Continuous flow nitration of salicylic acid

Article abstract of DOI:10.1021/op800112u

Continuous flow nitration of salicylic acid using HNO₃/AcOH was studied in the SS316 tubular microreactor. At specific reaction conditions, complete conversion of the reactant was achieved in less than 7 min. It yielded only mononitro derivatives with a higher selectivity of 5-nitrosalicylic acid. Presence of the lower amount of acetic acid in the reaction mixture was seen to be detrimental, leading to precipitation of the desired product (5-nitrosalicylic acid). Reaction at higher temperatures yielded byproducts. The continuous mode operation using the system comprising the microdevices was demonstrated for 2 h with consistent composition at the outlet

Full text of DOI:10.1021/op800112u

Organic Process Research & Development 2008, 12, 995–1000
Continuous Flow Nitration of Salicylic Acid
Amol A. Kulkarni,*,† Nayana T. Nivangune,† Vishwanath S. Kalyani,† Ramesh A. Joshi,‡ and Rohini R. Joshi‡
Chemical Engineering DiVision (CEPD), National Chemical Laboratory, Pune - 411 008, India, and DiVision of Organic
Chemistry, National Chemical Laboratory, Pune - 411 008, India
Abstract:
nitration, chlorination, biphasic Heck reaction, Wittig reaction,
Kumada-Suzuki coupling, etc. The high heat transfer area helps
to achieve a better control on the temperature variation in the
microreactor. This further helps to maintain the rates of reaction
in specific ranges and thus avoids byproduct formation. Also,
the small length scales help to achieve faster mixing, thereby
reducing the possibility of byproduct formation in fast reactions.
These advantages of microreactors are now known, and as a
result, they have been used for the continuous flow synthesis.1-8,10
A detailed discussion on the advantages of continuous flow
synthesis and a list of different classes of reactions that can be
carried out in continuous flow mode can be seen in a recent
review by Wiles and Watts.11 Also, since the concept of
continuous flow synthesis is feasible in the devices having
dimensions that do not offer any transport limitations, it is not
always necessary to carry out the flow synthesis using the
microdevices, and dimensions slightly greater than a millimeter
may also be feasible.
Nitration of aromatic compounds is usually exothermic and
fast, and in the presence of excess nitrating agent or at higher
temperatures, it can undergo polynitration.12,13 In many aromatic
nitrations to avoid the effect of high temperatures that may be
generated in the presence of excess nitrating agent, convention-
ally, the nitrating agent is slowly added to the aromatic substrate
(which can be in the form of solid or liquid). To overcome this
situation due to poor heat transfer in the system, in the past
few years, the advantage of high heat transfer area and better
mixing in microreactors is exploited to check the feasibility of
nitration of aromatic compounds. Some of the important initial
studies include the following: (i) Burns and Ramshaw14 have
reported the nitration of benzene and toluene using mixed acids
in a microreactor. In their experiments, the two-phase slugs were
generated using a “T”, and the reaction was studied in SS tubes
(127, 178, and 254 µm i.d., 1/16 in. o.d.) and PTFE tubes
(127-300 µm i.d., 1/16 in. o.d.). For all the experiments, acid
to organic flow ratio was maintained at 10.5, while the benzene
to nitric acid mole ratio was in the range of 0.8-1.8. With
similar experiments of toluene nitration in a 150 µm PTFE tube,
these authors have shown that higher concentration of H2SO4
pushes the reaction in mass transfer limited regime and smaller
diameter tubes yield better initial nitration rates. Also, higher
Continuous flow nitration of salicylic acid using HNO3/AcOH was
studied in the SS316 tubular microreactor. At specific reaction
conditions, complete conversion of the reactant was achieved in
less than 7 min. It yielded only mononitro derivatives with a higher
selectivity of 5-nitrosalicylic acid. Presence of the lower amount
of acetic acid in the reaction mixture was seen to be detrimental,
leading to precipitation of the desired product (5-nitrosalicylic
acid). Reaction at higher temperatures yielded byproducts. The
continuous mode operation using the system comprising the
microdevices was demonstrated for 2 h with consistent composition
at the outlet.
