OCR Text |
Show * .:• _ H . R ; 3 e fN.it - * ^3 3Q ; B _- 25 Weight Loss '0-1 5 5i (0 6 \ 3 -I 3_ \Tar V- •-OQO ----_•--fl ----------- 36 38 90 92 94 96 98 100 Carbon Content, daf wt. % =gure 3 Predicted devolatilization behavior of 17 petroleum cokes for heating at 10* K/s to 1600 K at 0.1 MPa. The nitrogen partitioning (top) resolves char-N (•) and tar-N (O), and the mass partitioning (bottom) resolves weight loss (•) and tar yieids(O). BIOMASS DEVOLATILIZATION Modeling Approach Previous attempts to adapt coal-derived depolymenzation •"odeis to piomass have so far met with limited success (Chen 1998) primarily for the following three reasons: (1) The multitude of relevant biomass fuels is usually charactenzed in terms of the cellulose fraction plus contnbutions from several poorly defined components, including iignm, hemi-celluloses, and xylans; (2) The macromolecular structure and compositions of biomass components contain few, if any, condensed polynuclear aromatic compounds; and (3) ash catalysis significantly affects both yields and product release rates, yet ash compositions are almost never reported. in addressing these difftculues to adapt FLASHCHAIN™ for piomass applications, w e were led to develop a model with far more differences than similarities to the basic reaction mechanism that is effective across the spectrum of coal and petroleum coke properties The new mechanism, called bio- F L A S H C H A I N ™ (bio-FC) is based on the following premises: (1) All biomass properties can be represented in terms of two sets of component properties, one set for cellulose and one set for a lignin-like component whose composition and mass fraction are assigned from the ultimate analysis of the whole biomass. Xylans and hemi-celluloses are ignored (2) FLASHCHAiN's straight-cham macrcr-oiec,ar configuration model is applied intact to biomass Chains are composed of two kinds of labile bridges, the original and dehydrogenated forms, plus refractory cnar links But the chains contain no aromatic nuclei or otherwise immutable units, so the connections between tne linkages are massless. (3) Ash catalyzes the conversion of bndges mto char imxs An abundance of ash tends to promote 'aster devolatilization rates but suppresses tar production while the total volatiles yields remain approximately constant. However, the extent of catalysis cannot pe correlated with total ash levels rather, it must ee inferred from a measured yield, sucn as the proximate volatile matter content The only mechanism in FLASHCHAIN™ that was unaltered for bio-FC is the flash distillation analogy for voiauies escape, in which a phase equilibrium relates the instantaneous mole fractions of like fragments m the tar vapor and condensed phase. No finite-rate mass transport phenomena are involved because all volatiles are presumed to escape in a convective flow that is initiated py the chemical production of noncondensibie gases The first premise is illustrated with the biomass procer.es in Table 2. These values were obtained by applying the submodel in bio-FC for fuel structure and composition to a database of 26 samples, including hardwoods and softwoods, vanous grasses and agncuitural residues, and one paper. The table reports the proximate volatile matter (PVM) and ash contents, both in daf wt. % Also shown are the number fraction of cellulose units and the number of H and O atoms per unit for the lignin-like component n all cases, the lignin-like component has 9 carbon atoms, and the cellulose composition is fixed by its molecular formula For biomass whose cellulose fraction is unity, the eeiiuicse composition was adjusted slightly to match the 'eocrec ultimate analysis. Table 2. Assigned compositions and cellulose number fractions for of a lignin-like eomporen: vanous forms of don-ass Form Wood Wood Wood Wood Wood Wood Wood Grass Grass Grass Grass Grass Grass Grass Paper Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. Ag. Res. PVM 86.1 91.3 87.1 86.1 867 83.5 87.3 81.5 82.4 82.3 81.4 80.2 85.6 839 89.3 890 842 82 9 80.2 73.3 854 80.2 88.0 83.0 80.7 81.6 rCEl 0.703 0.755 0530 0.437 0.564 0.582 0 640 0 778 0668 0.000 0.798 1.000 0 626 0.564 1 000 0 636 0 934 0.782 0.622 0535 0.292 1 000 1 000 0.655 1.000 0778 HUG 87 9.1 9.3 95 108 96 10 5 38 10.3 93 10.7 0.0 96 100 0.0 10 4 8 5 84 9.1 9.0 10.2 0.0 0.0 10.1 0.0 12.5 o L l G 3 0 30 30 40 30 30 33 4 0 30 5 3 20 0.0 50 3 0 00 50 50 4 0 40 60 60 00 00 30 00 1 0 Asr j 4 : 3 3 9 5 ' ] 3 : 9 • : c c : 5 4 5 5 8 • 9 a I 9 • ' 2 a - ] 9 2 9 : 9 i - - • 2 ; 5 : 5 a 5 9 • a • 13 5 |