Intensification of Nitrobenzaldehydes Synthesis from Benzyl Alcohol in a Microreactor

Article abstract of DOI:10.1021/acs.oprd.6b00426

A one-step process to produce 2- and 3-nitrobenzaldehyde isomers starting from benzyl alcohol in aqueous mixed nitric and sulfuric acid was developed as an inherently safe continuous flow process in a microreactor. The previously published kinetic model was validated in a microreactor and used to optimize operating conditions in silico to attain the desired product distribution. The molar fractions of nitric and sulfuric acids were increased up to 0.35, 0.45, and temperature to 68 °C, which was impossible to do under batch conditions. Experiments under continuous flow conditions have shown that yields of about 42% and 96% for the ortho- and the meta- isomers, respectively, can be achieved.

Full text of DOI:10.1021/acs.oprd.6b00426

Article
pubs.acs.org/OPRD
Intensification of Nitrobenzaldehydes Synthesis from Benzyl Alcohol
in a Microreactor
Danilo Russo,*,†,§ Ilaria Di Somma,‡ Raffaele Marotta,† Giovanna Tomaiuolo,† Roberto Andreozzi,†
Stefano Guido,† and Alexei A. Lapkin*,§
†Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale, Universita
̀
di Napoli Federico II, p.le V. Tecchio,
80−80125 Napoli, Italy
‡Istituto di Ricerche sulla Combustione (CNR), p.le V. Tecchio, 80−80125 Napoli, Italy
§Department of Chemical Engineering and Biotechnology, University of Cambridge, Cambridge CB2 1TN, United Kingdom
ABSTRACT: A one-step process to produce 2- and 3-nitrobenzaldehyde isomers starting from benzyl alcohol in aqueous mixed
nitric and sulfuric acid was developed as an inherently safe continuous flow process in a microreactor. The previously published
kinetic model was validated in a microreactor and used to optimize operating conditions in silico to attain the desired product
distribution. The molar fractions of nitric and sulfuric acids were increased up to 0.35, 0.45, and temperature to 68 °C, which was
impossible to do under batch conditions. Experiments under continuous flow conditions have shown that yields of about 42%
and 96% for the ortho- and the meta- isomers, respectively, can be achieved.
including, specifically, nitrated products.26−29 The peculiarity
of the small-diameter devices with embedded mixing and heat
exchange, is the ability to enhance heat and mass transfer,
INTRODUCTION
■
Nitrobenzaldehydes are important intermediates involved in
the synthesis of a wide range of bulk and fine chemicals.1,2 The
allowing to precisely control temperature and concentration
most recent literature shows an increasing interest in the
profiles. This allows to safely carry out reactions, adopting
synthesis of new high value molecules involving ortho-3−8 and
conditions under which a traditional batch reactor cannot
meta-nitrobenzaldehyde isomers,9−12 that have applications as
operate.30 Especially when involving fast exothermic reac-
anti-inflammatory, antibacterial, and antitumor agents, as well
as therapeutic candidates for cardiovascular diseases, fluores-
cence markers and optical materials. 3-Nitrobenzaldehyde is
tions.31,32 Several studies pointed out the possibility of
successfully suppressing hot spots, leading to improvements
in yield, selectivity, product quality, and safety, while enabling
traditionally produced via direct fed-batch nitration of
to perform hazardous reactions under unusual conditions.33−35
benzaldehyde using mixtures of concentrated nitric and sulfuric
Previous investigations also show an interest in increasing the
acids at 5−10 °C, leading to the formation of 2-nitro-
benzaldehyde as a side product with yields up to 20%.13,14 As
2-nitrobenzaldehyde yield during the direct nitration of
benzaldehyde in order to overcome safety problems associated
a consequence, different processes have been successively
developed in order to profitably synthesize the ortho-isomer,
such as cinnamic acid nitration, 2-nitrostyrene oxidation, and 2-
with some of the alternative proposed processes.2,36,37 These
studies demonstrated that an increase in nitric acid
nitrobenzyl bromide oxidation.13 However, most of these
concentration can increase the yield of 2-nitrobenzaldehyde,
but warn about a consequent increase in the reactivity and the
