Impurity Profiling of Chemical Reactions

Impurity profiling aims at identification and subsequent quantification of specific components present at low levels, usually less than 1% and ideally lower than 0.1%. Impurities are unwanted residuals that form during or after the course of the reaction. The residuals can be inorganic, organic chemicals or residual solvents that define the quality and properties of a specific product; for example, the efficacy of an active pharmaceutical ingredient (API).

Impurity profiling plays an important role during different stages of R&D in chemical development. Understanding impurity formation during chemical synthesis is essential in order to control and change reaction conditions in such a way that the formation of the impurity can be reduced to an acceptable level. Researchers can innovate methods to minimize impurity formation or concentrations. Various agencies, including the European Medicines Agency (EMA), the United States Food and Drug Administration (FDA) and The International Council for Harmonisation of Technical Requirements for Pharmaceutical for Human Use (ICH), regulate the rationale for the reporting and control of impurities for organic, inorganic and solvent residuals. Additionally, several authorities adopted Quality by Design (QbD) methodologies to drive improvement of product and process qualities for the discovery, development and manufacturing of APIs.

Every scientist takes samples throughout the duration of a chemical reaction. The sample or the aliquot represent the chemical composition of the reaction at a specific time point. Throughout the duration of the experiment, samples are taken and subsequently prepared for offline analysis, such as chromatography and Nuclear Magnetic Resonance (NMR). The information from offline analysis methods is then used to profile the transformation of a chemical reaction and link the concentrations to the conversion, yield and selectivity versus time. Therefore, samples are crucial in revealing reaction kinetics, impurity profiles, pathways and mechanistic aspects to gain fundamental insights within the research & development projects.

When taking a sample manually, there is a chance the sample is not representative of the specific time point. A sample does not necessarily take into account the reaction parameters and the chemical environment, which it must represent. Usually, the chemical reaction continues until the sample is further processed, modified or quenched. As a result, EasySampler was developed for reproducible, representative, automated and unattended sampling.

The automated and unattended sampling solution, EasySampler, provides accurate, reproducible and representative samples. At user-defined times, EasySampler automatically captures reaction samples and quenches samples immediately at reaction conditions. Finally, EasySampler dilutes the sample to a user-specified concentration. The innovative and patented combination of in-situ sampling, quenching and diluting guarantees high-quality samples, even from heterogeneous mixtures, air- and moisture-sensitive reactions, and reactions under pressure and toxic conditions.

Accurate and reproducible impurity profiling is critical for any new drug application. The purity of a product depends on several factors, such as raw materials, reaction type, reaction route and purification processes. In the early stages of chemical development, it is essential to measure and characterize impurity formation. Synthetic organic chemists change and control reaction conditions in such a way that the formation of the impurity can be reduced to an acceptable level, or even avoided. Drug authorities use the impurity profile as a fingerprint, indicating the robustness of the manufacturing process.

Sampling chemical reactions, including slurries, at elevated pressure, air- or moisture-sensitive reactions and multi-phase reactions, is often difficult or tedious. In addition, many reactions are long, or may be started late in the day, making impurity profiling and collecting samples throughout the entire reaction difficult. Consequently, there may be a limited number of samples collected, and/or sampling inconsistencies occur when collection methods are time-consuming, or when obtaining a representative sample is difficult. If samples are not continuously collected, the reaction profile including conversion, yield and impurity levels are not accessible. By using automated sampling to continually collect accurate and reproducible samples for analysis, chemists gain more quality information from fewer experiments, which leads to quick decisions to develop innovative chemistry.

Two application notes discuss how Janssen and Servier Pharmaceuticals sample reactions without affecting reaction progression:

• Reaction Profiling of a High Temperature Buchwald-Hartwig Slurry
• Endpoint Detection of an Air-Sensitive Ester Reduction

Oxygen-sensitive reactions are difficult and tedious to sample, and typically require repeat experiments to gain complete data sets. Manual sampling introduces oxygen into the reactor and stalls the reaction progression. Furthermore, samples are altered when they are introduced to air during manual sampling. These factors affect the reaction rate, accuracy, and integrity of the reaction sample and lead to poor and inconsistent data. Having the ability to capture a representative sample from these types of reactions in an automated manner provides a complete and accurate data set. This enables accurate kinetics studies and generates representative impurity profiles to understand the rate and mechanism of formation. Overall, the information-rich experiments improve productivity and cost savings, and lead to reduced development times.

David Place, Pfizer, presented a case study with a palladium catalyzed C-H Activation reaction, shown to the right. The reaction is oxygen-sensitive, and oxygen levels of 5000 ppm in the headspace would result in a 50% increase in reaction time. However, to develop the reaction it was necessary to understand the kinetics profile, the mechanism of impurity formation, and assign a reasonable endpoint; but sampling was difficult without introducing oxygen to the reaction.

