Kinetics of lab prepared manganese oxide catalyzed oxidation of benzyl alcohol in the liquid phase

Article abstract of DOI:10.1002/kin.20922

The oxidation of benzyl alcohol in the liquid phase was studied over manganese oxide catalyst using molecular oxygen as an oxidant. Manganese oxide was prepared by a mechanochemical process in solid state and was characterized by chemical and physical techniques. The catalytic performance of manganese oxide was explored by carrying out the oxidation of benzyl alcohol at 323-373 K temperature and 34-101 kPa partial pressure of oxygen. Benzaldehyde and benzoic acid were identified as the reaction products. Typical batch reactor kinetic data were obtained and fitted to the Langmuir-Hinshelwood, Eley-Rideal, and Mars-van Krevelene models of heterogeneously catalyzed reactions. The Langmuir-Hinshelwood model was found to give a better fit. Adsorption of benzyl alcohol at the surface of the catalyst followed the Langmuir adsorption isotherm. The heat of adsorption for benzyl alcohol was determined as -18.14 kJ mol^-1. The adsorption of oxygen followed the Temkin adsorption isotherm. The maximum heat of adsorption for oxygen was -31.12 kJ mol^-1. The value of activation energy was 71.18 kJ mol^-1, which was apparently free from the influence of the heat of adsorption of both benzyl alcohol and oxygen.

Full text of DOI:10.1002/kin.20922

Kinetics of Lab Prepared
Manganese Oxide
Catalyzed Oxidation of
Benzyl Alcohol in the Liquid
Phase
MUHAMMAD SAEED,1,2 MOHAMMAD ILYAS,2,3 MOHSIN SIDDIQUE2,4
1Department of Chemistry, Government College University Faisalabad, Faisalabad 38000, Pakistan
2National Centre of Excellence in Physical Chemistry, University of Peshawar, Peshawar 25120, Pakistan
3Department of Chemistry, Qurtuba University of Science and Information Technology Peshawar, Peshawar 25120, Pakistan
4Department of Chemistry, Bacha Khan University, Charsadda 24420, Pakistan
Received 4 September 2014; revised 16 January 2015; accepted 30 March 2015
DOI 10.1002/kin.20922
Published online 21 April 2015 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: The oxidation of benzyl alcohol in the liquid phase was studied over manganese
oxide catalyst using molecular oxygen as an oxidant. Manganese oxide was prepared by a
mechanochemical process in solid state and was characterized by chemical and physical tech-
niques. The catalytic performance of manganese oxide was explored by carrying out the oxi-
dation of benzyl alcohol at 323–373 K temperature and 34–101 kPa partial pressure of oxygen.
Benzaldehyde and benzoic acid were identified as the reaction products. Typical batch reactor
kinetic data were obtained and fitted to the Langmuir–Hinshelwood, Eley–Rideal, and Mars–
van Krevelene models of heterogeneously catalyzed reactions. The Langmuir–Hinshelwood
model was found to give a better fit. Adsorption of benzyl alcohol at the surface of the cat-
alyst followed the Langmuir adsorption isotherm. The heat of adsorption for benzyl alcohol
was determined as –18.14 kJ mol−1. The adsorption of oxygen followed the Temkin adsorption
isotherm. The maximum heat of adsorption for oxygen was –31.12 kJ mol−1. The value of acti-
vation energy was 71.18 kJ mol−1, which was apꢀpC arently free from the influence of the heat of
adsorption of both benzyl alcohol and oxygen. 2015 Wiley Periodicals, Inc. Int J Chem Kinet
47: 447–460, 2015
INTRODUCTION
The oxidation of organic compounds in the liquid
phase is one of the most important and widely used
reactions in laboratory-scale organic synthesis as well
as in large-scale chemical industrial processes [1,2].
Correspondence to: Muhammad Saeed; e-mail. pksaeed2003@
yahoo.com.
C
ꢀ
2015 Wiley Periodicals, Inc.

448
SAEED, ILYAS, AND SIDDIQUE
Many processes with different reagents and methods
are available for the oxidation of organic compounds.
Stoichiometric oxidants such as peroxides or high ox-
idation state metal compounds such as permanganate
and dichromate are frequently used reagents for oxida-
tion purposes. But these reagents are expensive, toxic,
and produce large amount of wastes and hence separa-
tion and disposal of the waste increases the number of
steps [3,4]. When these oxidation reactions are scaled
to tons instead of grams, the use of these stoichiomet-
ric oxidants is not an attractive option. For these kinds
of oxidation reactions, an alternative and environment
friendly oxidant is desirable. An ideal oxidant for any
large-scale oxidation reaction is the one which has low
price, high quality (purity), nontoxic, and easy avail-
ability. Molecular oxygen is the one which qualify this
criterion [5]. It is easily available as it is present in
air, and the only by-product produced from it is wa-
ter. There are a few points, however, which make the
use of molecular oxygen challenging: first, although
molecular oxygen has a high oxidation potential, it is
not very reactive toward organic molecules and sec-
ond the reactions where molecular oxygen is present
are often radical reactions, which are hard to control.
