(490c) Coupled Kinetics and Transport in the Mixing Region of a Hydrocarbon Reformer | AIChE

(490c) Coupled Kinetics and Transport in the Mixing Region of a Hydrocarbon Reformer

Authors 

Kim, S. - Presenter, Colorado School of Mines, KAIST
Zhu, H. - Presenter, Colorado School of Mines
Carstensen, H. H. - Presenter, Laboratory for Chemical Technology
Kee, R. - Presenter, Colorado School of Mines
Dean, A. - Presenter, Colorado School of Mines
Bae, J. - Presenter, Korea Advanced Institute of Science and Technology (KAIST)


Reformed liquid hydrocarbon fuels provide high energy densities for mobile power sources such as solid-oxide fuel cells in auxiliary power units. However there are numerous challenges in developing compact high-performance reformers. Achieving complete mixing of fuel and reforming agents (e.g., steam, CO2, air) is certainly one important challenge. Incomplete mixing with air can cause unacceptable temperature overshoots in the catalyst due to over-oxidation. Pyrolysis chemistry in the mixing region upstream of the reforming catalyst can lead to deleterious deposit formation. For example, ethylene is a known deposit precursor and even low levels will cause catalyst fouling. An important design objective is to achieve sufficiently complete mixing on a time scale that is short compared to gas-phase reaction times.

The objective of the present study is to develop and apply computational models to assist the design and development of gas-phase mixing chambers upstream of the reforming catalyst. A computational fluid dynamics model (ANSYS/FLUENT) is used to model the fluid flow and mixing for a class of forced-swirl mixing chambers. The model also includes detailed gas-phase chemical kinetics. For the purposes of this study, propane was used as a surrogate fuel. Although propane is somewhat less reactive than higher alkanes, the results provide quantitative insight about the performance of pre-vaporized liquid hydrocarbons.

            Figure 1 illustrates the general features of the mixing chamber. A compression-expansion section further enhances mixing prior to the flow exiting the mixing chamber through a porous ceramic foam. The chamber configuration was generated using ANSYS Design Modeler and the geometric characteristics were parameterized with 12 parameters, enabling efficient investigation of design alternatives.

            The chemical kinetics modeling began with a large reaction mechanism based on our earlier work but updated with improved hydrogen abstraction rate rules based on CBS-QB3 calculations. Additional improvements included updated potential energy surfaces for C3H7, C4H7 and C4H9. For these updates we performed electronic structure calculations at the CBS-QB3 level, calculated the high pressure rate constants with transition state theory and used QRRK/MSC theory to derive pressure-dependent rate expressions. However, because the FLUENT CFD model is limited to 50 species a reduced mechanism is required. As illustrated in Fig. 2, a mechanism consisting of 35 species and 237 reactions was found to be adequate. The flow-reactor modeling is accomplished with the CHEMKIN-Pro plug-flow reactor model. Figure 2 shows reasonable agreement between experiment and both the full and reduced reaction mechanisms.

            As illustrated in Fig. 1, flow exits the mixing chamber through a porous-foam structure. Thus, the foam characteristics are needed to develop the out-flow boundary condition in the CFD model. Using a modified Ergun equation, Lacroix et al.1 measured and modeled the pressure drop through SiC foams. The parameters developed by Lacroix et al. were implemented into the FLUENT model. Our predictions of pressure drop related to flow superficial velocity in their experiments showed good agreement.

Some preliminary CFD calculations were performed without propane homogeneous chemistry to examine the mixing behavior. Two results of CFD calculation are provided in Fig. 3. Propane distributions in designs A and B are dramatically different, and these results show the importance of the fluid dynamics within the mixing chamber. The major difference in the designs is the diameter of the entrance channel for the fuel/air stream (D1 in Fig. 1). The presentation will show the impact of mixing-chamber geometry and operating conditions on the extent of mixing and the pre-catalyst homogeneous chemistry. The simulations serve as a quantitative basis for developing general design guidelines.

References

1.      Lacroix M, Nguyen P, Schweich D, Huu CP, Savin-Poncet S, Edouard D. Pressure drop measurements and modelings on SiC foams. Chem. Eng. Sci. 2007;62:3259-3267.

    Figure 1. Sketch of a possible mixer design and its geometric parameterization.

   Figure 2. Product distribution of propane-oxidation experiments and modeling results.

    Figure 3. Propane mole fraction distribution for two different designs of the mixer

Topics