Kinetic Modelling of the catalytic oxidation of 2-(methylmercapto)-benzothiazole under mild conditions

Article abstract of DOI:10.1016/j.apcata.2017.07.001

2-(Methylmercapto)-benzothiazole oxidation was performed over Copper modified zeolites. The microporous materials were synthesized by the hydrothermal crystallization method and later modified with metal incorporation by wet impregnation. The solid cataly

Full text of DOI:10.1016/j.apcata.2017.07.001

Accepted Manuscript
Title: Kinetic Modelling of the catalytic oxidation of
2-(methylmercapto)-benzothiazole under mild conditions
´
Authors: A. Cordoba, C. Saux, L.B. Pierella
PII:
S0926-860X(17)30301-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2017.07.001
APCATA 16306
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
26-4-2017
23-6-2017
3-7-2017
´
Please cite this article as: A.Cordoba, C.Saux, L.B.Pierella, Kinetic Modelling of
the catalytic oxidation of 2-(methylmercapto)-benzothiazole under mild conditions,
Applied Catalysis A, Generalhttp://dx.doi.org/10.1016/j.apcata.2017.07.001
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Kinetic Modelling of the catalytic oxidation of 2-(methylmercapto)-
benzothiazole under mild conditions
A. Córdoba1*, C. Saux1 and L. B. Pierella1
1 Centro de Investigación y Tecnología Química (CITeQ), UTN – CONICET, Maestro
Marcelo Lopez y Cruz Roja Argentina, (5016) Córdoba, Argentina.
*Corresponding Author. Tel/Fax: +54 351 4690585; E-mail: acordoba@frc.utn.edu.ar
(A. Córdoba)
Keywords: Cu-Zeolites; 2-(methylmercapto)-benzothiazole; H2O2 oxidation; kinetic
modelling.

Graphical abstract
. Zeolite matrix
. Copper Content
. Reaction
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
[Cu(3)Y
[Cu-Y]: 0.0
(d)
[H2O2]: 3
2-MMBT
[H2O2]: 3.2
[2-MTBT
[2-MMBT]
2-MMBT
parameters
Zeolites -Cu+2
+
P2
P1
Sulfoxide (P1)
Reaction
mechanism
and
H2O2
00
0
2000 4000 6000 8000 10000120001400
kinetic modelling
Tiempo (seg)
Time (s)
Sulfone (P2)

Highlights
 Kinetic parameter estimation by a theoretical-experimental kinetic study is
necessary in order to design a heterogeneous catalytic system to mitigate the
environmental damage of benzothiazoles.
 Effects of matrix, Cu-charge, solvent and [H2O2] were evaluated for Cu-
zeolites in their catalytic performance. Experimental data fed the kinetic
modeling and the parametrization procedures.
 The proposed oxidation mechanism was confirmed by the kinetic modelling
and the parametrization results, being the Cu(II)/ H2O2 activation, the rate
controlling step.
 A synergic effect of zeolite Y matrix on 2-MMBT conversion was confirmed by
parametrization of homogeneous data.

Abstract:
2-(Methylmercapto)-benzothiazole oxidation was performed over Copper modified
zeolites. The microporous materials were synthesized by the hydrothermal
crystallization method and later modified with metal incorporation by wet
impregnation. The solid catalysts were characterized by means of X-ray diffraction,
surface
area
determinations,
inductively
coupled
plasma
emission
spectrophotometry, temperature programmed reduction and Fourier transformed
infrared spectroscopy. Reaction parameters (Copper content, nature of the solvent,
hydrogen peroxide concentration, reaction time, catalyst mass and reaction
temperature) were evaluated to reach the optimum reaction conditions. Kinetic
modelling and kinetic parameter estimation based on experimental kinetic data were
included. Cu(II) species were confirmed as the active sites of the catalysts, being Y
zeolite the optimal support. Thus, Fenton-like reaction mechanism with reactive
●
oxygen species as hydroxyl and hydroperoxyl radicals (HO●, HO2 ) was confirmed.
Reactions rates constants values and activation energy were determined by non-
linear least-square parameter estimation. Also, homogeneous catalytic system was
evaluated in order to discard diffusional limitations.
1.
Introduction
It is well known that benzothiazole and its derivatives are normally found in the
environment since they are used as fungicides, herbicides, corrosion inhibitors in
cooling water for automobiles and as vulcanization accelerators in rubber production
[1,2]. Considering their toxicity and extensive employment all around the world, their
degradation study is of great environmental concern.