Introduction
Continuous flow synthesis using microreactors is an upcom-
ing approach in the synthesis of key pharmaceutical intermedi-
ates and fine chemicals where either the reactions are highly
exothermic or there are situations where the selectivity of the
product is an issue.1-8 It allows high throughput screening of
catalysts and also of reactions. The subject of microreaction
technology has achieved significant attention in the laboratory-
scale research and development for the design of new devices
towards specific applications, viz. for carrying out specific
reactions, heat removal, attaining narrow residence time dis-
tribution, continuous synthesis of nanoparticles, and also in
checking their feasibility for multistep processes.9 In the past
few years, there have been reported in the literature, several
reactions that have better yield when carried out in microreactors
than when carried out in the conventional batch mode opera-
tion.10 Typical classes of reactions which are studied using
microreactors include fluorination, oxidation, hydrogenation,
* Corresponding author. Telephone: +91-20-25902153. Fax: 91-20-25902621.
E-mail: aa.kulkarni@ncl.res.in.
† Chemical Engineering Division.
‡ Division of Organic Chemistry.
(1) Jensen, K. F. Chem. Eng. Sci. 2001, 56, 293.
(2) Ehrfeld, W.; Hessel, V.; Lo¨we, H. Microreactors; Wiley-VCH:
Weinheim, 2000.
(3) Hessel, V.; Hardt, S.; Lo¨we, H. Chemical Micro Process Engineering
- Fundamentals, Modelling and Reactions; Wiley-VCH: Weinheim,
2004.
(4) Hessel, V.; Lo¨we, H. Chem. Eng. Technol. 2003, 26 (1), 13.
(5) Ja¨hnisch, K.; Hessel, V.; Lo¨we, H.; Baerns, M. Angew. Chem., Int.
Ed. 2004, 43, 406.
(10) Pennemann, H.; Hessel, V.; Löwe, H. Chem. Eng. Sci. 2004, 59, 4789–
4794.
(6) Pennemann, H.; Watts, P.; Haswell, H. J.; Hessel, V.; Lowe, H. Org.
Process Res. DeV. 2004, 8, 422.
(11) Wiles, C.; Watts, P. Eur. J. Org. Chem. 2008, 10, 1655–1671.
(12) Olah, G. A.; Malhotra, R.; Narang, S. C. Organic Nitro Chemistry
Series: Nitration Methods and Mechanisms; VCH: New York, 1989.
(13) Hoggett, J. G.; Monodie, R. B.; Penton, J. R.; Schofield, K. Nitration
and Aromatic ReactiVity; Cambridge at the University Press: London,
1971.
(7) Roberje, D. M.; Dukrey, L.; Bieler, M. Chem. Eng. Technol. 2005,
28 (3), 318.
(8) Geyer, K.; Code´e, J. D. C.; Seeberger, P. H. Chem. Eur. J. 2006, 12,
8434–8442.
(9) Sahoo, H. R.; Kralj, J. G.; Jensen, K. F. Angew. Chem., Int. Ed. 2007,
46, 57045708\.
(14) Burns, J. R.; Ramshaw, C. Chem. Eng. Commun. 2002, 189, 1611.
10.1021/op800112u CCC: $40.75
Published on Web 08/23/2008
2008 American Chemical Society
Vol. 12, No. 5, 2008 / Organic Process Research & Development
•
995

Scheme 1a
acid to organic ratio yielded lower rates while higher percent-
ages of H2SO4 yielded higher initial rates of nitration indepen-
dent of flow velocities and hence higher amounts of DNT. (ii)
Ducry et al.15 have reported the nitration of phenol in a glass
microreactor with a channel width <0.5 mm and 2.0 mL
internal volume. They have shown that the nitration reaction
in a microreactor yields a better fraction of the mononitrated
products and a reduction in the amounts of polymerized
products. (iii) Nitration of single-ring aromatic compounds in
a system comprising a Y-mixer followed by a PTFE capillary
(0.5-1 mm i.d. and 1-10 m length) immersed in a thermostat
are reported by Dummann et al.16 (iv) Panke et al.17 have used
the CYTOS microreaction system for the nitration of pyridine-
N-oxide (T ) 120 °C, yield of 78% vs 72% in the flask
experiment) and 4-nitropyridine-N-oxide. (iv) Antes and
co-workers18-20 have developed and analyzed continuous
processes for the nitration of aromatics. For the case of strongly
exothermic nitration of naphthalene using N2O5, these authors
have reported that, while the nitration in conventional batch
operation requires temperatures from -50 to -20 °C, the
continuous flow process in microreactors can be carried out at
30 °C with a flow rate of 1 mL/min at a residence time of 3 s
with large yields of mononitro derivatives with small amounts
of the dinitro compounds. They have also carried out the
nitration of urea derivatives in microreactors. The authors have
reported a new two-step route to the dinitrated products using
thiourea derivatives as the starting compounds was carried out
in the batch process and also in a microreactor. The nitration
of corresponding thioureas in a two-step procedure yielded the
mononitro urea derivative with nearly 100% selectivity. Thus,
although a few reports on the nitration in microreactors are
available in the literature, one of the reasons that might have
prohibited carrying out several exothermic nitration reactions
using microreactors is the possibility of precipitation of one of
the products. Since many of the pharmaceutical intermediates
are in the particulate form, and the reactions are exothermic
and also have issues of selectivity, it is necessary to identify
the avenues to exploit the advantages of microreactors for such
cases.
