The nitro derivatives of aromatic compounds find applications in the manufacture of dyes, active pharmaceutical ingredients (API), pesticides and fine chemicals. One of the most common ways for inserting a nitro group in the organic structure is by nitration using different nitrating agents, the nitrating mixture (HNO3+H2 SO4) being the most practiced. The synthesis of nitro aromatic compounds is usually associated with relatively high heat of reaction (usually of the order of -100 kJ/kmol per nitro group ). Also, since the presence of sulfuric acid largely drives the reaction by generating the nitronium ions, the concentration of sulfuric acid in the nitrating mixture also helps decide the isomer ratio. However, since the sulfuric acid usually does not participate in the reaction, it is not included in the reaction kinetics. Usually, the nitro aromatics are further reduced to obtain the respective amines which are relatively stable. Since the nitration using nitrating agents is usually nonspecific in terms of isomers, the mononitro derivatives primarily include a mixture of different isomers. Since the amines from individual isomers have different applications and are difficult to separate, the nitro derivatives are isolated and then reduced using a variety of reducing agents.
Continuous flow aromatic nitration using microreactors [2–6] is now a well-established fact and a comprehensive review on the topic can be found in Kulkarni . The advantages of miniaturized process devices or microreactors such as high heat transfer area, efficient mixing, better mass transfer rates and precise control on the residence time, etc., help achieve better conversion and much higher selectivity of the mononitro products. Several such examples from the literature on homogeneous and two phase nitration reactions in continuous flow miniaturized devices mainly aim at reducing the byproduct formation and enhancing the product yield [3–5, 8–15]. While the use of microfabricated devices is encouraged for laboratory scale syntheses, simpler approaches such as using simple tubular reactors [3–5, 8], which also offer the required heat transfer area per unit volume of the reacting mass, also help to carry out exothermic reactions. In continuation with our efforts to study the nitration of reactive aromatic substrates [3, 4], the nitration of acetophenone (1) was taken up in the microreactor system. The major product from mononitration is m-nitroacetophenone, which is an important raw material for the commercial production of fine chemicals and active pharmaceutical ingredients. The reduction of m-nitroacetophenone to m-aminoacetophenone, followed by diazotization and hydrolysis gives m-hydroxyacetophenone. The m-hydroxyacetophenone is the starting material in the production of important drugs like rivastigmine, which is used for the treatment of mild to moderate dementia of the Alzheimer’s type and dementia due to Parkinson’s disease [16, 17] and fenoprofen, used for symptomatic relief of rheumatoid arthritis, osteoarthritis and mild to moderate pain . In this paper, we bring out the lab scale approach that can be scaled up to a few kilograms from a single tubular reactor for the synthesis of 3. The system can be easily scaled up to achieve higher capacity. The entire process involves nitration followed by reduction; both of the steps can be made continuous, but separately. The paper also brings out the reasons that do not support a continuous process that comprises both the steps in a continuous manner.
The conventional route for the manufacture of 2 is through the nitration of 1 using a nitrating mixture . The reaction is extremely exothermic and needs to be carried out at 5°C with controlled addition of acetophenone over several hours. Typical production scales in the industry for m-nitroacetophenone in India are of the order of a few tens of tons/month (typically if converted in continuous mode, it comes between 5 and 25 g/min, depending upon the requirement of individual industries). Being a highly exothermic reaction, conventionally, addition of nitrating agent in a semi-batch manner is preferred. The addition time period is crucial because the additions over a longer time lead to the decomposition of raw material giving a poor yield. Also, very efficient stirring is required to achieve good mixing that helps achieve reasonable yields of the desired product. These constraints make the nitration of acetophenone difficult to scale up while retaining the stoichiometric selectivity. In this paper, we focus on the nitration and reduction steps (Scheme 1). In the first step of nitration, the literature reports indicate that second order kinetics apply for this reaction  and the rate constant decreases with increase in the concentration of nitric acid, thus needing an optimal concentration of nitric acid that would help achieve maximum reaction rate. The presence of sulfuric acid is necessary even to activate the organic substrate for nitration reaction, thereby helping to accelerate the reaction and hence sulfuric acid is used in excess.