To make an efficient use of molecular oxygen as an
oxidant, an appropriate catalyst is needed, which can
activate the oxygen molecules for an appropriate reac-
tion. Both homogeneous and heterogeneous catalysts
can be used for these transformations. However, with
homogeneous catalytic reactions both the reactants and
catalysts are present in one phase and from an engi-
neering viewpoint, a major disadvantage of this arises
from the difficulty in separating the products from the
catalyst, which increases the number of steps [6–9].
Therefore, the use of heterogeneous systems would
be superior to homogeneous counterparts due to the
easier separation of products and catalysts and reuse
of catalysts. Metal oxides and supported metal oxides
have been proposed as effective heterogeneous cata-
lysts for the oxidation of organic compounds. In many
cases, alumina- and zirconia-supported metal oxides
of precious metals such as Pt and Pd have shown high
activity for the oxidation of organic compounds. From
the economical point of view, using oxides of non-
precious transition metals like manganese (Mn), as
heterogeneous catalysts for the oxidation of organic
compounds, using clean oxidants such as molecular
oxygen is of great importance [10–13]. Many synthetic
routes can be applied for the preparation of Mn-based
catalysts such as pulsed laser deposition, sol–gel route,
reduction-oxidation route, gel hydrothermal oxidation,
homogeneous precipitation, staged oxidation process,
cobalt salt decomposition, and mechanochemical pro-
cess. The later one is a novel one for large-scale syn-
thesis, in which manganese oxide can be obtained in
the solid-state displacement chemical reaction, either
during milling or heat treatment. This mechanochem-
ical process for the synthesis of manganese oxide is
suitable for the large-scale synthesis due its simplicity
and low cost. In this work, the synthesis of the Mn cat-
alyst, which is much cheaper as compared to precious
metal catalysts, by the mechanochemical process and
its use as a catalyst for the oxidation of benzyl alcohol
in the liquid phase has been investigated.
EXPERIMENTAL
Preparation of Catalyst
For the preparation of catalyst, the method of Li et
al. [10] was modified [11]. Solid manganese chloride
and potassium permanganate were first ground at a
2:3 mol ratio in an agitated mortar at room tempera-
ture for 30 min and then heated for several hours and
ground again. After grinding, the reaction mixture was
kept at 373 K for 24 h to complete the reaction. The
resultant solid was washed with distilled water to re-
move the untreated precursor materials. After washing,
it was dried at 383 K for 24 h. The resultant black pow-
der was designated as the unreduced manganese oxide
catalyst. A portion of this sample was reduced at 573 K
under the flow of molecular hydrogen at a flow rate of
100 mL min−1, for 2 h. The resulted powder was des-
ignated as a reduced manganese oxide catalyst.
Characterization
The prepared catalyst was characterized by the de-
termination of oxygen content, surface area, particle
size, XRD, FTIR, and SEM analyses as described ear-
lier [11].
Oxidation Protocol
The liquid-phase oxidation of benzyl alcohol was car-
ried out in a magnetically stirred round-bottom Pyrex
glass three-necked batch reactor of 50 mL capacity,
provided with a reflux condenser and mercury ther-
mometer. Reaction temperature was maintained by us-
ing a hot plate. For a typical reaction run, the reactor
was charged with 2 mmol benzyl alcohol and 1.6 mmol
methyl benzoate as an internal standard, in 10 mL of
n-octane as a solvent. After getting the required tem-
perature (363 K), 0.1 g of catalyst was added to the
reactor. The flow of oxygen was kept at a flow rate of
60 mL min
−1, while stirring the reaction mixture con-
tinuously at the agitation speed of 800 rpm. Moisture
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
449
was removed from oxygen by passing it through a
moisture adsorbent. This drying of oxygen is neces-
sary because it affects the selectivity of the reaction
products. We have noted that if molecular oxygen is
not dried before entering to the reactor, formation of
benzoic acid in addition to benzaldehyde can be ob-
served even at the beginning of the reaction. In case of
dry conditions only, benzaldehyde was detected as the
reaction product at the start of reaction. Before entering
to the reactor, molecular oxygen was saturated with a
solvent by passing it through the saturator containing
a solvent at condenser temperature. This saturation of
oxygen with the solvent minimizes the loss of solvent
with flow of oxygen from the reaction mixture. The
catalyst was separated from the reaction mixture with
Whatman glass microfiber filter No. 1825 055 using
glass syringes. Analysis of the reaction mixture was
carried out by a UV–vis spectrophotometer (Shimadzu
UV-160A; Tokyo, Japan) and GC (Clarus 500; Perkin
Elmer, USA) equipped with a Flame ionization detec-
tor (FID).
of benzyl alcohol in the presence of nitrogen could be
due to the oxygen from the lattice taking part in the
reaction; however, this reduction of manganese oxide
could be limited to a few surface layers as there is no
difference in the XRD pattern of the unused and used
catalysts as described earlier [11].
