reached after 3.5 min of reaction
(
5
3. Results and discussion
time, while PA degradation required a double of reaction time.
These results indicate that benzoic acid is easily eliminated
with ozone, which agrees with the reaction rate constant
3.1 Degradation and mineralization of BA and PA by
conventional and catalytic ozonation
ꢁ
1
ꢁ1
25
As mentioned before, in the last decade, there has been (k = 1.2 M
s ) reported by P. Neta et al., and in the case
2
6
growing interest in the study of the heterogeneous catalytic of phthalic acid, G. Wen et al. determined a value of 0.092 ꢂ
ꢁ1
ꢁ1
ozonation as a new and efficient method for the removal of 0.042 M
s . Hence, the degradation efficiency is dependent
contaminants from drinking water and wastewater. Several on the organic acid chemical structure as well as the reaction
studies have indicated that metal oxides and supported metal time. It is worth noting that direct reaction with molecular
oxides are effective catalysts for ozonation (e.g. Al
2
O
3
, MnO
2
,
ozone was a feasible method to decompose BA and PA, achieving
7
TiO , among others). In particular, nickel oxide (NiO) is an up to 95% removal of both compounds in 20 min. Fig. 1A also
2
important transition metal widely used as a catalyst due to its displays the elimination of BA and PA in combination of ozone
extraordinary electrical, thermal, catalytic and redox properties, and NiO as catalysts. The integrated treatment did not present a
1
0
and its application in catalytic ozonation is scarce.
Fig. 1 shows the degradation profiles of both aromatic This result has been also observed by Q. Dai et al. in the
organic acids (BA and PA) during direct reaction with molecular degradation of acetylsalicylic acid using unsupported Fe and
significant difference in the degradation profiles of two acids.
2
7
3 4
O
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
2
supported on SiO , as catalysts. The total organic carbon (TOC) not a significant effect on the degradation profiles by varying
measurements showed that the mineralization degree was the catalyst concentration.
significantly improved in the catalytic ozonation mode com-
pared to the conventional ozonation (Fig. 1B). By using this late detected that PA degradation increased 30% at 1.0 g L
Concerning the degradation profiles of PA (Fig. 2B), it was
ꢁ
1
route, a 35% and 42% of mineralization degree was reached for catalyst concentration compared to the direct reaction with
BA and PA at 60 min, respectively. Despite the fact that ozone is a molecular ozone during the first 2 min. This fact is in agree-
strong oxidant agent but it is also selective, it does not remove ment with the results obtained by other researchers which have
2
8
the recalcitrant organic compounds in a reasonable time. In also reported that upon increasing the catalyst concentration, a
2
9,30
contrast, by the catalytic ozonation process, a significant better degradation rate of organic pollutants is obtained.
increase of TOC removal (around 95%) was attained at 60 min Note that the effect of catalyst concentration in both model
for both acids. In contrast to the fact observed in the degradation compounds was quite similar at long reaction times (410 min,
of organic acids (Fig. 1A), the mineralization rate of PA was see Fig. 2A and B).
higher than that of BA, probably due to an increase of the
generated oxidant species.
This result suggests that the reaction path for degrading BA
and PA could be forming similar byproducts before leading to
the mineralization step, which was not dependent on catalyst
dose. For this reason, after a careful analysis of the byproducts
generated in all experiments, we found that oxalic acid (OXA)
3.2 Effect of catalyst concentration
The heterogeneous catalytic ozonation is a complex process was the main recalcitrant byproduct and it was chosen to
which involves gas, liquid and solid phases where the catalyst continue the study of catalyst concentration influence.
