Optimal Design of Transient Microreactor Experiments

You will be able to download and print a certificate for these PDH credits once the content has been viewed. If you have already viewed this content, please click here to login.
Development of synthetic routes for active pharmaceutical ingredient (API) synthesis is increasingly performed using microreactors, because their use can reduce experimental cost (both time and material use), make hazardous reactions safer, and enable more precise control [1,2].  While microreactor experiments can be designed to find conditions maximizing yield or selectivity by adjusting processing conditions, others can be designed to discriminate between reaction mechanisms and identify rate parameters as precisely as possible.  The latter approach allows more flexibility because mechanisms and model parameters (such as kinetic and equilibrium constants) remain relevant for a range of reactor inlet conditions.  This is useful during iterations in a full process design, allowing a mathematical model for a whole process to be optimized.  With a dynamic process model in hand, the process analytical technology (PAT) and control systems can be designed as well.  Model-based optimal experimental design techniques have been used to maximize the information content of experiments, allowing a single mechanism to be selected and its parameters to be estimated precisely, and the results used to scale up to a 500x larger flow reactor [3].  McMullen and Jensen [3] used a sequence of steady-state experiments in a fully-automated microreactor configuration to choose the best reaction mechanism, then fit a preexponential factor and activation energy.  It used relatively little experimental time and material due to automation and the small size of the microreactor. Still, ever faster and cheaper process development is desired to scale up and certify processes, and enter the market as soon as possible.  Here we present an evolution of the above concept, in which transient experiments are designed rather than steady-state ones. By modeling the behavior of the system at all times instead of only at steady state, any physically-relevant input functions (such as ramp functions for temperature and flow rate) can be used and measurements can be taken at any time (subject to the limitations of the measurement technology).  This eliminates the waste of time and material which occurs between steady states in sequences of steady-state experiments, thus allowing more informative experiments to be performed at less cost.  We present case studies with a pharmaceutically-relevant reaction to show for a fixed experimental cost, the degree to which transient experiments reduce uncertainty in (i) parameter estimates and (ii) product purity.  We also show for a fixed parameter precision, the degree to which experimental resources can be reduced by optimally designing transient experiments.

[1] Roberge, D. M.; Ducry, L.; Bieler, N.; Cretton, P.; Zimmerman, B. Microreactor technology: a revolution for the fine chemical and pharmaceutical industries? Chemical Engineering and Technology 28 (3), 318–323, 2005.

[2] Ahmed-Omer, B.; Brandt, J. C.; Wirth, T. Advanced organic synthesis using microreactor technology. Organic and Biomolecular Chemistry 5, 733–740, 2007.

[3] McMullen, J. P.; Jensen, K. F. Rapid determination of reaction kinetics with an automated microfluidic system. Organic Process Research and Development 15, 398–407, 2011.

Presenter(s): 

Checkout

Checkout

Do you already own this?

Log In for instructions on accessing this content.

AIChE MEMBERS

AIChE Member Credits 0.5
AIChE Members $15.00
AIChE Undergraduate Student Members Free
AIChE Graduate Student Members Free
Non-Members** $25.00
  • Type:
    Conference Presentation
  • Duration:
    30 minutes
  • Skill Level:
    Advanced
  • PDHs:
    0.50

Share This Post: