Kinetic and safety characterization of the nitration process of methyl benzoate in mixed acid

Article abstract of DOI:10.1021/op300043x

The nitration of methyl benzoate is studied from a chemical and a kinetic point of view. The reaction network through which the system could evolve following the loss of thermal control or the uncorrect feed of the reagents is completely characterized. The data collected during the present investigation indicate that runaway phenomena can occur during the nitration of the substrate due to the development of side reactions. Isothermal experiments are thus carried out to estimate the unknown kinetic parameters through the adoption of a mathematical model able to predict the system behaviour upon variation of process parameters. The dependence of the acidity function H on the operating conditions is assessed. The proposed model and the estimated parameters are validated through the use of the results collected in a set of experimental runs performed at significantly different operating conditions from those adopted to identify them.

Full text of DOI:10.1021/op300043x

Article
pubs.acs.org/OPRD
Kinetic and Safety Characterization of the Nitration Process of
Methyl Benzoate in Mixed Acid
,†
‡
‡
‡
Ilaria Di Somma,* Raffaele Marotta, Roberto Andreozzi, and Vincenzo Caprio
†Istituto di Ricerche sulla Combustione (CNR), P. le V. Tecchio, 80, 80125, Napoli, Italia
‡Dipartimento di Ingegneria Chimica, Facolta
di Ingegneria, Universita di Napoli Federico II, P. le V. Tecchio 80, 80125 Napoli, Italia
̀ ̀
ABSTRACT: The nitration of methyl benzoate is studied from a chemical and a kinetic point of view. The reaction network
through which the system could evolve following the loss of thermal control or the uncorrect feed of the reagents is completely
characterized. The data collected during the present investigation indicate that runaway phenomena can occur during the
nitration of the substrate due to the development of side reactions. Isothermal experiments are thus carried out to estimate the
unknown kinetic parameters through the adoption of a mathematical model able to predict the system behaviour upon variation
of process parameters. The dependence of the acidity function H on the operating conditions is assessed. The proposed model
and the estimated parameters are validated through the use of the results collected in a set of experimental runs performed at
significantly different operating conditions from those adopted to identify them.
1. INTRODUCTION
2. EXPERIMENTAL SECTION
Methyl m-nitrobenzoate is a useful starting material for the
preparation of dyes and crop agent. This important
For all isothermal experiments a jacketed glass magnetically
1
a
−2
stirred (volume: 3.0 × 10 L) reactor has been used. The
temperature has been kept at desired value by using a Julabo
F32 refrigerated/heating circulator (cooling fluid: water). All
the runs have been carried out in batch mode. For each run 1.0
intermediate is currently obtained through the nitration of
2
methyl benzoate by means of mixed acid (an aqueous mixture
of nitric and sulfuric acids). Although this process is well-known
and it is currently used in school laboratories as a simple
−2
−2
×
10 −1.5 × 10 L of mixed acid with specific composition
3
,4
example of aromatic electrophilic substitution, there are no
indications in the literature about the kinetic parameters of
interest or about the network of reactions that can be involved
in it. Generally speaking, it is well-known that, being nitrations
exothermic reactions, runaway phenomena can occur during
these processes as the result of polynitration and/or side
reactions which take place when thermal control of the process
have been prepared adding, dropwise, a desired amount of
water and, successively, of nitric acid to sulfuric acid under
cooling, between 283 and 293 K. The temperature has been
then increased up to the desired value and an amount of
organic substrate has been instantaneously added to the
solution. In particular for each run, the concentration of the
involved species as a function of reaction time has been
recorded by submitting chemically quenched samples, with a
dilution in methanol (1.0 × 10− L of each sample withdrawn
from the reactor at varying reaction time were rapidly diluted in
5
is lost. With respect to this aspect it is useful to remember that
nitrations are along with polymerizations among the most
dangerous reactions at industrial level also for the inherent
4
6,7
thermal instability of the products. On the basis of these
considerations, the characterization of the reaction network