processes involve highly exothermic reactions, requiring
generated reaction heat. A recent kinetic study under batch
stringent control of operating conditions, evident by the low
conditions was undertaken in order to evaluate the kinetic
operating temperatures and the limited tolerance toward
parameters and predict the behavior of the reacting system at
changes in the nitrating agent composition. Specifically, during
the direct nitration of benzaldehyde the process could easily
varying operating conditions.2 Literature findings14,38,39 and
recent kinetic modeling40 also showed the possibility of
undergo explosive thermal runaway, due to the exothermic
selectively converting benzyl alcohol to benzaldehyde in the
oxidation and nitration reactions occurring in the acid mixture,
same nitrating mixtures in order to overcome the drawbacks of
if the reaction is performed in a conventional refrigerated batch
the traditional route of toluene oxidation to produce
reactor. In this respect, the possibility of performing the direct
benzaldehyde that requires expensive catalysts as well as
nitration process safely under various, and preferably more
temperatures and pressures up to 160 °C and 70 atm,
intensive, operating conditions is of great interest. Which is why
we turned our attention to compact and microreactors, which
respectively.14 In the present article a new process to synthesize
2- and 3-nitrobenzaldehyde isomers through benzyl alcohol
reportedly allow safe operation of the highly hazardous
processes, such as direct oxidation of hydrogen,15−17 oxidation
oxidation/nitration is proposed, adopting a commercial glass
microreactor. The advantage of using the microreactor chosen
of hydrocarbons with molecular oxygen at elevated pressures
and temperatures,18,19 and reactions with HF or hydrazoic
for this study, is that the scaled-up versions with similar
acid,20−22 as some examples.
Continuous flow microreactors have been applied in the
Received: December 16, 2016
Published: February 20, 2017
syntheses of highly value/low-throughput molecules23−25
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
functionalities are commercially available from Corning, in
glass, or Chemtrix, in silicon carbide. In this respect, the first
aim of this article is to validate the model prediction by
adopting experimental conditions different from those used to
develop the model in the previous batch experiments.
Furthermore, the kinetic model was used to identify the
experimental conditions to maximize the yield of 2- and 3-
nitrobenzaldehyde in a continuous process, and the results
confirmed the possibility of producing the two isomers with the
highest yield published so far.
times. The residence time in the reactor was changed varying
both the flow rate and the volume in the ranges 0.45−9.99 mL·
min−1 and 1.5−4.5 mL, respectively. The collected samples
were rapidly diluted and analyzed.
HPLC analysis was performed using a Shimadzu Prominence
HPLC, a UV DAD detector, and a Phenomenex Synergi 4 μm
polar RP/80A column, thermostated at 303 K. The mobile
phase (1.0 mL·min−1) was constituted of eluant (A) (buffer
solution: CH3OH 5% v/v; H3PO4 0.4% v/v; H2O 94.6% v/v)
and eluant (B) (acetonitrile). The gradient was as follows: 15%
B for 8 min, increased to 25% B in 10 min, and successively
decreased to 15% B in 5 min. The signals were acquired at
wavelength of 230, 250, and 265 nm.
EXPERIMENTAL SECTION
■
Materials. All reagents (acetonitrile ≥99.9%; methanol
≥99.9%; phosphoric acid 85 wt% in H2O; benzyl alcohol
anhydrous 99.8%; benzaldehyde ≥99%; 2-nitrobenzaldehyde
98%; 3-nitrobenzaldehyde 99%; 4-nitrobenzaldehyde 98%;
benzoic acid ≥99.5%; 2-nitrobenzoic acid 95%; 3-nitrobenzoic
acid 99%; 4-nitrobenzoic acid 98%; nitric acid fuming ≥99%;
sulfuric acid 99.999%; ethylene glycol technical with corrosion
inhibitor; urea 99%) were purchased from Sigma-Aldrich and
used as received. Water for experiments was obtained by means
of a Maxima (USF) Milli-Q system.
RESULTS AND DISCUSSION
■
Reaction Mechanisms. Based on the previously published
kinetics,2,40 the following simplified reaction scheme can be
proposed for the ortho- and meta-nitrobenzaldehyde syntheses
starting from benzyl alcohol, Scheme 1.