In another experiment, the amount of benzyl alco-
hol was doubled, almost same conversion of benzyl
alcohol was achieved under flow of oxygen in 2 h as
was in 1 h. These results strongly indicated that the
present manganese oxide acts as a catalyst for the oxi-
dation reaction between benzyl alcohol and molecular
oxygen. Furthermore, evidence for the catalytic role of
the manganese oxide was obtained by the dependence
of benzyl alcohol conversion on the partial pressure
of oxygen. The conversion of benzyl alcohol was re-
duced from 82.3% to 53% when the partial pressure of
oxygen was reduced from 101 to 34 kPa. Finally, the
used catalyst was washed, dried, and reused for the ox-
idation of benzyl alcohol under the flow of molecular
oxygen. It was observed that the used catalyst has same
catalytic performance as a fresh catalyst. From these
observations, it could be concluded that manganese ox-
ide in the present case acts as a catalyst rather than a
stoichiometric oxidant.
RESULTS AND DISCUSSION
Comparison of Reduced and Unreduced
Manganese Oxide as Catalyst
Effect of Mass Transfer
For comparison of catalytic activity of reduced and
unreduced manganese oxide, separate experiments
were carried out at 363 K by taking 2 mmol benzyl al-
cohol in 10 mL n-octane using reduced and unreduced
manganese oxides as catalysts. About 98% and 50%
conversion of benzyl alcohol was achieved at a batch
time of 2 h using reduced and unreduced manganese
oxide as catalysts, respectively. As reduced manganese
oxide was more active catalyst, therefore it was used
as a catalyst in all subsequent experiments.
The oxidation of benzyl alcohol using manganese ox-
ide as a catalyst in the present case is a typical slurry-
phase reaction having one liquid reactant (benzyl alco-
hol), a gaseous reactant (oxygen), and a solid catalyst
(manganese oxide). True kinetics of any multiphase re-
action is likely to be masked by the effect of mass trans-
fer. Hence, before investigating any oxidation kinetics,
it is needed to eliminate any mass transfer limitation
that might exist. The effect of mass transfer limitation
may either be external or internal mass transfer limita-
tion. The effect of external mass transfer on the rate of
reaction can be investigated by studying the effect of
speed of agitation on the rate of reaction (conversion).
The dependence of the rate (conversion) of the reac-
tion on speed of agitation shows the existence of mass
transfer limitation. To investigate the effect of speed of
agitation on the oxidation reaction, conversion of ben-
zyl alcohol at various speeds of agitation in the range
of 50–900 rpm at 363 K for 20 min was studied. The
effect of agitation on conversion is shown in Fig. 1,
which shows that initial conversion of benzyl alcohol
increases with increasing agitation speed, showing a
mass transfer regime from 50 to 450 rpm. The later
part of the graph shows that the increase in agitation
speed above 500 rpm has no effect on the conversion,
indicating that the oxidation of benzyl alcohol is most
Manganese Oxide as a Catalyst or Oxidant
As manganese oxide can be used as stoichiometric ox-
idants for the oxidation of alcohols, therefore it should
be confirmed that whether the manganese oxide in the
present investigation acts as a stoichiometric oxidant or
as a catalyst. For this purpose, separate reactions with
2 mmol benzyl alcohol were carried out under the flow
of molecular nitrogen and oxygen. Traces of oxygen in
nitrogen gas were removed by using specific oxygen
traps (C. R. S. Inc.; 202223). The results revealed that
conversion of benzyl alcohol under the flow of molec-
ular nitrogen was less (31.5% of 2 mmol) as compared
to the conversion under the flow of molecular oxygen
(82.3% of 2 mmol) in 1 h at 363 K. The low conversion
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

450
SAEED, ILYAS, AND SIDDIQUE
Time Profile Investigation
The time course study of manganese oxide catalyzed
oxidation of benzyl alcohol was monitored periodi-
cally. This investigation was carried out by suspend-
ing 0.1 g catalyst in 10 mL n-octane, 2 mmol benzyl
alcohol, 1.6 mmol methyl benzoate, and passing oxy-
gen through the reaction mixture with a flow rate of
60 mL min−1 at 1 atm pressure. It was observed that
only benzaldehyde was detected as the reaction product
in beginning of the reaction. Once the benzyl alcohol
was almost completely converted (ꢀ98%) to benzalde-
hyde, then formation of benzoic acid started (Fig. 2a).
The effect of reaction temperature on the progress of
the oxidation of benzyl alcohol was studied in the tem-
perature range of 323–363 K. Conversion of benzyl
alcohol increased from 39.4% to 95.6% in 120 min
with an increase in temperature from 323 to 363 K
for 0.1 g of manganese oxide and 2 mmol benzyl al-
cohol in 10 mL solvent. Time profile investigation at
various temperatures is shown in Fig. 2. In all experi-
ments, reaction conditions were kept the same except
temperature.
Figure 1 Effect of mass transfer on the oxidation of benzyl
alcohol. Reaction conditions: BzOH 2 mmol/10 mL solvent,
catalyst 100 mg, temperature 363 K, pressure of oxygen
101 kPa, time 20 min.
probably free from the external mass transfer limita-
tions, i.e. the reaction has taken place in a kinetically
controlled regime. This is the region of our interest,
and all the subsequent experiments were performed at
800 rpm. A Polytetrafluoroethylene (PTFE) coated stir-
rer bar was used for stirring.