Among them, 2-mercaptobenzothiazole (2-MBT) is worldwide used as accelerator in
vulcanization processes - since it catalyzes the formation of sulfide linkages between
unsaturated elastomeric polymers in order to obtain a flexible and elastic crosslinked
material [3] - and as a corrosion inhibitor in aqueous systems, as an ingredient in
cutting oils and petroleum products and as an additive in extreme-pressure greases
[4]. Its presence has been detected in wastewaters from rubber additive
manufacturing, effluents from treatment plants and in rivers near to the rubber or
tyre-manufacturing industries, but also in urban runoff, treated wastewater and even
in urban air, probably because of tyre abrasion [5,6]. According to Reemtsma et al
[7], 2-MBT may undergo biologically-mediated methylation to form 2-
(methylmercapto)-benzothiazole (2-MMBT), a typical breakdown product from
benzothiazoles compounds. Actually, 2-MMBT is not effectively removed by current
wastewater treatments and it has been found in groundwater and tap water, as a hint
that it is not efficiently biodegraded, whereby additional treatments should be done
[8]. In this sense, chemical treatments, in particular oxidation ones, seem to be the
most appropriate in order to reduce contamination, considering that oxidation
products (the corresponding sulfoxide and sulfone) are more biodegradable [9].
In “green” organic oxidations, modified zeolites have been employed as
heterogeneous catalysts with positive results [10]. So, we propose the use of
modified synthetic zeolites as active solid materials in order to mitigate this serious
issue. It is well known the ordered organization of these aluminosilicate materials
with a microporous structure consisting of a network of channels at a molecular
scale. The presence of framework aluminum atoms makes these materials solid-acid
catalysts. These properties make zeolites suitable for several catalytic applications
[11].

Kinetic studies are always necessary in order to improve catalytic reactions,
particularly for sulfides degradation, since they could help in the design of the
reactive system. Several oxidation mechanisms have been proposed for zeolites as
catalysts [12–15]. Reagents and products concentrations monitoring, allows inferring
some issues about different reactions pathways. Kinetic modelling and parameter
estimation by fitting experimental data (concentration vs. time) are useful tools in
order to discern between different proposed mechanisms [16]. In addition, different
approaches could be done to obtain the kinetics parameters-values, which could be
further applied for reactors design purposes. Ramli and coworkers have recently
reported a kinetic modelling of glucose decomposition catalyzed by Fe-Y zeolite [17].
The authors employed a pseudo first-order approach and linearization strategies in
order to obtain rate constants values. Temperature effect was also determined by
these approaches based on linear regressions procedures. Meanwhile, Hahn and
co-workers have modelled the NH3 oxidation on Fe-Beta zeolite employing stationary
stage approach [18]. Arrhenius plots were used to evaluate temperature effects.
Nevertheless, there are some assumptions which must be fulfill in order to obtain
appropriate results [19]. One of these assumptions is the normal distribution of
experimental errors. However, data transformations, as linearization strategies, do
not guarantee the normal distribution of errors [19]. In order to avoid this limitation,
non-linear least-square parameter estimation could be applied. Even though, some
of that linearization approaches could be used to obtain initial values for non-linear
least-square kinetics parameters estimation [20].
In this study, different Cu(II) modified zeolite matrices (ZSM-5, Y and Beta) are
tested as heterogeneous catalysts for 2-MMBT/H2O2 oxidation. The synthesis
procedure and post-synthesis modifications are described, as well as

