Using Microreactors in Chemical Synthesis: Batch Process versus Continuous Flow

Published on: 
Pharmaceutical Technology, Pharmaceutical Technology-09-01-2009, Volume 2009 Supplement, Issue 5

The authors discuss the advantages of microreactors and flow chemistry for various reaction types in achieving improved process economics and reaction efficiency.

This article is part of PharmTech's supplement "API Synthesis and Formulation 2009."

Organic chemists usually spend a substantial amount of time on the scale-up of synthetic routes to new materials. The best pathways need to be chosen, and reaction conditions must be optimized. Once a product is successful and demand for larger quantities is growing, the whole synthesis process needs to be revised again and readjusted for larger batch sizes.

In sharp contrast to the batch mode, chemical synthesis becomes a time-resolved process in flow chemistry. Reagent streams are continuously pumped into a flow reactor, where they are mixed and allowed to react. The product instantly leaves the reactor as a continuous stream. Only flow rate and operation time determine the scale of the synthesis. The same reactor with an inner volume of less than a milliliter will produce kilogram quantities of material.

Parameters other than size determine the performance of a microreactor and whether it will offer superior properties compared with a conventional batch reactor. The material that a reactor is made of—typically metal, glass, or silicon—is one such parameter. Each material offers advantages and disadvantages with regard to price, compatibility with reagents, and heat conductivity.

Figure 1: Heat distribution in a batch-synthesis reactor. (FIGURE COURTESY OF SAFC)

Efficient heat transfer

Because of their small surface-to-volume ratio, microreactors can absorb heat created from a reaction much more efficiently than a batch reactor can. Figure 1 shows the initial heat distribution for a neutralization-model reaction in a simulated 5-m3 batch reactor stirred at 500 rpm. The batch reactor is heated by an exothermic reaction. Cooling only takes place at the surface of the reactor. As a result, there is a strong temperature gradient of about 10 °C from the surface of the reactor to its center. In a microreactor, the heat created by mixing the two reagents is also detectable, but the temperature gradient is much smaller, only 3 °C (see Figure 2). Additionally, it only takes a few millimeters of path length for the reagent stream to cool down again to the temperature of the outside-cooling medium. The formation of hot spots or the accumulation of reaction heat may favor undesirable side reactions or fragmentation (see Figure 3). Microreactors, with their superior heat-exchange efficiency, present a perfect solution. Precise temperature control affords suppression of undesired byproducts.

Figure 2: Heat distribution in a microreactor. L is path length, and y is inner-channel diameter. (FIGURE COURTESY OF SAFC)

The following examples describe the use of microreactors in the industrial production of chemicals. In some cases, not all process details (e.g., catalysts, reagents, and structures) are described, but each example highlights specific advantages of microreactor technology and flow chemistry over batch-synthesis protocols.

Figure 3: Temperature control in a microreactor (MR) enhances product quality by suppressing side reactions. A is the reagent, B is the product, C is the undesired byproduct from a side reaction, and B and C are intermediate stages. (FIGURE COURTESY OF SAFC)

Ionic liquids


Precise control of the reaction temperature. The production of certain ionic liquids in high purity is a special challenge. The reaction between imidazole derivatives and sulfonates is extremely exothermic (see Figure 4). Conventional batch-synthesis procedures fail to yield products with purities higher than about 96%. Even with very slow dosing times and careful cooling, batch reactors are heated by the reaction. Investigations at Sigma Aldrich (Buchs, St. Gallen, Switzerland) showed that severe brown impurities are formed if the temperature rises above 20 °C in any part of the batch reactor, which leads to a miscolored product. Subsequent purification is troublesome and can only be achieved with huge efforts. Microreactors provide an ideal way out of this dilemma. Their superior heat-exchange capabilities make it possible to carry out the synthesis with neat reagents at an external cooling of only –3 °C. Under these conditions, the temperature inside the microreactor in the reaction zone is kept at a constant value of 8 °C, thus allowing ionic liquids to be produced with a purity exceeding 99% with only traces of undesired materials. Because neat reagents can be used without any solvent, the output of this procedure is very high.

Figure 4: Production of ionic liquids. (FIGURE COURTESY OF SAFC)

There is much discussion whether ionic liquids can generally be called "green." Although the answer to this question may be debated, it is quite obvious that the production of ionic liquids in a microreactor is as green as chemical synthesis can be. All reagents are quantitatively converted into the final product. No solvents are needed for the synthesis or purification. Absolutely no waste is generated.