plays a relevant role. Therefore, some experiments were carried
Fig. 2C displays the effect of catalyst dose on the concen-
out to study the effect of catalyst concentration on the elimina- tration profiles of oxalic acid generated by BA degradation. In
tion degree of model compounds and to find out the optimal the direct reaction with molecular ozone, OXA was accumulated
catalyst dose. Fig. 2A shows the degradation profiles of BA as a during the whole reaction time reaching a concentration of
ꢁ
1
ꢁ1
function of catalyst concentration (0.1, 0.5 and 1.0 g L ). 45 mg L . When NiO was introduced in the ozonation process,
According to these results, it can be observed that there was the concentration profile of OXA changed significantly. At a
Fig. 2 Effect of catalyst concentration on (A) elimination profiles of BA and (B) PA and (C and D) oxalic acid concentration coming from the degradation
of BA and PA, respectively.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
catalyst dose of 0.1 g L , the OXA concentration was about
Paper
ꢁ1
ꢁ
1
9
mg L which is almost 6 times lesser than conventional
ozonation, after 60 min of reaction time. A further increase of
ꢁ
1
the NiO loading (0.1 to 0.5 g L ) produced a minor amount of
ꢁ
1
OXA (E4 mg L ). However, the catalyst dose in the range 0.5 to
ꢁ
1
ꢁ1
1
1
g L slightly increased the OXA concentration (E5 mg L ).
ꢁ
These results exhibit that optimum NiO content (0.5 g L ) is
present in the catalytic ozonation for organic acid. Therefore,
due to the fact that the OXA concentration was lower in the
degradation of BA compared to the degradation of PA at the same
catalyst concentration, the amount of formed OXA will depend on
the chemical structure of the pollutants (Fig. 2C and D).
Our previous results clearly indicated that in the presence of
NiO as a catalyst, a significant reduction of the concentration of
OXA was observed, which did not occur in the conventional
ozonation. The kinetic study of the BA and PA decomposition
leads to a better characterization regarding the effect of catalyst
concentration. Table 1 enlists the reaction rate constants for BA
and PA decomposition, the OXA formation–decomposition
constants calculated on the basis of the proposed kinetic
models. According to the results, upon increasing the catalyst
Fig. 3 Mineralization degree of BA and PA without or with TBA addition.
3.4 FTIR and XPS of fresh and ozonated NiO
concentration, there was no effect in the rate constants for BA Previous publications have reported that the study of the
degradation, whereas, PA rate constants were slightly increased. catalyst in different stages of evaluation helps to explain the
1
0,18,33
With respect to the OXA reaction constants, k
f
(formation) and k
d
mechanism of the catalytic ozonation.
Based on the latter
(
decomposition), coming from BA and PA degradation, an and with the intention to conduct a careful analysis of NiO,
ꢁ1
optimal catalyst concentration of 0.5 g L was found.
FITR and XPS techniques were used as analytical tools to
provide information about functional groups, the oxidation
state and the chemical environment of the elements present
on the catalyst surface.
3.3 Effect of the hydroxyl radical scavenger
34
As is well known, tert-butyl alcohol (TBA) is a hydroxyl radical
According to the literature, ozone interacts with the hydroxyl
scavenger that is usually used to distinguish molecular ozone groups generated from water adsorption on Lewis acid sites,
oxidation from hydroxyl radical oxidation in ozonation pro- then, the production of oxidant species is initiated, favoring the
3
1,32
ꢁ1
cesses.
As shown in Fig. 3, the presence of TBA (200 mg L
)
pollutant degradation. Thus, it is worth investigating about the
significantly depressed the TOC profile of both organic acids. functional groups on the surface of NiO, which could be respon-
In the case of BA a reduction in comparison without TBA was sible for increasing the degradation rate. ATR-FTIR experiments
observed, while PA retards the mineralization degree from 95 to were carried out in the absence and the presence of BA at t = 0,
2
2%, during the 60 min of ozonation. These results showed 20 and 60 min of reaction time. As shown in Fig. 4A, the first
ꢁ
1
that TBA had a negative influence on the global efficiency of BA peak at 1350 cm corresponds to the rocking vibration of the
ꢁ1
and PA elimination, which means that the catalytic ozonation is CH groups while the second peak at 1570–1625 cm is a
2
followed by a concerted series of reactions including, very characteristic peak to the bending vibration of the free or adsorbed
importantly, hydroxyl radical generation in the aqueous phase. water. Upon checking the spectrum of BA after 20 min of reaction
On the other hand, it is feasible that the inhibition could be time an increase of the intensity of the H
ascribed to the adsorption of TBA on the catalytic surface. This observed. However, after 60 min of reaction time, the H
2
O peak is clearly
O peak
2
fact was confirmed by a surface chemical composition study of decreased due to the fact that surface chemisorbed water can
the spent catalyst carried out in the presence of TBA during the act as active sites participating with ozone to form oxidant species
ꢀ
BA and PA degradation by XPS.