characterization results (XRD, TPR, BET, ICP and FTIR). Effects of zeolite matrix,
Copper-charge, solvent and H2O2 concentrations are evaluated. A theoretical-
experimental kinetic study for the optimal catalyst is also carried out. An oxidation
mechanism is proposed and a dynamic modelling with parameter estimation by
fitting experimental data is included. Furthermore, temperature effect by Arrhenius
parameter estimation is evaluated. Finally, homogeneous vs heterogeneous catalytic
oxidations are studied in order to evaluate mass transfer problems.
2. Experimental
2.1. Catalysts preparation
Zeolitic microporous materials were prepared following the hydrothermal
crystallization procedure for ZSM-5 and Beta. The ammonium form of Y zeolite was
purchased from Sigma-Aldrich.
In the case of Na-ZSM-5, it was obtained in the Na2O-Al2O3-SiO2 system, using
TPAOH (Tetrapropylammonium Hydroxide, Fluka) as a structure directing agent by
known methods [21] with some modifications made by our group.
The Na-Beta was synthesized by means of the hydrothermal crystallization method
using Tetraethylammonium Hydroxide (Merck) as directing agent following the
method reported by Valencia et al. [22].
In both cases, the structure directing agent was desorbed in N2 atmosphere (20
ml/min) at programmed temperature (10 °C/min) from room temperature to 500 °C
for 8 h and then were calcined in air at 500 °C for 12 h. The ammonium forms of the
catalysts were obtained by ion-exchange of the Na-zeolites with NH4Cl (Fluka, 1 M)
for 40 h at 80ºC in a batch reactor with continuous mechanical agitation.

The ammonium matrices (NH4-zeolites) were suspended in a solution of
Cu(C2H3O2)2.H2O (Mallinckrodt) in the least amount of distilled water in order to
obtain the desired Copper content. Water was further evaporated using a rotator–
evaporator at 80°C under vacuum until complete dryness. After that, the samples
were dried at 110 ºC and desorbed in N2 flow (10 ml/min) at programmed
temperature (10 ºC/min) from room temperature to 500 ºC for 12 h followed by
calcination in air at 500 ºC for 12 h.
2.2. Catalysts characterization
Chemical composition was determined by using an inductively coupled plasma
emission spectrophotometer Varian 715ES. BET surface area determinations were
carried out with an ASAP 2000 equipment. Phase purity of the catalysts was
determined by X-ray diffraction (XRD) in a Philips X’Pert MPD diffractometer
equipped with a PW3050 goniometer (CuKα radiation, graphite monochromator),
provided with a variable divergence slit and working in the fixed irradiated area
mode. Diffraction data were recorded between 2θ= 4º and 40º at an interval of 0.05º.
A scanning speed of 2º/min was used.
Infrared measurements in the lattice vibration region (400-1800 cm-1) were
performed on a JASCO 5300 FTIR spectrometer using KBr 0.05% wafer technique.
In order to determine the acid sites of the samples self-supporting wafers were done
and a spectral scanning from 4600 to 400 cm-1 in 16 consecutive registers of 4 cm-1
resolution were performed. The wafers were heated at 400 ºC in vacuum (10-4 Torr)
for 6 h and then pyridine (3 Torr) was adsorbed at room temperature for 12 h and

desorbed for an hour at 350 ºC at 10-4 Torr. The experiments were carried out using
a thermostatized cell with CaF2 windows connected to a vacuum line.
Temperature programmed reduction (TPR) studies were done over calcined Cu-
containing zeolites in an Autochem 2910 Micromeritics equipment. Typically, 100-
160 mg of pelletized samples (0.2-0.4 mesh) were used in order to have a constant
weight of Copper in all the samples. Samples were firstly pre-treated under Argon
flow (50 ml/min) at room temperature during 15 min. Then, gas flow was changed to
10% H2/Ar (50 ml/min) and temperature was raised from 25 to 800 ºC at 10 ºC/min
using a computer-controlled program. The hydrogen uptake was monitored by
means of a TCD detector connected on-line, and data will present the evolution of
TCD signal in function of the increasing in temperature. Calibration of the equipment
was made under same experimental conditions using CuO as reference sample.
2.3. Catalytic activity
2-(Methylmercapto)-benzothiazole (2-MMBT) (97 %, Aldrich) oxidation reactions
were performed employing H2O2 (aqueous solution 30 wt %, Cicarelli) as oxidant
agent in a glass flask reactor (25 cm3) with magnetic stirring immersed in a
thermostatized bath, equipped with a reflux condenser. The standard catalytic tests
were done at 40 °C during 5 hours, in a 2-MMBT/ hydrogen peroxide molar ratio of
0.1, employing acetonitrile (99.5 %, Cicarelli) as solvent and 100 mg of the selected
catalyst. Nevertheless, temperature was varied from 25 to 60 °C, solvents of different
characteristics (polarity and proticity) were tested and different types of catalysts and
concentrations were also evaluated.

The reaction evolution was followed by taking aliquots of the solution at different
reaction times. Products were identified by mass spectrometry GC- Mass (Shimadzu
QP 5050 GC-17 A) and quantified by gas chromatography (Perkin Elmer Clarus
500). The solid catalysts were filtrated before the analysis of the products. The
substrate conversions are reported as mol %.
In order to determine diffusional problems of substrates into zeolites cavities, an
homogenous catalytic reaction was done employing Cu(CH3COO)2 as catalyst,
maintaining constant the same 2-MMBT/Cu ratio and the standard reaction
conditions.
2.4. Kinetic Modelling and parametrization
Based on catalytic activity results presented in section 3.2, Cu(3)Y was selected for
the Kinetic modelling and parametrization study. Initial rates analysis of 2-MMBT
oxidation catalyzed by Cu-Y were carried on. Experimental data (concentration vs.
time) has been fitted to double-exponential curves and the initial rates were
determined by (dC/dt)t=0.
The incorporation of different transition metals on zeolites matrices enables their use
in many kinds of reactions. Among them, heterogeneous Fenton and Fenton-like
catalysts could be obtained from transition metal-zeolites [23,24]. Fenton reaction
systems are based on HO● production from H2O2 in presences of Fe(II)/Fe(III)
species. However, Cu(I)/Cu(II) can catalyze hydroxyl radical formation -among other
oxygen reactive species (ROS)- in presence of H2O2 [24–26]. Organic compounds
can be oxidized by highly reactive hydroxyl radical (E°(HO●/H2O)=+2,8 VNHE) [27–29].
Wang et al have reported evidence of hydroxyl radical formation catalyzed by Cu-Y

zeolites in phenol hydroxylation with H2O2 [29]. Based on Cu(I)/Cu(II) Fenton-like
system, an oxidation pathway for 2-MMBT/H2O2/Cu-modified zeolites is proposed
(Scheme 1). The model considers Cu as the active specie of the catalyst and
●
formation of the hydroxyl and hydroperoxyl radical (HO●, HO2 ) over it.
Scheme 1. Proposed reaction mechanism for Cu/2-MMBT/H2O2 catalytic system.

Mathematical model (equations 1-10) was obtained by mass balance of all the
involved species (2-MMBT and its oxidized products -P1: sulfoxide, P2: sulfone-,
catalytic intermediaries and reaction intermediaries). Arrhenius model (equation 11)
was employed to evaluate the temperature effect, considering 40ºC as reference
temperature.
(1)
(2)
(3)
(4)
(5)
(6)

(7)
(8)
(9)
(10)
(11)
Multiresponse modelling of the proposed mechanism was carried on employing
commercial software g-PROMS 3.2 (General PROcess Modelling System, Process
System Enterprise Ltd., London, United Kingdom). g-PROMS parameter estimation
tool, which assume independent and normally distributed experimental errors, were
used. Kinetic and variance models were simultaneously optimized. A constant
variance model was selected (σ2=ω2). Initial guesses of rate constants were
obtained from literature. Also, intermediaries and reaction products initial

concentrations were set at zero. The |accuracy of the model was evaluated by Χ2
(95%) test.
In order to evaluate mass transfer limitations the parametrization of heterogeneous
and homogeneous experimental data were done in two steps and the resulting
kinetic constants optimized values were evaluated.
Finally, the parametrization, including temperature effect, was carried on to obtain
Arrhenius constants values.
3. Results and Discussion
3.1. Cu-zeolites characterization
Copper modified zeolites (Cu-ZSM-5, Cu-Beta and Cu-Y) were characterized
by different techniques. Table 1 summarizes physicochemical characterization
results. It is well known that when a metal is incorporated on a porous material, the
surface area could be diminished due to some pores blockage and metal sintering
during the calcination step [30]. Nevertheless, just a slight reduction in surface area
values calculated by the BET method were obtained.
In order to compare the zeolitic matrix effect on the catalytic activity, the same
Copper content was impregnated over the different porous solids. As it could be
seen from ICP results, about a 3 wt% was effectively loaded in each sample.
Table 1. Physicochemical analysis of Cu-zeolites
Sample
Si/Ala
Cu wt% a
Cu/Ala
SBET (m2/g)b H2 consumption c