Kinetics Analysis
For the internal mass transfer limitations, the Thiele
modulus (φ) can be used, which provides a convenient
nondimensional measure of the importance of inter-
nal diffusion on the rate of reaction (conversion). If
the Thiele modulus for a reaction is less than 1, it is
considered that the said reaction is free from internal
mass transfer limitations [4]. The Thiele modulus can
be calculated by the following expression:
The rates of heterogeneous catalytic reactions are com-
plex, because adsorption of the reacting species on the
surface of the catalyst, reactions between the adsorbed
species and desorption of the products take place si-
multaneously. Basically, the catalytic reactions are de-
scribed with surface concentrations; however, the ef-
fect of surface concentrations on rates of reactions can
be eliminated by using simplifying assumptions, such
as quasi-equilibrium and quasi–steady-state hypothe-
ses [14]. The kinetics of the oxidation of benzyl alcohol
in the present case can be expressed as the power rate
law,
ꢀ
k
φ = L
(1)
D
where L is the length of pore (in meters) of the bulk
catalyst, D is the coefficient of diffusion of benzyl
alcohol, and k is the initial rate for benzyl alcohol
disappearance. Taking the length of the pore (L) equal
to the maximum particle size (50 μm), the initial rate
(k) 4.55 × 10−3 min−1, and the diffusion coefficient
(D) for benzyl alcohol as 2.8 × 10−9 m2 s−1 [4], the
Thiele modulus (φ) is 0.06, which is much less than 1.
The small value of the Thiele modulus indicates that
the rate of reaction is not limited by internal diffusion.
Thus on the basis of the effect of agitation on the rate of
reaction (conversion) and the Thiele modulus, we can
confidently assume that in the present conditions the
oxidation of benzyl alcohol catalyzed by manganese
oxide is a kinetically controlled reaction.
Rate = kꢁ[BzOH]m[O2]n
(2)
where k is the rate constant and m and n are the reac-
tion orders with respect to benzyl alcohol and oxygen,
respectively. This representation has limited utility for
mechanistic elucidation; however, it can be used to ob-
tain a preliminary dependency of the rate of the reaction
on the concentration of benzyl alcohol and oxygen. At
constant partial pressure of oxygen, the rate expression
becomes as
Rate = kꢁ[BzOH]m
(3)
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
451
Figure 2 Time profile for the oxidation of benzyl alcohol (a) at 363 K, (b) at different temperatures. Reaction conditions:
BzOH 2 mmol/10 mL n-octane, catalyst 100 mg, agitation 800 rpm, pressure of oxygen 101 kPa.
where
the approach of fitting experimental data to one of the
three possible mechanisms of heterogeneous cataly-
sis [2,15–18]:
kꢁ = k[O2]n
(4)
Equation (3) can be integrated to the following equa-
tion:
i. The Langmuir–Hinshelwood mechanism,
ii. The Mars–van Krevelen mechanism,
iii. The Eley–Rideal mechanism.
−ln(1 − X) = kꢁt
(5)
According to Langmuir–Hinshelwood mechanism,
the reaction proceeds in two steps. In the first step, the
reactants adsorb on the surface of the catalyst followed
by a reaction between adsorbed reactants to form prod-
ucts in the second step. For manganese oxide catalyzed
oxidation of benzyl alcohol with molecular oxygen, the
rate expression can be written as
where X is conversion of benzyl alcohol at batch time
t. Equation (5) was applied to time profile data for
the oxidation of benzyl alcohol at various tempera-
tures as given in Fig. 3. The linearity of plots indicates
the first-order dependence of the oxidation rate on the
concentration of benzyl alcohol. The values of the rate
constants (k/) were calculated from the slopes of the
graphs and are listed in Table I. However, such repre-
sentation has limited utility. Hence, we have followed
Rate = kr θBzOHθO
(6)
2
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

452
SAEED, ILYAS, AND SIDDIQUE
Figure 3 Application of the first-order kinetic expression to time profile data at different temperatures from Fig 2.
Table I First-Order Rate Constants Determined from
Fig. 3
as
KBzOH[BzOH]KO [O2]gn
T (K)
kꢁ (min−1
)
R2
2
Rate = kr
(8)
(1 + KBzOH[BzOH] + KO [O2]ng)2
2
323
333
343
353
363
0.0102
0.0191
0.0385
0.0760
0.1483
0.9709
0.9974
0.8269
0.9288
0.9580
At constant partial pressure of oxygen, the rate ex-
pression can be written as
ab[BzOH]
(c + b[BzOH])2
Rate =
(9)
where θBzOH and θO represents surface covered by
2
benzyl alcohol and molecular oxygen, respectively.
Adsorption of benzyl alcohol and oxygen on the
surface of catalyst may take place either according to
Langmuir, Temkin, or Freundlich adsorption isotherm.