as OH. Furthermore, the apparition of a new peak at around
Table 1 Reaction rate constants of BA and PA decomposition, oxalic acid formation–decomposition for each model compound as a function of catalyst
concentration
Oxalic acid
Oxalic acid
ꢁ1
ꢁ1
ꢁ1
ꢁ1
0
ꢁ1
ꢁ1
ꢁ1
[
NiO], g L
Benzoic acid k, min
f
k , min
k
d
, min
Phthalic acid k , min
k
f
, min
k
d
, min
0
0
0
1
0.2069
0.1909
0.1847
0.1759
0.03851
0.1721
0.2146
0.03182
NA
0.1256
0.1328
0.1421
0.161
0.04856
0.04717
0.4299
—
NA
.1
.5
.0
0.009172
0.03182
0.01177
0.006173
0.02859
—
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
Fig. 4 ATR-FTIR spectra of NiO obtained during the catalytic ozonation of (A) BA and (B) PA at 0, 20 and 60 min of reaction time.
ꢁ1
1
025 cm (60 min) is attributed to the vibrational tension of and O –NiO samples has been fitted employing three basic
3
–
CQC–, which can be due to intermediate compounds contributions, situated at 285.3, 286.5 and 288.3 eV which
adsorbed on the NiO surface. correspond to –C–C–, CQO and –COOH, respectively. NiO-
In the case of PA, the results showed a similar behavior to fresh presents the lowest content in C–OH species whereas
those observed in BA (Fig. 4B). The only difference between BA –NiO exhibited decay tendency for C–C and CQO. Regarding
and PA spectra was observed at 20 min, the second peak at the samples which underwent BA removal processes, and
O
3
ꢁ1
1
625 cm of PA presented a lower intensity than that of BA besides the –C–C–, CQO and –COOH, the existence of addi-
ꢁ1
(1570 cm ). This previous result can be interpreted as the TOC tional chemical species is clearly noticeable at high binding
rate of PA was bigger than BA, as a consequence there is an energies and, as mentioned previously, it can be discerned
increment in the consumption of surface hydroxyl groups of from Fig. 5c–e. The latter was foreseeable due to the adsorption
NiO used during the PA elimination.
of some intermediaries onto the catalyst surface during BA
With the purpose of providing more information related to decomposition by ozone.
the nature of the ligands associated with the catalyst surface,
The analysis in the O1s region permitted agreeing the above-
XPS was used as a powerful tool to identify and follow-up the mentioned facts. In Fig. 5f–j, the passage from a two-
evolution of main species involved in the BA and PA elimina- component system, fresh nickel oxide, to a four-component
tion processes. Although, Ni species are considered as the only system by ozonation and used BA elimination catalysts can be
active sites, the diverse XPS spectra of Ni2p did not show elucidated. The fresh nickel oxide O1s XPS spectrum was
differences in all analyzed samples. For this reason, C1s and decomposed into two peaks located at 528.8 ꢂ 0.1 eV assigned
O1s XPS core lever spectra were analyzed for fresh and ozonated to lattice oxygen in Ni–O and 530.8 ꢂ 0.1 eV related to Ni–OH
2
NiO, as well as for the successive BA elimination processes. Raw bonds, specifically Ni(OH) (Fig. 5f). On applying the same
data were fitted with a combination of the Gaussian–Lorentzian model to the ozonated sample, a third contribution is needed
mix function and Shirley background subtraction (Fig. 5). It is to recover the experimental spectrum (Fig. 5g). An additional
noteworthy by direct observation of C1s curves that, Fig. 5a–e, contribution can be assigned to (O
the chemical evolution and/or emergence of carbon species 0.1 eV, but this species is less stable, so the signal likely
were produced during the different reaction stages. corresponds to H O. For samples of O –NiO–BA(20) and O –
2
)
2
-peroxy species, 532.3 ꢂ
2
3
3
From the C1s XPS core level spectra fitting, it was estab- NiO–BA(60), the presence of adsorbed organic groups on NiO
lished that, and for all cases, the main peak at 284.5 eV was established by the need of two additional peaks assigned to
corresponds to adventitious carbon which nearly presents a C–O, 530 ꢂ 0.1 eV and at 531.7 ꢂ 0.1 eV for CQO. The relative
constant concentrating in all samples. The C1s region for NiO amount of those signals increases as a function of the ozonation
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
Fig. 5 XPS high resolution spectra of the C1s (a–e) and O1s (f–j) regions for the NiO catalyst: (a and f) fresh, (b and g) ozonated, (c and h) at 20 min,
d and i) at 60 min and (e and j) at 60 min in the presence of TBA, after ozonation of benzoic acid.