a AS, salicylic acid; 5NSA, 5-nitrosalicylic acid; 2-NP, 2-nitrophenol; 3NSA,
3-nitrosalicylic acid.
one,21 which upon its further reduction yields 5-amino salicylic
acid (also known as mesalazine22). 5-ASA finds applications
in the treatment of ulcerative colitis. As of 2001 for the
pharmaceutical industry itself, the worldwide requirement of
5-ASA is about 300 tons/year. Thus, if proven efficient, a
technology for the synthesis of 5NSA based on continuous flow
chemistry using microreactors can be extended for further
continuous reduction to yield the desired 5-ASA. In this work,
we deal with the first step only, and we bring out several of
our observations that would be useful for the researchers who
would like to work on analogous systems. The second step of
continuous reduction of 5NSA will be discussed elsewhere. The
literature findings for the nitration of salicylic acid are based
on batch experiments23-25 and also one patent on continuous
nitration26 in the glass microreactor. In the rest of the manuscript,
we bring out the new findings from our studies and also
compare them with the relevant literature.
Experimental Section
The experiments were carried out in batch mode as well as
in continuous mode. The initial experiments in batch mode (20
°C) were mainly aimed to confirm our approach and the analysis
method based on the literature.23-25 In all the experiments, the
mixture of acetic acid and nitric acid was used as the nitrating
agent for the nitration of salicylic acid (all the chemicals were
of synthesis grade and procured from Merck). For the batch
experiments, the initial reactant composition was mole ratio SA:
HNO3 ) 1:5 and SA:AcOH ) 1:3. The reaction was carried
out in a 100 mL glass reactor (heat transfer area per unit volume
≈ 106.6 m2/m3) with a jacket for circulating cooling water.
Nitric acid was slowly added to the mixture of salicylic acid
and acetic acid with constant stirring. The nitric acid was
With the above discussion in background, in the present
work, we have demonstrated the nitration of salicylic acid to
selectively yield mononitro salicylic acid (5-nitro salicylic acid,
5NSA, and 3-nitro salicylic acid, 3NSA) (Scheme 1). Among
the synthesized mononitro products, isomer 5NSA is the desired
(15) Ducry, L.; Roberge, D. M. Angew. Chem., Int. Ed. 2005, 44, 7972.
(16) Dummann, G.; Quittmann, U.; Gro¨schel, L.; Agar, D. W.; Wo¨rz, O.;
Morgenschweis, K. Catal. Today 2003, 79-80, 433–439.
(17) Panke, G.; Schwalbe, T.; Stirner, W.; Taghavi-Moghadam, S.; Wille,
G. Synthesis 2003, 18, 2827.
(18) Antes, J.; Tuercke, T.; Marioth, E.; Schmid, K.; Krause, H.; Loebbecke,
S. Proceedings of the Fourth International Conference on Microre-
action Technology: IMRET 4 ; 2000; p 194; March, 2000, Atlanta,
U.S.A.
(21) Mayo, D. W.; Pike, R. M.; Trumper, P. K. Preparation of 5-Nitro-
salicylic Acid, 3rd ed.; Microscale Organic Laboratory, John Wiley
and Sons, Inc.: New York, 1994; pp 383-384.