The Langmuir adsorption isotherm may be either com-
petitive or noncompetitive. For competitive adsorption,
the rate expression can be written as
where a, b, and c are kr KO2[O2], KBzOH, and 1 +
KO2[O2], respectively.
Similarly, at the constant concentration of benzyl
alcohol, the rate expression is
ab[O2]
(c + b[O2])2
Rate =
(10)
KBzOH[BzOH]KO [O2]gn
2
Rate = kr
(1 + KBzOH[BzOH] + KO [O2]ng + KP[P])2
In case of noncompetitive adsorption of benzyl al-
cohol and oxygen, the rate expression can be written
as
2
(7)
where KBzOH, KO , and KP represent adsorption coeffi-
KBzOH[BzOH]KO [O2]gn
2
2
cient for benzyl alcohol, oxygen, and products, respec-
tively. The value of n can be 1 or 0.5 for molecular or
dissociative adsorption of oxygen, respectively.
Rate = kr
(11)
(1 + KBzOH[BzOH])(1 + KO [O2])
2
Adsorption/desorption processes in the above steps
are fast enough to be assumed at equilibrium. This
assumption helps to simplify the above expression
Equation (11) can be simplified by keeping the con-
centration of benzyl alcohol or oxygen constant. At
constant partial pressure of oxygen, Eq. (11) can be
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
453
transformed to Eq. (12)
oxidizes the substrate molecule and hence produces a
partially reduced catalyst. In the second step, the re-
duced catalyst is reoxidized by molecular oxygen. The
rate equation for the Mars–van Krevelen mechanism
can be given by the following expression:
ab[BzOH]
Rate =
(12)
1 + b[BzOH]
Similarly at the constant concentration of benzyl
alcohol, Eq. (11) can be transformed to Eq. (13):
k1[BzOH]k2[O2]ng
βk1[BzOH] + k2[O2]gn
Rate =
(20)
ab[O2]
Rate =
(13)
1 + b[O2]
In this equation, k1 and k2 represent the rate constant
for the oxidation of benzyl alcohol and reoxidation of
the partially reduced catalyst, respectively. β is the
stoichiometric coefficient for oxygen, which is consid-
ered as 0.5. The above expression can be simplified
by considering oxygen or benzyl alcohol constant as
discussed in the Langmuir–Hinshelwood mechanism.
For the constant oxygen and benzyl alcohol rate can be
expressed by expressions (21) and (22), respectively:
Considering the Temkin adsorption isotherm, the
rate expression becomes
−
−
Rate = kr (K1 ln K2[BzOH])(K1 ln K2[O2]) (14)
where K1 and K2 are constants related to heat of adsorp-
tion, which decreases linearly with surface coverage in
this case.
a[BzOH]
For the constant concentration of oxygen and benzyl
alcohol, Eq. (14) can be simplified to Eqs. (15) and
(16), respectively:
Rate =
(21)
b + c[BzOH]
−
a[O2]
Rate = kr (K1 ln K2[BzOH])
(15)
Rate =
(22)
b + c[O2]
According to the Eley–Rideal mechanism, the
gaseous reactant get adsorbed on the surface of the
catalyst while the second reacts with the adsorbed re-
actant from the fluid phase. In the present case, oxygen
is adsorbed at the surface while benzyl alcohol remains
in the fluid phase.
−
−
−
Rate = kr (K1 ln K2[O2])
(16)
A plot of ln[BzOH] or ln[O2] against the rate gives
a straight line.
Similarly, considering the Freundlich adsorption
isotherm, the rate expression becomes
The rate expression for the Eley–Rideal mechanism
can be given by Eq. (23)
1
1
/
/
n
n
Rate = kr KBzOH[BzOH] KO [O2]
(17)
2
Rate = kr θO [BzOH]
(23)
2
where KBzOH and KO2 are the adsorption coefficient for
benzyl alcohol and oxygen, respectively, and n (>1) is
a constant.
Considering the Langmuir adsorption isotherm, we
get
This equation can be simplified by keeping oxy-
gen/benzyl alcohol constant and lumping together all
the constants as below.
KO [O2]ng
2
Rate = kr
[BzOH]
(24)
1 + KO [O2]ng
2
−
At the constant partial pressure of oxygen, the above
equation can be transformed as
Rate = kr KBzOH[BzOH]1/n
(18)
Rate = a[BzOH]
(25)
−
Rate = kr KO [O2]1/n
(19)
2
Similarly in case of the constant concentration
of benzyl alcohol, Eq. (25) can be transformed to
Eq. (26):
A plot of ln[BzOH] or ln[O2] against ln rate gives a
straight line.
Like the Langmuir-Hinshelwood mechanism, the
Mars–van Krevelen mechanism also comprises two
steps. In the first step, the lattice oxygen of the catalyst
a[O2]
Rate =
(26)
1 + b[O2]
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

454
SAEED, ILYAS, AND SIDDIQUE
Figure 4 Applicability of noncompetitive Langmuir–Hinshelwood model (Eq. (12)) to time profile data by nonlinear least-
square analysis at 101 kPa pressure of oxygen and various temperatures: (a) 363 K, (b) 353 K, (c) 343 K, (d) 333 K, and
(e) 323 K.