through which the system of interest could evolve is attempted
−2
1
0
L), collected during the experiments, to HPLC analysis
using a Hewlett-Packard model 1100 II, equipped with an UV−
vis detector and a Phenomenex Synergi 4 μ polar RP/80A
column. During the preliminary work to tune all the
experimental procedures, it has been checked that no changes
in the concentration of the species analyzed were observed
repeating the HPLC analysis at different times on the same
solution kept at room temperature. The following operating
conditions have been adopted: the mobile phase has been
formed by 70% of a buffer solution (vol %: CH OH 5%, H PO
8
from chemical and kinetic point of view. Investigations are
carried out by considering, as process deviations, the loss of
thermal control and an uncorrected feed of the reagents. On
the basis of the structure of the substrate adopted in the present
study, the occurrence of both polynitration and hydrolysis
reactions are expected following these process deviations. In
particular, a relevant influence on the system reactivity of the
protonating power of the medium (acidity function) may be
3
3
4
9
−11
foreseen. Different relationships
for the acidity function
0.4%, H O 94.6%) and 30% of acetonitrile, the signals have
2
dependence on the temperature and the composition of the
mixed acid were found in the literature. Therefore, the present
investigations seeks at assessing the most reliable relationship to
be used for the estimation of the acidity function value related
to an adopted mixed acid. The ultimate goal of present work is
thus the collection of all the information necessary to develop a
complete kinetic model able to simulate the behaviour of the
studied system for different possible initial conditions.
been acquired at 210, 220, 230, 240 nm, the column
temperature has been kept at 298 K and the flow rate set at
−3
−1
1
× 10 L × min . For all the experiments analytical grade
Special Issue: Safety of Chemical Processes 12
Received: February 21, 2012
Published: June 25, 2012
©
2012 American Chemical Society
2001
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
reagents have been used (H SO 98% was obtained from Fluka
2
4
and the other reagents from Sigma Aldrich).
3. RESULTS AND DISCUSSION
It is well-known that the loss of thermal control is one of the
most frequent and dangerous process deviations in chemical
1
2,13
industry
which may result into a runaway event and, in
some cases, into a thermal explosion with severe damages to
people and plants. In particular, in the case of nitration
processes, it is clear that an unwanted increase of the
temperature can give rise to polynitrations and/or other side
1
4
reactions. On the basis of these considerations, during a safety
assessment for a chemical process, the complete reaction
network through which the system could evolve needs to be
Figure 1. Comparison between the starting temperatures and
generated heats for the considered reaction: mononitration and
dinitration.
1
5
identified. For these reasons preliminary experiments of
nitration of methyl benzoate with mixed acid at ambient or near
ambient temperature have been performed. The data obtained
indicated that the main product is methyl m-nitrobenzoate,
with a minor occurrence of methyl p-nitrobenzoate. Since it is
well-known that an increase of the reaction temperature favors
polynitration reactions of aromatics with the formation of di-
16
and trinitroderivatives, whereas an uncorrected feed of the
amount of the water can give rise to hydrolysis reactions of the
species involved in the process with the formation of the
17
corresponding organic acids, a complex reaction network can
be expected to originate from the successive reactions of the
above-reported species. To better understand the dangerous-
ness of the system of interest some simple calculations can be
done. On the basis of the results of preliminary experiments it
can be stated that the mononitration reactions start
approximately at room temperature, the ester hydrolysis at
temperatures around 320 K, whereas the dinitrations become
relevant near 340 K. Assuming a heat of reaction equal to 1060
J/g for the nitration of methyl benzoate (by means of a group
contributing calculations ), considering an amount of mixed
acid equal to two times the mass of the substrate (a common
choice in industrial applications) and a mean specific heat equal
to 2 J/g K, in the case of a total loss of the cooling power
Figure 2. Possible temperature rises due to the combination of the
considered reactions.