(22) Zaiyou, T.; Tiansui, L.; Gengxin, Z.; Zhufen, L.; Xiaobin, X.; Lianfang,
X.; Bin, L. Guangzhou Huagong 2003, 31, 37–38.
(19) Loebbecke, S.; Antes, J.; Tuercke, T.; Marioth, E.; Schmid, K.; Krause,
H. 31st International Annual Conference: Energetic Materials-Analysis,
Diagnostics and Testing, Karlsruhe, 2000.
(20) Antes, J.; Tuercke, T.; Marioth, E.; Lechner, F.; Scholz, M.; Schnoerer,
F.; Krause, H. H.; Loebbecke, S. Proceedings of the Fifth International
Conference on Microreaction Technology: IMRET 5; Matlosz, M.,
Ehrfield, W., Baselt, J. P. Eds.; Springer: Berlin, 2002; p 446; May,
2001, Strasbourg.
(23) Andreozzi, R.; Caprio, V.; Di Somma, I.; Sanchirico, R. J. Hazard.
Mater. 2006, 134, 1–7.
(24) Andreozzi, R.; Canterino, M.; Caprio, V.; Di Somma, I.; Sanchirico,
R. J. Hazard. Mater. 2006, 138, 452–458.
(25) Andreozzi, R.; Canterino, M.; Caprio, V.; Di Somma, I.; Sanchirico,
R. Org. Process. Res. DeV. 2006, 10, 1199–1204.
(26) World Patent, WO2007087816A1, 2007.
996
•
Vol. 12, No. 5, 2008 / Organic Process Research & Development

Figure 1. Proton spectra for (A) 5NSA and (B) a mixture of 3NSA and 5NSA.
completely added within 6 min after starting the agitation. The
samples were withdrawn at different time intervals and analyzed
using HPLC (Agilent model 1100 II, equipped with a UV-vis
detector and a Phenomenex Synergi 4 µ polar RP/80A column).
To the withdrawn samples, a solution of urea in methanol was
added for denitrification.27 The mobile phase used for analysis
consisted of 80% buffer (vol %: CH3OH 5%, H3PO4 0.4%, H2O
94.6%) and 20% of acetonitrile. For all analyses HPLC grade
reagents were used. The HPLC analysis was carried out at three
different wavelengths, viz. 240, 280, 350 nm; the column
temperature was maintained at 25 °C; and the flow rate was
set at 1 mL/min. As a part of the analysis procedure, mononitro
derivatives were obtained (A solution of 2.5 g of salicylic acid
in 40 mL of acetic acid was heated at 50 °C. Eleven milliliters
of concentrated HNO3 was slowly added to the above solution
and stirred for 10 min. The reaction was quenched by pouring
the reaction mixture on 50 g of crushed ice.) and used for
calibration of the HPLC response. Upon quenching of the
reaction on ice, a yellow solid was obtained. This solid was
filtered and washed with cold water. The mother liquor was
kept at 0-5 °C for over 12 h, and the second crop was isolated
after further filtration. The solid products isolated in these two
steps were analyzed using NMR (NMR spectra were recorded
analysis. On confirming the entire procedure and the analysis
method, we extended it for the continuous flow experiments.
For the continuous flow experiments, typically the experi-
mental setup involved two HPLC pumps (Laboratory Alliance,
U.S.A.) followed by a micromixer, which was then connected
to a 12 m long stainless steel (SS316) tube (1/16 in. o.d. and
1.38 mm i.d.). The SS tube was immersed in a thermostat
(Julabo - ME12, Germany), and the samples were collected at
the outlet of the tube. The residence time was varied by
changing the flow rates. Experiments were carried out at
different conditions as shown in Table 1. The samples were
collected in glass vials immersed in an ice bath (to quench the
temperature of the reacting mixture and thereby avoid any
further reaction), and after getting volume sufficient for analysis,
the unreacted nitric acid was removed by adding urea in
methanol solution. The supernatant was subjected to analysis
after further dilution. It is necessary to note that the results from
the HPLC analysis were based on calibration of the column
response for different concentrations of the reactants and the
products. A schematic of the continuous flow setup is shown
in Figure 2.
