The ACS 12 Principles of Green Chemistry [1] are largely dependent on gaining comprehensive knowledge and understanding about molecules, reactions, and the chemical and manufacturing processes that underlie sustainable, high-quality, and cost-efficient products. Process analytical technology (PAT) from METTLER TOLEDO provides precise control of reaction variables, real-time analytical information, and advanced data modeling to support these goals of sustainability through a thorough understanding of reaction kinetics, thermodynamics, and how process variables directly affect yield and by-product waste.
The following examples from the recent literature showcase how scientists and engineers apply PAT and related technology to drive green chemistry practices in support of sustainability goals and a healthier, safer environment.
Qin, Y., Mattern, K. A., Zhang, V., Abe, K., Kim, J., Zheng, M., Gangam, R., Kalinin, A., Kolev, J. N., Axnanda, S., Dance, Z. E. X., Ayesa, U., Ji, Y., Grosser, S. T., Appiah-Amponsah, E., & McMullen, J. P. (2024). Evolution of a green and Sustainable Manufacturing Process for Belzutifan: Part 4─Applications of Process Analytical Technology in heterogeneous biocatalytic hydroxylation. Organic Process Research & Development, 28(2), 432–440. https://doi.org/10.1021/acs.oprd.3c00419
Biocatalytic reactions are advanced through the use of PAT
Scientists at Merck are developing and utilizing biocatalytic-based syntheses in the production of key intermediates in pharmaceutical processes. Biocatalysis has numerous advantages relative to classical chemistry synthesis with the goal of fewer reaction steps, milder catalytic conditions, better cost-effectiveness, and the use of aqueous reaction media, all resulting in an overall improved process sustainability. Though the application of biocatalysis is promising, numerous developmental challenges must be addressed concerning cost, robustness, and scalability.
In the work described in this article, Merck scientists tackle these issues to establish a new route for the drug Belzutifan, which is used in the treatment of renal cancer. They apply biocatalysis in the development of a single-step hydroxylation reaction, which replaces five chemical steps required in the original route [2]. To form the desired hydroxyindanone molecule, indanaone was treated with engineered L-pipecolic acid 4 hydroxylase. Accomplishing this required overcoming specific problems. For example, since the biocatalytic reaction is performed in aqueous media, the reagents and products had variable, limited solubility, and formed slurries. In addition to the physical engineering aspects of handling slurries, reaction kinetics and the ability to acquire representative samples for analysis are significantly affected.
A suite of PAT was used to develop the knowledge necessary for the successful execution of this biocatalytic step, including in-situ FTIR (ReactIR), in-situ imaging (EasyViewer), and in-situ sampling (EasySampler) for offline HPLC measurements. The biocatalytic reactions were performed in an automated laboratory reactor (EasyMax) to ensure precise control of reaction variables. The multi-phase, dense slurries that were required for the manufacturing scale presented significant analytical challenges. Offline sampling was challenging due to sample inhomogeneity, and in-situ sampling was difficult due to clogging. Furthermore, because the hydroxyindanone product precipitates from solution, direct measurement by in-situ FTIR became problematic. To overcome this latter issue, the scientists developed a method to enable in-situ FTIR to measure the soluble succinate by-product, quantitively relate it to hydroxyindanone product formation, and thereby accurately track the reaction progress.
Another interesting, initially poorly understood observation was that the addition of 1-octanol in place of an anti-foam agent substantially improved reaction conversion. EasyViewer imaging studies performed under reaction conditions clearly demonstrated the difference between reactions with and without 1-octanol. These studies revealed that when 1-octanol is absent, the indanone aggregates and forms large particles. When 1-octanol is present, the hydroxyindanone product crystallizes out of solution faster, and the crystals present with different morphology. Furthermore, ReactIR tracking shows that the reaction progresses more rapidly in the presence of the 1-octanol. The studies revealed that these effects likely result from the amphiphilic nature of 1-octanol and its ability to wet the surface of the indanone particles for improved solubility. The particle size of the indanone reagent proved to substantially affect reaction conversion.