2 Materials and methods
The experiments were performed in two steps, as shown in Scheme 1. Individual steps were studied separately. Batch, as well as continuous flow experiments were performed to synthesize 3 and were later optimized for the highest quantitative yield at lab scale. Details of the experimental procedure are given as follows.
2.1 Nitration of acetophenone: experimental set-up and experimental procedure
2.1.1 Preparation of solutions
A solution of acetophenone (Loba Chemie, India) in concentrated sulfuric acid (Thomas Baker, India) was prepared in a beaker in an ice-salt bath, at a temperature between -5°C and 0°C. The addition of acetophenone to sulfuric acid was done dropwise, such that the mixture temperature did not rise above 2°C. Because of the low melting point of acetophenone (19°C), during the addition it was kept at room temperature to retain it in liquid state. Addition of acetophenone at a lower temperature to sulfuric acid is not advisable, as the heat of dissolution creates hotspots and even the solid powder gets accumulated on the stirrer, thereby causing significant variability in the dissolution time. At a low temperature the dissolution becomes liquid-solid mass transfer controlled. With higher mixture volumes, lumping of solids was seen to yield larger flocks, which take a long time to dissolve/react. Hence, mixing of acetophenone at room temperature (>25°C) in cold sulfuric acid was preferred and was found to be very exothermic. The heat generation rate was in the range of 3.9°C/ml to 4.2°C/ml, which limits the mixing volumes and also indicates the need for having either a longer mixing time to avoid rapid heat generation, or rapid heat removal from the system. The nitrating mixture was prepared by mixing concentrated nitric acid with concentrated sulfuric acid and was maintained at 0°C in a v/v ∼40/60 proportion. The effect of composition of nitrating mixture was also studied.
2.1.2 Batch experiments
In a typical batch experiment, the nitrating mixture was added dropwise to the mixture of acetophenone in sulfuric acid over 30–45 min and care was taken to maintain the temperature at 0°C. This was done in a jacketed stirred glass reactor. After addition was complete, the solution was stirred for 15 min and added to a 500 ml beaker containing 150 g of crushed ice. This resulted in quenching of the reaction and a yellow solid product was precipitated. The product was filtered (which contains almost >90% of the meta isomer and the rest was the ortho isomer), dried and later recrystallized in ethanol to get the pure meta isomer. At higher temperatures phenolic byproducts as well as the byproducts due to cleavage of the ketone group were observed. The melting point of the dried product was checked for several batches and was ∼76°C–77°C; it was confirmed by nuclear magnetic resonance (NMR) as 2 (the value from the literature is 77.8°C). The above procedure for isolation was also followed for the continuous flow experiments.
2.1.3 Continuous flow experiments
The typical experimental setup consisted of two syringe pumps (Longer Pumps, China), a constant temperature bath (Julabo GmbH, Germany), a micromixer and a tubular reactor. The residence time was controlled by choosing the flow rates and the reactor tube outside the constant temperature bath (the outlet portion about 60 mm) was maintained at a constant temperature by insulating it using glass-wool. Typical experimental set-ups are shown in Figure 1.
After preparation of solutions, both the solutions were taken in glass syringes. For the continuous flow experiments, tubular reactors of three different sizes (outer diameter (o.d.) ~ 1.58 mm Hastelloy tube, 3.16 mm and 6.32 mm SS316 tubes) were used. While the first one resulted in a clogged reactor due to precipitation of the product/polynitrated byproduct the latter two tubes did not clog. In all cases, flow rates were adjusted to achieve the desired residence time. For the selection of a suitable micromixer for this reaction, a simple T-mixer, a split and recombine type planar micromixer and a caterpillar IMM micromixer were used. Unless explicitly stated, most of the experiments were carried out at 5°C. The samples were collected on crushed ice and the product was filtered and washed twice with 20 ml of ice-cold water. Then, residue was dried for 1 h and the filtrate was extracted by using diethyl ether. From that extracted layer, 1 ml was taken in a sample vial and analysis was done with gas chromatography (GC) (Thermo Trace Ultra GC) with a HP5 column (30 m×0.25 mm inner diameter (i.d.), 0.2 μm film). Experiments were also carried out without filtration and the product was extracted using diethyl ether. After extraction with ether, the mixture was concentrated on Rotavapor (Buchi, Germany) and the product weight was monitored until it reached a constant value. Experiments were also carried out by mixing 1 and sulfuric acid continuously at 0°C and then further mixing it with the nitrating mixture using another micromixer immersed in the constant temperature bath just before it reaches the tubular reactor. A few experiments were also carried out for the nitration of p-chloroacetophenone and m-chloroacetophenone.