The time profile data at various temperatures in the
range of 323–263 K at 101 kPa partial pressure of
oxygen from Fig. 2 was subjected to kinetic analysis
using a nonlinear regression analysis according to the
above models, using Curve Expert 1.3 software.
Equation (12) fitted best to the data at constant par-
tial pressure of oxygen (101 kPa) and different temper-
atures. This equation was applied to time profile data
in an integral form (Eq. (27)) as given in Fig. 4.
The values of noncompetitive Langmuir–
Hinshelwood constants were determined using
Curve Expert 1.3 software and are listed in Table II.
Effect of Partial Pressure of Oxygen
The effect of the partial pressure of oxygen on the
oxidation of benzyl alcohol was investigated in the
range of 34–101 kPa with a constant initial 0.2 M
benzyl alcohol concentration at various temperatures
in the range of 323–363 K. For the determination
of various partial pressure of oxygen, nitrogen was
admixed with oxygen having the total flow rate of
t = (b ∗ (0.2 − x) ∗ (4 + b ∗ (0.2 − x))
+ 2 ∗ ln(0.2/x))/(2 ∗ a ∗ b)
(27)
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
455
Table II Kinetic Constants Determined by Nonlinear
where Et and ꢀHad represents the true activation en-
ergy and heat of adsorption, respectively. Since the
adsorption is exothermic in nature in this case, there-
fore the apparent activation energy is less than the true
activation energy by the amount of the heat of adsorp-
tion. Obviously, it can be concluded that the heat of
adsorption of oxygen decreases with an increase in the
partial pressure of oxygen, a situation that arises due
to the nonhomogeneous nature of the oxygen adsorp-
tion sites. It was proved that the Langmuir adsorption
isotherm for oxygen adsorption was not successful.
Similarly, the Freundlich isotherm was also not appli-
cable to the data. The Temkin adsorption isotherm was
applicable to the data. Insertion of the Temkin adsorp-
Least-Square Analysis According to
Langmuir–Hinshelwood, Noncompetitive Model
(Eq. (12)) at the Constant Oxygen Partial Pressure of
101 kPa
kr KO2 pO2
a =
= krꢁ
1+KO2 pO2
b = KBzOH
,
T (K)
(mol L−1 min−1
)
(L mol−1
)
R2
323
333
343
353
363
0.0102
0.0191
0.0385
0.0760
0.1483
0.4271
0.3529
0.2979
0.2615
0.2237
0.998
0.997
0.999
0.999
0.997
tion isotherm in Eq. (29) for θO , gives
2
60 mL min−1 as described earlier [4,11,16,17]. It
was found that with the increase in partial pressure
of oxygen conversion of benzyl alcohol increases as
shown in Fig. 5. The time profile data at various par-
tial pressures of oxygen from Fig. 5 was subjected to
kinetic analysis according to Langmuir–Hinshelwood,
Mars–van Krevelen, and Eley–Rideal mechanism as
discussed earlier. The Langmuir–Hinshelwood mech-
anism was applicable where adsorption of benzyl al-
cohol at the surface of catalyst takes place according
to the noncompetitive Langmuir adsorption isotherm,
from which rate coefficients and adsorption coeffi-
cient of benzyl alcohol were determined (Table III).
It is evident from Table III that the values of krꢁ
depends upon the oxygen partial pressure and temper-
ature as well. Using the Arrhenius equation by plot-
ting ln (krꢁ ) versus 1/T at various oxygen partial pres-
sures (Fig. 6) gives the apparent activation energies
(Table IV), which are free from the effect of the heat of
adsorption of benzyl alcohol (ꢀHBzOH). The apparent
activation energies depend on the partial pressure of
oxygen.
krꢁ = kr K1 ln(K2pO )
(31)
2
which could be changed to
krꢁ = a ln(bpO )
(32)
2
Values of a and b were determined by applying
Eq. (32) to the data by nonlinear least-square fit analy-
sis using Curve Expert 1.3. Values of a, b, and regres-
sion coefficient are listed in Table V. It shows that the
Temkin Isotherm is the best fit for oxygen adsorption
in this case. Furthermore, recalculation of the krꢁ val-
ues using the constants a and b from Table V shows
excellent agreement with the krꢁ values in Table V, as
given in Fig. 7. By using the values obtained from the
Temkin isotherm the apparently true activation energy
(free from the heat of adsorption of both benzyl alco-
hol as well as the heat of adsorption of oxygen) can be
calculated. It can be seen that the constant a in Eq. (32)
is
a = kr K1
(33)
At constant partial pressure of oxygen, Eq. (6) can
also be written as
RT
Considering that K1 =
, where as ꢀH0 is the
maximum value of the heaαtꢀoHf0adsorption obtained at
the lowest surface coverage, i.e. as θ → 0 and α is the
proportionality constant used for estimating the de-
crease in the heat of adsorption with surface coverage.
