Qualitative treatment of catalytic hydrolysis of alkyl formates

Article abstract of DOI:10.1016/j.apcata.2010.06.002

Liquid-phase hydrolysis of alkyl formates was performed in a stirred batch reactor using formic acid as a homogenous catalyst, cation exchange resin as heterogeneous catalyst and an additive as a complexation agent. The catalysts increased the rate of the reaction considerably, but the equilibrium conversion was slightly suppressed by the homogenous catalyst. The additive not only accelerated the reaction rate, but also improved the yield significantly. The effect of external and internal mass transfer limitations present in the heterogeneous reaction steps was investigated and it was observed that there is the existence of internal diffusion limitation for the largest catalyst particles. Other parameters such as temperature, catalyst pre-treatment, catalyst loading and stirring speed were investigated in order to optimize the process. The experiments also demonstrated that the ion exchange resin can be reused more than once.

Full text of DOI:10.1016/j.apcata.2010.06.002

Applied Catalysis A: General 384 (2010) 36–44
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Qualitative treatment of catalytic hydrolysis of alkyl formates
Olatunde Jogunolaa,∗, Tapio Salmia, Kari Eränena, J.-P. Mikkolaa,b
a Åbo Akademi, Process Chemistry Centre, Laboratory of Industrial Chemistry and Reaction Engineering, FI-20500 Åbo/Turku, Finland
b Umeå University, Technical Chemistry, Chemical-Biological Centre, Department of Chemistry, SE-90187 Umeå, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
Liquid-phase hydrolysis of alkyl formates was performed in a stirred batch reactor using formic acid as a
homogenous catalyst, cation exchange resin as heterogeneous catalyst and an additive as a complexation
agent. The catalysts increased the rate of the reaction considerably, but the equilibrium conversion was
slightly suppressed by the homogenous catalyst. The additive not only accelerated the reaction rate, but
also improved the yield significantly. The effect of external and internal mass transfer limitations present
in the heterogeneous reaction steps was investigated and it was observed that there is the existence
of internal diffusion limitation for the largest catalyst particles. Other parameters such as temperature,
catalyst pre-treatment, catalyst loading and stirring speed were investigated in order to optimize the
process. The experiments also demonstrated that the ion exchange resin can be reused more than once.
© 2010 Elsevier B.V. All rights reserved.
Received 22 March 2010
Received in revised form 31 May 2010
Accepted 1 June 2010
Available online 10 June 2010
Keywords:
Additive
Alkyl formate hydrolysis
Amberlite IR-120 resin
Formic acid
1. Introduction
ation of the acid product with an additive. The reaction involved in
this complexation step is denoted as
The hydrolysis of alkyl formates (particularly methyl formate,
MeFo) has been of particular interest to the fine chemical indus-
try because one of the products, formic acid (FA) has found a
wide market in agricultural sector as silage additives [1]. Other
major applications include leather tanning, textile dyeing and fin-
ishing, carpet printing, chemical synthesis and pharmaceuticals,
food chemicals, formate salts, rubber chemicals (antiozonants and
coagulants), catalysts, plasticizers, fuel cell (direct formic acid fuel
cell), chemical pulping and regulation of pH in chemical pro-
cesses [1–3]. Alkyl formates hydrolysis process is an endothermic
reversible reaction with a low equilibrium constant (e.g. KC ≤ 0.2 for
ethyl or methyl formate). The hydrolysis equilibrium is relatively
unfavourable and thus excess water is required to shift the equi-
librium towards the products. This leads to operational problems
in terms of cost in removing the excess water and finding a cost
effective corrosion-resistant material for process equipment. Other
methods to improve the equilibrium conversion are the removal of
the methanol product from the reaction system and the complex-
where R = CH3 or C2H5, X = additive and HCOOH–X = acid–additive
complex
The additive is an organic base such as an imidazole derivative
[4]. It shifts the equilibrium conversion towards the product as well
as increases the reaction rate when added to the reaction mixture. It
does not react with the reactants but forms a not too strong adduct
with the carboxylic acid product so as to enhance the easy recovery
of the acid and facilitate a more straightforward recycling of the
organic base. Hence, with a stronger acid–additive complex, the
acid will not be easily recovered, while a weaker adduct is incapable
of shifting the equilibrium towards the product.
Furthermore, the alkyl formate hydrolysis reaction is slow at
neutral pH but can be enhanced by an acid catalyst. A typical reac-
tion involving the acid-catalyzed hydrolysis of alkyl formate is
expressed as
Abbreviations: AMR, amberlite IR-120 resins; CX, adduct (complex formed
between formic acid and additive); DVB, divinyl benzene; EFH, ethyl formate hydrol-
ysis; MFH, methyl formate hydrolysis; EtFo (A), ethyl formate; EtOH (D), ethanol;
FA (C), formic acid; FAC, formic acid-catalyzed; FID, flame ionization detector;
GC, gas chromatograph; H2O (B), water; MeFo (A), methyl formate; MeOH (D),
methanol; meq, milliequivalent; NC, non-catalytic; PS, polystyrene; rpm, revolution
per minute; TOF, turnover frequency; X, additive.
+
HCOOR + H2OHꢀHCOOH + ROH
Both homogenous and heterogeneous catalysts can be used to
speed up the reaction rate. Typical homogenous catalysts include
mineral acids such as H2SO4 [5] and HCl [6–8]. However, H2SO4
promotes the decomposition of the formic acid produced to car-
∗
Corresponding author. Tel.: +358 2 2154431; fax: +358 2 2154479.
E-mail address: jolatund@abo.fi (O. Jogunola).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.06.002

O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
37
to catalyze its own reaction and there have been reports of such
process in the chemical industry [9,10]. Furthermore, heteroge-
neous catalyst provides an attractive alternative to homogenous
catalyst. The main advantage of a heterogeneous catalyst is that a
catalyst-free product can be obtained simply by filtration, which is
incorporated in the process.