Tat maximum (°C)
202 (326)
220 (300)
(ml/g)
10.01
13.34
12.76
Cu-ZSM-5
Cu-Beta
Cu-Y
19.93
10.51
2.41
2.99
3.14
3.38
0.68
0.39
0.15
337
525
655
300 (201)
a Measured by inductively coupled plasma emission spectrophotometer (ICP).
b Determined by BET.
c
Obtained from TPR measurements.(T): Temperature at maximum of the minor
signal.
H2-TPR results presented in Table 3 shown that Copper species in the matrices
could be divided into two groups based on the reducibility of Cu ions. In the literature
[31,32], it was suggested that Cu2+ is reduced to Cu+ at 200 °C, while Cu+ is reduced
to Cu0 at 315 °C for ZSM-5 and Beta matrices. In the case of Cu-Y zeolite, the two
reduction peaks observed at 201 °C and 300 °C are only attributed to the reduction
of Cu2+ to Cu+ ions inside faujasite supercages (201 °C) and inside sodalite cages
(300 °C), as found in literature [32,33]. It should be mentioned that Cu-Y sample
presented the highest Cu2+ concentration of the evaluated matrices, according to
TPR results. This is consistent with ICP results presented in Table 1, considering the
fact that each isolated Cu2+ must be charge-compensated by two framework Al
tetrahedral sites [34]. So, the density of these species is inversely proportional to
Cu/Al molar ratio, that is the lowest for Cu-Y sample. This hypothesis was confirmed
by the studies of Paolucci et al. for Cu exchange zeolite SSZ-13 where authors
reported single metal ions, or [Cu-OH]+ species over zeolite surface [35,36]. Even

more, Göltl and coworkers have recently published studies focus in the active sites
nature in Cu- zeolite SSZ-13. They showed that Copper can be as monovalent or
bivalent cation. However, at low Si/Al ratios, the principal metal active site is Cu2+
because of the abundant presence of tetrahedral Al [37,38].
According to XRD patterns (Figure 1), zeolites structures were preserved after
Copper incorporation and all the thermal treatments applied to the materials, as
evidenced by the characteristic reflexes presence in each sample. The absence of
signals corresponding to crystalline Copper oxide phases could be due to small
particle sizes (< 4 nm) or that the ions are at exchanged positions in the zeolites.
Cu-ZSM-5
Cu-Y
Cu-Beta
5
10
15
20
25
30
35
40
2(degree)
Figure 1. X-ray diffraction patterns of fresh Cu-zeolites.

Cu-ZSM-5
Cu-Beta
Cu-Y
1400
1200
1000
800
600
Wavenumber (cm-1)
Figure 2. FTIR spectra of Cu-modified zeolites (fresh samples) in the fingerprint
region.
FTIR spectra of the samples in the fingerprint zone (1500-500 cm-1) are shown in
Figure 2. As could be seen, Copper incorporation did not alter the structures of the
zeolites, confirming XRD results. The well-defined bands at 800 and 1000 - 1300
cm-1 are assigned to the ʋs(T-O) and ʋas(T-O) stretching of Si-O(Al) vibration in
tetrahedra, respectively [24,39,40], characteristics of these kind of structures.
It is well known the identification and quantification of acid sites by the adsorption of
a probe molecule, as pyridine, followed by FTIR spectroscopy. Figure 3 shows the IR
spectra of the Cu-zeolites after pyridine adsorption and further desorption at 350 °C
for an hour at 10-4 Torr of vacuum. It is possible to observe the bands corresponding
to ʋ 19B mode (1456 cm-1) and ʋ 8A mode (1612 cm-1), corresponding to pyridine
interaction with Lewis acid sites, and the bands indicative of the interaction with

Brönsted acid sites in the ʋ 19B mode (1547 cm-1) and ʋ8A mode (1635 cm-1) [41].
According to the spectra, Cu-Y presents a higher concentration of Brönsted acid
sites than the other matrices, while Lewis acid sites are similar to that obtained for
Cu-Beta.
Cu-Beta
Cu-Y
Cu-ZSM-5
1700
1650
1600
1550
1500
1450
1400
Wavenumber (cm-1)
Figure 3. FTIR spectra of pyridine adsorbed at room temperature and desorbed at
350 ºC and 10-4 Torr over Cu-zeolites (fresh samples).
3.2. Catalytic Activity
The oxidation of 2-MMBT employing hydrogen peroxide as oxidant was selected as
model reaction in order to evaluate this kind of contaminant degradation. The
selection was the result of studying Cu-zeolites catalytic behavior and reaching the
optimal reaction conditions for this class of oxidation. Thus why, all the Copper
modified matrices were tested at 40 °C, employing 0.1 g of catalyst, acetonitrile as
solvent and a substrate/oxidant molar ratio of 0.1, as standard conditions. It should

