A Computational Model of Catalyzed Carbon Sequestration: Back To Exploration - 2008 CSPG CSEG CWLS Convention
A Computational Model of Catalyzed Carbon Sequestration: Back To Exploration - 2008 CSPG CSEG CWLS Convention
A Computational Model of Catalyzed Carbon Sequestration: Back To Exploration - 2008 CSPG CSEG CWLS Convention
472
injection well drilling expense and maintenance, public objection to being in proximity of a storage
field of a toxic gas that may leak, and the expense of developing the technology to treat a leaking
storage field.
In this paper we describe a computation-based investigation into strategies for the removal of
carbon dioxide and methane directly from the point of emission, a process known as direct carbon
sequestration. We approach this problem using standard and innovative numerical methods to
efficiently solve the potentially very high-dimensional systems of differential equation (e.g., models
of realistic systems might simulate the evolution of a thousand or more chemical species
simultaneously).
In direct carbon sequestration, the chemical inputs must be abundant on-site, and the products
must be environmentally benign or have commercial value. For example, carbon dioxide from
combustion, atmospheric nitrogen, and methane from coal are readily available at coal-fired
electrical generating stations. Generation of thermogenic methane from coal begins in the higher
ranks of the high-volatile bituminous coals, and at about 250F, generation of methane exceeds
generation of carbon dioxide. Maximum generation of methane from coal occurs at about 300F.
With even higher temperatures and higher-rank coals, methane is still generated, but at lower
volumes [1, 2]. The volatile component of coal can be 15 35%, most of which is methane [3].
Thus in the case of coal, direct carbon sequestration appears feasible.
Theory
The reaction between methane and carbon dioxide yielding soot or graphite and water is
thermodynamically favorable [4]:
CO2 + CH4 2C(gr) + 2H2O
(1)
Hydrogen generation is also possible from a direct carbon sequestration reactor designed for
coal-fired electrical generating stations [4]:
CO2 + 3CH4 4C(gr) + 4H2 + 2H2O
(2)
Equation 2 above is more thermodynamically favorable than the traditional synthesis gas reaction
which runs at 700 1000 oC over a nickel catalyst [4]. The temperature of the flue gas at the Otter
Tail coal-fired electrical generating station in Big Stone City, South Dakota, is normally between
300500 oF depending on the electrical load [5]. The temperature of the flue gas is relatively
constant from the bottom to the top of the stack due to the relatively high velocity of the exhaust
gases. This waste heat can be used to vary the temperature of a catalytic reactor without adding an
additional energy expense.
A commercially valuable synthesis gas can also be produced in a direct carbon sequestration
reactor:
CH4 + CO2 2CO + 2H2
(3)
This reaction is only slightly less thermodynamically favorable than the traditional synthesis gas
reaction.
Back to Exploration 2008 CSPG CSEG CWLS Convention
473
The carbon monoxide produced can be handled in a number of ways. For example, it can be used
in synthesis gas applications to produce specialty chemicals, it can be polymerized forming
polycarbonate, or it can be further hydrogenated or reacted with carbonaceous radicals on the
catalytic surface. Soot or graphite formation is desirable because both forms of carbon sequester
essentially just carbon, which could be compressed and shipped back to the coal mine on the coal
trains return trip. The synthesis of other commercially valuable chemicals is also possible.
Method
Equation (4) shows the general form of the differential equations that we want to solve where C is
the vector surface concentrations for all of the possible chemical reactants in the system, M is the
matrix of rate coefficients that govern the rate at which any two reactants merge to form a third, and
N is the matrix of rate coefficients for the rate at which individual reactant decompose to create two
reactants:
G
G
dc G G
= c Mc + N c
dt
(4)
For this system, we select the reactants that we want to consider in the system, calculate the rate
coefficients using methods previously developed [4] and solve the system using a stiff solver to
determine the systems capacity for sequestering carbon. The parameters depend on temperature,
pressure and the ambient reactant ratios in the surrounding gas. By modifying these parameters
we can modify the efficacy of the sequestration.
Experimental Results
The graphs below display some preliminary results. The first plot illustrates the coverages of
OCOH, which is an important reactant intermediate, and water, which is an indicator of the
chemistry occurring as a function of temperature. The horizontal axis is temperature in Kelvin
beginning at 300 K (approximately room temperature) to 550K in increments of 50 Kelvin. The
second plot shows CH3 coverage and the amount of water that has desorbed. These data were
collected after a simulation time of 0.1 microsecond. Data such as these allow us to study the
chemical processes that are occurring and make changes in the system to optimize the distribution
of reaction products.
500
400
OCOH
300
200
H2O
100
0
1
Temperature (Kelvin)
474
150000
100000
50000
H2O
0
1
Temperature
Conclusions
We have proposed a novel approach to perform direct carbon sequestration and have argued that
this approach appears feasible for coal-fired power plants. We have described a general
mathematical formulation of the problem which is flexible enough to apply to a wide variety of
scenarios. We have implemented this formulation and have run computer simulations that
demonstrate the kind of behavior we expect from a bare bones model of a simple system. We
have shown for a small system that the Arrhenius rate constant parameters for a set of chemical
reactions on a metal surface can be determined in a timely fashion. In future work, we will scale this
problem up to more physically realistic and more complex systems and develop the tools we need
to optimize the sequestration efficiency of such systems. We believe that large systems of such
equations can be solved efficiently on modern supercomputers to long time-scales, and that these
techniques can be applied to model catalyzed carbon sequestration.
References
1.
2.
3.
4.
5.
Rightmire, C.T., 1984., Coalbed methane resources, in Rightmire, C.T., Eddy, G.E., and Kirr, J.N., editors, Coalbed
methane resources of the United States: American Association of Petroleum Geologists Explorer, v. 21, no.4 (April,
2000), p. 16, 18-20, 22-23.
Rice, D.D. and Claypool, G.E., 1981, Generation, accumulation, and resource potential of biogenic gas: American
Association of Petroleum Geologists Bulletin, v. 65, no. 1, p. 5-25.
Ramage, J., 1997, Energy: A Guidebook, Oxford, Oxford University Press.
Sellers, H., 2007, Modeling the kinetics of large sets of reactions on metal surfaces: RussianJournal of Physical
Chemistry B, v. 1, no. 4, p. 377-393.
Private communications, Otter Tail Power Company engineering staff of the coal-fired electrical generating station in Big
Stone City, South Dakota.
475