The nitration of methyl benzoate (MB) has been
investigated, under isothermal conditions, at different temper-
atures and nitrating mixture compositions. The operating
conditions adopted are reported in table 1.
18
Table 1. Operating conditions adopted in the experiments
performed on methyl benzoate
(
adiabatic conditions), it is possible to calculate an adiabatic
temperature rise due to the complete mononitration of the
substrate as:
T
(K)
[MB]
(mol/L)
[ HNO3]
(mol/L)
[H2SO4]
(mol/L)
[H2O]
(mol/L)
2
2
83
88
0.224
0.224
0.224
0.270
4.52
4.52
4.52
5.06
9.14
9.14
9.14
8.72
16.59
16.59
16.59
16.05
ΔHnit
·cpM
ΔT = 3
= 117K
ad
(1)
293
93
2
Following the results of this calculation and considering a
process temperature for mononitration equal to 290 K, it can
be stated that-under adiabatic conditions- the side reactions,
polynitrations and hydrolysis of the products, can surely start
The data obtained in a run performed at 288 K are reported
in figure 3; it is clear that the main product was methyl m-
nitrobenzoate (3NMB), with a significant occurrence up to
around a yield of 20% of methyl p-nitrobenzoate (4NMB).
Similar results have been obtained in the other runs performed.
Successively, the reactivity of methyl m-nitrobenzoate has
been investigated (initial system composition, units mol/L:
C3NMB = 0.27, CHNO3 = 6, CH2SO4 = 12.2, CH2O = 5.5; T=343
K).
The gathered information (Figure 4) pointed out that the
compound was converted into methyl 3,5-dinitrobenzoate
(3,5DNMB) through the successive nitration and into 3-
nitrobenzoic acid (3NBA) through the hydrolysis. Both species
can be converted into 3,5-dinitrobenzoic acid (3,5DNBA):
3,5DNMB through a hydrolysis reaction whereas 3NBA
through a nitration step (scheme 1). It is important to stress
(
8
Figure 1). Moreover, considering a heat of reaction equal to
03 J/g for the nitration of the mononitroderivatives and
keeping constant the ratio between the total mass and the
reagent mass (in order to make the calculation easier), it is clear
that the temperature rise in the reactor could become
dangerously relevant. In fact, for the complete nitration of
the mononitrobenzoate an additional temperature rise of 134 K
b
is obtained. The final temperature due to the unwanted side
reactions will be thus between the following two calculated
values: 470 and 604 K (Figure 2). Moreover, it is foreseeable
that in this temperature range the thermal decomposition of
products starts. On the basis of the considerations done, a set of
experiments has been thus planned and performed to identify
the reactions really occurring when the temperature increases.
2
002
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
characterization of the investigated system is among the most
important requirements for the adoption of safe operating
conditions devoted to prevent the occurrence of unwanted
events. Therefore, with the aim of estimating the kinetic
parameters of the reactions involved when the nitration of
methyl benzoate is carried out in the presence of a loss of the
thermal control or with an uncorrected feed of the reagents, a
new set of isothermal experiments has been performed. Four
subsystems obtained respectively by the reactions involving the
species identified in the scheme 1 (3NMB, 3,5DNMB and
3
NBA) have been taken into account for further investigations;
in particular, in the case of 3NMB, being this specie involved in
both the reactions, nitration and hydrolysis, the two different
pathways have been thus separately studied. In other words, the
four reactive steps, that is, the hydrolysis of 3NMB, the
nitration of 3NMB, the hydrolysis of 3,5DNMB and the
nitration of 3NBA, have been separately investigated carrying
out some experimental runs at different temperatures and
different nitrating mixture compositions. In table 2 the
operating conditions adopted in the experiments performed
on 3NMB have been reported; it is important to stress that an
increase of the amount of the water in the nitrating mixture
favors the hydrolysis. This last sentence becomes clear by
Figure 3. Methyl benzoate nitration at 288 K. Initial system
composition: CMB = 0.224 mol/L, CHNO3 = 4.52 mol/L, CH2SO4
=
9.14 mol/L, CH2O = 16.59 mol/L.
c
comparing figures 4 and 5 ; in fact, the results reported in figure
5
are related to a run performed on the system at the same
temperature of that whose data are shown in figure 4, 343 K,
but with a different mixture composition, in particular with a
higher ratio water/substrate with respect to the other (see table
2
).