Nomenclature
a
ap
C
Di
Dei
dp
Ea
shape factor
interfacial area-to-volume (m−1
concentration (mol/kg)
)
molecular diffusion coefficient (m2 s−1
)
Over the past few years, considerable research efforts have been
devoted to the liquid-phase hydrolysis of esters (particularly alkyl
acetates) using hydrophobic solid acid catalysts such as cation
exchange resins, heteropolyacids, zeolites and zirconium phos-
phonates [11]. This is due to the fact that many inorganic solid
acid catalysts are hydrophilic, thus limiting the accessibility of the
organic substances to the acid sites. However, there are few reports
on the catalytic hydrolysis of alkyl formates and some of these
reports used catalysts such as activated charcoal [12,13] and ion
exchange resins [14–16]. Ion exchange resins are highly ionic, cova-
lently cross-linked, insoluble polyelectrolytes and a typical one is
the PS-DVB resin, in which sulphonic acid is fixed to a polymer
carrier such as polystyrene (PS) cross-linked with divinylbenzene
(DVB). These solid resins as catalyst have a lot of advantages: the
problems associated with separation and corrosion encountered in
homogenous catalysts is eliminated and they can be used in con-
tinuous mode to achieve high product purity [11,17]. More so, the
resins can discriminate between small and large molecules [18].
Most of the research work found in the open literature on
hydrolysis of alkyl formates using activated charcoal and ion
exchange resins have been performed in a chromatographic reactor
[12,13,16,17]. This is due to the fact that alkyl formate hydrol-
ysis is an equilibrium limited reaction and the properties of the
heterogeneous catalysts as an adsorbent can be utilized in a chro-
matographic reactor to separate different components and achieve
higher conversion of the alkyl formate. Thus, the reverse reaction
is prevented and the reaction is shifted towards the products by
removing the product components from the reaction mixture [13].
There have been very few reports in the literature concerning
the study of alkyl formate hydrolysis using a heterogeneous catalyst
in a stirred batch reactor. This might be due to the fact that in a batch
reactor, the property of an ion exchange resin to separate the reac-
tion products cannot be utilized. Therefore, no conversion greater
than the equilibrium conversion, is possible. One representative
report (Metwally et al. [19]) used cation resin for the hydrolysis
of ethyl formate (EtFo). They reported the activation parameters
for the resin-catalyzed ethyl formate hydrolysis in water–acetone
mixture. Nevertheless, no detailed study has been published in
the literature concerning the kinetics of alkyl formate (particu-
larly methyl formate) resin-catalyzed hydrolysis in a stirred batch
reactor. Furthermore, to the best of my knowledge, there has not
yet been a report in the open literature concerning alkyl formate
hydrolysis in the presence of formic acid catalyst or an additive.
In this work, we have studied the thermodynamics and kinet-
ics of alkyl formate hydrolysis using formic acid and Amberlite
IR-120 resin as acid catalysts, and an organic base as an additive
in a stirred batch reactor. Furthermore, the three processes were
compared qualitatively. Also, the mass transfer limitations using
heterogeneous catalyst were investigated and the effect of other
parameters such as catalyst pre-treatment, stirring speed, catalyst
loading and temperatures on both kinetics and thermodynamics
equilibrium were included in the study.
effective diffusion coefficient (m2 s−1
particle diameter (mm)
)
activation energy of the formic acid-catalyzed reac-
tion (kJ mol−1
activation energy for the uncatalyzed reaction
(kJ mol−1
activation energy for the autocatalytic reaction
(kJ mol−1
activation energy for the additive-enhanced step
(kJ mol−1
)
Eao
Eaꢀ
Eaꢀꢀ
)
)
)
Eq
f
activation parameter of the equilibrium constant
rate function
KC
concentration-based equilibrium constant
rate constant due to dissociation of
k◦
B
(kg mol−1 min−1
)
k
rate constant of the formic acid-catalyzed reaction
(kg2 mol−2 min−1
rate constant of the autocatalytic reaction
(kg2 mol−2 min−1
rate constant of the complexation step
(kg2 mol−2 min−1
molar mass of the solvent (kg mol−1
flux (mol m−2 s−1
)
kꢀ
kꢀꢀ
)
)
MB
N
pH
R
)
)
power of hydrogen
gas constant (J mol−1 K−1
)
Rj
r
T
radius of the particle (mm)
reaction rate (mol kg−1 min−1
temperature (K)
)
t
V
time (min)
volume (cm3)
VA
XA
X
molar volume of the solute (m3 mol−1
equilibrium conversion
dimensionless coordinate
)
Y
xj
frequency function for particle size distribution
ratio of particle radius to average radius
Greek letters
˛
ˇLS
ꢀ
stoichiometric ratio of the additive
liquid–solid mass transfer coefficient (m s−1
effectiveness factor
)
εp
Â
ꢁp
ꢂp
ꢃm
porosity of the particle
association factor of the solvent
tortuosity of the particle
density of the particle (kg m−3
)
viscosity of the mixture (g cm−1 s−1
)
Subscripts and superscripts
DR
dry resin
eqm
equilibrium
swollen resin
initial value
component
SR
0
i
j
particle size fraction
2. Experimental section
bon monoxide and water. Nevertheless, HCl has been reported to
catalyze the reaction efficiently without decomposition. Another
disadvantage of homogenous catalysts is their miscibility with the
reaction medium, which might lead to separation and corrosion
related problems. This can be overcome by the use of formic acid
2.1. Experimental set-up, procedure and matrix
All experiments were carried out in a conventional 500 ml Parr
autoclave made of zirconium metal. The reactor consists of feeding
and reaction vessels, a heating unit, a sampling line and a stirrer.

38
O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
Table 1
Experimental details.