In summary, the researchers observed that the PAT used in this work was not only effective for monitoring reaction progress but also provided significant mechanistic insight, thereby enabling the effective development of this biocatalytic reaction
Murbach-Oliveira, G., Akturk, I., Nagy, Z., & Thompson, D. H. (2024). Process development and kinetic analysis in the green synthesis of lomustine. Organic Process Research & Development, 28(5), 1910–1916. https://doi.org/10.1021/acs.oprd.3c00463
Developing a greener process for a flow reaction
Improved environmental and health safety (EHS) characteristics are an important advantage for the continuous flow, telescoped process developed in 2019 for lomustine [3], a DNA alkylating agent used in treating brain cancer. Even so, the process uses dichloromethane as a solvent, which the FDA has classified as a Class 2 solvent that should be limited in pharmaceutical products. With this in mind, Purdue University chemists and engineers initiated a project to modify the process by using two biorenewable green solvents, 2-methyltetrahydrofuran (2-MeTHF) and acetic acid to replace dichloromethane and to determine the reaction kinetics.
In initial experiments designed to replicate the original flow process but using 2-MeTHF as a solvent, the concentration of the cyclohexylamine intermediate had to be reduced to avoid precipitation. This, in turn, resulted in a reduced product yield relative to the original process. An approach using a cosolvent with the 2-MeTHF was investigated, and solubility studies using several different green solvents were initiated. From these studies, it was determined that a 7:3 2-MeTHF/acetic acid mixture was effective at eliminating the precipitate, but only after the reactor configuration was changed so that the intermediate was produced in 2-MeTHF and then solubilized after formation by the addition of the 7:3 2-MeTHF/acetic acid cosolvent via a T-mixer. The unreacted starting material was then removed by liquid-liquid extraction. The intermediate was passed on to a second reactor and reacted with tert-butyl nitrite (TBN) to form lomustine. This new approach provided yields that were nearly equivalent to the original dichloromethane solvent process with even greater productivity.
With this new modified process using green cosolvent, the researchers investigated the reaction kinetics of the first reaction of cyclohexylamine and 2-chloroethyl isocyanate to form the lomustine intermediate and the second reaction, where the intermediate reacts with TBN to form lomustine. The kinetic experiments were performed in an EasyMax 102 automated laboratory reactor with ReactIR measurements via a diamond ATR insertion probe.
To determine the kinetics, calibration experiments were performed to determine the concentration of key species as a function of time. The reaction was performed in batch mode, and the IR data was used to obtain the estimated rate constant. To replicate the conditions used in the flow experiment, the first reaction experiments were performed in 2-MeTHF and the second in 7:3 2-MeTHF/acetic acid cosolvent. For reaction 1, the data was treated as a second-order reaction with respect to isocyanate (i.e., rate = k[ISO]2) due to the fact that the cyclohexylamine did not have unique characteristic IR bands. Reaction 1 proceeded rapidly, and the calculated rate constant was 10–60 times higher than previously reported, likely due to the change in solvents and different experimental configurations. Reaction 2 was found to proceed far slower with a calculated rate constant 4 orders of magnitude lower than reaction 1.
In conclusion, it was determined that using a cosolvent greatly improves the solubility of the intermediate, resulting in improvements in both the yield and productivity of lomustine and achieving a greener process. Kinetic rate parameters determined for both reactions that form lomustine revealed that reaction 1 proceeds far more rapidly than reaction 2. The work demonstrates the dependence of reactant solubility and reactor configuration in optimizing a continuous flow process.
Stamberga, D., Einkauf, J. D., Liu, M., & Custelcean, R. (2024). Direct air capture of CO2 via reactive crystallization. Crystal Growth & Design. 24(11), 4556–4562. https://doi.org/10.1021/acs.cgd.4c00201
Reactive Crystallization enables direct CO2 absorption from air
To address the issue of climate change, a method being investigated for removing CO2 from the atmosphere is through direct air capture (DAC) technology. Since CO2 concentrations in air are low, it is necessary to have strong binding of the CO2 for effective removal. One method uses aqueous alkaline solvents to bind the CO2 as bicarbonate/carbonate anions, which are then crystallized as insoluble salts such as MgCO3 or CaCO3.