Benzyl alcohol (1) is completely converted to benzaldehyde
through the formation of protonated benzyl nitrite (2) as a key-
intermediate. Moreover, benzaldehyde can be either directly
nitrated on the aromatic ring to give 3-nitrobenzaldehyde (3)
or can coordinate a nitronium ion (4) on the aldehyde
group2,36 (6) and internally rearrange to form the ortho-isomer
(5). The internal rearrangement occurs when the coordinated
species (6) is attacked by another nitronium ion.2,36 Nitronium
ion concentration is affected by nitric acid concentration in
mixed acid due to the following equilibrium:
Procedures. The experiments were carried out in both
batch and continuous modes. The adopted batch reactor is a
magnetically stirred jacketed glass reactor (3.0 × 10−2 L)
refrigerated using a circulator (Thermo-Scientific DC30 K20;
cooling fluid: ethylene glycol). A scheme of the reactor can be
found elsewhere.40 Nitric acid, sulfuric acid, and water in
appropriate amounts were previously mixed and equilibrated at
the operating temperature. Benzyl alcohol was then added and
samples (10−4 L) were collected at different reaction times and
rapidly quenched by diluting (1:100) with a solution of urea in
acetonitrile (6.66 × 10−2 M), and analyzed by HPLC.
The continuous flow experimental runs were carried out in a
commercial glass microreactor with embedded static mixer,
heat exchange, and three sampling ports along the reactor
length (Little Things Factory XXL-ST-04). A scheme of the
reactor is shown in Figure 1.
HNO3·H2O + H2SO ↔ NO+2 + HSO− + 2H2O
4
4
Russo et al.,2 pointed out that direct nitration of the aromatic
ring to form 2-nitrobenzaldehyde is negligible compared to the
latter mechanism of internal rearrangement. It is worth pointing
out, that according to the published kinetic model2 the internal
rearrangement is only possible in ortho-position because of its
spatial closeness to the aldehydic group, whereas the meta-
isomer (3) is formed by the aromatic ring direct nitration of
both benzaldehyde and species (6); the latter can lose the
coordinated −NO2 on the aldehydic group after direct nitration
in meta-position.2 In the previous investigations1,2 there was no
significant evidence of 4-nitrobenzaldehyde and dinitrated
products formation.
For the sake of clarity, the oxidation pattern leading to the
formation of benzoic acid, 2- and 3-nitrobenzoic acid has been
discarded in Scheme 1. However, under most of the adopted
conditions these secondary products were not detected, or were
measured at trace levels.
Batch Reactions and Evaluation of Reaction Hazard.
Some preliminary batch experimental runs were carried out in a
homogeneous phase to confirm the possibility of producing 2
and 3-nitrobenzaldehydes from benzyl alcohol. Figure 2 shows
results of an experimental run carried out by adding benzyl
alcohol (C0 = 45 mM) into the standard mixed acid (HNO3 =
20% w/w; H2SO4 = 60% w/w; H2O = 20% w/w).
Figure 1. Scheme of the microreactor. (1) Coolant inlet. (2) Coolant
outlet. (3) Mixed acid inlet. (4) Benzyl alcohol inlet. (5) Sampling
points.
Mixed acids and benzyl alcohol were separately fed by means
of a peristaltic pump (V-3 pump, Vapor tec Ltd.) and a syringe
pump (PhD Ultra Harward Apparatus), and preheated to the
operating temperature. Temperature was kept constant in the
reactor during each experimental run by using the thermostated
circulator connected to the embedded heat exchanger. Samples
were collected at three different outlets at different residence
The results clearly show that it is possible to obtain 2- and 3-
nitrobenzaldehydes starting from benzyl alcohol. Moreover,
under the adopted conditions, the rate of benzyl alcohol
oxidation to benzaldehyde is much faster than that of nitration
so that no traces of benzyl alcohol are detected even for the
sample collected immediately after addition of the alcohol.
Benzaldehyde was instantaneously formed and slowly nitrated.
B
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 1. Simplified Scheme of Nitrobenzaldehydes Synthesis from Benzyl Alcohol in Mixed Acids
Figure 2. Oxidation/Nitration of benzyl alcohol. (a) Reaction performed by adding benzyl alcohol to acid. (b) A comparison of two runs starting
■ □
○
▲ △
, ) 3-
from benzaldehyde (full symbols) or benzyl alcohol (empty symbols). Figure legend: (•, ) Benzaldehyde; ( , ) 2-nitrobenzaldehyde; (
◇
nitrobenzaldehyde; ( ) benzyl alcohol; T = 20 °C; (lines are for convenience of the eye).
results shown in Figure 2 were obtained at low concentrations
of the reactants and long reaction times, to allow for heat
transfer, thus operating at isothermal conditions. An increase in
concentrations of either the acids or the reactants will
significantly increase the demand for heat removal. Figure 3
shows temperature profile of the reacting mixtures resulting
from the addition of benzyl alcohol to the standard mixture of
acids using an organic/nitrating mixture ratio 0.4 w/w at T = 23
°C. The results clearly show that temperature can very rapidly
increase up to 83 °C.