Therefore, dividing the value of a by the temperature
T will give
Rate = krꢁ θBzOH
(28)
From the comparison of Eqs. (6) and (28), it is
obvious that
krꢁ = kr θO
(29)
2
a
T
R
= kr
(34)
This is the reason that the apparent activation en-
ergies (Eap) increase with the increase in the oxygen
partial pressure (Table IV). Considering that the appar-
ent activation energy is
αꢀH0
A plot of ln(Ta ) versus 1/T will give the apparently
true activation energy, free from the heat of adsorption
of both oxygen and benzyl alcohol as given in Fig. 8,
giving the Et = 71.8 kJ mol−1, which is almost equal to
Eap = Et + ꢀHad
(30)
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

456
SAEED, ILYAS, AND SIDDIQUE
Figure 5 Time profile data for oxidation of benzyl alcohol at various temperatures and partial pressure of oxygen. Reaction
conditions: BzOH 2 mmol/10 mL n-octane, catalyst 100 mg, agitation 800 rpm.
due to the fact that bꢁ has some contribution from the
entropy of adsorption, which has more pronounced
effect at the highest temperature.
the value obtained previously for the vapor-phase de-
hydrogenation of cyclohexanol to cyclohexanone cat-
alyzed by Y2O3/ZrO2, using the Temkin isotherm [19].
On the other hand, the constant b in Eq. (32) is
A reasonably good straight line (R2 = 0.923) is
obtained for the other four temperatures (323, 333,
343, and 353 K) giving a value of –31.12 kJ mol−1
for the heat of adsorption (ꢀH0) of oxygen (Fig. 9a).
Similarly for the determination of heat of adsorption
of benzyl alcohol, Van’t Hoff equation was applied to
ꢀH
0
b = bꢁe , where bꢁ is a constant. If bꢁ is independent
RT
of temperature then a plot of ln(b) versus I/T would
give the value of ꢀH0. However, a plot of ln(b) versus
1/T (Van’t Hoff equation) shows that the value at 363 K
has a large deviation as given in Fig. 9a. It is probably
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
457
Figure 6 Application of Arrhenius equation to the data
from Table III.
Table IV Apparent Activation Energies Calculated
from Fig. 6
pO2 (kPa)
Eaap (kJ mol−1
)
34
51
68
84
101
52.51
60.30
63.07
64.50
65.41
Table V Constants Determined by Applying the
Temkin Isotherm by Nonlinear Least-Square Method of
Analysis
a = krK1
T (K)
(mol L−1 min−1
)
b = K2
R2
323
333
343
353
363
0.004
0.010
0.022
0.050
0.082
0.115
0.061
0.052
0.041
0.060
0.951
0.941
0.971
0.973
0.992
Note that bothK1 and K2 are unitless quantities.
the adsorption constant of benzyl alcohol at various
temperatures as in Table III as given in Fig. 9b (in
this figure average values of KBzOH at various partial
pressures of oxygen were used). Heat of adsorption
of benzyl alcohol (ꢀH0) was –18.14 kJ mol−1. Gibbs
free energy (ꢀG) and entropy (ꢀS) of adsorption for
benzyl alcohol were –35.36 kJ mol−1 and 58.45 J
mol−1, respectively. Now considering that Et = Eap
–
ꢀHad, where ꢀHad is the heat of adsorption. Using the
values of apparent true activation energy (Et = 71.81
kJ mol−1) and apparent activation energies (Eap) from
Table IV, the heat of adsorption at various oxygen
partial pressures was calculated as given in Table VI.
According to the Temkin isotherm, heat of adsorption
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

458
SAEED, ILYAS, AND SIDDIQUE
ꢁ
Figure 7 Comparison of the apparent rate constant kr from Table V (observed) with those calculated by using the Temkin
isotherm (the constant was obtained by nonlinear least-square fitting).
(ꢀHad) is decreased linearly with surface coverage (θ)
according to
of ꢀHad versus ln(p) is linear with a negative slope. It
shows that the Temkin isotherm in this case is appli-
cable. However, in this case we are unable to calculate
the value of ꢀH0 and/or α from the intercept and slope
as the constants A and B in Eq. (36) becomes more
complicated when these are inserted in Eq. (35). It is
also important to remember that the Temkin isotherm
is applicable only in the middle range of surface cover-
age (θ ꢀ 0.3–0.6); therefore, deviation in the lower
and upper regions of surface coverage is expected.
Furthermore, the apparently true activation energy (Et
= 71.81 kJ mol−1) is free from the effect of heat of
ꢀHad = ꢀH (1 − αθ)
(35)
◦
However, considering that according to the Temkin
isotherm
θ = A + B ln(p)
(36)
Therefore, a plot of ꢀHad versus ln(p) must be lin-
ear with a negative slope. Figure 10 shows that the plot
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

KINETICS OF MANGANESE OXIDE CATALYZED OXIDATION OF BENZYL ALCOHOL
459
The Final Kinetic Equation
From these analyses, it is evident that the following
kinetic equation (Eq. (37)) is the best fit as evident
from Tables II and V:
ꢁ
ꢂ
KBzOH[BzOH]
Rate = kr
(K1 ln(K2pO ))
2
1 + KBzOH[BzOH]
(37)
This is based upon the Langmuir–Hinshelwood
mechanism, where the adsorption of benzyl alcohol
takes place according to the Langmuir adsorption
isotherm whereas adsorption of oxygen follows the
Temkin adsorption isotherm.
