(540b) Toward Optimal NGL Conversion to Olefins: Advances in Steam Cracking Optimization

Authors: 
Onel, O., Princeton University
Niziolek, A. M., Princeton University
Floudas, C. A., Texas A&M University
Recent advances in the shale gas industry resulted in a significant increase in natural gas liquids (NGL) production [1]. NGL production increased by more than 100% since 2008 and is expected to grow by another 27% until 2040 [2,3]. In some wet shale plays, the combined ethane and propane composition can exceed 30% [4]. These recent trends motivate industrial efforts to build several new ethane crackers with a combined ethylene production capacity of 12.5 million tonnes/year in the United States [5]. Similar to ethane, other natural gas liquids can undergo steam cracking to produce olefins [6,7] with different underlying kinetics and mechanisms. This suggests that each hydrocarbon has a unique set of optimal operating conditions for the steam cracking process.

Although the steam cracking kinetics are well-established [6,7], several side reactions exist and the underlying reactor model contains a complex set of ordinary differential equations. In a stiff system such as this one, there is a need to develop a mathematical optimization framework to simultaneously address all trade-offs for optimal sizing operation. Therefore, the differential equations regarding mass, energy, and momentum balances are transformed using orthogonal collocation on finite elements [8]. In this work, the proposed reactor model also incorporates the kinetics of coke formation as well as decoking downtime. Finally, the reactor length is optimized to minimize the reactor volume and the associated investment costs. The resulting mixed integer nonlinear optimization model (MINLP) is solved to optimize the inlet flowrates, external heat flux, and reactor length (volume) to maximize the profit. The optimal reactor topologies and configurations are compared with the industrial cases, since the overall profit increased significantly. Finally, novel reactor designs are proposed that input a mixture of hydrocarbon feeds as opposed to a pure feed (ethane, propane, or butane). These optimal designs unravel the untapped opportunity of optimal allocation of NGLs as petrochemical feedstocks, while establishing a systematic framework for kinetic reactor design and optimization.

In order to illustrate the practical benefits of our systematic framework over a complete shale gas to olefins process, the optimal cracking designs are investigated in an integrated process superstructure [9]. Through the use of optimal steam cracking reactors that we developed, the net present values of these systems are significantly improved (>10%).

[1]: Floudas, C. A.; Niziolek, A. M.; Onel, O.; Matthews, L. R. Multi-scale systems engineering for energy and the environment: Challenges and opportunities. AIChE Journal 2016, 62 (3), 602-623.

[2]: Energy Information Administration, Annual Energy Outlook, 2015

[3]: Energy Information Administration, Petroleum and Other Liquids, 2015

[4]: Hill, R. J.; Jarvie, D. M.; Zumberge, J.; Henry, M.; Pollastro, R. M.; Oil and gas geochemistry and petroleum systems of the Fort Worth Basin, 2007, AAPG Bulletin, 91(4), 445-473.

[5]: Chang, J.; New projects may raise US ethylene capacity by 52%, PE by 47%, 2014, ICIS Petrochemicals.

[6]: Sundaram, K. M.; Froment, G. F.; Modeling of thermal cracking kinetics â?? I: Thermal cracking of ethane, propane and their mixtures, 1977, Chemical Engineering Science, 32(6), 601-608.

[7]: Sundaram, K. M.; Froment, G. F.; Modeling of thermal cracking kinetics â?? II: Cracking of iso-butane, of n-butane, and of mixtures ethane-propane-n-butane, 1977, Chemical Engineering Science, 32(6), 609-617.

[8]: Cuthrell, J. E.; Biegler, L. T.; On the optimization of differential-algebraic process systems, 1987, AIChE Journal, 33(8), 1257-1270.

[9]: Onel, O.; Niziolek, A. M; Floudas, C. A Optimal Production of Light Olefins from Natural Gas via the Methanol Intermediate. Industrial & Engineering Chemistry Research 2016, 55 (11), 3043-3063.