Figure 3. Temperature profile of the reacting mixture. Organic/
nitrating reagents ratio = 0.4 w/w. Standard mixed acids. T0 = 23 °C.
These results suggest that the process could not be safely
carried out adopting a standard batch reactor in which the
prevention of the thermal runaway explosion cannot be
guaranteed.
In Figure 2b the normalized concentrations measured in two
different runs carried out under the same conditions but
starting either from benzyl alcohol or from benzaldehyde, are
reported. The concentrations profiles are coincident, suggesting
that the nitration process is not affected by the starting material,
since oxidation rate of benzyl alcohol is much higher than
nitration rate of benzaldehyde.
Tandem oxidation-nitration process starting from benzyl
alcohol seems an attractive option, but it presents a serious
safety concern due to both the very fast rate of the highly
exothermic oxidation step and the consequent nitration. The
Moreover, assuming a heat of reaction of about ΔHox = 1100
J·g−1 and ΔHn = 1185 J·g−1 (estimated by group contribution
method41) for the oxidation and the nitration reactions
42
respectively, and a mean specific heat cp of 2 J·g−1·K−1 it is
possible to estimate the adiabatic temperature rise for complete
oxidation/nitration of the organic substrate as follows (eq 1):
ΔHox + ΔHn
mAr
mAr + mmix
ΔT =
·
ad
cp
(1)
C
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
trinitration, side oxidations) leading to the formation of
unstable intermediates with release of gaseous species and
vessels pressurizations and/or explosion.43
Microreactor Performance. To safely carry out the
process under harsher conditions, namely higher temperature
and stronger acidic media, the glass microreactor with
embedded heat transfer and static mixer was adopted.
The previously obtained kinetic parameters2,40 were used to
adapt the mathematical model to a continuous flow reactor
based on the assumption of a plug flow reactor (PFR). For each
species the mass balance has been described as (2)
dci
dτ
=
r
j
∑
j
(2)
where ci is the concentration of the i-th species, τ is the mean
residence time in the reactor (ratio between the volume and the
flow rate), and rj is the j-th reaction rate of formation (positive)
or consume (negative) of the i-th species. Details of
experimental conditions for all the microreactor runs reported
in Figures 4−8 are summarized in Table 1. The model was
validated by comparing its prediction to the experimental
results for the different experimental conditions showing a good
predictive capability of the kinetic model, Figure 4.
Once validated, the model was used to define the best
conditions to maximize the yield of the two isomers. Figure 5
shows the effect of temperature and mixed acid composition on
the yield of 2-nitrobenzaldehyde in the neighborhood of the
local optimum. An upper limit temperature was fixed at 70 °C
to avoid a significant decomposition of nitric acid in mixed
acids at higher temperatures (bold vertical line).
Figure 4. Concentration vs residence time. Calculated (lines) and
■
●
experimental data (symbols). ( ) Benzaldehyde; ( ) 2-nitro-
▲
benzaldehyde; ( ) 3-nitrobenzaldehyde. (a) xn (nitric acid molar
fraction) = 0.197; xs (sulfuric acid molar fraction) = 0.347; T = 6 °C;
C0 = 0.05 M. (b) xn = 0.199; xs = 0.348; T = 25 °C; C0 = 0.2 M.
An optimal temperature, which maximizes the yield of ortho-
nitrobenzaldehyde can be found at a fixed mixed acid
composition. An increase in acid concentration shifts the
maximum point to higher temperatures and is reflected in an
increase of the yield. The effect of nitric acid, Figure 5a, and
sulfuric acid, Figure 5b, are qualitatively similar to each other.