Heterogeneous catalyst C0A (mol/kg)
Homogenous catalyst C0A (mol/kg)
Organic base C0A (mol/kg)
Methyl formate hydrolysis
[MeFo]0
[H2O]0
[MeOH]0
[FA]0
10.49
18.87
0.94
9.91
17.83
1.03
0.00–1.11
4.42–7.94
7.91–17.43
0.46–0.82
0.49–0.89
0.00–4.43
[X]0
Catalyst loading
0.00–16.67 g/kg on dry basis
Ethyl formate hydrolysis
[EtFo]0
[EtOH]0
[H2O]0
[FA]0
[X]0
9.04
0.44
16.28
8.64
0.42
15.56
0.00–0.97
4.15–8.64
0.2–0.42
7.47–15.90
0.46–0.97
0.00–4.15
Catalyst loading
0.00–66.67 g/kg on dry basis
The sampling line was equipped with a cooling bath and a filter.
A precise temperature control required two thermocouples and
cooling air. The reaction was performed at a nitrogen-pressure of
20 bar and isothermally at temperatures of 60 and 90 ◦C for hetero-
geneous catalyst (Amberlite IR-120 and Dowex 50Wx8), 80–110 ◦C
for the homogenous catalyst (formic acid) and 80–110 ◦C for the
additive-enhanced reaction, respectively. Distilled water with or
without the catalyst was placed in the reaction vessel. In the case of
the additive-enhanced reaction; distilled water, the additive ((Alfa
Aesar, ≥98.87) and formic acid (Sigma–Aldrich, 99.5%) were placed
in the vessel. Methanol (Baker, >99 wt.%) and methyl formate
(Sigma–Aldrich, 97 wt.%) in the case of methyl formate hydrolysis
(MFH) or ethanol (Altia, 99.5%) and ethyl formate (Sigma–Aldrich,
97 wt.%) in the case of ethyl formate hydrolysis (EFH) were dis-
charged into feeding vessel with the aid of a dropping funnel.
Nitrogen line to the reactor and the vent lines were connected to
the system. The stirred reaction vessel was heated until it reached
the desired temperature. The content in the feeding vessel was
discharged into the reaction vessel as quickly as possible using
nitrogen-pressure and the reaction commenced immediately. The
initial total amount of the hydrolysis mixture was close to 0.3 kg,
while a stirring rate of 300 rpm was used for all the experiments
except the one involving the investigation of external mass transfer
limitation in which 700 rpm was used instead. Table 1 shows the
experimental matrix.
of 50 (dp = 0.3–0.84 mm); 100 (dp = 0.15–0.3 mm); and 400 mesh
(dp = 0.04–0.08 mm).
The catalyst was pre-treated by washing with distilled water,
decanted and then dried in an oven at 99 ◦C for 2 days until a con-
stant mass was obtained. Our group has shown that the nature of
the pre-treatment has no effect on the reaction kinetics [20]. After
the hydrolysis reaction, the catalyst was washed with water and
dried at 99 ◦C for subsequent reuse.
The swelling ability of the resin in different liquids was studied
at 25 ◦C. About 1 g of the catalyst was placed in a bottle and 25 ml
of the liquid was added to it. The mixture was shaken intermit-
tently and allowed to settle overnight so that equilibrium would be
reached. The liquid not absorbed was separated from the catalyst
and the volume occupied by the catalyst was measured accordingly.
The acidic capacity was measured by a conventional titration
method [21]. About 0.5 g of the resin was added to 50 ml of NaCl
solution (200 g/l) and stirred. The system was allowed to stand for
1 day so that cation exchange between the H+ and Na+ could take
place. The mixture was titrated with a 0.1 N NaOH solution.
3. Mathematical modelling
A reversible reaction similar to the hydrolysis of alkyl formate
can be represented as
A + BꢁC + D
2.2. Gas chromatographic analysis
where A = ethyl formate (EtFo) or methyl formate (MeFo), B = water,
C = formic acid (FA) and D = ethanol (EtOH) or methanol (MeOH).
In the modelling, the system is considered as a pure liquid-
phase. The vapourization of the components was surpassed by
carrying out the reaction under pressure, so the role of the gas-
phase, including the volatilization of the components can be
ignored. The gas volume is quite small compared to the liquid vol-
ume in the reactor.
Samples were withdrawn at defined sampling intervals (5, 10,
15 or 20 min) using the reactor system internal pressure and were
analyzed off-line with a gas chromatograph (GC): 6890N, injec-
tion port temperature 150 ◦C, oven temperature 150 ◦C, HP-PLOT
U column, 250 m × 530 ␮m × 20 ␮m, carrier gas helium (15 ml/min
at 1 min), detector FID (250 ◦C, H2 flow 40 ml/min, air flow
450 ml/min). The calibration was performed using acetonitrile
(Labscan, >99%) as an internal standard. The calibration of the
internal standard solution was always done immediately before
the analysis of any sample. The analysis time for a sample was
between 6 and 11 min. The experimental results were based on
alcohol analysis (methanol in case of MFH and ethanol in case of
EFH) due to the volatility of methyl formate and the difficulty in
getting a reliable analysis method for ethyl formate and formic
acid.
Table 2
Properties of the solid catalysts as given by the manufacturer.
Properties
Amberlite
IR-120
Dowex 50Wx8-50,
-100, -400 mesh
Bead type
Gel
Gel
Cross linking (%)
8
8
Particle size range (␮m)
Moisture content (% mass)
Density of wet catalyst (g/cm3)
Max. operating temperature (◦C)
Capacity by dry weight (meq/g)
300–1200
45
1.26
120
4.4
Depend on the mesh
54
–
150
4.8
2.3. Catalyst properties and characterization
The properties of the solid catalysts supplied as beads by the
manufacturer are shown in Table 2. The particle diameter, dp of
Amberlite IR-120 is 0.3–1.2 mm, while Dowex 50Wx8 with a mesh
Max. = maximum.