be noted that when the reaction was done without catalyst, the final sulfide
conversion after 5 hours of reaction was just 9 mol%. Figure 4 shows the sulfide
conversion results obtained for the different zeolite matrices evaluated and the non-
catalytic reaction. Cu-Y is the most active material in order to convert 2-MMBT in its
corresponding sulfoxide and sulfone, the major obtained products, reaching a 98
mol% of conversion at the end of the reaction.
Figure 4. Zeolite matrices evaluation for 2-MMBT conversion. Reaction conditions:
catalyst mass (0.1 g), acetonitrile as solvent, sulfide/H2O2 molar ratio: 0.1, after 5 h
at 40 ºC.
From the above results, Y matrix was selected as the optimal support for Copper.
Considering that Cu-Y showed the highest total acidity, a stabilizing effect of acid
sites for Copper species could be proposed, as previously reported by Wang and
coworkers [42], that favors the Fenton-like oxidation mechanism (Scheme 1). In this
sense, under the conditions specified in Figure 4, the evolution of the oxidation as
function of reaction time was evaluated and the results are presented on Figure 5.

100
80
60
40
20
0
100
80
60
40
20
0
Conv2-MMBT
SSulfoxide
SSulfone
0
1
2
3
4
5
Reaction Time (h)
Figure 5. Reaction evolution when Cu-Y was employed as catalyst. Reaction
conditions: acetonitrile as solvent; catalyst mass: 0.1 g; sulfide/H2O2 molar ratio: 0.1;
40 ºC.
As could be seen in the figure, as 2-MMBT reacts and its conversion increases with
reaction time, the sulfoxide is obtained as primary oxidation product (P1), which is
further re-oxidized to form the sulfone (P2), as proposed in the reaction Scheme 1.
Finally, a 98 mol% of 2-MMBT conversion is reached with selectivities of 33 and 48
mol% to P1 and P2, respectively, after 5 h of reaction. Looking forward to confirm
this proposed mechanism, where Copper species seem to be the active sites, the
reaction was catalyzed by the protonic form of the zeolite matrix (H-Y) and just a 14
mol% of 2-MMBT conversion was obtained after 5 h of reaction, confirming that
dispersed Copper species are the active sites of the catalyst. Nevertheless, this high
surface area material is an optimal support that cooperates in the catalytic activity,

since the final conversion obtained employing H-Y was a 5 mol% higher than the non
catalytic reaction.
Many authors have studied solvent effect on heterogeneous catalysis [14,43,44],
since its characteristics would affect in a significant way the reaction development. In
this sense, a group of common organic molecules of different polarity and proticity
properties (Acetonitrile, Ethanol, n-Hexane, 2-Butanol) were tested as solvents,
maintaining the other reaction variables in the standard conditions, and the results
after 5 hour of reaction are presented in Figure 6 as function of the Dielectric
Constant of each solvent.
A linear enhancement on 2-MMBT conversion values was obtained as the polarity of
the solvent increases. In fact, the highest conversion values were achieved
employing Acetonitrile as solvent (highest Dielectric Constant). This aprotic solvent
favors reactant interaction with the hydrophilic surface of the catalyst, facilitating its
access to the Copper active sites.

100
80
60
40
20
0
acetonitrile
ethanol
2-butanol
n-hexane
0
5
10
15
20
25
30
35
40
Dielectric Constant (D)
Figure 6. Solvent effect on 2-MMBT conversion. Reaction conditions: sulfide/H2O2
molar ratio: 0.1, Cu-Y (0.1 g), after 5 h at 40 ºC.
When reaction temperature effect on sulfide conversion and products
selectivities was studied (Figure 7), a considerable increase on 2-MMBT conversion
(27 mol%) was observed in the variation from room temperature to 40 °C. A similar
behavior was obtained for sulfone selectitvity at the cost of sulfoxide. Liu and
coworkers attributed this behavior to the increase in H2O2 decomposition into highly
reactive OH● radicals with increase in temperature [45]. However, when a further
increment on temperature was done to 60°C, just a slight variation on sulfone
selectivity was reached.