(
2 3 3 3
time including the peak of Ni(OH) , see Fig. 5h and i. On the other catalyst (NiO, O –NiO, O –NiO–BA(20), O –NiO–BA(60) and
hand, the samples O –NiO and O –NiO–BA(60)-TBA showed a O –NiO–BA(20)–TBA) excluding the adventitious carbon, which
3
3
3
decrease in intensity for the Ni–O bond contribution while for remained constant in all the studied catalysts. As can be
the rest samples no significant change was perceived (Fig. 5g and j). observed, all functional groups were detected, in greater or
Furthermore, both samples presented a bigger intensity of lesser intensity in all catalysts, except for the –CQO– group,
the peak attributed to Ni(OH)
contributing to the generation of active oxidizing species. As the higher amount of –C–C– species was found after 60 min of
previously mentioned, it was identified by FTIR, the OH group catalytic ozonation of BA (i.e. O –NiO–BA(60)), however, around
2 3
which acts as Lewis acid sites which was not negligibly detected in NiO or O –NiO. Note that
3
also favoring the generation of these species, for instance 12% intensity for –C–OH, –CQO and –COOH groups was
hydroxyl radicals which improved the efficiency in BA decom- identified for the same catalyst.
position and the subsequent increase of the mineralization
degree of the intermediates. In view of these results, it can be
Importantly, in the case of the catalyst without BA (NiO and
3
–NiO), the functional groups –C–C, –COH and COOH were
O
established that NiO (when interacted with ozone) favored the also identified, which was attributed to the interaction of
production of Ni(OH) , Fig. 5g. Nonetheless, in the presence of adventitious carbon with ozone. As expected, TBA addition
organic matter, Ni(OH) reacted with the compounds generat- decreased the intensity of the majority of functional groups,
2
2
ing organic acid for the case of BA. When TBA was introduced except for –CQO, which was attributed to a decrease in the
ꢀ
in the BA elimination, it was blocking the active centers (Ni–O) amount of OH radicals responsible for the BA oxidation. As a
on the catalyst surface, therefore the XPS spectra were similar preliminary conclusion, these results are indicative that the
to O –NiO, for the peak of Ni–O and Ni(OH) .
NiO surface was involved in the BA degradation due to the
Fig. 6 summarizes the type of surface-bound functional presence of organic matter (byproducts) shown in Fig. 6A.
group on NiO under different reaction conditions identified Furthermore, a careful analysis of the O1s region, Fig. 6B,
by XPS. In the case of the region C1s, Fig. 6A shows the intensity showed the presence of –O–C, Ni(OH) , –CQO and H O (O–H)
of –C–C–, –C–OH, –CQO and –COOH, for each analyzed signals for all the studied catalysts, with a predominance of
3
2
2
2
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
Fig. 6 Relative intensity of functional groups estimated from: (A) C1s and (B) O1s for NiO, under different reaction conditions.
hydroxylated Ni species. As can be seen, the ozonation of NiO
Finally, Fig. 7 illustrates the proposed mechanism for the
produced the highest Ni(OH) (65%) and O–H (12%) intensity catalytic ozonation of BA in the presence and the absence of
2
signals, which can be considered as the creation of surface TBA. NiO promotes the ozone decomposition and byproduct
active sites. On the contrary, in the presence of BA (O
BA(60)), the Ni(OH) and O–H intensity signals were decreased sent in the reaction, it blocks part of active sites then the BA
at around 50% and 100%, respectively after 60 min of ozona- elimination grade was reduced (as previously demonstrated in
tion in comparison with O –NiO. In addition, the presence of Fig. 3).