The same procedure has been adopted for the other
Figure 4. Methyl m-nitrobenzoate nitration and hydrolysis at 343 K.
Initial system composition: C3NMB = 0.27 mol/L, CHNO3 = 6 mol/L,
CH2SO4 = 12.2 mol/L, CH2O = 5.5 mol/L.
subsystems identified in the previous step; in other words.
the runs have been first performed at fixed initial concen-
trations of both substrate and nitrating mixed, varying the
temperature and successively, at temperature values already
tested, the experiments have been repeated with different
reactive mixture compositions (data not reported).
On the basis of the results obtained during this investigation
a mathematical model aiming at describing the system
behaviour under the different operating conditions analyzed
has been developed.
Scheme 1. Reaction network identified during the
investigations
In detail, for each species included in the scheme 1, a material
balance equation has been written:
n
h
dCi
2
=
∑ ±k* ·C ·C
± ∑ ±k* ·C ·C
N,p
j
HNO3
H,l
j
H O
2
dt
p=1
l=1
(2)
The subscript i indicates the i-th species, whereas the subscript j
the j-th species participating in the p-th nitration or in the l-th
hydrolysis, including the possibility that i is equal to j. The
parameter k* represents a gross kinetic constant, which makes
possible to use a simple power law to describe the
disappearance and formation of involved species. On the
basis of the reaction mechanisms, known from the literature for
both nitration and hydrolysis, the true kinetic laws are reported
that the selectivity between the two different reactions
(
nitration and hydrolysis) depends upon the composition of
d
below :
mixed acid, an increase of the water concentration favoring the
hydrolysis.
⎛
⎜
−E ⎞
N
r = k ·C
+
·CArH = k0N·exp
⎟
·CNO2
+
·CArH
·CArH
N
N
NO2
⎝
⎠
(3)
(4)
R·T
3.1. KINETIC ASSESSMENT
The results previously collected and discussed clearly indicate
that a loss of the thermal control or an uncorrected feed of the
reagents during the nitration process of methyl benzoate can
give rise to a complex reaction network which could result into
a thermal explosion. It is well-known that a thermokinetic
= k0H·exp⎛−E ⎞
H
r = k ·C
·CArH
+
⎜
⎝
⎟
⎠
·C
+
H
H
H O
H O
2
2
R·T
were CArH is the concentration of the organic substrate
+
participating in the nitration and CArH is the concentration
2
003
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
Table 2. Operating conditions adopted in the experiments performed on methyl m-nitrobenzoate
T [K]
reaction N = nitration H=hydrolysis
[3NMB] (mol/L)
[HNO ] (mol/L)
[H SO ] (mol/L)
[H O] (mol/L)
2
[H O]/[3NMB]
2
3
2
4
3
3
3
3
3
3
3
3
3
23
23
38
43
43
13
23
43
23
N
N
N
N
N
H
H
H
H
0.27
0.27
0.28
0.27
0.48
0.20
0.18
0.20
0.25
6.00
6.31
6.00
6.00
6.80
4.48
4.48
4.48
5.30
12.20
12.73
12.20
12.20
11.70
9.05
5.50
3.30
20.37
12.22
19.64
20.37
11.46
82.50
91.67
82.50
35.20
5.50
5.50
5.50
16.50
16.50
16.50
8.80
9.05
9.05
10.63
f
0
E = E + ΔH
R
R
(j)
(9)
(10)
f
f
f
(−E /R·T)
R
k = k ·e
R
0R
where j is referred to the equilibrium (a) for the nitration and
to equilibrium (b) for the hydrolysis.
