Download Network Devolatilization Models - Coal Combustion - Lecture Slides and more Slides Geochemistry in PDF only on Docsity!
Network Devolatilization Models
Class 7
NMR Data on Char
• Aromatic C/Cluster increases slightly
• Char aromaticity increases to 90%
• Aliphatic C decreases
• Attachments constant
• Attachments per cluster constant
• Bridges and loops increases at end of
devolatilization
• Crosslinking!
(from Fletcher et al., 23rd^ Combustion Symposium, 1990)
NMR Data on Tar
• Tar C aromaticity increases to 95%
at 1250 K
• Not as much H aromaticity increase
at 1050 K
• Decrease in 1-ring compounds in tar
at 1250 K
• Increase in 3+ ring compounds
• Minimal change at 1050 K
(from Fletcher et al., 23rd^ Combustion Symposium, 1990)
Tar MW Dist
0.0E+
5.0E-
1.0E-
1.5E-
2.0E-
2.5E-
3.0E-
3.5E-
4.0E-
MW
Mass (g)
Coal Structure
Pyrrolic Nitrogen
Pyridinic Nitrogen
Bridge Structures
Side Chain
Loop Structure
Aromatic Cluster
Mobile Phase Group
Bi-aryl Bridge
H
C
H 2
HO C
H 2
N
R
C
R
O
H
S
H 2
OH
C
H 2
H 2 OH
H 2
OH
CH 2
O
O
CH 3
C OH
O
R
C
H 2
HN
H H
H
HH
H 2
H 2
H 2
H 2 O
CH 3 O
C
H
O H 2
H
H 2
C
H H
H H
Primary Coal Pyrolysis
H
N
R
OH
C
CH 3
H 2
H 2
H 2
R
CH 3
H
O
HC H
CH 3
O S
C
CH 3
O
H 2 OH
H 2
H 2
H 2
N
CH 3
H H
Tar
R
CO (^2)
H 2 O
H 2 O
CO (^2)
CH 3
Tar
Network Devolatilization Models
• FG-DVC, FLASHCHAIN, CPD Models
• Parent coal structure
• Statistical network model
• First order rate expressions with
distributed activation energies
• Correlation of vapor pressure with tar
molecular weight
FG-DVC
FG-DVC Flashchain
Niksa & Kerstein, Energy & Fuels, 5, 647 (1991)
Flashchain
Chemical Percolation
Devolatilization (CPD) Model
Joint research effort
Sandia National Labs
University of Utah
Brigham Young University
Vapor-Liquid Equilibrium and Crosslinking
Finite Fragments (Metaplast)
Infinite Coal Matrix
Tar Vapor
Reattached Metaplast
Crosslinking
Vapor-Liquid Equilibrium
Labile Bridge Scission
MW
f
MW
f
MW
f
How Do You Treat Vapor Pressures
of Coal Fragments?
Generalized Hydrocarbon Vapor
Pressure Correlation for the CPD Model
Vapor Pressure (atm)
1000/Temperature (K
)
MW = 315 285
(667 K) (500 K) (400 K) (333 K)
P c cMW T
c i
vap
i exp^ /
3
Data taken from Gray et al. (Ind. Eng. Chem. Process Des. Dev., 1985) for
12 narrow boiling point fractions of coal liquids from a Pittsburgh seam coal
Vapor Pressure Model Compares Well with Pure
Component Data
800
700
600
500
400
300
200
100
Boiling Point (K)
50 100 150 200 250 300
Molecular Weight
Fletcher-Grant Unger-Suuberg .007 atm .08 atm 1 atm 10 atm
Input Parameters Required by the CPD Model
• Number of attachments per cluster (+1)
(i.e., coordination number)
• Fraction of attachments that are bridges (p 0 )
(bridges/bridges+side chains)
• Molecular weight per aromatic cluster (Mcl)
• Molecular weight per side chain (M)
• Fraction of bridges that are stable (c 0 )
Other Parameters
(not usually adjusted)
• Rate coefficients
– Assumed to be coal-independent
– Set based on extensive comparisons with data
– Uses sequential (not parallel) distributed activation energy
– A b , E b , b , A g , E g , g , A cr , E cr , (ratio of 2 A’s)
• Vapor pressure coefficients
– Assumed to be coal-independent
Good Agreement with Tar and Total Volatile Yields
70
60
50
40
30
20
10
0
Predicted Yield (CPD)
0 10 20 30 40 50 60 70
Measured Yield
Maximum Tar Total Volatiles
Coal-dependent input coefficients taken directly from NMR data for 16 coals (0.5 to 1000 K/s, 1000 to 1300 K)
Total Volatile Yield Increases with Increasing Heating Rate
Argonne Premium coals heated to 700 oC in helium with 30 s hold (Gibbins and Kandiyoti, Energy & Fuels,
70
60
50
40
30
20
10
0
Total Volatiles (% of daf coal) 1 10 100 1000
Heating Rate (K/s)
Illinois No. 6 hv bituminous
Wyodak Subbituminous
Pocahontas No. 3 lv bituminous
Reaction Temperature Increases with Increasing Heating Rate
Pittsburgh No. 8 hv bituminous coal in Helium (Gibbins and Kandiyoti, E&F, 1989). Lines are CPD model predictions (Fletcher, et al., E&F 1992)
Total Volatile and Tar Yields Decrease with Increasing Pressure for hv Bituminous Coals
Pittsburgh hv bituminous coal data from heated grid experiments, Anthony (1974) and Suuberg (1977), 1000 K/s to 1000 oC. CPD model predictions from Fletcher, et al. (1992)
Effect of Pressure on Low Rank Coal Devolatilization is Small
Zap lignite data from heated grid experiments, Anthony (1974) and Suuberg (1977), 1000 K/s to 1000 oC. CPD model predictions from Fletcher, et al. (1992)
What if the NMR parameters are
not measured for your coal?
Application in CPD Model (Sandia)
70
60
50
40
30
20
10
0
% Mass Release (daf)
Beulah Zap Blue #1 Illinois #6 Pittsburgh #8 Pocahontas #
Measured Correlations (^13) C NMR
105 K/s, 0% post flame O 2 , 13 C NMR data available
Application in CPD Model (Xu and Tomita)
60
50
40
30
20
10
0
%
Yield (daf)
65 70 75 80 85 90 95
% Carbon (daf)
limit of data used to make correlations
CPD mass release measured mass release CPD tar yield measured tar yield
17 non - U.S. coals, 3000 K/s to 1037 K,
No 13 C NMR data available
Summary of NMR Correlation
- Correlations work well for most coals
- Not an adequate replacement for detailed
13 C NMR analysis
- Reasonable predictions of tar and light gas
release may be expected when using
correlated chemical structure parameters