Figure 7. Reaction Temperature Effect. Reaction conditions: 2-MMBT/H2O2 molar
ratio: 0.1, Cu-Y: 0.1 g, acetonitrile as solvent, reaction time: 5 h.
3.3. Kinetic modelling and parametrization
Effects of Copper loading (1-10 wt %), hydrogen peroxide concentration (2-5 M) and
catalyst mass (0.02-0.2 mg) are summarized in Figure 8 for Cu-Y. Figure 8.a shows
the non-linear dependency of -r2-MMBT with metal loading. The initial reaction rate
shows a lineal dependency with Copper content at loadings lower than 3 wt %. At
higher metal loading, reaction rate remains almost constant. These results could be
assigned to the nature of Copper species over zeolite matrix. Paolucci and co-
workers have studied Copper active sites and concluded that in catalysts with Cu/Al<
0,2 all Copper sites are equivalents. Beyond, at high Cu content the authors
assigned the catalytic activity decrease of Cu-zeolite to the presence of Cu-OH
species at Copper content >3 % [35]. Nevertheless, higher size Copper clusters on

the zeolite surface, when higher metal content is incorporated, may produce mass
transfer problems by the blockage of some pores mouths [24,30]. Thus, 3 wt% is the
optimal metal incorporation in order to improve catalytic activity and it was selected
for further analysis.
Catalytic nature of the oxidation was confirmed by the linear dependency of –r2-MMBT
with Cu-Y mass, as expected (Figure 8.b). Nevertheless, a low non-catalytic
oxidation rate was measured, corresponding to 0 mg of Cu-Y.
The linear dependency of –r2-MMBT with [H2O2] indicates a first order reaction (Figure
8.c). This result is in agreement with the proposed oxidative mechanism (Scheme 1).
In order to confirm the radical oxidation pathway proposed, a radical scavenger
(isopropyl alcohol) was added to the reaction mixture after 30 minutes of reaction
and 2-MMBT conversion was followed as function of time. It was noticeable that
when the radical scavenger was incorporated, 2-MMBT conversion practically did not
vary, showing just a 10 mol% of increment in the next 270 minutes of reaction.

(a)
(b)
(c)
Figure 8. Initial rates plots. (a) 2-MMBT oxidation rate vs Cu amount (wt%), (b) 2-
MMBT oxidation rate vs catalyst mass, (c) 2-MMBT oxidation rate vs [H2O2].

Table 2 presents the obtained results for the parametrization procedure of the
proposed mechanism (Scheme 1). The model accuracy to fit experimental data was
demonstrated by χ2(95%) > weighed residuals. Furthermore, the parameter
estimation gave meaningful values for k1, k3 and k4. The obtained value of k1
indicates that global reaction rate is controlled by the Cu(II) activation step. In
addition, constants of 2-MMBT and sulfoxide oxidation steps (k3 and k4) are in
accordance to the high reactivity of the OH● radicals [27–29].
Table 2. Initial guesses and optimized rate constants values.
(M-1 s-1)
Initial guess
(M-1 s-1)
Optimized
(M-1 s-1)
k1
0.002a
0.0016 ±0.0003
7.6 x 10 6
4.0 x 10 9
k2
7.6 x 10 6 a
4.0 x 10 9 b
4.0 x 10 9 b
1.2 x 10 6 a*
3.2 x 10 5 a*
5.0 x 10 8 a*
3.1 a*
3
3
k3
±6.5 x 10
k4
4.0 x 10 9
±6.0 x 10
k5
-
-
-
-
-
k6
k7
k8
k9
2.1 x 10 9 a*
Ea
15000
465.49±136 J mol-1
Initial guess of σ2
Optimized σ2
Residuals
Χ2 (95%)
0.01
0.04 ±0.003
158.5
183.96
*Fixed values during parametrization. a From [46]. b From [47].
Figure 9 shows modelled and experimental concentrations profiles of 2-MMBT and
its oxidized products. Model ability to fit experimental data could be observed in

