BIFURCATION ANALYSIS OF METHANE

OXIDATIVE COUPLING WITHOUT CATALYST

Vemuri Balakotaiah1*, Arun Kota

1, Sagar Sarsani

2 and David H. West

2

1University of Houston, Houston, TX 77204, USA

2SABIC Corporate Research & Development, Sugarland, TX-77478, USA

Abstract

We present a detailed bifurcation analysis of methane oxidative coupling in the gas phase (without

catalyst) using a global kinetic model for the various oxidation, reforming and dehydrogenation

reactions. The model predictions are compared with literature and new experimental data to validate the

approach. The model is used to determine the methane conversion (X) and C2+ products selectivity (Y)

under various feed and operating conditions. It is found that (X+Y) can vary in the range 0 to 200% with

the C2 product being mostly C2H2 at highest values of X+Y. The ratio C2H4/C2H2 decreases with higher

X and lower values of CH4/O2 ratio. The best C2+ selectivity of about 80% is obtained at methane

conversions of around 30% for adiabatic operation in a temperature window of 1200-1300K and on

ignited branches close to the extinction point.

Keywords

Oxidative coupling, ethylene, ethane, methane, reforming, dehydrogenation, adiabatic operation

Introduction

* To whom all correspondence should be addressed

Since the pioneering work of Keller and Bhasin [1],

oxidative coupling of methane (OCM) has been

investigated for more than thirty years both experimentally

and theoretically through kinetic modeling and reactor

simulations. However, most of these studies dealt with

OCM in the presence of a catalyst and only a few

considered the same in the absence of catalyst. Most prior

studies dealing with gas phase chemistry were also

restricted to reactor simulations for isothermal operation.

To our knowledge, there are no prior studies that presented

a complete bifurcation analysis (ignition, extinction and

autothermal behavior) of the gas phase OCM process. This

is the main goal of the present work.

It is often reported that the first step in OCM chemistry is

generation of methyl radicals on catalyst surface, while

subsequent reactions happen in gas phase [2]. While

debate continues on the extent of catalytic versus

homogeneous contribution in the subsequent reactions, it

is generally accepted that gas phase reactions may become

significant in OCM when the temperature exceeds about

873K but certainly important (and may be even dominant)

for temperatures above 1173K. Since the oxidation

reactions occurring in OCM are highly exothermic leading

to adiabatic temperature rise of 300 to 1200K (depending

on the methane to oxygen ratio in the feed), a detailed

understanding of the catalytic OCM also requires an

understanding of the contribution of the gas phase

chemistry to the overall process. This is the main

motivation for this study.

We consider various limiting homogeneous reactor

models (adiabatic CSTR, isothermal and adiabatic PFR

models) with global kinetics for various oxidation,

dehydrogenation and reforming reactions and present a

detailed bifurcation analysis. The focus here is on the

determination of the methane conversion and selectivities

of various C2 products (ethane, ethylene and acetylene)

and how these vary with the feed composition, operating

conditions (inlet temperature, space time) and the reactor

type.

Global Kinetic Model

While detailed models consisting of many species and

several hundred reaction steps have been proposed for gas

phase OCM, such models are not suitable for bifurcation

analysis. Here, we consider a simplified model consisting

of nine gas phase species (CH4, O2, CO, CO2, C2H6, C2H4,

C2H2, H2, H2O) and use a global kinetic model consisting

of three groups of reactions as follows:

0

4 2 2 6 2

0

4 2 2 4 2

0

4 2 2

2 2

(i) Parallel oxidation reactions

1(1) 2 42.26 /

2

(2) 2 2 67.38 /

(3) 1.5 2 124.1 /

(4) 0.5

R

R

R

CH O C H H O H kcal mole

CH O C H H O H kcal mole

CH O CO H O H kcal mole

CO O CO

+ → + ∆ = −

+ → + ∆ = −

+ → + ∆ = −

+ →0

0

4 2 2 2

0

2 4 2 2 2

0

2 6 2 4 2

67.64 /

(ii) Dehydrogenation/Pyrolysis reactions

(5) 2 3 89.943 /

(6) 32.772 /

(7) 41.692 /

R

R

R

R

H kcal mole

CH C H H H kcal mole

C H C H H H kcal mole

C H C H H H kcal mol

∆ = −

= + ∆ =

= + ∆ =

= + ∆ =

0

4 2 2

0

2 2 2

(iii) Reforming/Shift reactions

(8) +H O=CO 3 49.269 /

(9) + =CO 9.84 /

R

R

e

CH H H kcal mole

H CO H O H kcal mole

+ ∆ =

+ ∆ =

The global rate equations used for the last five (reversible

reactions) satisfy the thermodynamic constraints. The

kinetic constants for the oxidation reactions were taken

from the combustion literature and minor adjustments were

made based on experimental data on carbon selectivity at

low methane conversions. A similar approach was

followed for dehydrogenation (pyrolysis) and shift

(reforming) reactions.

Results and Discussion

In order to validate our kinetic model as well as

determine the conversion at equilibrium for isothermal as

well as adiabatic operation, the predicted conversion and

selectivity at large space times were compared with the

code of McBride and Gordon [3]. For example, figure 1

shows the equilibrium calculations for an isothermal case

with methane to oxygen feed ratio of 16. Similar

calculations and comparisons were done for adiabatic case

and other feed compositions.

Figure 1. Equilibrium conversion and selectivity of

various products for isothermal operation at 1 bar

(carbon/graphite formation excluded)

The validated model is then used to compare the

selectivity to various species for nearly isothermal reactor experiments. One such comparison is shown

in figure 2 [The experimental data shown in figure 2 is

obtained at SABIC]. Other similar comparisons with

literature data [4] were also made to validate the kinetic

model.

Figure 2. Comparison of model predictions with

new (isothermal) data at various methane to oxygen

ratios.

The validated global kinetic model is used with three

different types of reactor models (mentioned earlier) to

determine bifurcation diagrams of reactor steady-states as a

function of inlet fluid temperature for a fixed residence

time. The bifurcation diagrams were also determined when

residence time was taken as the bifurcation variable (and

inlet temperature is fixed). The main results, which will be

presented in the full manuscript may be summarized as

follows: (i) unlike prior literature claims, it is found that

the sum of methane conversion (X) plus C2+ product

selectivity (Y= sum of ethane, ethylene and acetylene

selectivity), X+Y can vary in the range 0 to 200% for

isothermal operation (which is difficult to achieve in

practice), (ii) at high values of X+Y, the main C2 product

is C2H2 , (iii) the ratio C2H4/C2H2 decreases with higher X

and lower values of CH4/O2 ratio in the feed, (iv) for the

adiabatic case, the best C2+ selectivity of about 80% is

obtained at methane conversions of around 30% for inlet

temperature window of 1200-1300K and on ignited

branches close to the extinction point, and (v) the

exothermic chemistry dominates at short residence times

and the endothermic at longer times.

Acknowledgements:

The work was at University of Houston was supported

by grant from SABIC Global Technologies.

References

[1] G. E. Keller & M. Bhasin, J. Catalysis, 73, 9 [1982].

[2] Tomoyasu I., Ji-Xiang Wang, Chiu-Hsun Lin & J. H.

Lunsford, J. Am. Chem. Soc., 107, 5062 [1985]

[3] B. J. McBride & S. Gordon, NASA-Lewis Publication 1311

[1996].

[4]. Chen, Q., Hoebink, H.B.J., & G. B. Marin, Ind. Eng. Chem.

Res. 30, 2088 [1991].