However, according to the model prediction, the effect of an
where mAr and mmix are the masses of the aromatic substrate
and mixed acid, respectively. Assuming a ratio mAr/mmix of 0.5
(a common choice in industrial applications)43 the adiabatic
temperature rise is estimated to be ΔTad ≈ 380 K. Under
adiabatic condition, even adopting lower mAr/mmix ratios, the
increase in temperature can trigger side reactions (dinitration,
a
Table 1. Detailed Experimental Conditions Adopted for Microreactor Runs
run
xn
xs
T (°C)
C0 (M)
Qmix (mL·min−1
)
Qorg (μL·min−1
2.317
)
V (mL)
reported in
Figure 4a
1
0.197
0.197
0.199
0.199
0.250
0.450
0.250
0.350
0.350
0.300
0.134
0.134
0.130
0.130
0.130
0.130
0.130
0.200
0.080
0.347
0.347
0.348
0.348
0.450
0.450
0.350
0.350
0.450
0.400
0.350
0.350
0.318
0.318
0.450
0.270
0.320
0.320
0.320
6
6
0.05
0.05
0.2
0.45
1.5, 3.0, 4.5
1.5, 3.0
1.5
2
1.50
7.724
Figure 4a
Figure 4b
Figure 4b
Figure 5a
Figure 5a
Figure 5b
Figure 5b
Figure 5a-b, Figure 6
not shown
Figure 7a
Figure 7a
Figure 7b
Figure 7b
Figure 8a
Figure 8a
Figure 8a
Figure 8b
Figure 8b
3
25
5.00
103.0
4
25
0.2
1.50
30.90
1.5, 3.0, 4.5
1.5
5
50, 60, 68
68
0.05
0.05
0.05
0.05
0.05, 0.5
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
9.99
51.44
6
9.99
51.44
1.5
7
50, 60, 68
57, 65
68
9.99
51.44
1.5
8
9.99
51.44
1.5
9
9.99
51.44, 514.4
51.44
1.5, 3.0
1.5, 3.0
1.5, 3.0, 4.5
3.0
10
11
12
13
14
15
16
17
18
19
52
9.99
59
3.00
15.45
59
7.50
38.62
68
5.63
28.97
3.0, 4.5
1.5
68
9.99
51.44
45, 60
68
3.00, 7.5
1.67
15.45, 38.62
8.599
3.0
3.0
45, 55
45, 55, 68
55, 68
0.20, 0.75
1.029, 3.862
4.5
1.50, 3.00, 9.99
0.28, 1.23
7.724, 15.45, 51.44
1.448, 6.333
3.0
4.5
a
xn = nitric acid molar fraction; xs = sulfuric acid molar fraction; C0 = benzyl alcohol initial concentration; Qmix = nitrating mixture flow rate; Qorg
benzyl alcohol flow rate; V = microreactor volume.
=
D
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 5. Percentage yield of 2-nitrobenzaldehyde. (a) The effect of nitric acid at fixed xs (sulfuric acid molar fraction) = 0.450; (b) the effect of
sulfuric acid at fixed xn (nitric acid molar fraction) = 0.350. Model prediction (continuous lines); experimental data (symbols); theoretical limit by
model predictions (dashed lines).
Figure 6. Concentration vs residence time in the neighborhood of the optimum point. xn (nitric acid molar fraction) = 0.350; xs (sulfuric acid molar
■
●
▲
fraction) = 0.450; T = 68 °C; calculated (lines) and experimental data (symbols). ( ) Benzaldehyde; ( ) 2-nitrobenzaldehyde; ( ) 3-
nitrobenzaldehyde. (a) C0 = 0.05 M. (b) C0 = 0.5 M. (c) zoom in, C0 = 0.5 M.
increase in acids concentration in the reported ranges does not
seem to significantly affect the yield at temperatures below 45
°C and for molar fractions higher than 0.350 and 0.450 for
nitric acid (xn) and sulfuric acid (xs), respectively (limit dashed
line). For this reason, the experimental runs were carried out
under these conditions: xn = 0.350; xs = 0.450; T = 68 °C.
The experimental results and the comparison with the model
prediction are reported in Figure 6. It is worth noting that the
reaction is very fast and a complete conversion is achieved in
less than 2 ms according to the model simulation shown in
Figure 6c. However, the lowest residence time attainable in the
microreactor was about 9 s. Under these conditions, the side
reactions of oxidation were too slow to affect the yield and no
differences were measured in the samples collected with
residence times up to about 18 s (Figure 6a,b). No significant
variations in the yield were recorded increasing the initial
concentration of the organic substrate from 0.05 to 0.5 M,
Figure 6a,b. However, for the highest concentration a
significant volume of the gas phase formed was evident through
the reactor glass. Even though bubbles were efficiently carried
by the liquid stream, the presence of the gas phase will
necessarily affect the residence time distribution. In this specific
case this did not affect the yield. In fact, as shown in Figures
6a,b, the yields of the two isomers are constant at different
residence times. The average experimental 2-nitrobenzaldehyde
percentage yield was 41.6% while the space-time yields were
0.33 and 3.29 g·L−1·s−1 for the runs reported in Figure 6a,b,
respectively.
