(651c) Reduced-Order Kinetic Models and Reactor Optimization for Modular Shale Gas Utilization | AIChE

(651c) Reduced-Order Kinetic Models and Reactor Optimization for Modular Shale Gas Utilization


Ghosh, K. - Presenter, University of Notre Dame
Dowling, A., University of Notre Dame
In recent years, microkinetic modeling has grown in popularity as a means to elucidate surface catalytic reactions [1 – 3]. The molecular-level detail embedded in these mechanisms suggests that microkinetic models can facilitate reaction engineering, equipment design, and process intensification [4]. However, microkinetic models of catalytic processes are not fully utilized for design, optimization, and scale-up of industrial reactors. This is due to inherent difficulty of developing kinetic schemes, determining model parameters, validating resulting kinetic models, and integrating the key features into rigorous process simulations for reactor design. There remain opportunities for the development of multiscale computational approaches to link quantum chemistry, microkinetic reaction mechanisms, and rigorous nonlinear reactor optimization for systematic design of integrated chemical processes. The abundance of shale resources in the US presents one such unique opportunity. The stranded location of shale wells coupled with the uncertainty of gas supply from the wells necessitates research at molecular and process scales for the development of novel modular gas processing plants in order to realize the full potential that the shale resources present. The NSF Center for Innovative and Strategic Transformation of Alkane Resource (CISTAR) aims to answer this multiscale engineering challenge through the development of efficient, modular, and highly networked gas processing plants for the transformation of light hydrocarbons from shale resources to chemicals and transportation fuels.

Existing bottom-up multiscale modeling strategies are predominantly used to predict reactor behavior from microscopic scale calculations [5] which departs significantly from the empirical process design and control strategies of the past, whereby fitting to experimental data was essential to model building. These approaches naturally lead to unprecedented accuracy in process design, control, and optimization. Current efforts are focused on incorporating the molecular scale aspects of microkinetic model development into process scale aspects including catalyst material design and reactor design only for single-product reaction networks. Owing to the variability associated with feed supply and composition of shale resources, microkinetic modeling efforts in CISTAR for oligomerization of light olefins to heavier and high-value olefins [6] takes into consideration all possible reaction scenarios yielding multiple products from a given reactant to facilitate robust catalysis research. The size of the resultant kinetic differential equation system, however, is 2 orders of magnitude larger than system sizes that can be solved using current NLP solver capabilities.

In this work, we explore reduced-order kinetic model forms [7, 8] to incorporate molecular-level detail in reactor design to provide bottom-up and top-down systems analysis for novel shale gas processing technologies. As part of CISTAR, we seek to establish strong feedback loops between catalysis, separations, and systems engineering researchers and develop a multi-scale optimization framework for detailed reactor optimization and intensification that leverages microkinetic modeling [6], process synthesis [9], and systems analysis [10] to systematically help guide catalyst research.

We first seek to determine if existing oligomerization reduced-order kinetic models from literature adequately emulate microkinetic model predictions in an effort to reduce the size and complexity of the kinetic differential equation model to be used for detailed reactor design. We use weighted least-squares approach [11] to reparametrize the reduced-order model using a library of 64 microkinetic model simulations over a range of temperatures and feed conditions. Statistical uncertainty quantification is performed to determine the quality of fit, assess model uniqueness, and approximate model uncertainty. Using the fitted microkinetic model, we perform sensitivity analysis of product conversions to different reactor design conditions and compare the results to a Gibbs free energy minimization model. Finally, we use nonlinear optimization to compute the optimal reactor temperature profiles, catalyst loading, and feed compositions to maximize the production of high-value olefins to be used as gasoline and diesel additives. Orthogonal collocation on finite elements is applied to discretize the differential algebraic equation model into sparse algebraic constraints that are efficiently solved with Ipopt in the Pyomo modeling environment. As future work, we plan to embed the reactor design in a larger flowsheet optimization problem. Our goal, ultimately, is to design modular shale conversion systems that are robust to kinetic model uncertainty and operate over a wide range of feed compositions. Such modeling capabilities will help reinforce the feedback loops between systems analysis and catalysis and separations research in CISTAR.


  1. Salciccioli, M., Stamatakis, M., Caratzoulas, S., & Vlachos, D. G. (2011). A review of multiscale modeling of metal-catalyzed reactions: Mechanism development for complexity and emergent behavior. Chemical Engineering Science, 66(19), 4319-4355.
  2. Stamatakis, M., & Vlachos, D. G. (2012). Unraveling the complexity of catalytic reactions via kinetic Monte Carlo simulation: current status and frontiers. ACS Catalysis, 2(12), 2648-2663.
  3. Stamatakis, M. (2014). Kinetic modelling of heterogeneous catalytic systems. Journal of Physics: Condensed Matter, 27(1), 013001.
  4. Prasad, V., Karim, A. M., Ulissi, Z., Zagrobelny, M., & Vlachos, D. G. (2010). High throughput multiscale modeling for design of experiments, catalysts, and reactors: Application to hydrogen production from ammonia. Chemical Engineering Science, 65(1), 240-246.
  5. Raimondeau, S., & Vlachos, D. G. (2002). Recent developments on multiscale, hierarchical modeling of chemical reactors. Chemical Engineering Journal, 90(1-2), 3-23.
  6. Vernuccio, S., Bickel, E. E., Gounder, R., & Broadbelt, L. J. (2019). Microkinetic model of propylene oligomerization on Brønsted acidic zeolites at low conversion. ACS Catalysis, 9(10), 8996-9008.
  7. Oliveira, P., Borges, P., Pinto, R. R., Lemos, M. A. N. D. A., Lemos, F., Vedrine, J. C., & Ribeiro, F. R. (2010). Light olefin transformation over ZSM-5 zeolites with different acid strengths–a kinetic model. Applied Catalysis A: General, 384(1-2), 177-185.
  8. Young, C. T., von Goetze, R., Tomov, A. K., Zaccaria, F., & Britovsek, G. J. (2020). The Mathematics of Ethylene Oligomerisation and Polymerisation. Topics in Catalysis, 1-25.
  9. Ridha, T., Li, Y., Gençer, E., Siirola, J., Miller, J., Ribeiro, F., & Agrawal, R. (2018). Valorization of Shale Gas Condensate to Liquid Hydrocarbons through Catalytic Dehydrogenation and Oligomerization. Processes, 6(9), 139.
  10. Pacsi, A. P., Alhajeri, N. S., Zavala-Araiza, D., Webster, M. D., & Allen, D. T. (2013). Regional air quality impacts of increased natural gas production and use in Texas. Environmental science & technology, 47(7), 3521-3527.
  11. Rubert‐Nason, P., Mavrikakis, M., Maravelias, C. T., Grabow, L. C., & Biegler, L. T. (2014). Advanced solution methods for microkinetic models of catalytic reactions: A methanol synthesis case study. AIChE Journal, 60(4), 1336-1346.