So the eqs 3 and 4 became:
γ
HNO ·H O
1
f
3
2
·10−H· C ·CHNO3·CArH
r = k ·
N
N
γ
H O
H2O
(11)
(12)
2
r = k ·γ ·10−H·C
f
2
H
H
H O
H O
·CArH
2
2
Therefore, the eq 2 can be just obtained assuming:
Figure 5. Methyl m-nitrobenzoate hydrolysis at 343 K. Initial system
γ
HNO ·H O
1
CH2O
f
3
2
·10−H
·
composition: C3NMB = 0.20 mol/L, CHNO3 = 4.48 mol/L, CH2SO4
=
k*N = k ·
N
9.05 mol/L, CH2O = 16.50 mol/L.
γ
H O
(13)
(14)
2
k*H = k ·γ ·10−H
f
H
H O
2
of protonated substrate involved in the hydrolysis reaction.
CNO2 and CArH , in the eqs 3 and 4, can be derived from the
equilibria established in the reactive systems:
To solve the obtained differential equations system, the
knowledge of the activity coefficients for the ternary system
+
+
9
HNO /H O/H SO is required along with a suitable relation-
3
2
2
4
ship for H. From a literature survey different and conflicting
C
+
≅ aNO2
+
NO2
9−11
relationships
have been found to describe the dependence
Keq(a)·a
·a
−
HNO ·H O H SO
4
of H on the temperature and on the composition of the mixed
acid. Therefore, in the present investigation, an equation of the
following type has been adopted:
3
2
2
=
=
2
a
Keq(a)·a
a
H O·HSO
2
4
·a
HNO ·H O H SO
4
3
2
2
⎛
⎜
g ⎞
2
−
−H
=
(a·C
+ b·CHNO3)·
z +
H O·HSO
(5)
H2SO4
2
4
⎝
⎠
T
(15)
CArH
+
≅ aArH
+
in which the coefficients a, b, z and g have been considered as
unknown parameters to be estimated.
Keq(b)·aArH·a
H SO
2
4
=
The experimental results obtained during the runs performed
on each reactive subsystem identified in the reaction scheme 1
have been simultaneously used in a single optimization
procedure to estimate the unknown gross kinetic parameters
a
−
HSO4
Keq(b)·CArH·γArH·γ
C
H SO
2
H SO
2
4
4
=
·
γ
−
CHSO4
−
f
f
k
and E , defined in the eqs 8 and 9, and the coefficients a, b,
HSO4
(6)
0R
R
z and g in the eq 15. The adopted procedure has been based on
the minimization of an objective function defined as:
It is possible to assume that small concentrations CNO2+ and
CArH+ are equal to the corresponding activities. Introducing the
acidity function H, defining a gross pre-exponential factor k
and a gross activation energy E (where R is equal to N for the
m
n
m
f
0R
φ = ∑ ∑ ∑ (yi,j,l − Ci,j,l)2
f
R
i=1 j=1 i=1
(16)
nitration and H for the hydrolysis) and considering an
Arrhenius-like equation for the global gross kinetic constant kR:
In which the terms y and C are the calculated and
experimental concentration whereas m, n, and h are the
number of experimental data recorded in each experiment, the
number of the substances and the number of the experiments
used in the optimization procedure respectively. Some
examples of comparison of calculated (continuous lines) and
a
H SO
2
4
H = −log a
2
·aHSO4
−
(7)
(8)
H O
k0f = k ·e
(ΔS(j)/R)
R
0R
2
004
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
experimental (symbols) concentration are shown in Figures
e
6
−10 for each subsystem identified in the reactive scheme:
Figure 9. Comparison between experimental (symbols) and calculated
(continuous lines) data for 3-nitrobenzoic acid nitration at 343 K.
