Bioreactor Design and Control in the Biopharmaceutical Industry

By Benedette Cuffari, M.Sc.Reviewed by Danielle Ellis, B.Sc.

Principles of bioreactor design
Control strategies in bioreactor systems
Technological advancements
Challenges in bioreactor design and control
Case studies 
Future directions 
References
Further reading

Bioprocessing is an umbrella term that describes the research, development, manufacturing, and commercialization of products derived from or used by biological systems. For industrial purposes, bioprocessing is widely incorporated into pharmaceutical, nutraceutical, food, and beverage production processes.

Within the pharmaceutical industry, bioreactors refer to vessels and containers used to store microorganisms such as bacteria, cells, and algae. Typically, these organisms are used to produce certain biomolecules or biomaterials that can subsequently be incorporated into vaccines, medications, or genetic engineering tools¹.

Principles of bioreactor design

Several different types of bioreactors have been developed for pharmaceutical applications, the most common of which include stirred tanks, bubble columns, airlift, fluidized beds, and fixed bed bioreactors². The design of each of these bioreactors depends on various factors, including the type of biocatalyst being used.

When a bioreactor is used to support cell culture growth, the continuous flow of media is essential to supporting the proliferation and survival of these cells. Stirred tank bioreactors, for example, are steel-based systems equipped with an impeller that provides mechanical agitation to homogenize media containing cells, antibodies, or enzymes within the system.

Other cell culture projects, such as those that rely on immobilized cells, may rely on fluidized bed bioreactors, in which media passes through the distributor to the cells. Within this system, velocity allows media to reach, subsequently suspend, and mix with the cells.

Within the biopharmaceutical industry, single-used bioreactors have become increasingly relied upon as they can range in size from 50 liters up to 2,000 liters². Additional advantages of these systems include the ability to incorporate different filters, valves for pressure and flow control, and other ports for sensors to monitor them, particularly when used for large-scale production processes.  

Control strategies in bioreactor systems

The biological nature of the microorganisms cultured in bioreactors leads to inherent variability and unpredictability. Despite these characteristics, precise control strategies, algorithms, and processes can be used to maintain bioreactor operations within desired ranges.

In almost all bioreactors, actuator levels can be controlled through various devices ranging from pumps, valves, and heaters to electric voltages and stirrer speeds. For large-scale industrial processes, proportional integral derivative (PID) controllers have been successfully used for electrical, aerospace, and mechanical single-input single-output linear systems². Typically, PID controllers are used to control a single variable, such as the bioreactor's temperature or acidity, as they are not ideal for monitoring complex bioprocesses.

Since their initial development in the 1970s, distributed control strategies (DCSs) have enabled various advanced process control strategies to be controlled within a single framework. DCSs allow operators to supervise and control an entire industrial plant from a centralized control station while simultaneously gathering and storing data for control and process analytics.

As technology has advanced, control devices within DCSs have become more digital. For example, many DCSs are now equipped with smart transmitters and actuators, each with its own microprocessor, which allows them to perform highly complex tasks ranging from autocalibration and signal conditioning to self-diagnosis on-site.  

Technological advancements

Sensing technologies have become an essential aspect of modern industrial bioreactors. These sensors facilitate the non-contact and automated monitoring of numerous variables, such as pH, temperature, dissolved oxygen (DO), glucose, and lactate levels, that must be precisely maintained to ensure the sterility and quality of final products.  

To monitor the pH of media, for example, porous glass electrode-based electrolyte-filled sensors, as well as optical property-based and electrochemical sensors, have been incorporated into industrial bioreactors. Maintaining temperature within a bioreactor with a precision of 0.5 °C or better is also optimal²; therefore, many bioreactors are equipped with temperature sensors such as thermocouples, resistance temperature detectors (RTDs), or thermistors.

Impeller speed can also be controlled and closely monitored in both single- and multiple-impeller bioreactors. Tachometers, for example, can be installed into bioreactors to sense whether desired impeller speeds are within the optimal range³. Importantly, uneven shear characteristics or energy dissipation of impellers can lead to the destruction of cells and microorganisms.

Challenges in bioreactor design and control

Pharmaceutical-grade industrial bioreactors are associated with numerous challenges due to the inherent complexity of biological systems housed within these devices, sterility requirements, as well as high process variability.