O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
39
The reaction rate in the absence of catalyst (neutral medium) is
expressed as
ꢀ
ꢁ
1
KC
ri = f CACB
−
CCCD
(1)
where f = k◦ + kꢀCC
For the heterogeneous catalyst, during the course of study of the
hydrolysis process, internal diffusion affected the kinetics. A math-
ematical model, which incorporates the particle size distribution of
the solid catalyst, was developed to reveal the kinetics and internal
mass transfer effects of the porous particles. The mass balance for
the bulk phase is
ꢂ
∂Ci
= ap
yjNijx2 + ri
(2)
∂t
j
The mass balance for the catalyst particles is
ꢃ
ꢄ
Fig. 1. The effect of the catalyst pre-treatment on the kinetics of alkyl formate
∂Cpi
∂t
∂2Cpi
∂X2
∂C
pi
r˙iꢂp
εp
hydrolysis at 60 ◦C; B/A molar ratio = 1.8.
Dei
εpRj2
a − 1
= ri +
+
+
(3)
X
∂X
Eq. (5) becomes:
with the boundary conditions Ci(X = 1) = Cpi at the particle sur-
face and (∂Cpi/∂X)X=0 = 0 at the centre of the particle. ri is the
reaction rate of the homogenous catalytic part of the system in
mol min−1 m−3 and r˙i is the reaction rate of the heterogeneous
1
KC
rꢀ = kCCe−E /RT (CACB
−
CCCD)
(8)
(9)
a
i
system in mol min−1 kg−1
.
The additive enhanced reaction rate Eq. (6) takes the form:
The shape factor, a is 3 (sphere) for the resin. The effective diffu-
sion coefficient is given as: Dei = εpDi/ꢁp. The diffusion coefficient,
Di can be calculated from Wilke–Chang equation
ꢅ
In the presence of formic acid catalyst, we assumed that the
rate constant of the non-catalyzed reaction is negligible compared
to the rate constant of the catalytic reaction. Thus, we have
o
a
ꢀ
ꢀꢀ
1
KC
/RT
/RT
riꢀꢀ = koe−E
+ kꢀCCe−E
+ kꢀꢀCXe−E /RT (CACB
−
CCCD)
a
a
The temperature dependence of the equilibrium constant can
be described as:
7.4 × 10−12 ÂMB
Di =
ꢃmVA0.6
Keqi = Koe−Eq /RT
(10)
i
4. Results and discussion
ꢀ
ꢁ
CCCD
KC
rꢀ = kCH
+
CACB
−
(4)
4.1. Catalyst characterization and properties
i
Since formic acid catalyzed its own reaction, CH+ = CC. It is
important to note that CC comprises the added acid and the acid
generated due to autocatalysis. Thus Eq. (3) becomes
4.1.1. The effect of the pre-treatment
Often, dried catalysts are necessary for enhanced reactivity and
conversion rates or if the addition of water is a problem. In our case,
the aim of the pre-treatment was to remove water, corresponding
to 45% of the total weight (see Table 2) of the wet Amberlite IR-120
resins. Therefore, the exact amount of water in the reaction system
is known since water is one of the reactants. The pre-treatment of
the resins modified its structure and increased its activity slightly
because in the dry form, the diffusion properties of these resins are
quite different from the wet type [20]. This is exemplified by the
differences in kinetic experiments carried out with the dry and wet
resins as shown in Fig. 1.
ꢀ
ꢁ
CCCD
KC
rꢀ = kCC CACB
−
(5)
i
For the additive-enhanced reaction, the reaction involved two
steps: the slow uncatalyzed process and the fast complexation step.
The reactions involved in the process are expressed as:
4.1.2. The swelling ratio of the resins
where A = ethyl or methyl formate, B = water, C = formic acid,
D = ethanol or methanol, X = additive and CX = acid–additive com-
plex (adduct). The stoichiometric coefficient of the additive ˛ is not
unity.
The reaction rate for the alkyl formate hydrolysis in the presence
of an additive can be expressed as
When polymer materials such as ion exchange resins come in
contact with a liquid, they absorbed the liquid and swelled. The
swelling kinetics is linked with the diffusion of liquids. The nature
of the solvent determines the extent of swelling and the swelling
degree decreases with the reduction in polarity of the solvent [22].
The resin was left in water overnight (24 h) for it to attain equilib-
rium. The extent to which Amberlite IR-120 resin swelled in pure
solvents at 25 ◦C was investigated using the concept of swelling
ratio:
ꢀ
ꢁ
1
KC
riꢀꢀ = f CACB
−
CCCD
(6)
where f = k◦ + kꢀCC + kꢀꢀCX
The hydrolysis reaction depends on temperature, thus Arrhe-
nius equation can be used. The reaction rate in neutral medium
(Eq. (1)) becomes:
Volume of swollen resin at equilibrium
Swelling ratio =
Volume of dry resin
eqm
SR
V
o
a
ꢀ
1
KC
=
(11)
ri = koe−E
+ kꢀCCe−E /RT (CACB
−
CCCD)
(7)
/RT
a
VD0R

40
O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
Fig. 2. The catalytic activity of different dry resins at 60 ◦C.
Fig. 3. The effect of external mass transfer around the catalyst particles at 60 ◦C;
H2O/MeFo molar ratio = 1.8.
The swelling ratio of the resin at 25 ◦C after equilibrium has been
attained was found to be: H2O 2.94; MeFo 1.66; EtFo 1.60; MeOH
2.33, EtOH 2.37 and FA 2.00. Water has the highest swelling ratio,
while the two esters (MeFo and EtFo) have the lowest swelling ratio.
The same sequence was obtained by Mai et al. [15]. Since water is
one of the reactants, the swelling behaviour of the liquid mixture
was assumed to follow that of water for the sake of simplicity.
should offer stronger external mass transfer resistances. This is
illustrated in Fig. 3.
In fact, we can conclude that no external mass transfer limi-
tations were present in the case of the dry resin at the stirring
speed range of 300–700 rpm (Fig. 3). Also, no attrition of the cat-
alyst occurred under the chosen operating conditions. Therefore,
300 rpm was used for all additional experiments to avoid external
mass transfer limitations and to suppress attrition. Also, a stirring
speed of 300 rpm was enough for the hydrolysis of the two alkyl
formates using a homogenous catalyst and an additive.