O–C, and –CQO signals confirmed the presence of organic XPS spectra and analysis of functional groups in C1s and
3
–NiO– absorption generated by BA decomposition, when TBA is pre-
2
3
–
byproducts adsorbed on the NiO surface generated by BA O1s regions of PA elimination presented a similar behavior and
decomposition. It is worthwhile mentioning that the active site for this reason were not shown here.
consumption was different under all reaction conditions, for
instance, in Fig. 6b it is clearly shown that the intensity of two acids is shown in Fig. 8. It is clearly observed that the
Ni(OH) and O–H signals was only reduced by 15% and 5%, atomic percentages did not change significantly after 20 min of
respectively in O
and the –O–C and –CQO signals were not detected.
For this reason only the analysis of atomic percentages or
2
3
–NiO–BA(60)-TBA, in comparison with O
3
–NiO ozonation for two acids (BA and PA). Nevertheless, at 60 min
phthalic and benzoic acids increased the atomic percentage of
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
Fig. 7 Proposed reaction mechanism for BA elimination in the presence of NiO with or without TBA.
Fig. 8 Surface elemental composition obtained from the XPS spectra of the NiO catalyst at different reaction times: (a) benzoic acid and (b) phthalic acid.
carbon that indicated the adsorption of the organic matter on unsupported NiO is a feasible candidate to mineralize an
ꢁ
1
the catalyst surface during the ozonation process. Hence, the average of 93% of 80 mg L of organic acids in short reaction
number of carboxylate groups presented in the organic com- times (1 h). The faster decomposition of both organic acids was
ꢁ1
pound (benzoic and phthalic acids) did not influence the achieved with an optimal NiO of 0.5 g L . The mineralization
reaction mechanism in the catalytic ozonation.
degree of the organic acids was significantly increased with the
presence of NiO in comparison with the conventional ozona-
tion, which was attributed to the combination of two consecu-
tive reactions: (a) direct reaction with molecular ozone and (b)
4
. Conclusions
ꢀ
indirect decomposition by OH radicals. The mainly identified
The degradation of benzoic acid and phthalic acid by ozone in byproduct (OXA) was only eliminated in the presence of NiO.
aqueous solution in the presence of nickel oxide as a catalyst Furthermore, the presence of the diverse group in the region
was around 100% at 20 min of reaction time. It was found that O1s detected by XPS confirmed that ozone was adsorbed on the
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
catalyst surface forming hydroxyl radicals which served to 15 C. Hu, S. Xing, J. Qu and H. He, J. Phys. Chem. C, 2008, 112,
eliminate the oxalic acid byproduct.
5978–5983.
1
1
1
1
2
6 F. J. Beltr ´a n, F. J. Rivas and R. Montero-de-Espinosa, Appl.
Catal., B, 2002, 39, 221–231.
7 S. M. Avramescu, C. Bradu, I. Udrea, N. Mihalache and
F. Ruta, Catal. Commun., 2008, 9, 2386–2391.
8 T. Zhang and J.-P. Crou ´e , Appl. Catal., B, 2014, 144,
Acknowledgements
The authors thank the Department of Graduate Study, Investi-
gation of the National Polytechnic Institute of Mexico (Project:
831–839.
20150428).
9 T. Velegraki, E. Nouli, A. Katsoni, I. V. Yentekakis and
D. Montzavinos, Appl. Catal., B, 2011, 101, 479–485.
0 M. M. Abeldaiem, J. Rivera, R. Ocampo and J. M ´e ndez,
J. Environ. Manage., 2012, 109, 164–178.
References
1
C. C. Chang, C. Y. Chiu, C. Y. Chang, C. F. Chang, 21 M. I. Pariente, F. Mart ´ı nez, J. A. Melero, J. A. Botas,
Y. H. Chen, D. R. Ji, Y. H. Yu and P. C. Chiang, J. Hazard.
Mater., 2009, 161, 287–293.