Considering that the heat of adsorption of benzyl
alcohol (ꢀH0 ꢀ –18 kJ.mol−1) is low; therefore, the
adsorption is relatively weak. In this case, Eq. (37) can
Figure 8 Determination of apparently true activation en-
ergy, free from the heat of adsorption of both oxygen and
benzyl alcohol.
adsorption of benzyl alcohol and oxygen only. The ef-
fect from the heat of adsorption of the solvent and other
products (i.e., benzaldehyde, benzoic acid, and water)
could not be ignored.
Figure 9 Van’t Hoff plot for the determination of heat of adsorption of (a) oxygen and (b) benzyl alcohol.
International Journal of Chemical Kinetics DOI 10.1002/kin.20922

460
SAEED, ILYAS, AND SIDDIQUE
Table VI Activation Energies and Heat of Adsorption
of Oxygen at Various Partial Pressure of Oxygen
and oxygen, are adsorbed at the surface of catalyst,
whereas in second step the adsorbed reactants re-
act and give the final products. Adsorption of ben-
zyl alcohol and oxygen has taken place according to
the noncompetitive Langmuir adsorption isotherm and
Temkin adsorption isotherm, respectively. It shows that
the adsorption sites for benzyl alcohol are homoge-
neous in nature (ꢀHBzOH` = –18.14 kJ mol−1). The
adsorption sites for oxygen are heterogeneous in na-
ture, whereas the heat of adsorption varies with surface
coverage by oxygen. The maximum value for ꢀHO2 is
–31.12 kJ mol−1 at θ → 0.
pO2 (kPa)
Ea, (kJ mol−1
)
ꢀHad (kJ mol−1
)
34
51
68
84
101
53.28
59.72
62.33
63.64
65.38
18.53
12.09
9.48
8.17
6.43
BIBLIOGRAPHY
1. Ilyas, M.; Siddique, M.; Saeed, M. Chinese Sci Bull
2013, 58, 2354.
2. Ilyas, M.; Sadiq, M. Chem Eng Technol 2007, 10, 1391.
3. Ilyas, M.; Saeed, M. J Chem Soc Pak 2009, 31, 5263.
4. Ilyas, M.; Saeed, M. Int J Chem Reac Eng 2010, 8, A77.
5. Puzari, A.; Jubaraj, B. J Mol Catal, A: Chem 2002, 187,
149.
6. Punniyamurthy, T.; Asthana, P.; Kalara, S. J. Proc Indian
Acad Sci (Chem Sci) 1995, 107, 335.
7. Su, Y.; Wang, S. L.; Liu, Y.; Cao, Y.; He, H.; Fan, K.
Catal Commun 2007, 8, 2181.
8. Ji, H.; Wang, T.; Zhang, M.; She, Y.; Wang, L. Appl
Catal, A 2005, 282, 25.
Figure 10 Heat of adsorption ꢀHadb versus ln(pO2).
9. Punniyamurthy, T.; Asthana, P.; Kalara, S.; Iqbal, J. J
Chem Sci 1995, 107, 335.
be modified to Eq. (38):
10. Li, Q.; Luo, G.; Li, J. J Mater Proc Technol 2003, 137,
25.
11. Ilyas, M.; Saeed, M. Int J Chem React Eng 2011, 9, A75.
12. Mondelli, C.; Grunwaldt, J. D.; Ferri, D.; Baiker, A.
Phys Chem Chem Phys 2010, 12, 5307.
Rate = kr (KBzOH[BzOH])(K1 ln(K2pO )) (38)
2
as the KBzOH[BzOH] term in the denominator could be
ignored in comparison to 1.
13. Villa, A.; Janjic, N.; Spontoni, P.; Wang, D.; Su, D.;
Prati, L. Appl Catal, A 2009, 364, 221.
14. Salmi, T.; Warna, J.; Toppinen, S.; Ronnholm, M.;
Mikkola, J. P. Braz J Chem Eng 2002, 17, 1023.
15. Ilyas, M.; Khan, H. React Kinet Catal Lett 2007, 92, 75.
16. Saeed, M.; Ilyas, M. Appl Catal, B 2013, 129, 247.
17. Saeed, M.; Ilyas, M.; Siddique, M. Arab J Sci Engin
2013, 38, 1739.
18. Saeed, M.; Ilyas, M.; Siddique, M. J Chem Soc Pak
2012, 34, 626.
19. Ilyas, M.; Ikramullah, Catal Commun 2004, 5, 1.
CONCLUSIONS
The oxidation of benzyl alcohol catalyzed by man-
ganese oxide in the present case is purely taken place
in a kinetically controlled region, where Langmuir–
Hinshelwood type of mechanism is operative. Accord-
ing to this mechanism, the reaction proceed in two
steps. In first step, both the reactants, i.e. benzyl alcohol
International Journal of Chemical Kinetics DOI 10.1002/kin.20922