Results and Discussions
1
in D2O/NaOD. The H measurements were carried out on a
As mentioned previously, in the case of continuous experi-
ments, the outlets from the two pumps were connected to a
micromixer. Initially the experiments were carried out using
HPIMM interdigital micromixer at 20 °C. It is worth mentioning
that, because of the lower solubility of 5NSA in the aqueous
medium, upon mixing and subsequent reaction the product
precipitated. This resulted in the blocking of the HPIMM
micromixer and also a very high pressure drop. In order to avoid
such a situation, we used a simple T-mixer (800 µm i.d.) for
Bruker AV 400 NMR spectrometer). The proton spectrum
(shown in Figure 1) confirmed the first crop to be 5NSA (also
confirmed by measuring its melting point ∼230-232 °C),and
the second crop of nitrated product was found to be a mixture
of 3- and 5-mononitro salicylic acid (1:1). The samples from
these two crops were used for the calibration needed for HPLC
(27) Kondo, Y. J. Radioanal. Nucl. Chem. 1999, 242 (2), 515–526.
Vol. 12, No. 5, 2008 / Organic Process Research & Development
•
997

Table 1. Experimental conditions
authors
parameters
this work
Andreozzi et al.25
ref 26
mode of operation
system
batch and continuous
batch: 40 mL volume in a
glass reactor
batch
batch: 20 mL glass
reactor
batch and continuous
batch: 100 mL glass
reactor
continuous: tubular reactor
in SS316 (1.58 mm o.d., 1.38 mm i.d.
and length ) 12 m)
20, 30, 40, 50
continuous: Corning Glass
microreactor (Corning Inc.)
temperature (°C)
SA:AcOH
15, 20, 25
1:2.43
45, 75
1:5-1:16
1:5-1:20
1: 1.1-1:2
10-20 s
SA:HNO3
1:1-1:10
1:5
batch time or
residence time
6-25 min
15-40 min
mixing of the fluids. It was observed that, although replacing
the HPIMM micromixer with the T-mixer helped in eliminating
the blockage in the mixer itself, for the SA:AcOH mole ratio
in the range of 1:3 to 1:9, the product clogged the tube
completely, and practically nothing came out of the tube. This
was also observed for different orientations of the coiled tube
(vertical, horizontal, etc.) and also for the tube of larger diameter
(3.175 mm o.d. and 2.77 mm i.d.) even in a serpentine
configuration oriented horizontally. On removing the clogged
compound, it was confirmed to be 5NSA. In order to avoid
such situations, the mole ratio of SA:AcOH was taken as 1:16,
and the experiments were carried out without any blockage
problems in the entire setup. In all these experiments, the mole
ratio of SA:HNO3 was maintained at 1:5. The observed outlet
yields at 20 °C at different residence times (for continuous
mode) or sampling times (for batch mode) are shown in Figure
3. The results from the batch experiments (mainly carried out
to verify the analysis approach quantitatively) were compared
with the reported data in Andreozzi et al.25 at identical conditions
and were seen to agree well. It can be seen that the percent
yield of 5NSA observed in the continuous experiments in the
tube of 1.38 mm i.d. was better than that of the batch
experiments. It needs to be noted that, although in the batch as
well as the continuous experiments the SA:HNO3 mole ratio
was maintained the same, the continuous flow experiments
contained very high amounts of AcOH, resulting in lowering
the overall concentration of the reactants. Importantly, the
continuous flow experiments yielded only the mono-NSA as
products, while in the batch experiment we observed 3,5-
dinitrosalicylic acid as a byproduct along with the mono-NSA.
It needs to be mentioned that the measured area available for
heat transfer per unit volume of fluid was 3318.63 m2/m3, which
was sufficiently high to maintain the constant temperature in
the entire reaction tube. Further, the selectivity of 5NSA, which
was the desired compound was close to 60%. Details of a few
of the continuous flow experiments are given in Table 2. The
residence time in the tubular reactor as well as the value of the
ratio of the outlet mole fraction of the nitrated isomers
corresponds to 100% conversion of the salicylic acid for the
conditions mentioned in Table 2. The literature25 data on the
Figure 3. Comparison of the experimental % yields of 5NSA
and 3NSA obtained from the batch and continuous experiments
at 20 °C and with the data reported by Andreozzi et al.25 The
experimental conditions are identical, except that for the
continuous experiments the acetic acid was taken in excess to
avoid the precipitation of 5NSA in the tube and its subsequent
blocking. ∆ - 5NSA continuous, ]- 5NSA batch, O - 3NSA
continuous, ( - 5NSA batch, 2- 5NSA batch,25 b - NSA batch.25
Table 2. Inlet composition, experimental conditions, and
results for continuous flow experiments
inlet mole ratio
SA: SA:
sr. no. AcOH HNO3 T (°C) time (min)
outlet mole ratio
residence
5NSA:
3NSA
1
2
2
3
5
6
7
1:16
1:16
1:16
1:16
1:16
1:16
1:10
1:5
1:3
1:3
1:3
1:3
1:5
1:3
20
20
30
40
50
50
50
44
44
21
7
1.66
1.62
1.65
1.70
1.78
1.77
1.85
6
4
7
Figure 2. Schematic of the experimental setup.