Products obtained from bioreactors are initially developed in smaller laboratory settings. Therefore, when scaling up these systems to meet the greater demand of pharmaceutical industry production requirements, physiological similarity must be achieved within the bioreactor. Although complete physiological similarity is not a practical goal, partial similarity in biomass growth, feeding, accumulation, and removal rates, as well as micro- and macronutrient concentrations², is often the goal of bioreactor designers.

Each type of biopharmaceutical-grade bioreactor is associated with different design and control challenges. Batch bioreactors, for example, experience fluctuating process conditions that require highly sophisticated control algorithms to maintain the integrity of batch reaction processes. Since each batch will begin with different initial conditions, quality control is essential to optimize process yields and titers.

Case studies

Cedarstone Industry, a Houston, Texas-based manufacturing company, offers a comprehensive range of bioreactors and fermenters for use at any stage of the biopharmaceutical development process.

To meet the highest industry standards, the Cedarstone Bioreactor / Fermentation Tank is equipped with temperature control sensors that maintain an accuracy of 0.2 °C, in addition to automated sensors that monitor pH, DO, stirring, aeration, defoaming, rehydration, and inoculation³. Four different types of ventilation devices can also be incorporated into these systems, depending on the chemical gas requirements, without compromising the sterility of the system.

Likewise, the Austrian technology company Zeta offers several magnetic mixers that have successfully been incorporated into different bioreactors to overcome some of the challenges associated with traditional mixing technologies. More specifically, the magnetic coupling technology of these products ensures the transmission of materials at a very high torque while reducing the risk of potential contamination to extremely low levels⁴.  

Future directions

As artificial intelligence (AI) has advanced at an unprecedented rate over the past several years, it has inevitably been incorporated into bioprocesses to enhance their performance and optimize their control for a wide range of applications. For example, deep learning approaches trained on anaerobic digestion (AD) sensor data have effectively predicted critical process parameters (CPPs)⁵.

Artificial neural networks (ANNs) have also been explored for their ability to monitor, simulate, optimize, and control bioreactors. More specifically, ANNs appear to be ideal solutions for monitoring and optimizing the complexity of these systems while simultaneously identifying any errors to increase process reliability and performance⁶. In the near future, ANNs will likely be integrated with Internet of Things (IoT) devices and advanced sensors to transform modern bioreactors into smart, automated, and self-adaptive systems.

References

1. “Bioreactors in the Pharmaceutical Industry” [Online]. Available from: https://resources.arcmachines.com/bioreactors-in-the-pharmaceutical-industry-ami/.
2. Mitra, S., & Murthy, G. S. (2022). Bioreactor control systems in the biopharmaceutical industry: a critical perspective. Systems Microbiology and Biomanufacturing 2(1); 91-112. doi:10.1007/s43393-021-00048-6.
3. “Bioreactor & Fermenter” [Online]. Available from: https://cedarstoneindustry.com/product-category/reactor/bioreactor-fermenter/?gad_source=1&gclid=CjwKCAjwmYCzBhA6EiwAxFwfgEeICC_XxTXbyKgsMSf1_0r24EoAdH0gvRJY8Y0pfYypFq7ehOZmMBoC6qsQAvD_BwE.
4. “Magnetic Mixing Agitator” [Online]. Available from: https://www.zeta.com/en/product/magnetic-mixing-technology.html?utm_term=pharmaceutical%20bioreactor&utm_campaign=US+ZETA+Products/Services&utm_source=adwords&utm_medium=ppc&hsa_acc=8410811225&hsa_cam=20325007906&hsa_grp=150920673979&hsa_ad=664029421907&hsa_src=g&hsa_tgt=kwd-1467697451892&hsa_kw=pharmaceutical%20bioreactor&hsa_mt=e&hsa_net=adwords&hsa_ver=3&gad_source=1&gclid=CjwKCAjwmYCzBhA6EiwAxFwfgP0-aHrqSBhA7Yu9Iq07IWsSiHflxe8-z9xFEu7fAQCNNrhgIDfeMhoC8bcQAvD_BwE.
5. Cheng, Y., Bi, X., Xu, Y., et al. (2023). Artificial intelligence technologies in bioprocess: Opportunities and challenges. Bioresource Technology 369. doi:10.1016/j.biortech.2022.128451
6. Frontistis, Z., Lykogiannis, G., & Sarmpanis, A. (2023). Artificial Neural Networks in Membrane Bioeractors: A Comprehensive Review – Overcoming Challenges and Future Perspectives. Science 5(3); 31. doi:10.3390/sci5030031.