You are on page 1of 1

Reaction Constants in Underground Coal Gasification Simulation

Mojtaba Seifi, Zhangxing Chen, and Jalal Abedi Department of Chemical and Petroleum University of Calgary

1 BACKGROUND
Coal as one the fossil fuels currently provides 25% of the total energy demand of the world. It is being used either directly as a fuel in the furnaces or being gasified to a mixture of flammable gases that are mostly composed of H2, CO, and CH4. Most current technologies of coal gasification such as entrained flow, fluidized bed, and moving bed use a surface reactor for gasification. An alternative for surface gasifiers is an underground coal gasifiers. Several field designs, such as linked vertical well, steeply dipping coal seams, and retracting injection point (CRIP) have been operated in order to make UCG functional for different operating aspects such as a linking technique, avoiding heat and gas loss into adjacent formation, producing constant high quality gas, and so on [1]. A typical UCG process and relevant phenomena is depicted in Fig. 1. As shown, a UCG process is composed of several underground reactors. Cavities are filled with a mixture of produced and injected gases. Bottom of the cavity is filled with rubble materials. Moreover, there are several processes taking place around the surface of cavity.
Air/Oxygen Syngas Overburden

3 METHODOLOGY
HOMOGENEOUS REVERSIBLE REACTIONS
Modeling of homogeneous reversible reactions in porous medium approach is handled by two reactions, one for forwards and the other for backward direction, which requires application of two sets of kinetics. In this approach, equilibrium constant is not considered directly [4]. The proposed method to obtain the kinetics for the endothermic direction of these reactions requires:
Fixing the cavity pressure at 90% of hydrostatic pressure at the average depth of coal seam Providing the pressure-based equilibrium constant values as function of temperature Curve fitting of data of step 2 as an exponential form Providing the frequency factor and activation energy for exothermic direction of the reaction Finally, obtaining the kinetics of the endothermic direction using Eq. 1 and 2
(1)

HETEROGENEOUS REACTIONS
Kinetics of char with gaseous species is handled in different ways in literature, in the form of Eq. 3 [6], and in porous medium approach. In porous medium, it is defined per bulk volume of the domain, while in literature is based on surface area of char. Using the following steps, the kinetics of the heterogeneous reactions can be obtained:
Obtaining frequency factor and activation energy of reaction from literature in the form of Eq. 3 Determining the value of pore structure parameter (e.g. 5) Determining the value of surface area of char at initial state (complete pyrolysis condition), Fig. 4 Obtaining the average value of surface area during the whole process Finally, estimating the frequency factor using Eq. 4. Activation energy will remain the same
(3)

PYROLYSIS REACTION
Pyrolysis process can be modeled using simultaneousindependent reactions or single-step decomposition methods. Porous medium approach uses the latter method for pyrolysis modeling in which the whole process is simulated only with a single reaction. Applying the following steps, the realistic kinetics of the pyrolysis reaction can be estimated:
Obtaining a set of graphs for various values of kinetics using Eq. 5, e.g. as in Fig. 5, which indicates the relationship between frequency factor and activation energy Simulating pyrolysis test results using Eq. 5 and estimating optimized kinetics of pyrolysis [8], e.g. as in Fig. 6 Conducting several simulation runs for different proper sets of kinetics at different values of permeabilities Finally, selecting the set that produces realistic length of pyrolysis active region (less than one meter)

Original Coal

(5)
Fig. 1: Schematic of linear CRIP and underground cavities with relevant phenomena

(2)

(4)

2nd Reactor Original Wet Coal Evaporation Pyrolysis and Gasification Ash Layer Gas Layer (Film) Bulk Gas

1st Reactor

Rubbles

4 RESULTS
4500 CH4 4050
3

2 WHY / PURPOSE
Cumulative Produced Species, sm

7000

0.022

2.25
0.020

0.056 A0 (1/min) 1.00E+05 1.00E+06 1.00E+07 1.00E+08 1.00E+09 1.00E+10 1.32E+11 1.00E+12 1.00E+13 0.048 Seam Leopold (Upper Part) Experimental Simulated

CH4 CO CO H2 H2 CO2 CO2

6000
3

3600 3150 2700 2250 1800 1350 900 450 0 25

2.00

0.018 0.016

Cumulative Produced N2, sm

Weight Loss Rate, gr/C

5000

1.75 1.50 1.25

= 20

0.014 0.012 0.010 0.008 0.006 0.004 0.002

Weight Loss Rate, gr/C

0.040

= 5 C/min

From Simulation
0.032

The modeling of the UCG process has mostly been done by using surface gasifier approaches such as CFD method [1]. An appropriate alternative to the current models is application of a porous medium simulator. Due to the different natures of the processes of heavy oil in-situ TEXT combustion and coal gasification, there are variations in the implementation of the reactions in the corresponding models. In prior approach all reactions are treated as first order and Arrhenius type which implies that only a set of kinetic parameters, pre-exponential factor and activation energy, mimics the effect of all phenomena, such as bulk diffusion, gas film, intraparticle diffusion, and intrinsic reactivity [2]. The porous medium approach models the pyrolysis process as a single reaction. It takes into consideration the rate of the heterogeneous reactions per bulk volume of the whole domain, whereas literature data are available based on the surface area of the solid materials. Moreover, hydrocarbon reservoir simulators do not directly account for the equilibrium state of the reversible reactions [3]. In this study, procedures are proposed to obtain the frequency factor and activation energy for the pyrolysis reaction using the weight loss rate test on coal, the conversion of frequency factor data from the literature into simulator format for heterogeneous reactions, and a method that considers the equilibrium constant in the homogeneous reversible reactions. Proposed methods have been investigated through numerical simulation using STARS (Steam, Thermal, and Advanced Processes Reservoir Simulator) software of Computer Modeling Group Ltd.

4000

= 10

Ea = 105.0 kJ/gmol A0 = 1.00E+05 min-1

N2 N2 3000

S/S0

=5

Ea = 150.0kJ / gmol

0.024

1.00 0.75 0.50 0.25 0.00

0.016

2000

=1
=0

0.008

1000

0.000 200 300 400 500 600 700 800 900 1000 1100 1200 0.000 300 400 500 600 700 800 900 1000

0 50 75 100 125 150

Temperature, C

Temperature, C

Time, day

Fig. 5: Weight loss rate for activation energy of 150kJ/gmol at various frequency factors
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
1e+5 8e+4 6e+4

Fig. 6: Total degassing curve

Fig. 2: Comparison of cumulative production of species using proposed method (symbols) and Hossein et al (lines)

0.0

Conversion (Xc )
70 60 50 40 30 35

CH4

CO
30 25 20 15 10 5 0

Fig. 4: Specific surface area of coal for various pore structure parameters
Coal and Char Concentration, gmol/m 3

4e+4 2.5 day 2e+4 0 1e+5 8e+4 6e+4 4e+4 10.0 day 2e+4 0 1e+5 8e+4 6e+4 0 1 2 3 4 5 12.5 day 15.0 day 5.0 day 7.5 day

Species Rate, sm 3/day

20 10 0 40 35 30 25 20 15 10 5 0 0 25 50 75 100 125 150

Table 1: Frequency factor of heterogeneous reactions for porous medium approach of UCG simulation

40

H2

35 30 25 20 15 10 5 0 0

CO2

Reaction Type kg-char/m2-s-[atm]eij Oxidation Boudouard Steam Gasification


25 50 75 100 125 150

Pressure Order, eij mole-char/day-kPa 1.0 0.5 0.5 1.0 2.494E+14 2.469E+12 2.469E+13 2.329E+06

3.050E+03 3.000E+00 3.000E+01 2.848E-05

Distance, m
Coal (Fast Pyrolysis) Coal (Slow Pyrolysis) Char (Fast Pyrolysis) Char (Slow Pyrolysis)

4e+4 17.5 day 2e+4 0 20.0 day

Time, day

Time, day

Methanation

5 0

Fig. 2: Comparison of production rate of species using proposed method (dashed-lines) and Hossein et al (solid-lines)

Distance, m

Distance, m

Fig. 7: Coal and char concentration at various times for fast- and slow- pyrolysis models

Fig. 2 and 3 illustrate the cumulative and the rate of production of various species for two cross-sectional models. Activation energy of the homogeneous reversible reactions for these models are according to the proposed method and the method by Hossein et al [5]. in which activation energy was assumed to be the same for both forward and backward directions. As can be seen, there is higher production of methane, CH4, for former case, whereas carbon monoxide, CO, and hydrogen, H2, were produced in lower amount. Higher activation energy of backward direction of WGS and forward direction of MSR reactions prevents the conversion of CH4 and CO. Therefore, using the proposed method MSR reaction will have less effect on conversion of CH4 into CO and H2. The produced amount of nitrogen, N2, and carbon dioxide, CO2, were nearly the same.

Typical values of frequency factors for heterogeneous reactions using the proposed method are summarized in Table 1. The estimated values of frequency factors are in the same order of magnitude of those obtained by Hossein et al. Char density, char molecular weight, and average surface area (nitrogen based) are assumed to be 9 m2/gr, 12.01 kg/gmol, and 1200 kg/m3, respectively, which are typical values of Albertas coal seam [7].

Fig. 7 shows the coal and char concentration profiles at various times for two one-dimensional simulation models using different frequency factors for pyrolysis reaction. The activation energy for both models is 150.0kJ/gmol. Applied frequency factors are 1.00E+06min-1 and 1.00E+13min-1 for first (fastpyrolysis) and second (slow-pyrolysis) models, respectively. As can be seen, the whole domain was completely pyrolyzed for first case before time 2.5day, due to lower onset temperature, narrow temperature range, and higher rate of evolution. Whereas for second case, approximately first half meter of the domain was partially pyrolyzed. Moreover, the length of currently active pyrolysis region in second case is about one meter, determined from zero to maximum value of coal concentration, which indicates that the applied kinetics for slow-pyrolysis can be considered as more reliable kinetics for pyrolysis of the specified coal.

1. G. Perkins, Mathematical modeling of underground coal gasification. PhD Dissertation, The University of New South Wales, 2005. 2. Computer Modeling Group. STARS Technical Manual. 3. M. Seifi, Z. Chen, and J. Abedi, Numerical simulation study of underground coal gasification using the CRIP method, The Canadian Journal of Chemical Engineering, Dec. 2011, 89- 6. 4. J. M.Smith, H. C. Van Ness, and M. M. Abbott, Introduction to chemical engineering thermodynamics, Sixth edition, The McGraw-Hill Companies, Inc., 2001.

5. H. Nourozieh, M. Kariznovi, Z. Chen, and J. Abedi, Simulation study of underground coal gasification in Alberta reservoirs: geological structure and process modeling, Energy Fuels, Vol. 24, pp. 3540-3550, 2010. 6. S. Parkash and S. K. Chakrabartty, Microporosity in Alberta plains coals, International Journal of Coal Geology, Vol. 6, pp. 55-70, 1986. 7. R. J. Richardson and P. Geol. CanZealand, Albertas 2 trillion tonnes of unrecognized coal, Alberta Innovates-Energy and Environmenta Solutions, Aug. 2010. 8. M. Mostade, The Thulin underground coal gasification field test in Belgium: Summary of resources and gasification data, Oct. 2012.