The researchers from Oak Ridge National Laboratories have previously demonstrated a DAC process usingaqueous bis-iminoguanidines (BIG) to form insoluble carbonate salts [4]. They found that when an amino acid such as potassium glycinate or sarcosinate was used in the initial CO2 absorption step, there was a very rapid build-up of bicarbonate/carbonate anions, which were subsequently precipitated from the aqueous solution by BIG. They described this process as reactive crystallization of CO2 in which a chemical reaction leads to supersaturation and subsequent crystallization. The DAC process continues as the BIG carbonate salts crystallize, and additional CO2 from the air is drawn into the solution.
In initial studies, a DAC reactive crystallization system consisting of aqueous methylglyoxal-bis(iminoguanidine) (MGBIG) and sarcosine (SAR) appeared promising. Two crystalline carbonate phases were identified from atmospheric CO2 drawn using this system. Phase 1 (P1), with the molecular formula (MGBIGH+)2(CO32-)(H2O)2, forms from more concentrated solutions, while phase 3 (P3), with the molecular formula (MGBIGH22+)(CO32-)(H2O)2, crystallizes from more dilute solutions. They found that only small amounts of SAR added to the system significantly increased the amount of crystals formed.
Based on this information, their goal was to more fully investigate the MGBIG/SAR system to better understand the reaction mechanism, reaction variables that favor the formation of P3 vs. P1 with regard to the amount of CO2 absorbed, and why the SAR amino acid so greatly influences the outcome. For a deeper understanding of this reaction crystallization, several in-situ PAT methods were applied, including using EasyViewer for imaging crystal formation and ReactRaman to monitor the CO2 absorption, both in real time. These probes were inserted directly into the reservoir of a humidifier, used as an air-liquid contactor in this work. A total carbon analyzer system (TIC) was used ex-situ to measure the total amount of CO2 absorbed into the solution.
The EasyViewer system revealed that initially, 3D-shaped crystals formed were primarily consistent with the P1 form, with some of the P3 needle crystals also present. After 6 hours, the P3 needles began to predominate, and after 24 hours, the P3 needles were virtually the only form present in the MGBIG-CO3 product. The ReactRaman measurements showed HCO3-, CO32-, and carbamate species with spectral peaks at 1015 cm-1, 1065 cm-1, and 1163 cm-1, respectively.
The results of the in-situ EasyViewer, ReactRaman, pH measurements, and ex-situ TIC, XRD, and NMR measurements enabled the proposal of a likely reaction mechanism for the reactive crystallization-direct air capture (RC-DAC) system. The reactions of CO2, H2O, and MGBIG generate both the carbonate anions and the MGBIGH+/MGBIGH22+ cations in situ. Even though SAR is present in stoichiometric quantities, it acts catalytically since it is not consumed. It accelerates atmospheric CO2 absorption by converting it into carbamate, which is rapidly hydrolyzed to bicarbonate/carbonate, regenerating the amino acid. When supersaturation is reached by accumulation of carbonate anions and the pH is reduced sufficiently to allow protonation of MGBIG, the carbonate anions crystallize with the MGBIGH+/MGBIGH22+ cations into P1 followed by P3 forms, respectively.
In summary, the researchers suggest that reactive crystallization is an effective means of DAC, combining CO2 absorption and carbonate precipitation into a single step. As crystallization removes carbonate salts from the solution, atmospheric CO2 is drawn into the solution, driving the process. Process costs may be reduced since the RC-DAC method is intensified by combining the CO2 absorption and MGBIG-CO3 crystallization, and the SAR is continuously regenerated.
Drive sustainable development through green chemistry practices by achieving efficiency and reducing waste in chemical process development through reaction control, analysis, and modeling.
Read this white paper for tools, trends, and tactics to drive green chemistry and sustainable development.