2.2 Reduction of m-nitroacetophenone
2.2.1 Batch experiment
Reaction of m-nitroacetophenone (2 g) with granulated tin (Thomas Baker, India) (4 g) dissolved in 40 ml of 10% HCl (Thomas Baker, India) in a batch reactor at 95°C for 2 h was monitored. The reaction mixture was brought to room temperature and was filtered to remove any undissolved tin. A 40% NaOH (SD Fine Chemicals) solution (24 ml) was added to the filtrate with stirring and cooling. The resultant yellow precipitate was filtered and washed with water (20 ml). After drying on suction pump for 2 h, it was recrystallized from water. The melting point of the product was 96–97°C, which matches with the melting point from literature and was confirmed by NMR as 3.
2.2.2 Continuous flow reduction
The reaction was carried out in a silicone tube (1/8” o.d.). 1 g acetophenone in 20 ml methanol (Thomas Baker) and 5 g SnCl2·2H2 O (SD Fine Chemicals) in 20 ml 10% HCl were pumped using syringe pumps with a residence time of 22 min at 100oC. However, since the tin chloride based reduction is not a green approach, all further experiments were carried out using sodium sulfide (Na2 S) as a reducing agent. Details are given in the next section.
3 Results and discussion
3.1 Nitration of acetophenone
As mentioned previously, for the continuous flow experiments, tubular reactors of three different sizes were used. While the smallest tube diameter (1.38 mm i.d.) got clogged due to precipitation of the product, the latter two tubes did not clog for a range of residence times studied in these experiments. Hence, further experiments were carried out using a 1/8” tube. For the selection of the right micromixer, experiments were carried out at 50°C and residence time of 10 min with 2 m long, 1/8” o.d. tubular reactor (i.d.=2.7 mm). The micromixer and the reactor were immersed in a constant temperature bath. The observations on the yield of the meta and ortho isomers are shown in Figure 2. The reaction needed excellent mixing between the two reactants and the conversions were 78%, 86% and 100%, respectively, with T-mixer (0.8 mm bore size), planar split and recombine type of mixer (0.5 mm characteristic channel dimension) and the IMM Caterpillar micromixer (R600). The planar split and recombine mixer was an in-house design with obstacles in the flow path that would just split and recombine the fluid streams in the same plane. The mixtures of acetophenone in sulfuric acid and the nitrating agent have different density and viscosity values, which makes this reactive mixing challenging. At flow rates that need about 10 min residence time in the tubular reactor, mixing of these two streams was seen to be the limiting factor. Better mixing will always help in completion of the reaction. In all of the subsequent experiments, the IMM Caterpillar micromixer (R600) was used before the tubular reactor.
The effect of residence time was studied at a reaction temperature of 5°C. The higher yield of the desired meta isomer was obtained at a residence time of 5 min. A longer residence time yielded impurities, which resulted in reduction in the yield of the desired meta product. The effect of volume of nitrating mixture (at identical composition) was studied at 10°C and 10 min residence time. The product yield increased with the increasing volume of nitrating mixture. However, the maximum yield was still lower than that of the previous experiment at 5°C with 5 min residence time, indicating the formation of impurities (mainly deacetylation followed by dinitrobenzene derivative and was confirmed by GC-mass spectrometry) at higher temperatures. The effect of reaction temperature was studied by carrying out the reaction at 0–25°C. The observations are given in Table 1. Lower temperatures were seen to reduce the yield in a given reaction time, while the higher temperature yielded more impurities.