Ring closure in synthesizing exomethylenecyclopentane

Complete suppression of side reactions. The cyclization of 1,5-hexadiene to exomethylenecyclopentane poses multiple challenges. The batch reaction affords an inseparable mixture of the endo- and exo- cyclization product (see Figure 5). The desired kinetically favored product isomerizes under reaction conditions in the presence of the catalyst. Separation of the two isomers is difficult because both have virtually the same boiling point. In addition, the transition-metal catalyst used for this transformation requires a minimum working temperature of 55 °C. At this temperature, the starting material evaporates.

Figure 5: Production of exomethylenecyclopentane in a microreactor. DIBAH is diisobutylaluminium hydride. (FIGURE COURTESY OF SAFC)

Microreactor technology solved all these problems. By inserting a simple pressure valve into the product channel, the reaction was performed at an elevated pressure of 2 bar. This higher pressure raised the boiling point of the reactant sufficiently to 58 °C, a perfect working temperature for the catalyst, and the reaction proceeded without gas formation. Isomerization of the desired product was completely suppressed, and a highly pure product (> 99%) was formed at a multikilogram scale.

Staudinger hydration

Efficient phase-transfer catalysis, safe and scale-independent handling of explosive intermediates in small hold-up volumes with instant conversion. Organic azides are high-energy compounds. Substrates with multiple azide moieties and low-carbon content are potentially explosive. For example, the azide intermediate shown in Figure 6 has a potential energy of 3.8 kJ/g at 114 °C, as measured by differential scanning calorimetry.

Figure 6: Two-stage microreactor assembly for the safe production and immediate conversion of organic azides. (FIGURE COURTESY OF SAFC)

A two-stage microreactor with a continuous-flow system allows for the safe handling of organic azides in small hold-up volumes with immediate conversion in the second microreactor. In the first microreactor, the azide intermediate is formed by nucleophilic substitution. The azide intermediate undergoes Staudinger hydration with triphenylphosphine in a subsequent second microreactor. Notably, it is possible to work with biphasic systems in microreactors. A phase-transfer catalyst moderates the initial reaction with efficient mixing by the microreactor.

Production capacity of o-xylylenediamine in two subsequent glass microreactors exceeds 1 kg/day with an overall yield of 60%. The work-up and isolation of the potentially hazardous intermediate is completely avoided between reaction steps, thereby saving time and money while offering the utmost scale-independent safety level.

Grignard reactions

Accelerated and economic process development for large-scale synthesis. Microreactors have simple and pragmatic advantages over traditional batch vessels. They take up less space on the factory floor, but can be used to make large quantities of product. 2-Benzoyl pyridine is an important building block and has annual demand of about 15 metric tons. It is synthesized through a Grignard reaction (see Figure 7). In a microreactor, this reaction takes less than one minute. The Grignard microreaction is followed by two on-line quench modules. Precise reaction control leads to a highly pure crude product, thereby making a distillation step unnecessary, as would be the case for purification of the lower quality batch product.

Figure 7: Grignard microreaction with two subsequent quench modules. (FIGURE COURTESY OF SAFC)

This reaction was performed in a stainless-steel plate reactor (ART, Alfa Laval, Lund, Sweden). The reactor took up only 30 X 50 cm of space on the bench. Rapid flow rates were achieved with this reactor, thus allowing the continuous production of 200–300 kg of 2-benzoyl pyridine per day.


These reactions show the versatility and utility of microreactor technology and flow chemistry in chemical synthesis. Such an approach offers many advantages over traditional batch-mode manufacturing. Depending on the reaction, improved process economics, greater reaction efficiency, and waste reduction may be achieved. Problems inherent in scale-up are eliminated or reduced, making microreactor technology a viable tool in the synthesis of APIs.

Andreas Weiler*, PhD, is global business director of SAFC Pharma, Industriestrasse 25, 9470 Buchs, Switzerland, tel. 41 81 755 2405, fax 41 81 755 2584, Matthias Junkers, PhD, is product manager of chemistry at Sigma-Aldrich.

*To whom all correspondence should be addressed.


1. T. Schwalbe, V. Autze, and G. Wille, "Chemical Synthesis in Microreactors," Chimia 56 (11), 636–649 (2002).

2. T. Schwalbe et al., "Novel Innovation Systems for a Cellular Approach to Continuous Process Chemistry from Discovery to Market," Org. Process Res. Dev. 8 (3), 440–454 (2004).