Safe Automated Dosing with Exothermic Reactions - Case Study
Add bookmark
Based on work by Dr. Carsten Griebel, Labhead Chemical Development, Grunenthal.
In the pharmaceutical and fine chemical industries , reactions are often exothermic. Though they have inherent safety hazards at small scale, these risks can be alleviated by operating in semi-batch mode with solid or reagent dosing. While monitoring batch temperature can be used to control thermal risk, upon scale-up to larger jacketed glass reactors, heat removal becomes limited due to decreasing surface area to volume ratios, and this introduces new risks. These include: miscalculating dosing rates or heat removal capacity, and equipment failure producing a loss of cooling capacity. Due to these risks, either constant manual supervision or automated and safe dosing control is required.
In order to increase productivity without increasing staff, the Research & Development group at Grünenthal GmbH, Germany, emphasize that experiments are run safely and without manual supervision. At Grünenthal, finding a way to perform unattended and safe dosing is one of the challenges in scale-up projects. Developing a programmable logic control (PLC) scheme is not practical with tight project timelines, leading researchers to use safe, plug-and-play automated reaction control.
Introduction:
The Grünenthal Group, headquartered in Aachen, Germany, is present in 32 countries, their products are sold in more than 155 countries , and they employ approximately 5,300 individuals . They are an international, research-based pharmaceutical company; their fully integrated R&D has a long track record with bringing innovative pain treatments and state-of-the-art technologies to patients.
To improve the production process of an active pharmaceutical ingredient (API), a commercial process had been scaled down and optimized on the 100 milliliter scale . However, when preparing for scale-up and production transfer, new risks and challenges arose, requiring attention.
Read the report to learn the importance of using data to create scientific value and how to digitize data for more efficient lab work. Download the Report.Pharma IQ Special Report: Unlock the Power of Data to Advance Pharma Businesses
Experimental Setup:
A 2 liter jacketed lab reactor was connected to a thermostat and stirrer in the initial step of the scale-up, with no interconnection among the equipment. Doing such created safety risks. For example, should the thermostat fail and lose cooling power, reagent dosing would not be stopped without an operator intervening. In order to avoid safety risks, experiment steps (such as dosing), could not run without manual supervision.
A challenging scale-up step within this process contained a Grignard reaction. The high exothermicity of the reaction, and the limited possibilities of the production facility in regard to heat removal, required reagent dosing times between 8 and 16 hours . Such long execution times for the dosing exceeded the length of a normal scientist working day. If dosing was to run unattended, and overnight, without violated safety guidelines , a solution had to be found.
Knowing the relationship between key process parameters, such as accumulation caused by dosing, is important when understanding the events in the reaction. This is particularly true for unattended reactions when scientists need a way to verify the proper execution of the experiment, and quickly spot undesired events. A delayed reaction initiation after the start of a dosing must be avoided to circumvent reactant accumulation and a potentially catastrophic event on scale-up. At Grünenthal, it was also important that data from their larger jacketed lab reactor volumes was easily compared with their small-scale 50-400 milliliter automated synthesis reactor data files .
Method:
In the scale-up of a highly exothermic Grignard reaction, Grünenthal used the following approach to implement safe and unattended dosing on a 2 liter jacketed lab reactor:
1. An automated reactor control unit [METTLER TOLEDO RX-10] was connected to the thermostat [Julabo], stirrer [IKA], and the syringe pump [METTLER TOLEDO DU SP-50] through plug-and-play connections.
2. Safety settings were defined to avoid dangerous situations in case of unexpected events.
a. The automated reactor control unit has an integrated safety program that activates automatically if a safety limit is violated, or if a failure occurs in any integrated equipment (e.g. a thermostat failure that results in a lack of cooling capacity). The following actions are triggered when the safety program gets executed:
i. The temperature is set to the pre-defined safe temperature (T safe).
ii. All running dosings are aborted.
iii. The stirrer speed remains unchanged.
3. A task sequence on the touchscreen was used to pre-program simple recipe steps such as heat, cool, wait, or dose.
4. For dosing steps, a maximum reactor temperature was specified to suspend dosing if the reaction temperature exceeds the defined threshold due to accumulation. Dosing resumes as soon as the temperature drops back to the defined maximum value.
5. After experiment completion, all experimental data was exported directly from the automated reactor control unit on a USB stick and opened on a PC in Microsoft Excel format. The captured data was added to the Electronic Lab Notebook (ELN) to ensure complete experiment documentation, including measured data and details on all executed recipe steps.
Experimental Results:
1. During the workup of the Grignard reaction , the temperature was held below the defined limit by suspending quench dosing at the moment the reactor exceeded the temperature threshold (Figure 3).
2. Tr-Tj tracked the difference between the temperature of the reaction (Tr) and the temperature of the jacket (Tj) (Figure 4). The Tr-Tj trend was an indicator for a transformation or change that occurred in the reactor during the experiment and indicates a physical or chemical change. It indicates the start and end of the reaction. By tracking Tr-Tj, scientists applied a real-time heat-rate meter that helped them make quick informed decisions on when to take samples, when to initiate the next step, and when to end the experiment.
3. Once data was recorded, data sets from multiple experiments (the same or different vessels) were easily compared with the iControl Software trend viewer , which helped to identify differences between batches (Figure 5).
4. During the process development, scale-up, and piloting of the project, no experiment failed due to issues related to applied automation tools (either through human errors or technical limitations). This was a proof for the easy operation as well as of the stability of the system.
42% of the industry claim efficient lab data management will have the biggest impact on industry growth in the next five years. Find out how to get ahead of the curve. Download the Pharma IQ Report Now.Achieve Digital Business Success in Pharma with Data-Driven Approaches
Carsten Griebel of Grünenthal said: “The automated reactor control unit is a versatile solution to optimize scale-up processes. In our lab, it has closed the gap between small scale synthesis reactors and multi-liter jacketed lab reactors. As main benefits, it provides a consistent lab workflow to the user and gives documentation of experiments along the whole scale-up workflow."
Conclusion:
Chemists and engineers at Grünenthal focus on chemistry and scale-up, not on set-up of complex automation systems. Therefore, it was key that any automation in their laboratory was set up, understood, and operational within minutes, without the need for longer pieces of trainings. Automation helped Grünenthal increase lab efficiency and safety, while reducing complexity for researchers. It is well accepted by chemists and engineers, and easily integrates with the existing 2 to 20 liter scale vessels within Grünenthal process R&D. Automation with controlled dosing not only provides a safe environment for reactions to be run unattended, it also increases productivity and the ability to gather knowledge from each experiment. Continuous data collection and reporting improves the ability to make informed evaluations during scale-up. Consistent data sets are easily compared between small scale synthesis reactor systems, and large scale jacketed lab reactors . This consistency makes it simple to evaluate and compare experiment results from 50 milliliter to 20 liter and make quick decisions to speed process development.
Image: sspopov