figure 9.c and d. Moreover, experimental profiles corresponding to higher Copper
concentrations had been successfully fitted (Figure 9.e). However, reaction rate is
over-estimated at low catalyst concentration (Figure 9.a). Probably, non-catalytic
oxidation pathways, not included in the proposed model, can become more
important at these conditions. Some authors have modelled the effect of acid
catalysts in oxygen atom transfer from H2O2 to a nucleophile as H2S or (CH3)2S.
They conclude that weak acid catalysts can acts on the activation of O-O bond
cleavage and favours the sulfides oxidation [48]. It is possible that at low metal
content, the acid sites of the zeolitic matrix, catalyse H2O2 activation.
In order to evaluate the temperature effect over k1, Arrhenius equation has been
included into the model parametrization. Activation energy value (Ea) was
determined (see Table 2). The obtained value is lower than those reported for
homogeneous Fenton reactions witch vary from 8000 to 70000 J mol-1 as function of
the reaction medium and the organic substrate that was oxidized [49–51]. However,
this result is consistent with the proposed radical oxidation mechanism [52,53],
improved by Copper stabilization through the acid matrix.
Homogeneous catalytic system
In order to evaluate the effect of Copper on catalysts activity, and the potential
diffusion problems, activities measurements and rate constants estimations were
carried on for systems catalyzed by soluble Cu(II) (Cu(CH3COO)2). Rate constant
k1hom was estimated by the parametrization procedure for homogenous experiments.
Figure 9.f and g show the agreement between experimental vs modelled 2-MMBT
and its oxidized products evolution. The obtained k1hom rate constant (k1hom
:
0.0008±0.0001) is lower than the optimized k1 value for the heterogeneous system.
This result allows us to discard any diffusion limitations. Also, confirming the

proposed oxidation model in which the small size H2O2 molecules are the species
that diffuse into the zeolite channels. Moreover, the lower value obtained for the
homogeneous rate constant, suggests that there is a synergic effect of the zeolite
matrix over Copper catalytic activity, as probed when H-Y was evaluated as catalyst.
Similar results were obtained in partial methane oxidation catalyzed by Cu-SSZ-13
zeolites. In fact, authors assigned this behaviour to the environment generated by
Copper confinement effect into the zeolite channels [37,38].

0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.5
0.4
0.3
0.2
0.1
0.0
[Cu(3)Y]: 0.0015 M
[Cu-Y]: 0.0015 M
(a)
[Cu(3)Y]: 0.008 M
[Cu-Y]: 0.008 M
[H2O2]: 1.98 M
[2-MMBT]:0.39 M
[H2O2]: 3.23 M
(b)
[H2O2]: 1.98 M
[2-MTBT]: 0.39M
[H2O2]: 3.23 M
[2-MTBT]: 0.33 M
[2-MMBT]:0.33 M
0
2000 4000 6000 8000 1000012000140001600018000
0
2000 4000 6000 8000 1000012000140001600018000
Tiempo (seg)
Tiempo (seg)
Time (s)
Time (s)
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
[Cu(3)Y]: 0.008 M
[Cu-Y]: 0.008 M
(c)
[Cu(3)Y]: 0.015 M
[Cu-Y]: 0.015 M
(d)
[H2O2]: 3.23 M
[H O ]: 3.23 M
[H2O22]: 23.23 M
[H2O2]: 3.23 M
[2-MTBT]: 0.33M
[2-MTBT]: 0.33 M
[2-MMBT]:0.33 M
[2-MMBT]:0.33 M
0
2000 4000 6000 8000 1000012000140001600018000
0
2000 4000 6000 8000 1000012000140001600018000
Tiempo (seg)
Tiempo (seg)
Time (s)
Time (s)
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
[Cu]
: 0.008 M
[Cu]: 0.008 M
(e)
SOLUBLE
[H O ]: 3.23 M
[H2O2]: 3.23 M
(f)
2
2
[2-MTBT]: 0.33M
[2-MMBT]:0.33 M
[Cu]SOLUBLE: 0.008 M
[Cu]: 0.008 M
[H2O2]: 1.62 M
[H2O2]: 1.62 M
[2-MTBT]: 0.33M
[2-MMBT]:0.33 M
0
2000
4000
6000
8000
0
2000
4000
6000
8000
Tiempo (seg)
Time (s)
Tiempo (seg)
Time (s)
Figure 9. Experimental (points) and modelled (lines) concentration profiles of 2-
MMBT (continues lines) and its oxidized products (sulfoxide: dotted lines, sulfone:
dashed line).
4.
Conclusion