(424h) Characteristics of Ethylene Generation in the Mixing Region of Autothermal Reforming Applications

Mixture
preparation prior to a reforming catalyst is one of the important challenges in
developing reliable hydrocarbon reformers. Reactants should be mixed properly
before supplying it to the downstream catalyst, and unwanted gas-phase
reactions in the mixing region should be suppressed at the same time. Ethylene,
a well-known deposit precursor, is one of the major products from gas-phase
reactions, which can cause deleterious catalyst failure even with a small
amount as ~ 1,000 ppm¹. Therefore, it is essential to characterize
ethylene generation from gas-phase reactions before designing the mixing
chamber of a hydrocarbon reformer. Previous works^2-5 showed that
gas-phase reactions are closely related to mass and heat transfer; thus, the
coupled transport-kinetics was developed to describe the gas-phase reactions in
the mixing region of a hydrocarbon reformer⁶.

The mixing
chamber design described previously⁶ was used for the CFD calculations.
A mixture of fuel/steam vapor and air are supplied separately and mixed before
entering a downstream reforming catalyst. The initial autothermal
reforming (ATR) conditions had an O/C ratio of 1.6 and a
S/C ratio of 1.5. The flow rates of reactants were calculated with the
assumption that the reformer is designed for a 5kW_e-class system.

N-heptane was selected as a surrogate liquid fuel. The
gas-phase kinetic mechanism for n-heptane
used for the CFD calculations was based on that developed by Westbrook
and co-workers^7-9 and consists
of 654 species and 2827 reactions. The mechanism was reduced in size because the coupled transport-kinetics model is limited to 50 species. As illustrated in Figure 1, a reduced mechanism
consisting of 50 species and 98 reactions was found to be reasonably consistent
with the predictions of the original mechanism. This figure shows the
decrease in reactivity as the temperature increases from 450 to 550 ^oC. This region of decreasing reactivity
with increasing temperature is known as the Negative Temperature Coefficient
(NTC) region. The reasons for this unusual behavior are
discussed in previous
work¹⁰. Applying the reduced reaction kinetics, the
coupled transport-kinetics analysis was performed, and the calculations showed
that significant ethylene generation
is predicted when the inlet temperature is either below
450 ^oC or above 625 ^oC as described in Figure 2.

Figure 1 Product distribution of n-heptane oxidation modeling results under the nominal ATR condition (O/C=1.6, S/C=1.5): solid line (LLNL
mechanism) and dashed line (a reduced mechanism)

Figure 2 Predicted distributions of ethylene concentration
on the cross section shown in (a) at inlet temperature of (b) 450 ^oC
and (c) 625 ^oC

At lower temperature
(450 ^oC), the local O/C ratio (~1.9) is
higher than the overall O/C ratio (1.6) at the location of maximum ethylene
production. Therefore, the more oxidizing environment
promotes ethylene production. Figure 3 describes the results of the perfectly mixed
kinetic study using the reduced mechanism. The rate of ethylene
production increases as O/C increases. In this temperature range, most of the
ethylene is produced by the reaction C₂H₅+O₂ = C₂H₄+HO₂.

Figure 3 Perfectly-mixed
predictions of the reaction
rate profiles for ethylene production at
the inlet temperature of 450 ^oC (S/C=1.5)

At higher temperature (625 ^oC), the rate of
ethylene generation and the temperature change are much lower compared to the
lower inlet temperature case. The rate of ethylene generation peaks at two
points (before and after the nozzle). The results showed that the heat release rate is
moderate but coincident with the ethylene generation rate. Figure 4 describes the results of perfectly mixed kinetic calculations also using
the reduced mechanism. The effect of O/C is opposite to that predicted at
the lower temperature. At the higher temperature, pyrolysis chemistry is dominant, so
that the rate of
ethylene production decreases as O/C increases. In this temperature range, high-temperature chemistry is activated and alkyl radical
decompositions begin to contribute to ethylene production by the following
reactions: n-C₃H₇ = CH₃+C₂H₄ and n-C₄H₉
= C₂H₅+C₂H₄. Ethylene
production peaks in two locations in the coupled transport-kinetics
analysis. At the peak
upstream of the nozzle, the local reactant composition is O/C = 0.7. The more pyrolytic environment stimulates the ethylene production;
thus the finitely mixed case (266 ppm) showed more
ethylene production than the perfectly mixed case (114 ppm).

Figure 4 Perfectly-mixed
predictions of the reaction
rate profiles for ethylene production at
the inlet temperature of 625 ^oC (S/C=1.5)

The coupled transport-kinetics
analysis was performed to investigate the characteristics of ethylene
production in the mixing region of a hydrocarbon reformer. At lower
temperature, most of ethylene is produced by oxygen addition; thus, ethylene
production is stimulated with efficient mixing of fuel/steam vapor stream with
the air stream. On the other hand, pyrolysis chemistry dominates ethylene
production near 625 ^oC, and ethylene
generation can be suppressed by better mixing. The results suggest that the
complexities of the gas-phase kinetics demand that careful attention be paid to
design of the mixing region upstream of an ATR reformer to avoid undesirable
gas-phase reactions that have the potential to interfere with proper catalytic
operation.  

References

1.         Kang
IY, Carstensen HH, Dean AM. Impact
of gas-phase reactions in the mixing region upstream of a diesel fuel autothermal reformer. Journal of
Power Sources. Feb 2011;196(4):2020-2026.

2.         Yoon
S, Kang I, Bae J. Suppression of ethylene-induced
carbon deposition in diesel autothermal reforming.
Int. J. Hydrog. Energy. Feb
2009;34(4):1844-1851.

3.        Lindstrom
B, Karlsson JAJ, Ekdunge P,
et al. Diesel fuel reformer for automotive fuel cell applications. Int. J. Hydrog. Energy. May 2009;34(8):3367-3381.

4.         Porš Z, Pasel J, Tschauder A, Dahl R, Peters R, Stolten
D. Optimised Mixture Formation for Diesel Fuel
Processing. Fuel Cells. 2008;8(2):129-137.

5.         An
SM, Kim WS, Lee SY. Spraying of liquid fuel for improvement
of reforming performance for hydrogen generation. Int. J. Hydrog. Energy. 2011;36(9):5342-5349.

6.         Kim
S, Zhu H, Carstensen HH, Kee
R, Dean AM, Bae J. Coupled kinetics and transport in
the mixing region of a hydrocarbon reformer. AIChE. Salt Lake City,
UT2010:280.

7.         Mehl M, Pitz WJ, Sjöberg M, Dec JE. Detailed kinetic modeling of
low-temperature heat release for PRF fuels in an HCCI engine. SAE 2009 International Powertrains, Fuels and Lubricants Meeting.
Florence, Italy2009.

8.         Mehl M, Pitz WJ, Westbrook CK,
Curran HJ. Kinetic modeling of gasoline surrogate components and mixtures under
engine conditions. Proceedings of the Combustion Institute.
2011;33(1):193-200.

9.         Curran
HJ, Gaffuri P, Pitz WJ,
Westbrook CK. A comprehensive modeling study of n-heptane oxidation. Combustion and Flame. Jul 1998;114(1-2):149-177.

10.       Kim
S, Carstensen H-H, Dean AM, Bae
J. Investigation of gas-phase reactions in the mixing region for hydrocarbon autothermal reforming applications. Int. J. Hydrog. Energy. 2012;37(9):7545-7553.