It is important to report that a relatively high percentage
yield of 38.9% was measured under milder conditions (i.e., T =
52 °C; xn = 0.300; xs = 0.400) (data not shown).
Furthermore, the MATLAB fmincon optimization routine
was used to maximize the yield of 3-nitrobenzaldehyde yield. In
Figure 7 are reported the results and the prediction for the best
measured percentage yield of 93% and 96%, corresponding to
space-time yields of 0.078 and 0.12 g·L−1·s−1, respectively.
A similar study of the influence of nitric acid, sulfuric acid,
and temperature on the yield of 3-nitrobenzaldehyde was
carried out. In Figure 8 the maximum predicted obtainable
yields of the meta-isomer are plotted at varying nitric acid
sulfuric acid molar fractions and different temperatures. At the
lowest temperatures and acid concentrations the significant
E
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
the yield, and an optimum nitric acid molar fraction exists at
fixed temperature and sulfuric acid molar fraction, Figure 8c.
This is in accord with the previous investigations36,37 that
pointed out an increase in 2-nitrobenzaldehyde yield (and,
hence, a decrease in 3-nitrobenzaldehyde yield) for the nitric
acid molar fractions above 0.14, Figures 8b,c. The effect of an
increase in either nitric or sulfuric acid molar fraction in the
neighborhood of the optimum point are qualitatively similar
(but quantitatively different) as shown by the comparison in
Figure 8a,b. It is important to stress that the decrease in
temperature and acid concentration results in the increase in
the residence time necessary to achieve the maximum yield by 2
orders of magnitude. In fact, a decrease in the yield is often
associated with an increase in the residence time to achieve it.
Further experiments carried under the same experimental
conditions adopted for the runs of Figure 7b, but with higher
concentrations of 3-nitrobenzaldehyde (up to 0.5 M) resulted
in formation of liquid−liquid heterogeneous flow. In fact, even
though the nitrated products are still soluble in the acidic
aqueous homogeneous phase, the formed benzaldehyde is
poorly soluble under less acidic conditions such that the
amount of converted benzaldehyde is limited by the solubility
of the organic in the aqueous phase, in which the reaction takes
place. As a result, a significant amount of unconverted
benzaldehyde was collected at the outlet of the reactor at
residence times longer than 1 min. The same experiments were
run again starting from benzaldehyde instead of benzyl alcohol
to rule out the possibility that this could be entirely ascribed to
the significant gas formation during the instantaneous oxidation
reaction of benzyl alcohol (data not shown). However, for
these experimental runs the average measured percentage
selectivity for 3-nitrobenzaldehyde was 96.3%, in agreement
with the one reported for the homogeneous phase experiments
(selectivity = yield for a complete conversion). The measured
selectivities are reported in Table 2.
Figure 7. Concentration vs residence time. Calculated (lines) and
■
●
experimental data (symbols). ( ) Benzaldehyde; ( ) 2-nitro-
▲
benzaldehyde; ( ) 3-nitrobenzaldehyde. (a) xn = 0.134; xs = 0.350;
T = 59 °C; C0 = 0.05 M. (b) xn = 0.130; xs = 0.318; T = 68 °C; C0 =
0.05 M.
occurrence of the oxidation of 3-nitrobenzaldehyde to 3-
nitrobenzoic acid reduces the yield. As a result, the depend-
ences at low temperatures are significantly different from the
case of 2-nitrobenzaldehyde, Figure 5, and no limit linearity can
be observed. Moreover, unlike the ortho-isomer, an increase in
the acid concentration does not always result in an increase in
Figure 8. Percentage yield of 3-nitrobenzaldehyde in the neighborhood of the optimum point. (a) The effect of sulfuric acid at fixed xn = 0.130. (b)
The effect of nitric acid at fixed xs = 0.320. (c) The effect of nitric acid at fixed xs = 0.320 and T = 68 °C.
F
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Table 2. Measured 3-Nitrobenzaldehyde Selectivities under Heterogeneous Conditions
run
1
2
3
4
5
6
7
8
9
10
average
96.3
selectivity (%)
95.5
92.4
95.1
97.9
94.5
97.4
97.2
97.8
97.2
97.7
a
xn = 0.130; xs = 0.318; T = 68 °C; C0: 0.3−0.5 M.