4.1.3. The catalytic activity of different resins
The catalytic activity of 5 g Amberlite IR-120 resin on dry basis
was compared with three different size ranges of the same amount
of Dowex 50Wx8 at 60 ◦C using the concept of turnover frequency
(TOF). Both catalysts are microporous cation exchange resins with
the same percentage of cross-link.
4.1.4.2. The effect of catalyst particle sizes. Kinetic experiments were
carried out for 2.5 g of dry Dowex 50Wx8 using three different parti-
cle size ranges (dp = 0.04–0.08, 0.15–0.3 and 0.3–0.84 mm) in order
to ascertain the influence of the internal mass transfer limitation.
The effect of the particle size on hydrolysis of methyl formate is
presented in Fig. 4.
The observed difference in the reaction rate under identical
experimental conditions was due to the internal diffusion present
inside the catalyst particles, provided that the concentration of the
acid sites does not depend on the particle size. It was assumed that
accessibility of these sites do not depend on the particle size since
the resin swelled well in water before the reaction commenced.
Thus, the smallest beads (dp = 0.04–0.08 mm) exhibited the fastest
rate at 60 ◦C. Assuming these particles have the limiting rate (with-
out internal diffusion limitation); the effectiveness factor ꢀ can be
estimated as
Initial rate of FA formation
TOF =
No. of Brønsted acid sites
The number of Brønsted acid sites was estimated based on the
cation exchange capacity of the resin. The initial TOF for the differ-
ent resins is displayed in Fig. 2.
It can be deduced that catalysts with the same particle size has
almost the same initial reaction rate. For example, Amberlite IR-120
(dp = 0.3–1.2 mm) has similar catalytic activity as Dowex 50Wx8-50
(dp = 0.3–0.84 mm), even though their capacities according to the
manufacturer is slightly different (see Table 2). However, Amberlite
IR-120 was selected for further experimental matrix because of its
affordability and wider particle size range. The exchange capacity of
the resin on dry basis, determined by titration method was found
to be 4.7 meq/g. This value was obtained as the average of three
different measurements. This is different from the one obtained by
the manufacturer (4.4 meq/g) because of experimental error and
probably the non-ideality of the titration mixture. The apparent
Brønsted acid concentration [H]+ is calculated based on the cation
exchange capacity by dry weight (meq/g):
Observed rate of MeFo disappearance
ꢀ =
Rate in the absence of internal diffusion
4.7 × 10−3 × mass of catalyst
+
H
[
=
]
Volume of solution
4.1.4. Mass transfer limitation effects (batch-mode)
4.1.4.1. The effect of stirring speed. Two experiments were per-
formed at stirring speeds of 300 and 700 rpm, respectively in order
to investigate the effect of agitation speed and external mass trans-
fer on the overall reaction rate of methyl formate hydrolysis using
2.5 g of Amberlite IR-120 resin. The anticipated external mass trans-
fer limitations were characterized by a liquid–solid mass transfer
coefficient ˇLS, which depends mainly on the diameter of the cat-
alyst particle; thus Amberlite IR-120 with a larger particle size
Fig. 4. The effect of internal diffusion inside the catalyst particles at 60 ◦C; H2O/MeFo
molar ratio = 1.8.

O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
41
Table 3
Effectiveness factor for different size ranges.
Particle size (mm)
ꢀ
t = 0 min
t = 15 min
0.04–0.08
0.15–0.3
0.3–0.84
1.00
0.75
0.43
1.00
0.98
0.75
ꢆ
dCA/dt
particle -size in question
ꢆ
ꢀ =
(12)
dCA/dt
d
p
=0.04−0.08 mm
From Eq. (12), one can estimate the effectiveness factor for the
particle size ranges in question. This is presented in Table 3.
Thus, one can deduce that internal diffusion limitation effects
were significant in these particles. The internal mass transfer
affected the observed overall reaction rates at the beginning
of the reaction and the effect diminished as the equilibrium is
approached (see Fig. 4). The same is true for Amberlite IR-120 resin
(dp = 0.3–1.2 mm) too.
Fig. 6. Deactivation studies of Methyl formate hydrolysis over Amberlite IR-120 at
60 ◦C; H2O/MeFo molar ratio = 1.8.
48 h until a constant mass is obtained. The result is depicted in
Fig. 6.
It can be concluded that the deterioration of the catalyst activ-
ity was negligible after three batches since the conversion curves
are practically overlapping one another. All curves in Fig. 6 are
the same, within experimental accuracy. Similar results have been
reported by Altiokka and C¸ itak [23] for Amberlite IR-120 during
the esterification of acetic acid with isobutanol. So, the resin may
be used repeatedly since the capacity of the catalyst remains vir-
tually the same. Indeed, that the catalyst can be used repeatedly is
one of the most important prerequisites for green chemistry.
4.1.5. The effect of catalyst loading
The amount of catalyst was varied over a range of 1.67–66.67 g of
dry catalyst per kg of the solution, while keeping other parameters
constant in order to determine the effect of catalyst loading on the
hydrolysis kinetics. The procedure was applied for both ethyl and
methyl formates. The results are shown in Fig. 5.
It can be observed that the initial reaction rate is linearly increas-
ing with the catalyst loading for both ethyl formate and methyl
formate hydrolyses, since the active site, the sulphonic acid groups
(–SO3H) are directly proportional to the amount of catalyst present
in the solution. Thus, more suphonic acid group is readily available
as the amount of catalyst increases, which leads to the formation
of larger amount of carbonium ions per unit time. This in turn,
increases the reaction rate. However, the reaction rate curve of MFH
is not as perfectly linear as that of EFH.
4.2. Qualitative analysis of reaction rate and chemical
equilibrium
4.2.1. The effect of water on equilibrium conversion
The presence of excess water in alkyl formate hydrolysis
improves formic acid yield in the absence of any additive. This is
also true in the presence of a suitable additive. The purpose of this
study is to find the optimum molar ratio of water that will be needed
to give a reasonable conversion. The molar ratio of water-to-ester
was varied from 1.0 to 3.0 at 90 ◦C, while additive-to-ester molar
ratio was kept constant at 0.40. The value of the equilibrium con-
version was obtained by taking the average of the experimental
conversions when the reaction had attained chemical equilibrium.