T. Velegraki, N. P. Xekoukoulotakis and D. Mantzavinos,
Appl. Catal., B, 2008, 85, 24–32.
2
3
H. Yan, P. Lu, Z. Pan, X. Wang, Q. Zhang and L. Li, J. Mol. 22 V. Gandhi, M. Mishra, M. Rao, A. Kumar, P. Joshi and
Catal. A: Chem., 2013, 377, 57–64. D. Shah, J. Ind. Eng. Chem., 2011, 17, 331–339.
P. R. Gogate and A. B. Pandit, Adv. Environ. Res., 2004, 8, 23 S. Garc ´ı a-Segura, R. Salazar and E. Brillas, Electrochim. Acta,
01–551. 2013, 113, 609–619.
A. Lunhong and Y. Zeng, Chem. Eng. J., 2013, 21–216, 269–278. 24 T. Velegrakia, G. Balayiannisb, E. Diamadopoulosa,
5
4
5
S.-P. Tong, W.-W. Li, S.-Q. Zhao and C.-A. Ma, Ozone: Sci.
Eng., 2011, 33, 441–448.
A. Katsaounisa and D. Mantzavinos, Chem. Eng. J., 2010,
538–548.
6
7
8
9
C. Hu, S. Xing, J. Qu and H. He, J. Phys. Chem. C, 2008, 112, 25 P. Neta, R. E. Huie and A. B. Ross, J. Phys. Chem. Ref. Data,
978–5983. 1988, 17, 1027–1284.
B. Kasprzyk-Hordern, M. Zi o´ lek and J. Nawrocki, Appl. 26 G. Wen, J. Ma, Z. Liu and L. Zhao, J. Hazard. Mater., 2011,
Catal., B, 2003, 46, 639–669. 195, 371–377.
R. Gracia, S. Cortes, J. Sarassa, P. Ormad and J. L. Ovelleiro, 27 Q. Dai, J. Wang, J. Yu, J. Chen and J. Chen, Appl. Catal., B,
Water Res., 2000, 34, 1525–1532. 2014, 144, 686–693.
P. Konova, M. Stoyanova, A. Naydenov, S. Christoskova and 28 Y. Lee and U. Von Gunten, Water Res., 2010, 44, 555–566.
D. Mehandjiev, Appl. Catal., A, 2006, 298, 109–114. 29 H. Y. Jung and H. C. Choi, Appl. Catal., B, 2006, 66, 288–294.
0 J. L. Rodr ´ı guez, M. A. Valenzuela, T. Poznyak, L. Lartundo 30 M. Muruganandham and J. J. Wu, Appl. Catal., B, 2008, 80,
and I. Chairez, J. Hazard. Mater., 2013, 262, 472–481. 32–41.
1 A. Ikhlaq, D. R. Brown and B. Kasprzyk-Hordern, Appl. 31 R. Andreozzi, V. Caprio, A. Insola and R. Marotta, Catal.
Catal., B, 2015, 165, 408–418. Today, 1999, 53, 51–59.
2 T. Zhang, W. Li and J. P. Crou ´e , Environ. Sci. Technol., 2011, 32 T. Zhang, W. Li and J.-P. Crou ´e , Appl. Catal., B, 2012,
5, 9339–9346. 121–122, 88–94.
3 N. G. Cho, I.-S. Hwang, H.-G. Kim, J.-H. Lee and I.-D. Kim, 33 L. Yang, C. Hu, Y. Nie and J. Qu, Appl. Catal., B, 2010, 97,
Sens. Actuators, B, 2011, 155, 366–371. 340–346.
4 C. Bradu, L. Frunza, N. Mihalche, S.-M. Avramescu, M. Neat- 34 L. Zhao, Z. Sun and J. Ma, Environ. Sci. Technol., 2009, 43,
5
1
1
1
1
1
4
˘
a and I. Udrea, Appl. Catal., B, 2010, 96, 548–556.
4157–4163.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015