Initial system composition: C3NBA = 0.26 mol/L, CHNO3 = 6.02 mol/L,
CH2SO4 = 12.20 mol/L, CH2O = 5.60 mol/L.
Figure 6. Comparison between experimental (symbols) and calculated
continuous lines) data for methyl benzoate nitration at 288 K. Initial
system composition: CMB = 0.224 mol/L, CHNO3 = 4.52 mol/L, CH2SO4
9.14 mol/L, CH2O = 16.59 mol/L.
(
=
Figure 10. Comparison between experimental (symbols) and
calculated (continuous lines) data for methyl 3,5-dinitrobenzoate
hydrolysis at 343 K. Initial system composition: C3,5DNMB = 0.19 mol/
L, CHNO3 = 4.63 mol/L, CH2SO4 = 9.50 mol/L, CH2O = 16.70 mol/L.
Figure 7. Comparison between experimental (symbols) and calculated
continuous lines) data for methyl m-nitrobenzoate nitration and
(
hydrolysis at 343 K. Initial system composition: C3NMB = 0.27 mol/L,
CHNO3 = 6 mol/L, CH2SO4 = 12.2 mol/L, CH2O = 5.5 mol/L.
Table 3. Best estimated values for the unknown kinetic
parameters identified in scheme 1
reaction
Ef [kJ/mol]
kf [reaction dependent]
0R
R
(
(
(
(
(
(
9.61±0.4)
7.06±0.3)
8.25±0.5)
5.94±0.3)
7.24±0.4)
5.25±0.3)
1
2
3
4
5
6
79.1 ± 4.2
68.2 ± 5.0
107.2 ± 7.9
86.2 ± 6.2
92.9 ± 5.4
87.9 ± 7.1
10
10
10
10
10
10
used in previous experiments (0.18 mol/L÷0.48 mol/L). Also
in this condition the system kept at constant temperature
during the runs and the proposed model well predicts its
behaviour (Figures 11 and 12). Therefore, it can be stated that
the model is capable to predict the system behaviour at varying
its composition provided that it is perfectly homogeneous; that
is, the developed model can be applied for experimental
conditions in which no immiscible phases are present, since no
equations to account for material transport are included in it.
Figure 8. Comparison between experimental (symbols) and calculated
(
continuous lines) data for methyl m-nitrobenzoate hydrolysis at 343
K. Initial system composition: C3NMB = 0.20 mol/L, CHNO3 = 4.48
mol/L, CH2SO4 = 9.05 mol/L, CH2O = 16.50 mol/L.
The estimated values of the unknown kinetic parameters are
4
. CONCLUSIONS
reported in table 3, whereas for a, b z and g coefficients the
−1
following values have been evaluated: a=0.56 ± 0.028 L mol ,
The behaviour of the system methyl benzoate/mixed acid has
been studied when a loss of the thermal control or an uncorrect
feed of the reagents happens. In these conditions, the possibility
of occurrence of runaway phenomena and thermal explosion
has been demonstrated as a result of a process development
−1
b=0.52 ± 0.022 L mol , z=0.33 ± 0.015 and g=49.0 ± 2.1 K.
f
To validate the proposed model , two additional nitration
runs have been performed on 3NMB starting from initial
concentrations higher (0.95 mol/L and 1.14 mol/L) than those
2
005
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
T
R
Temperature K
Universal constant of gas J mol− K−1
1
2
−2
k*H,i Gross kinetic constant for i-th hydrolysis L mol
−1
min
k*N,i Gross kinetic constant for i-th nitration L mol min−1
kR
−1
Kinetic constant for R-th reaction L mol min−1
−1
Velocity of R-th reaction mol L− min
1
−1
r
k
R
0
Pre-exponential factor for R-th reaction L mol min−1
−1
R
Activation energy for R-th reaction kJ mol−
1
ER
t
ai
γi
Reaction time min
Activity of i-th species mol L−
1
Activity coefficient of i-th species dimensionless
Acidity function dimensionless
H
f
2
−2
Figure 11. Comparison between experimental (symbols) and
calculated (continuous lines) data for methyl m-nitrobenzoate
nitration and hydrolysis at 343 K at initial concentrations higher
than those used previously. Initial system composition: C3NMB = 0.92
mol/L, CHNO3 = 6.75 mol/L, CH2SO4 = 11.00 mol/L, CH2O = 5.90 mol/
L.