Based on these observations, the optimized conditions for the reaction were 10°C, 10 min residence time, w/v ratio of 1 to sulfuric acid of 1:2.5, v/v ratio of substrate mixture to nitrating mixture of 1:1.66 (with standard nitrating mixture), that yields 98.55% of the expected yield of the mononitro derivative and complete conversion of acetophenone.
A few experiments were also carried out for nitrating acetophenone with fuming nitric acid. It was observed that 1.8 mole equivalents of fuming nitric acid gave around 80% conversion in 20 min residence time at 20°C. However, the selectivity of the meta isomer decreased with increasing temperature as well as increasing reaction time. Similarly for the above conditions, nitration of p-chloroacetophenone using 10 moles of fuming nitric acid yielded over 99% conversion in 10 min. Nitration of m-chloroacetophenone, even with a large excess of fuming nitric acid (7 moles), resulted in only 85% conversion at temperatures as high as 70°C. Experiments at higher temperatures were not pursued due to safety concerns. A detailed analysis of these nitrations using fuming nitric acid, including the estimation of kinetic parameters, was carried out over a wide range of conditions and will be reported separately. The E-factors for these nitrations were estimated and were found to vary significantly depending upon the excess nitrating agent used in the reactions, and the amount of water used for quenching by dilution.
3.2 Reduction of m-nitroacetophenone
The reduction in batch mode used m-nitroacetophenone 2 (2 g), 4 g granulated tin metal, 40 ml 10% HCl and 24 ml of 40% NaOH (used for work-up). At 85–95°C, the reduction over 2 h yielded 74% of 3 (melting point ∼96–97°C). The product was confirmed by NMR. In the presence of substrate, tin was soluble in 10% HCl at 85°C and the reaction mixture became homogeneous within 45 min. The reaction was conducted by following the above dissolution procedure and a sample was taken after 24 h. Eventually, the solution was in slurry form with some amount of undissolved tin. The experiment was repeated and six samples were taken with 15 min intervals after the solution became homogeneous. The samples were quenched with 40% NaOH and then extracted by ether and the analysis showed that all of the samples contained only the product. Additional experiments were carried out using 1.9 g of the catalyst SnCl2.2H2 O, methanol and 10% HCl. The reaction was kept for reflux at 78°C for 4 h and the reaction was still incomplete. Higher catalyst quantity (3.87 g) and reflux at 110°C yielded 95% conversion of 2 in 30 min.
A typical experimental set-up for continuous flow reduction consisted of two syringe or peristaltic pumps and a thermostat. Experiments were carried out at 100°C in a 1/16” o.d. tube (1.38 mm i.d.) with 8.5 ml volume. Experiments yielded 100% reduction for a residence time of 22 min. Since the use of an SS316 tubular reactor may give a colored product, all of the experiments were carried out in a silicone tube. The experiments in a silicone tube (4.4 ml volume, 1.5 mm i.d.) at 100°C yielded 70% yield of the product. The reaction was further optimized to yield 100% yield in 5 min using a higher catalyst quantity. Since the process of reduction using tin chloride is not economical and also environmentally challenging at large scale due to sludge formation, all further experiments were carried out using sodium sulfide as the reducing agent.
The reduction reaction was carried out in an SS316 1/8” o.d. tube (2.8 mm i.d., 8.5 ml volume) with 2 (2 g) in ethanol (37 ml) and Na2 S (2.5 g in 8 ml water) at 70°C. Residence time was maintained at 20 min. The reaction achieved complete conversion of 2, yielding only a single product at the outlet. Yield of the 3 in the crystalline form from the first stage of crystallization was 89%. Remaining was obtained as the second crop. In another set of experiments, both the precursors (2 and Na2 S) were mixed in an aqueous solution of ethanol and pumped using a single peristaltic pump either using a silicone tube or an SS316 tube immersed in a constant temperature bath at 70°C. The ethanol-water mixture was recycled and reused. This gives a perfect reduction reaction where only selective reduction of the nitro group takes place, giving an E-factor of 0, as the salt formed in the reaction can be used for many applications.