998
•
Vol. 12, No. 5, 2008 / Organic Process Research & Development

Figure 4. (A) Effect of temperature on the outlet composition of mononitro salicylic acid in the case of 100% conversion and at O
SA:HNO3 ) 1:5, 0 SA:HNO3 ) 1:3. (B) Effect of mole ratio on the outlet composition and the residence time required for complete
conversion at 50C. b - residence time, 0 - outlet composition (mole ratio 5NSA:3NSA). In all these experiments SA:AcOH ) 1:16.
batch experiments show that, irrespective of the different
reaction temperatures, at the complete consumption of SA (for
example, at temperature of 25 °C the complete conversion takes
place in almost 15 min) the maximum selectivity for 5NSA
does not exceed 55%. Even the composition of the mixture at
the end of a batch reaction has been reported to contain
byproducts, viz. 2-NP, 3,5-dinitrosalicylic acid, etc.
positive relationship between selectivity of 5NSA and temper-
atures can be the relatively higher rates of the nitration favoring
5NSA. This is usually found due to higher rate constants for
the favored reaction. The data from the literature21 show that
the pre-exponential factor (k0) for the reaction yielding 5NSA
is 1.83 times higher than that of for the 3NSA, while the
activation energies of these nitrations are of similar magnitude.
As a result with sufficient excess of nitric acid, 5NSA formation
is always favored. Further, at any given reaction temperature,
the rate constants for the consecutive reactions involving 3NSA
are higher than that of the reactions involving 5NSA. As a result,
the possibility of extent of reduction in the concentration of
3NSA by successive nitration or decarboxylation is higher than
that for the 5NSA. This situation even prevails for the lower
SA:AcOH ratio as the 3NSA remains dissolved in the solution
while 5NSA gets precipitated as a result of the competitive
solubility. This overall situation favors increase in the selectivity
of 5NSA with increasing temperature.
In an analogous approach of using continuous flow for
nitration of aromatic compounds, a recent patent26 illustrates
that the reaction time for nitration of salicylic acid can be
reduced to as low as 10-30 s using a glass microreactor at 75
°C with other conditions given in Table 1. In the reported
experiments, although the conversion is not complete, the ratio
of 5NSA:3NSA was in the range of 1.6-1.7 with many
additional byproducts. In comparison to that, we had observed
complete conversion of the reactant (SA) and very high yield
of 5NSA with practically no byproducts as reported in the
literature. One of the main reasons was that the use of a metal
tube (which has much higher thermal conductivity thereby
achieving better heat transfer) for this reaction helped to
maintain almost isothermal conditions in the reactor. The
simulations of this system as a plug flow reactor (not shown
here) clearly indicated that, with the inlet composition and the
inlet temperatures, the extent of variation in the temperature in
the tube was less than 0.5% of the temperature of the fluid
surrounding the tube. The batch experiments carried out at
identified conditions showed the selectivity of 5NSA to be closer
Since no byproducts were observed from the continuous
experiments at 20 °C, further experiments were carried out at
higher temperature (30, 40, and 50 °C) by increasing the flow
rates (Table 2). The analysis showed that for the experiments
at 50 °C, the complete conversion of salicylic acid was observed
in less than 7 min of residence time in the tubular reactor and
also with the higher selectivity for the desired product, we
repeated the experiments with lower mole ratios of SA:HNO3
≈ 1:3 and 1:1.1, and the results were seen to be unchanged. In
order to further reduce the quantity of AcOH such that the
desired product is completely soluble in it, experiments were
carried out at SA:AcOH ) 1:10. The outlet composition at the
completion of reaction showed the complete conversion of
salicylic acid with selectivity value of 64.9% for 5NSA with
no additional byproduct except 3NSA. At constant temperature
and constant mole ratio of SA:AcOH, the selectivity of 5NSA
was seen to increase with the concentration of nitric acid. Also,
the time required for complete consumption of salicylic acid
was seen to decrease with increasing concentration of nitric acid.