k
Gross kinetic constant for R-th reaction (L mol
R
−1
−1
min ) or (min )
f
−1
E
k
Gross activation energy for R-th reaction kJ mol
R
f
2
Gross pre-exponential factor for R-th reaction (L
0
R
−2
−1
−1
mol min ) or (min )
−
1
yi
ΔS
ΔH
Calculated concentration of i-th species mol L
Entropy change for the j-th equilibrium J K− mol−1
1
(
j)
0
−1
Entalpy change for the j-th equilibrium J mol
(
j)
ADDITIONAL NOTES
Generally speaking an efficient stirring is really important to
carry out a nitration process in which two immiscible phases
■
a
(
organic and aqueous ones) are contacted. In the present work,
since the primary interest was that of achieving a basic kinetic
characterization of the studied system, we investigated the
nitration of methylbenzoate under experimental conditions in
which only a single phase was present (in other words, the
system was perfectly homogeneous!). In this case, the sole
mixing that could be critical was that of the substrate and mixed
acid, at the beginning of the run, when the first was poured in
the second one. Direct observations done for many runs
indicated that with the magnetic mixing we have chosen, after a
very short time (20−30 s), an homogeneous single phase was
obtained.
Figure 12. Comparison between experimental (symbols) and
calculated (continuous lines) data for methyl m-nitrobenzoate
nitration and hydrolysis at 345 K at initial concentrations higher
than those used previously. Initial system composition: C3NMB = 1.14
mol/L, CHNO3 = 8.50 mol/L, CH2SO4 = 9.00 mol/L, CH2O = 4.50 mol/
L.
b
Calculations have been done with a conservative approach.
Therefore, the final temperatures have been calculated by
considering a closed system. Consequently, no water vapor-
ization heat has been considered in the thermal balance.
through a complex reaction network. Kinetic assessments have
been performed for each reaction of the network allowing the
assessment of the dependence of gross kinetic parameters upon
temperature and the protonating power of the medium.
Satisfactory results have been obtained in a validation
procedure of estimated kinetic parameters, which have been
used to simulate the data collected during additional nitration
experiments, still under isothermal conditions, on more
concentrated systems than those used to develop the model.
c
At high water content and 70 °C the nitration of 3NMB does
not occur.
d
In the eqs 2 and 4 the concentration of water is present with
two different orders. Obviously this is not a mistake! In fact, eq
4
is obtained by considering the rate-determining step in the
hydrolysis process which is represented by the attack of water
on protonated ester molecule (water concentration present at a
first order). In the eq 2 the final form is reported which can be
obtained by making the appropriate substitutions in eq 4 (water
concentration present at second order).
AUTHOR INFORMATION
Corresponding Author
Telephone: +39 081 7682225. Fax: +39 081 5936936. E-mail:
idisomma@unina.it.
e
■
*
With reference to Figure 6 the reviewer asks about the origin
of the parameters used for profile calculation. These parameters
have been estimated by means of an optimization procedure
which is based on the minimization of the objective function
(eq 16) and makes use of all the experimental data collected
during the runs at the same time. In other words, the values of
the parameters have been estimated as those which make
possible to calculate the concentration profiles which show the
best fitting with the experimental data.
Notes
The authors declare no competing financial interest.