For this reaction (scheme above), twelve samples were acquired using EasySampler in-situ sampling over a 24-hour period and analyzed by UPLC (see graph). The data gathered over the 24-hour period provides insight to the interconnectivities of the impurities monitored in the reaction, and especially highlights the need for a tight reaction cycle time. The conversion and impurity profiles show that when the product reaches a maximum at 8 hours, the reaction system must be immediately cooled from 102 °C to 20 °C, to avoid the Des-CN impurity (impurity 1) formation. If the mixture continues stirring too long, the by-product forms suddenly, and cannot be easily removed in subsequent workup and reaction steps. Based on this information, researchers quickly set in-process controls to correlate reaction time with impurity levels. 

Automated chemical reactors with unattended, representative sampling change the way chemists work in the laboratory. Setup is intuitive, and experiments can begin at any time and be safely left to run overnight. With unattended control and continuous data collection, scientists gain an advantage: to understand, innovate, and make informed decisions. 

Below is a list of publications from peer-reviewed journals on impurity profiling by researchers to support data-rich experimentation to advance their research.

Featured Impurity Profiling Citations

• Wang, K., Han, L., Mustakis, J., Li, B., Magano, J., Damon,D., Dion, A., Maloney, M.T., Post, R.J., Li, R., "Kinetic and Data Driven Reaction Analysis for Pharmaceutical Process Development", Ind. Eng. Chem. Res. (2019), Advance article
• Beutner, G.L., Coombs, J.R., Green, R.A., Inankur, B., Lin, D., Qiu, J., Roberts, F., Simmons, E.M., Wisniewski, S.R., "Palladium-Catalyzed Amidation and Amination of (Hetero)aryl Chlorides under Homogeneous Conditions Enabled by a Soluble DBU/NaTFA Dual-Base System", Org. Process Res. Dev. (2019), 23, 1529-1537.
• Mennen, S.M., Alhambra, C., Allen, C.L., Barberis, M., Berritt, S., Brandt, T.A., Campbell, A.D., Castañón, J., Cherney, A.H., Christensen, M., Damon, D.B., de Diego, J.E., García-Cerrada, S., García-Losada,P., Haro, R., Janey, J., Leitch, D.C., Li, L., Liu, F.,  Lobben, P.C., MacMillan, D.W.C., Magano, J., McInturff, E., Monfette, S., Post, R.J., Schultz, D., Sitter, B.J., Stevens, J.M., Strambeanu, I.I., Twilton, J., Wang, K., Zajac, M.A., "The Evolution of High-Throughput Experimentation in Pharmaceutical Development and Perspectives on the Future", Org. Process Res. Dev. (2019) 23, 1213−1242.
• Carter, H.L.,  Connor, A.W., Hart, R., McCabe, J., McIntyre, A.C., McMillan, A.E., Monks, N.R., Mullen, A.K., Ronson, T.O., Steven, A., Tomasia, S., Yates, S.D., "Rapid route design of AZD7594", React. Chem. Eng. (2019), Advance Article.
• Zawatzky, K., Grosser, S., Welch, C.J., "Facile Kinetic Profiling of Chemical Reactions Using MISER Chromatographic Analysis", Tetrahedron, (2017) 73, 5048-5053.
• Gurung, S.R., Mitchell, C., Huang, J., Jonas, M., Strawser, J.D., Daia, E., Hardy, A., O’Brien, E., Hicks, F., Papageorgiou, C.D., "Development and Scale-up of an Efficient Miyaura Borylation Process Using Tetrahydroxydiboron", Org. Process Res. Dev. (2017) 21, 65-74.
• Rougeot, C., Situ, H., Cao, B.H., Vlachos, V., Hein, J.E., "Automated Reaction Progress Monitoring of Heterogeneous Reactions: Crystallization-induced Stereoselectivity in amine-catalyzed aldol reactions", React. Chem. Eng. (2017), 2, 226-231.
• Malig, T.C., Koenig, J.D.B., Situ, H., Chehal, N.K., Hultinb, P.G., Hein, J.E., ”Real-time HPLC-MS reaction progress monitoring using an automated analytical platform", React. Chem. Eng. (2017) 2, 309-314.
• Duan, S., Place, D., Perfect, H.H., Ide, N.D., Maloney, M., Sutherland, K., Price Wiglesworth, K.E.; Wang, K., Olivier, M., Kong, F., Leeman, K., Blunt, J., Draper, J., McAuliffe, M., O’Sullivan, M., Lynch, D., Palbociclib Commercial Manufacturing Process Development. Part I: Control of Regioselectivity in a Grignard-Mediated SNAr Coupling", Org. Process Res. Dev. (2016), 20, 1191−1202.