The increment is the difference between the succeeding and pre-
ceding values of the equilibrium conversions. The result is shown
in Table 4 for both methyl formate and ethyl formate hydroly-
ses.
It can be deduced that the equilibrium conversion increases as
the amount of water increases in the presence of the additive. How-
ever, the increment increases initially but decreases as more water
is added to the system. Therefore, it is not economically viable to
increase the amount of water just to get a minor change in the con-
version. So, there should be a point of process optimization in terms
of capacity and energy requirement in water removal. From Table 4,
the optimum water-to-ester molar ratio is between 1.5 and 2.0. It
can be concluded that less excess water is needed in the hydrolysis
4.1.6. Catalyst reuse
The main obstacle encountered in the study of catalyst deacti-
vation was the repeatability of the experimental conditions. The
main goal of this study was to determine whether the resin can
be used more than once without a significant decrease in activ-
ity. The apparent Brønsted acid concentration [H+] was calculated
based on the dry weight capacity (meq/g) of the catalyst, given
by the manufacturer and this capacity was assumed to remain
constant from batch-to-batch. The catalyst deactivation was inves-
tigated using 2.5 g of Amberlite IR-120 four times consecutively
at 60 ◦C under the same reaction conditions. After each run, the
catalyst was washed with water and dried at 99 ◦C for almost
Table 4
The effect of excess water on equilibrium conversion at 90 ◦C.
H2O/ester molar ratio
Equilibriumconversion (mol%In)crement
MeFo
EtFo
MeFo
EtFo
1.00
1.50
2.00
2.50
3.00
–
0.51
0.56
0.62
0.64
0.65
–
–
0.08
0.04
0.02
–
0.56
0.64
0.68
0.70
0.05
0.06
0.02
0.01
Fig. 5. The effect of catalyst loading on alkyl formates hydrolysis at 60 ◦C; B/A molar
ratio = 1.8.

42
O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
Table 5
The effect of the initial FA charge on equilibrium composition and conversion.
Temperature (◦C)
C0A (mol/kg)
Equilibrium concentration (mol/kg) [FA]0 = 0.0 mol/kg
C0A (mol/kg)
Equilibrium composition (mol/kg) [FA]0 = 1.11 mol/kg
MeFo
H2O
FA
MeOH
XA
MeFo
H2O
FA
MeOH
XA
90
100
110
10.49
10.49
10.49
6.853
6.760
6.843
15.237
15.147
15.067
3.637
3.727
3.807
4.577
4.667
4.750
.347
.355
.363
9.907
9.907
9.907
6.882
6.859
6.840
14.808
14.785
14.766
4.135
4.158
4.177
4.053
4.076
4.093
.305
.308
.31
Temperature (◦C)
C0A (mol/kg)
Equilibrium concentration (mol/kg) FA = 0.0 mol/kg
C0A (mol/kg)
Equilibrium composition (mol/kg) FA = 1.11 mol/kg
EtFo
H2O
FA
EtOH
XA
EtFo
H2O
FA
EtOH
XA
90
100
110
9.043
9.043
9.043
5.653
5.613
5.583
12.892
12.853
12.822
3.391
3.430
3.460
3.828
3.865
3.895
.375
.379
.383
8.643
8.643
8.643
5.900
5.851
5.818
12.813
12.764
12.731
3.707
3.757
3.789
3.159
3.208
3.241
.317
.323
.327
A = MeFo or EtFo; X = conversion.
of alkyl formate in the presence of an additive in order to achieve
a reasonable conversion.
rium conversion, since its increment decreases as the molar ratio of
X/ester increases. Therefore, there should be an optimum amount
of X in the system for an efficient process. This could be between
the additive-to-ester molar ratios of 0.40–0.60.
4.2.2. The effect of the additive or formic acid catalyst on the
equilibrium composition and reaction rate
Also, the reaction rate increases as the amount of the additive
increases (see Fig. 7) especially when the molar ratio, X/EtFo = 0.2
and 0.4 compared to when there was no additive (X/EtFo = 0.0). The
difference in kinetics is relatively small at the initial stage of the
reaction, but becomes prominent as the reaction proceeds towards
chemical equilibrium. At the initial stage of the reaction, less FA
is formed and the reaction is slow. But, as the reaction progresses,
more H+ accumulate and it is being taken up as quickly as possible
by X. Thus, the frequency of reactants collision increases and the
reaction rate increases. The noticeable difference between the two
reaction rates as the reaction progresses might be due to the fact
that at X/EtFo = 0.2, there might be some free (uncaptured) formic
acid left in the mixture due to the unavailability of enough X to
complex the acid product in the mixture. However, as more X is
added (X/EtFo = 0.4), the percentage of formic acid left unattached
to the additive reduced drastically.
The effect of the FA-catalyzed hydrolysis reaction on the chem-
ical equilibrium was investigated by maintaining the molar ratio
of water-to-ester constant at 1.8, while the molar ratio of acid-to-
ester is 0.0 for the neutral aqueous solution hydrolysis (uncatalyzed
reaction) and 0.1 for the acid-catalyzed reaction. The results at three
different temperatures can be seen in Table 5.
It was observed that the initial FA charge acting as a catalyst
decreased the equilibrium conversion slightly for both ethyl for-
mate and methyl formate hydrolysis. For methyl formate hydrolysis
at 90 ◦C, the equilibrium conversion decreased from 34.7 to 30.5%.
This is due to the presence of more formic acid as catalyst and
product in the reaction solution and this accumulation of the acid
favours the backward reaction. Therefore, it is imperative to use
moderate amount of formic acid as catalyst in the system in order
not to reduce its yield drastically.