NOMENCLATURE
■
ΔT
Adiabatic temperature rise K
ad
−
1
f
ΔH
Heat of R-th reaction J g
In the text it has been reported that some validation runs have
R
−
1 −1
c
pM
Mean specific heat J g K
Concentration of i-th species mol L
been performed. These have been done mainly to demonstrate
that the estimated values of the parameters may allow to predict
−1
Ci
2
006
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007

Organic Process Research & Development
Article
the behaviour of the system also at varying the initial
conditions. In particular, two validation runs (Figure 11 and
1
2) have been carried out at starting concentrations of the
substrate significantly higher than those adopted in the other
experiments whose results have been used to estimate the
values of the unknown parameters. Generally, we can state that
the model is capable to predict the system behaviour provided
that this is perfectly homogeneous, no matter the starting
composition of mixed acid or the concentration of the
substrate; that is, the developed model cannot be applied for
conditions in which two immiscible phases are present, since no
equations to account for material transport are included in it.
REFERENCES
■
(
1) Hackenberger, A.; Patsch, M.; Gaeng, M. Preparation of Methyl m-
nitrobenzoate, U.S. Patent 4,506,089, 1985.
(
(
2) Kamm, O.; Segur, J. B. Org. Synth. 1941, 1, 372.
3) Clarke, D. D. Abstracts, 42nd Middle Atlantic Regional Meeting
of the American Chemical Society, College Park, MD, United States,
May 21−24, 2011.
(
4) Hudson, M. J.; Isaacs, N. S.; Wheeler, J. G. School Science Review.
1
(
980, 62 (219), 330−332.
5) Kozak, G. D.; Raikova, V. M. Cent. Eur. J. Energetic Mater 2010, 7
1), 21−32.
6) Maddison, N.; Rogers, R. L. Chemical Technology Europe 1994, 1
5), 28−31.
7) Kozak, G. D.; Vasin, A. Ya.; Dyachkova, A. V. New Trends in
(
(
(
(
Research of Energetic Materials, Proceedings of the Seminar, 11th,
Pardubice, Czech Republic, Apr. 9−11, 2008, 381-386.
(
8) Andreozzi, R.; Canterino, M.; Caprio, V.; Di Somma, I.;
Sanchirico, R. J. Hazard. Mater. 2006, A138, 452−458.
9) Zaldivar, J. M.; Barcons, C.; Molga, E.; Snee, T. J. Chem. Eng. Sci.
992, 47, 2517−2522.
10) Cook, M. J.; Dassanayake, N. L.; Johnson, C. D.; Katrizky, A. R.;
Toone, T. W. J. Am. Chem. Soc. 1975, 97, 760.
11) Kanhere, J. M.; Chandalia, S. B. Ind. Eng. Chem. Process. Des. Dev.
981, 20, 407−409.
12) Casson, V.; Lister, D. G.; Milazzo, M. F.; Maschio, G. J. Loss
(
1
(
(
1
(
Prev. Process Ind. 2012, 25 (1), 209−217.
(
13) Gustin, J. L. J. Loss Prev. Process Ind. 1993, 6 (5), 275−291.
(
14) Andreozzi, R.; Caprio, V.; Di Somma, I.; Sanchirico, R. J.
Hazard. Mater. 2006, A134, 1−7.
15) Stoessel, F. Thermal Safety of Chemical Processes; Wiley-VCH:
New York, 2008.
16) Raikova, V. M.; Ivanova, O. A.; Shraiber, G. A. New Trends in
(
(
Research of Energetic Materials, Proceedings of the Seminar, 10th,
Pardubice, Czech Republic, Apr. 25−27, 2007, 866-875.
(
17) Gunaydin, H.; Houk, K. N. J. Am. Chem. Soc. 2008, 130 (46),
1
5232−15233.
18) Medard, L. A. Accidental explosions Vol. 1; Ellis Horwood
Limited: Chichester, 1989.
(
2
007
dx.doi.org/10.1021/op300043x | Org. Process Res. Dev. 2012, 16, 2001−2007