(11) Alvarez-Rodriguez, N. V.; Dos Santos, A.; El Kaim, L.; Gamez-
Montano, R. Synlett 2015, 26, 2253−2256.
CONCLUSIONS
■
In this article the production of 2- and 3-nitrobenzaldehydes
from benzyl alcohol by means of mixed acid was studied under
homogeneous conditions. The adoption of a commercial
microreactor enabled us to carry out the process under unusual
conditions of high temperatures and strongly acidic media,
identifying the conditions to maximize the yield of the two
isomers. The latter are significantly higher than the currently
accessible through the direct nitration of benzaldehyde in
traditional refrigerated batch reactors. The adoption of more
acidic media and higher temperature is also convenient to
increase the solubility of the organics in the mixtures. However,
a significant increase in the organic substrate concentration can
easily lead to the formation of a biphasic system, affecting the
residence time necessary to achieve complete conversion.
When adopting less acidic mixtures this could affect the yields
because of the significant occurrence of the undesired oxidation
product, mainly 2- and 3-nitrobenzoic acids. Further study of
the solubilities of the organics, namely benzaldehyde, nitro-
benzaldehydes, and nitrobenzoic acid isomers, in the nitrating
mixtures is in progress. Despite the promising results, more
effort must be made in the future to investigate the possibility
of safely carrying out the process under heterogeneous
conditions and to integrate the current kinetic results in a
comprehensive model taking into account the occurrence of
demixing.
(12) Cai, J.; Liu, L.; Hong, K. H.; Wang, P.; Li, L.; Cao, M.; Sun, C.;
Wu, X.; Zong, X.; Chen, J.; Ji, M. Bioorg. Med. Chem. 2015, 23, 657−
667.
(13) Bruhne, F. Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-
VCH: New York, 2003.
(14) Joshi, S. R.; Kataria, K. L.; Sawant, S. B.; Joshi, J. B. Ind. Eng.
Chem. Res. 2005, 44, 325−333.
(15) Veser, G. Chem. Eng. Sci. 2001, 56, 1265−1273.
(16) Janicke, M. T.; Kestenbaum, H.; Hagendorf, U.; Schuth, F.;
Fichtner, M.; Schubert, K. J. Catal. 2000, 191, 282−293.
(17) Dittmeyer, R.; Grunwaldt, J. D.; Pashkova, A. Catal. Today
2015, 248, 149−159.
(18) Roduner, E.; Kaim, W.; Sarkar, B.; Urlacher, V. B.; Pleiss, J.;
Glaser, R.; Einicke, W. D.; Sprenger, G. A.; Beifuß, U.; Klemm, E.;
Liebner, C.; Hieronymus, H.; Hsu, S. F.; Plietker, B.; Laschat, S.
ChemCatChem 2013, 5, 82−112.
(19) Bavykin, D. V.; Lapkin, A. A.; Kolaczkowski, S. T.; Plucinski, P.
K. Appl. Catal., A 2005, 288, 175−184.
(20) Zakrzewski, J.; Smalley, A. P.; Kabeshov, M. A.; Gaunt, M. J.;
Lapkin, A. A. Angew. Chem., Int. Ed. 2016, 55, 8878−8883.
(21) Gutmann, B.; Roduit, J. P.; Roberge, D.; Kappe, C. O. Angew.
Chem., Int. Ed. 2010, 49, 7101−7105.
(22) Palde, P. B.; Jamison, T. F. Angew. Chem., Int. Ed. 2011, 50,
3525−3528.
(23) Zhang, X.; Stefanick, S.; Villani, F. J. Org. Process Res. Dev. 2004,
8, 455−460.
(24) Narkevich, I. A.; Tarasov, I. N.; Golant, Z. M.; Alekhin, A. V.
Pharm. Chem. J. 2016, 49, 760−764.
AUTHOR INFORMATION
(25) Porta, R.; Benaglia, M.; Puglisi, A. Org. Process Res. Dev. 2016,
20, 2−25.
■
Corresponding Authors
*E-mail: danilo.russo3@unina.it.
*E-mail: aal35@cam.ac.uk.
(26) Kulkarni, A. A.; Nivangune, N. T.; Kalyani, V. S.; Joshi, R. A.;
Joshi, R. R. Org. Process Res. Dev. 2008, 12, 995−1000.
(27) Kulkarni, A. A. Beilstein J. Org. Chem. 2014, 10, 405−424.