Furthermore, as more additive is added to the system (i.e. when
X/EtFo = 0.60, 0.75 and 1.0), there is no appreciable increase in the
reaction rate since the curves have almost the same gradient. This
can be attributed to the fact that there is no more free formic acid
left to be captured by the additive in the reaction mixture. The same
effect is observed in methyl formate hydrolysis.
In the case of the additive-enhanced hydrolysis process, the
additive-to-ester molar ratio was varied from 0.0 to 1.0, while the
others parameters were kept constant. The results for ethyl formate
hydrolysis are depicted in Fig. 7.
The equilibrium conversion increases as the amount of the addi-
tive in the mixture increases. The additive forms a complex with
formic acid by removing the free acid as it is being formed, thus
preventing the backward reaction (re-esterification of the acid).
However, it is not economically viable to increase the amount of
additive without getting an appreciable increase in the equilib-
4.2.3. The effect of temperature
The influence of temperatures was investigated by performing
experiments under the same reaction conditions (e.g. water-to-
ester ratio of 1.8) but at temperatures of 60 and 90 ◦C in the case
of the heterogeneous catalyst and 80–110 ◦C for the formic acid-
catalyzed (FAC) reaction. 1.11 mol of FA/kg of solution and 0.1 mol
SO3H+/kg solution were used as catalysts, respectively. As expected,
the rate of reaction increases as temperature increases for both
methyl formate and ethyl formate hydrolyses. Some of the results
for the purpose of clarity are illustrated in Fig. 8 for the homogenous
and heterogeneous catalysts.
Methyl formate hydrolysis was found to be faster than ethyl
formate hydrolysis under similar reaction conditions for both
homogenous and heterogenous catalysts. This is due to steric
effects present in ethyl formate. In acid-catalyzed hydrolysis, the
susceptibility of the reaction to polar effects is very low and the
relative rates are determined by steric factors only [24]. There is
the presence of steric hindrance in the intermolecular motion of
ethyl formate hydrolysis.
Also, an increase in the temperature enhances the reaction rate
of the additive-enhanced process for alkyl formate hydrolysis. This
is displaced in Fig. 9 for methyl formate hydrolysis.
Fig. 7. The effect of the additive on equilibrium conversion and reaction rate at
90 ◦C; H2O/EtFo molar ratio = 1.8.

O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
43
Table 6
pH of the hydrolysis processes.
Catalyst type
MFH
EFH
t = 0
t = teqm
t = 0
t = teqm
FAC
AMR
X
1.87
4.26
6.64
0.96
1.29
5.20
1.66
3.65
6.46
0.87
1.24
4.80
AMR = Amberlite IR-120 resin; FAC = formic acid-catalyzed reaction; X = additive.
0 = initial; eqm = equilibrium.
MFH and EFH, respectively at the end of the reaction. Also, the
additive-enhanced hydrolysis process took place under acidic con-
ditions.
4.2.4. The effect of the additive or catalysts on initial rate and
equilibrium time
Fig. 8. The effect of temperatures on the kinetics of alkyl formate hydrolysis for
acid-catalyzed reaction; B/A molar ratio = 1.8.
Nevertheless, it is difficult to compare formic acid as a cata-
lyst with Amberlite IR-120 resins since their intrinsic acidities are
at different levels. Sulphonic acid is a much stronger acid than its
corresponding carboxylic acid. Arbitrarily, for the purpose of this
work, we consider the optimum molar ratios of acid-to-ester of
0.1 (0.33 mol H+) and water-to-ester of 1.8 at 80–110 ◦C, which
enhances the reaction rate without drastically affecting the formic
acid yield for the homogenous catalyst. For the heterogeneous cat-
alyst, we use 2.5 g (0.031 mol SO3H+) of Amberlite resins with the
same water-to-ester ratio at 60 and 90 ◦C but without initial FA
charge. In case of the additive-enhanced reaction, we consider
the following optimum amount: X/MeFo of 0.6 (1.02 mol X/kg),
FA/MeFo of 0.1 with the same water-to-ester ratio at 80–110 ◦C.
This was done for the purpose of comparison (see Table 1). All other
experimental conditions remain unchanged. The results of the ini-
tial reaction rate ri0 per mole of H+, SO3H+ or X are presented in
Table 7.
From Table 7, we can observe that the reaction rate was quite
slow in the absence of the catalyst especially at the lowest tem-
perature (60 ◦C) and increased significantly as expected when
catalyst was added to the system. Also, the time required to attain
chemical equilibrium reduced drastically from 8 h for the non-
catalytic reaction to an hour for the resin-catalyzed reaction. At
higher temperature (e.g. 90 ◦C), the hydrolysis rate was faster for
the Amberlite-catalyzed reaction compared to the formic acid-
catalyzed reaction and additive-enhanced system. In addition, the
time needed to attain equilibrium reduced significantly especially
for the resin-catalyzed reaction. Therefore, with the solid catalyst,
one has the luxury of saving time and operating at milder reac-
tion conditions (e.g. lower reaction temperature), which leads to a
reduction in energy cost.
However, there is the problem of formic acid separation from
water in the two catalytic processes because formic acid formed an
azeotropic mixture with water. And separation cannot be effected
by simple distillation. This problem is overcome by the use of the
additive X, since formic acid binds weakly to the additive and the
acid can be recovered and the additive recycled by simple distilla-
tion.
Fig. 9. The effect of temperature on kinetics of methyl formate hydrolysis in the
presence of the additive; X/MeFo molar ratio = 0.6; H2O/MeFo molar ratio = 1.8.
At higher temperature (e.g. 110 ◦C), there is more energy in the
system and this results in earlier accumulation of H+, which is taken
up by X as fast as it is being formed and this leads to more reactants
particle collisions. However, these reactions cannot be performed
at a very high temperature because it favoured the decomposition
of FA. Furthermore, it can be deduced from Fig. 9 that tempera-
ture has no effect on equilibrium conversion since alkyl formate
hydrolysis has weak temperature dependence.
