OCR Text |
Show following thirteen dissociating species: H2, H2 0 , N2, C02' CO, C, N02, NO, 02, OH, H, ° and N. The computer program solves a system of 13 non linear equations with 13 unknowns (molar fraction of the different species); 9 of these equations come from the chemical equilibrium relations (with their equilibrium constants as function of temperature), 3 consider the mass conservation for carbon C, nitrogen N and hydrogen H and the last equation the conservation of mass of the total system (addition of the mole fraction of the different species should be equal to unity} The algorithm for solution is similar to that described by Agrawal and Gupta~ A trial value of the equilibrium temperature Te is assumed and the equilibrium constants are evaluated at this temperature. The mole fractions of the different species are expressed in terms of the mole fractions of 4 species, namely, H2, H20, C02 and N2; then, to find the composition of all the products at a given temperature, the mole fractions of H2, H20, C02 and N2 are initially assumed; expressing the mass conservation equations for C, N, H and the total mass conservation equation in terms of XH2, XH20' XC02 and XN2 and linearizing them results in a system of 4 linear equation with 4 unknown; the unknowns are dXH2, JXH20, ~XC02 and JXN2~ and the results expressed as absolute values of (Xi + JXi) are confirmed when the addition of the molar fractions of all the different species is equal to 1 or when dXi/xi < 10-7 where i represents the species taken as four variables. For a given set of conditions of the reactants (preheated air temperature, fuel temperature, excess air and air composition) and a given fuel composition, in order to obtain the adiabatic flame temperature, the program calculates the enthalpy of the reactants, then assumes a value of temperature for the products, and with that value calculates the molar fraction and number of moles of the different species, evaluates the enthalpy of the products and then verifies if this value matches the enthalpy of the reactants; if it does, the assumed value of temperature would be the adiabatic flame temperature, if the entha1pies' values are different, the program will assume a new value for temperature and repeat the described process again. The procedure outlined above was used to obtain the adiabatic flame temperatures and composition of combustion products for a variety of fuels and oxidant compositions ranging from 21% to 50% oxygen in air and preheat temperatures ranging from 100°F to 2000°F. The fuels reported onS were Birmingham natural gas, No.6 fuel oil and coal. However, the computer program permits the evaluation of the combustion of various gaseous fuels, such as methane, natural gas, coke oven gas, blast furnace gas, 189 low BTU gas, synthesis gas and other waste gases, various liquid fuels, such as No.2 oil, No.6 oil, methanol, etc., and various coals of different compositions. In order to validate the computational procedure and the system of equations employed, the mole fractions of the combustion species, obtained from the program were compared with those reported by Marteney6 for the stoichiometric combustion of methane-air mixtures at one atmosphere and were found to be valid. In the case of No.6 oil, limited information on adiabatic flame temperatures as a function of oxygen enriched air without preheat were checked against the results of Garrido et a17 and compared very well. The only reported work in the combustion of solid fuels is that of Samui10v et a18 who looked at the temperature dependence of the combustion species in the combustion of coke dust with air and pure oxygen. The comparison of the data with their work indicated that results were reasonably close. The variation of adiabatic flame temperature as a function of excess air and the preheat temperature for the combustion of regular air with Birmingham natural gas, No. 6 fuel oil and coal are presented in Figures 1, 2 and 3. Similar curves are available for 25%, 30%, 35%, 40% and 50% oxygen enriched air. As expected, the adiabatic flame temperature increases substantially with degree of oxygen enrichment. For example, with 25% oxygen enriched air preheated to 700°F with 10% excess air the adiabatic flame temperature is 43S0°F as compared with 3900°F for regular and at the same conditions. The only other observation from the data is that while the slopes of these curves are linear with regular air, they tend to increase as the preheat values reach above lS00°F with oxygen levels of 30% and above. This trend is illustrated in Figure 4. Note that at preheat temperatures of 2000°F and oxygen levels of 30% or higher, the flame temperatures reach values in excess of 6500°F. How these high temperatures can be handled by the burners and furnaces may be questionable. It must be pointed out that in the computer codes, the equation taken for the variation of specific heat with temperature was valid up to 6000°F. As the temperature of 6000°F is approached and beyond this temperature, errors could be introduced into the computations but since specific heat data was not available above 6000°F there is no way to determine the extent of any error. Figure 5 shows results of equilibrium values of NO as a function of the excess air for different preheat air temperatures in a combustion of natural gas with stoichiometric amounts of regular air. It could be seen that for a preheat temperature of 100°F,~he NO peaks at 12%, then NO diminishes as the percentage of excess air increases. This peak is not so noticeable with preheat air temperatures above 1000°F. Note that with 2000°F air preheats and 10% excess air, the equilibrium levels of NOx are |