In all these cases, at the suitable residence time when the
salicylic acid is completely consumed, the products were seen
to be only the mononitro derivatives of the salicylic acid.
The continuous flow experiments further showed that, at
constant mole ratio of salicylic acid to the nitrating agent, the
selectivity of 5NSA increased with increasing temperature.
Since the data shown in Figure 4A belongs to the complete
conversion of salicylic acid, the higher concentration of nitric
acid was seen to affect both the residence time required to
achieve complete conversion in the reactor tube and also the
selectivity of the desired product. This particular observation
is depicted in Figure 4B. One of the reasons for having a
Vol. 12, No. 5, 2008 / Organic Process Research & Development
•
999

to 55% with complete conversion of the reactant. The LC/MS
analysis of the products showed the existence of a byproduct
3,5-DNSA. Moreover, some amount of SA was seen to remain
unconverted. This was clearly due to poor mixing and heat
transfer that can be achieved in batch mode.
Thus, the continuous flow nitration of salicylic acid in a
SS316 tubular reactor yielded only the mononitro products at
the outlet with higher selectivity for 5NSA. The better con-
ductivity of metal, high heat transfer area, and the specific
residence time helped to practically bypass the byproduct
formation in this process. The experiments at different temper-
atures showed a positive relationship between the selectivity
of 5NSA and temperatures. Also, a higher amount of the
nitrating agent was seen to reduce the reaction time and achieve
higher selectivity for the desired product. Lower amounts of
acetic acid were seen to lead to precipitation of 5NSA in the
tubular reactor, causing blockage. A continuous flow experiment
at 50 °C with 7 min residence time for 2 h yielded consistent
outlet composition. Although the higher amount of acetic acid
is needed to keep the desired product in solution, complete
recycle of acetic acid would help to reduce the acetic acid
inventory. The economics of this continuous process, the
optimization of the operating conditions and the design param-
eters (heat transfer area and heat transfer coefficient) will be
discussed elsewhere with the help of systematic modeling and
simulations for this system.
Further to this, we operated the continuous flow microplant
for the specific conditions (SA:HNO3 ) 1:3, SA:AcOH ) 1:10,
T ) 50 °C) continuously for 2 h. The residence time in the
tube was maintained at 7 min, and samples were taken
intermittently to check the composition and analyzed using
HPLC. The results were seen to be totally consistent and yielded
only the mononitro salicylic acid. These experiments were also
repeated with a higher amount of acetic acid (SA:AcOH ) 1:16)
and were seen to yield identical results. Thus, in order to avoid
the possible precipitation of the desired product in the tube and
still get better selectivity at complete conversion, it was
necessary to work with higher amounts of acetic acid (SA:
AcOH ) 1:10). Although higher amounts of acetic acid are
needed, the separation of acetic acid from the product needs to
be dealt with separately as its recycle would be cost-effective.
With a suitable recycle mechanism, with a suitable numbering-
up strategy (to produce as much as a few kilograms per hour)
it would certainly be possible to use this approach for the
continuous production of mononitro derivatives of salicylic acid.
Further, we would extend the existing experimental system to
continuous selective reduction of 5NSA to 5-amino salicylic
acid. The intricacies involving continuous removal of HNO3,
selective separation of 3NSA from the mixture, feasibility of
reduction in the presence of excess acetic acid, and continuous
three-phase reduction under pressure will be dealt with separately.
Acknowledgment
We acknowledge the financial help from the Consortium
on Microreaction Technology (www.ncl-india.org/cmr/). A.A.K.
also acknowledges the in-house funding from NCL. The
immense help of Dr. R. M. Deshpande and Dr. A. A. Kelkar
in HPLC analysis is gratefully acknowledged.
Received for review May 12, 2008.
OP800112U
1000
•
Vol. 12, No. 5, 2008 / Organic Process Research & Development