The pH of the hydrolysis process at 90 ◦C in the presence of
the additive, formic acid and Amberlite IR-120 resin (AMR) was
monitored during the course of the reaction. As expected, the pH
decreases as formic acid is formed and remains virtually constant
at equilibrium. This is shown in Table 6 for both methyl formate
hydrolysis (MFH) and ethyl formate hydrolysis (EFH) under similar
reaction conditions.
The pH of the homogenous catalyst was as measured about 0.33
and 0.37 units lower than that of the heterogeneous catalyst for
Table 7
The effect of the additive or catalysts on the initial reaction rate and the equilibrium time for MFH.
Temperature (◦C)
ri0 (10−2 mol/kg min)
ri0/mol H+
Time to reach equilibrium (min)
r0/mol SO3H+
ri0/mol X
ri0 NC
FAC
AMR
X
NC
i
60
80
90
100
110
–
369.4
–
0.10
0.27
0.53
1.50
4.33
–
150
70
50
40
60
–
25
–
480
200
120
80
7.21
11.72
19.22
34.45
–
1008
–
1.88
2.29
8.05
16.88
190
100
70
–
50
55
NC = non-catalytic reaction; AMR = Amberlite IR-120 resin; FAC = formic acid-catalyzed reaction; X = additive.

44
O. Jogunola et al. / Applied Catalysis A: General 384 (2010) 36–44
5. Conclusions
Programme (2006–2011) by the Academy of Finland. Financial
support from the Åbo Akademi Foundation is gratefully acknowl-
edged.
The hydrolysis of alkyl formate was studied using formic acid as
the homogenous catalyst, Amberlite IR-120 resin as the heteroge-
neous catalyst and an additive as the complexation agent and the
three systems were compared arbitrarily under the same exper-
imental conditions. Both catalysts and the additive improved the
reaction rate significantly and reduced the time required to reach
the equilibrium. The equilibrium conversion remains unchanged in
the case of the heterogeneous catalyst but it was slightly reduced
when formic acid was used as the catalyst. However, the equilib-
rium conversion improved significantly when the hydrolysis reac-
tion was carried out in the presence of the additive. Furthermore,
the need for the separation of the formic acid product from water is
eliminated if the additive is used instead of the catalysts. Neverthe-
less, the three hydrolysis processes took place in an acidic medium.
Also, the catalytic activity of Amberlite IR-120 resins was com-
pared with different size ranges of Dowex 50Wx8 with the same
percentage of cross link and it was observed that Dowex 50Wx8-50
has almost the same activity as Amberlite IR-120. In addition, there
was the presence of internal mass transfer limitation for the larger
catalyst particles of Amberlite IR-120 resins and it was found that
the resins can be reused more than once without any significant
reduction in its activity. Thus, the cost of the resins per product
kilogram is significantly reduced. Furthermore, a reduction in the
energy cost was noticeable by operating at a lower temperature
with more catalyst.
References
[1] Kirk-Othmer Encyclopaedia of Chemical Technology, vol. 11, John Wiley & Sons,
New York, 1993, pp. 951–959.
[2] Ullmann’s Encyclopaedia of Industrial Chemistry, vol. 15, Wiley-VCH, Wein-
hein, 2003, pp. 49–70.
[3] A. Aguilo, T. Horlenko, Hydrocarb. Process 59 (1980) 120–130.
[4] H. Hohenschutz, J.E. Schmidt, H. Kiefer, EP 1432 (1978), to BASF (translated
from German to English).
[5] J.F. Eversole, US Patent 2,160,064 (1939), to Carbide and Carbon Chemicals.
[6] F.A. Long, M.A. Paul, Chem. Rev. 57 (1957) 946–950.
[7] R.P. Bell, A.L. Dowding, J.A. Noble, J. Am. Soc. 9 (1955) 3106–3110.
[8] W. Newling, C. Hinshekwood, J. Chem. Soc. (1936) 1357–1361.
[9] J.D Leonard, EP 5998 (1979).
[10] J.B. Lynn, O.A. Homberg, A.H. Singleton, US Patent 3,907,884 (1975), to Bethle-
hem Steel Corp.
[11] Y. Izumi, Catal. Today 33 (1997) 371–409.
[12] B.K. Cho, W.R. Carr, R. Aris, Chem. Eng. Sci. 35 (1980) 74–81.
[13] R. Wetherold, E. Wissler, K. Bischoff, Adv. Chem. Ser. 133 (1974) 181–190.
[14] K. Saari, E. Tirronen, A. Vuori, M. Lahtinen, US Patent 6,429,333 B1 (2002), to
Kemira Chemicals Oy.
[15] P.T. Mai, T.D. Vu, K.X. Mai, A. Seidel-Morgenstern, Ind. Eng. Chem. Res. 43 (2004)
4691–4702.
[16] C.W. Davies, G.G. Thomas, J. Chem. Soc. (1952) 1607–1610.
[17] T. Falk, A. Seidel-Morgenstern, Chem. Eng. Sci. 57 (2002) 1599–1606.
[18] R. Roy, J. Bhatia, J. Chem. Tech. Biotechnol. 37 (1987) 1–10.
[19] M.S. Metwally, A. Abel Razik, M.F. El Hadi, M.A. El-wardany, React. Kinet. Catal.
Lett. 49 (1) (1993) 151–159.
[20] S. Leveneur, J. Wärnå, T. Salmi, D. Murzin, L. Estel, Chem. Eng. Sci. 64 (2009)
4101–4114.
[21] Z.P. Xu, K.T. Chuang, Chem. Eng. Sci. 52 (1997) 3011–3017.
[22] A.A. Zagorodini, Ion Exchange Materials Properties and Application, Elsevier,
Amsterdam, 2007.
Acknowledgements
[23] M.R. Altiokka, A. C¸ itak, Appl. Catal. A: Gen. 239 (2003) 141–148.
[24] K.J. Laidler, Chemical Kinetics, Harper & Row, New York, 1987.
This work is part of the activities of the Åbo Akademi Process
Chemistry Centre (PCC) within the Finnish Centre of Excellence