These studies indicate that, while nitration of an aromatic substrate can be done efficiently in continuous mode, reduction in the continuous mode is feasible only when the mononitro derivatives are used in isolated/separated form. While this reduction may not be greener than the catalytic hydrogenation, using simple reducing agents viz. Na2 S with a completely recyclable solvent make it an economically viable option. The product from this reduction is sodium sulfate, which is used in large quantities for the manufacture of detergents, in the Kraft process of paper pulping, drying and storage of moisture sensitive items and in the manufacture of glass. Being a noncorrosive salt, it is used in large quantities in textiles. Thus, although the use of Na2 S may not seem as approach like hydrogenation, the use of a recyclable solvent and consumption of the generated salt makes it a greener and more economical approach.
The recent trends of development of integrated processes [21, 22] are definitely a forward going approach. After every reaction, separation of isomers is critical and should be done at the stages where the separation is possible without much hassle (viz. different melting points, very different boiling points, very selective solubility in solvents, etc.). Incidentally, the products from all reactions do not show such features and the integration of reaction and separation remains limited. In such cases, while individual steps in the syntheses do show merits of going continuous, the feasibility of separation and the economics will drive such recommendations.
Two-step, discontinuous, flow synthesis of m-aminoacetophenone is demonstrated using simple tubular reactors (SS316 and silicon tube). The yields from continuous flow reactions were comparable with the batch reactions. The flow synthesis approach for the nitration step yields a safer process, even at enhanced temperatures with shorter reaction times, to achieve consistent performance. Both of the steps involved homogeneous solutions (in contrast to typical two phase aromatic nitration). Using a good micromixer is essential to achieve the activated aromatic substrate with the nitrating agent, while for the reduction, a simple T micromixer was sufficient to achieve the desired mixing. The formation of the dinitro derivative as an impurity was seen to have a strong dependence on temperature, residence time and the internal composition of the nitrating mixture. The reduction with sodium sulfide is recommended due to lower costs, which makes the process economical. From the simple set-up as described here, both of the steps can be scaled to make a few 100 g quantities of 3 at lab scale in a single day. Since the solutions are homogeneous, scaling up of this process will be relatively easy if the necessary heat transfer area is made available to control the reaction. Continuous flow nitration of other aromatic ketones using a variety of nitrating agents, including fuming nitric acid, is also reported and more work on its scale up is in progress.
Panke G, Schwalbe T, Stirner W, Taghavi-Moghadam S, Wille G. Synthesis 2003, 2003, 2827–2830.Google Scholar
Kulkarni AA, Nivangune NT, Kalyani VS, Joshi RA, Joshi RR. Org. Process Res. Dev. 2008, 12, 995–1000.Google Scholar
Kulkarni AA, Kalyani VS, Joshi RA, Joshi RR. Org. Process Res. Dev. 2009, 13, 999–1002.Google Scholar
Brocklehurst CE, Lehmann H, La Vecchia L. Org. Process Res. Dev. 2011, 15, 1447–1453.Google Scholar
Gage JR, Guo X, Tao J, Zheng C. Org. Process Res. Dev. 2012, 16, 930–933.Google Scholar
Kulkarni AA. Beilstein J. Org. Chem. 2014, 10, 405–424.Google Scholar
Burns JR, Ramshaw C. Chem. Eng. Commun. 2002, 189, 1611–1628.Google Scholar
Ducry L, Roberge DM, Angew. Chem., Int. Ed. 2005, 44, 7972–7975.Google Scholar
Halder R, Lawal A, Damavarapu R. Catal. Today 2007, 125, 74–80.Google Scholar