(28) Lewin, J. J.; Choi, E. J.; Ling, G. Am. J. Health-Syst. Pharm. 2016,
73 (2), 45−54.
ORCID
Danilo Russo: 0000-0003-1809-7309
Notes
The authors declare no competing financial interest.
(29) Gage, J. R.; Guo, X.; Tao, J.; Zheng, C. Org. Process Res. Dev.
2012, 16, 930−933.
(30) Lowe, H.; Ehrfeld, W. Electrochim. Acta 1999, 44, 3679−3689.
(31) Wang, K.; Lu, Y. C.; Xia, Y.; Shao, H. W.; Luo, G. S. Chem. Eng.
J. 2011, 169, 290−298.
(32) Rossetti, I.; Compagnoni, M. Chem. Eng. J. 2016, 296, 56−70.
(33) Jensen, K. F. Chem. Eng. Sci. 2001, 56, 293−303.
(34) Jahnisch, K.; Hessel, V.; Lowe, H.; Baerns, M. Angew. Chem., Int.
Ed. 2004, 43, 406−446.
(35) Loebbecke, S.; Antes, J.; Ferstl, W.; Boskovic, D.; Tuercke, T.;
Schwarzer, M.; Krause, H. IChemE Symposium Series; Institution of
Chemical Engineers, 2007; Vol. 153.
REFERENCES
■
(1) Kulkarni, A. A.; Kalyani, V. S.; Joshi, R. A.; Joshi, R. R. Org.
Process Res. Dev. 2009, 13 (5), 999−1002.
(2) Russo, D.; Onotri, L.; Marotta, R.; Andreozzi, R.; Di Somma, I.
Chem. Eng. J. 2017, 307, 1076−1083.
(3) Boratynski, P. J.; Nowak, A. E.; Skarzewski, J. Synthesis 2015, 47,
3797−3804.
(4) Bartee, D.; Morris, F.; Al-khouja, A.; Freel Meyers, C. L.
ChemBioChem 2015, 16, 1771−1781.
(36) Sainz-Diaz, C. I. Monatsh. Chem. 2002, 133, 9−22.
(37) Di Somma, I.; Marotta, R.; Andreozzi, R.; Caprio, V. Chem. Eng.
Trans. 2014, 36, 181−186.
(5) Wang, L.; Tai, X. Crystals 2016, 6, 57.
(6) Mahdavi, M.; Lotfi, V.; Saeedi, M.; Kianmehr, E.; Shafiee, A. Mol.
Diversity 2016, 20, 677−685.
(38) Ross, D. S.; Gu, C. L.; Hum, G. P.; Malhotra, R. Int. J. Chem.
Kinet. 1986, 18, 1277−1288.
(7) Higaki, T.; Kitazawa, H.; Yamazoe, S.; Tsukuda, T. Nanoscale
2016, 8, 11371−11374.
(39) Gasparrini, F.; Giovannoli, M.; Misiti, D.; Natile, G.; Palmieri,
G. Synth. Commun. 1988, 18, 69.
(8) Wang, X.; Xie, X.; Cai, Y.; Yang, X.; Li, J.; Li, Y.; Chen, W.; He,
M. Molecules 2016, 21, 340.
(40) Di Somma, I.; Russo, D.; Andreozzi, R.; Marotta, R.; Guido, S.
Chem. Eng. J. 2017, 308, 738−744.
(41) Medard, L. A. Accidental explosions; Ellis Horwood Limited:
Chichester West Sussex, P019 1EB, England, 1989.
(42) Grewer, T., Rogers, R. Thermal hazards of chemical reactions,
second ed.; Vol. 4: Industrial Safety Series, Elsevier, 2015.
(9) Chen, Q.; Hou, Y.; Hou, G.; Li, N.; Cong, W.; Zhao, F.; Li, H.;
Wang, C.; Sun, J. J. Chem. Res. 2016, 40, 400−403.
(10) Neelakandan, K.; Manikandan, H.; Prabhakaran, B.; Santosha,
N.; Chaudhari, A.; Kulkarni, M.; Mannathusamy, G.; Titirmare, S. Der
Pharmacia Sinica 2014, 5 (1), 11−17.
G
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(43) Olah, G. A.; Malhotra, R.; Narang, S. C. Nitration methods and
mechanisms; VCH: New York, 1989.
H
DOI: 10.1021/acs.oprd.6b00426
Org. Process Res. Dev. XXXX, XXX, XXX−XXX