Roberge D, Ducry L. The autocatalytic nitration of activated heteroaromatic or aromatic compounds in micro reactors. WO/2007/087816, 2007.Google Scholar
Braune S, Pochlauer P, Reintjens R, Steinhofer S, Winter M, Lobet O, Guidat R, Woehl P, Guermeur C. Chim. Oggi 2009, 27, 26–29.Google Scholar
Pelleter J, Renaud F. Org. Process Res. Dev. 2009, 13, 698–705.Google Scholar
Shen JN, Zhao YC, Chen GW, Yuan Q. Chin. J. Chem. Eng. 2009, 17, 412–418.Google Scholar
Yang JL, Li JF, Lu Y. Acta Phys.-Chim. Sinica 2009, 25, 2045–2049.Google Scholar
Arava VR, Gorentla L, Dubey PK. Int. J. Org. Chem. 2011, 1, 26–32.Google Scholar
Breuer M, Pletsch A, Hauer B, Siegel W. (BASF SE, Germany) WO2010031776, 2010.Google Scholar
Jinqing HLL, Guangxiang, Z. Huagong Shengchan Yu Jishu 2008, 15, 20–22.Google Scholar
Corson BB, Hazen RR. Org. Synth. 1930, 10, 74–77.Google Scholar
Moodie RB, Penton JR, Schofiel K. J. Chem. Soc. B 1969, 578.Google Scholar
Mascia S, Heider PL, Zhang HT, Lakerveld R, Benyahia B, Barton PI, Braatz RD, Cooney CL, Evans JMB, Jamison TF, Jensen KF, Myerson AS, Trout BL. Angew. Chem., Int. Ed. 2013, 52, 12359–12363.Google Scholar
Liu XY, Jensen KF. Green Chem. 2013, 15, 1538–1541.Google Scholar
About the article
Jagdish Tibhe is pursuing his PhD at Eindhoven University in the group of Professor Volker Hessel. He obtained his MSc in Organic Chemistry (2010) from the Department of Chemistry, University of Pune, Pune, Maharashtra (India) and subsequently worked as a Project Assistant at the National Chemical Laboratory, Pune.
Yachita Sharma is a PhD student at CSIR-National Chemical Laboratory, Pune (India). She received her MSc in Applied Organic Chemistry in 2010. Her work focuses on exploring the continuous flow synthesis involving exothermic reactions and their integration.
Ramesh A. Joshi
Ramesh A. Joshi is a Chief Scientist in the Division of Organic Chemistry of the National Chemical Laboratory, Pune (India). His research interests focus on the development of continuous flow processes for pharmaceuticals intermediates and dyes, and process development for APIs and fine chemicals. He obtained his PhD in Synthetic Organic Chemistry from the National Chemical Laboratory, Pune (1980). He has 30 years of experience in major projects in drugs and pharmaceutical sciences and has developed several drug technologies which are commercialized by Indian Pharmaceutical Industries. He is recipient of many awards for his recognition in the field of organic process development. He has 30 publications in peer-reviewed international journals and 22 international patents.
Rohini R. Joshi
Rohini R. Joshi is a Scientist in the Organic Chemistry Division of the National Chemical Laboratory, Pune (India). Her research interests are on process development for pharmaceuticals intermediates and fine chemicals and biotransformation. She has a PhD in Synthetic Organic Chemistry. She has developed several drug technologies which are commercialized by Indian Pharmaceutical Industries. Her efforts in the area have been recognized through many awards for organic process development. She has 15 publications in international peer-reviewed journals and 18 international patents.
Amol A. Kulkarni
Amol A. Kulkarni is a scientist in the Chemical Engineering Division at the National Chemical Laboratory (NCL), Pune. He obtained his B Chem Eng (1998) and his PhD in Chemical Engineering (2003) from the Institute of Chemical Technology, Mumbai (formerly UDCT). He works in the area of design of microreactors, continuous flow syntheses of pharmaceutical intermediates, dyes and nanoparticles and design of multiphase reactors. He has published 50 papers in international peer-reviewed journals and has filed 12 patents. He is an active industry consultant.
Published Online: 2014-08-13
Published in Print: 2014-08-01