химия
АТМОСФЕРЫ
541.124
SAMPLING MEASUREMENTS OF NITRIC OXIDE IN METHANE-OXYGEN-CARBON DIOXIDE FLAMES WITH ADMIXTURE OF AMMONIA
© 2004 I. V. Dyakov, A. A. Konnov*, J. De Ruyck
Department of Mechanical Engineering, Vrije Universiteit Brussel, Brussels, Belgium
E-mail: akonnov@vub.ac.be Received 01.10.2003
Probe sampling measurements of the concentrations of nitric oxide in the post-flame zone of methane-oxygen-carbon dioxide flames with admixture of ammonia (0.5% of the fuel) are reported. A Heat Flux method was used for stabilization of non-stretched flames on a perforated plate burner at atmospheric pressure. Dilution ratios of oxygen, O2/(O2 + CO2), were varied from 0.26 to 0.35. Measured burning velocities for these flames agree well with those of the flames of methane-oxygen-carbon dioxide within an experimental accuracy. The concentrations of NOx were measured by means of a non-cooled quartz probe at different axial distances from the burner. The data are compared to modeling. The modeling over-predicts the measured concentrations of NOX however the experimental data trends are well reproduced. The effect of dilution of the flames by ambient air is stronger at smaller dilution ratios of oxygen. Including the downstream heat losses to the modeling does not remarkably alter the predicted concentrations of NOX.
ХИМИЧЕСКАЯ ФИЗИКА, 2004, том 23, № 8, с. 19-24
yffK
INTRODUCTION
Utilization of nitrogen-containing fuels leads to side effects particularly relating to the nitric oxide pollution. The concentration of fuel nitrogen, for instance in biomass, can be rather significant reaching up to 2 wt.%. Through the combustion process fuel nitrogen releases as volatile components mostly in form of ammonia and hydrogen cyanide and, in turn, converts to NOX. Fractions of different volatile nitrogen compounds during this process can vary depending on the sort of the fuel. The analyses of the product gases in the pressurized fluidized-bed air gasification of a number of the biomass feedstocks [1] indicate that the ammonia concentrations significantly exceed those of hydrogen cyanide.
Fuel nitrogen is more chemically active than atmospheric nitrogen [2]. During the combustion the routes of conversion of fuel nitrogen become dominant over those of the atmospheric nitrogen and thus, the formation of nitric oxide considerably increases [3]. Recent experimental researches using different techniques and conditions including premixed [4-8] and diffusion [3, 9] flames doped with different nitrogen compounds provide possibility of prediction for some particular cases [3], however the governing mechanism for fuel nitrogen conversion in flames is not yet completely understood [10].
A validation of detailed reaction mechanisms requires accurate measurements of NO* concentration in well-defined flame conditions. Particularly, temperature profiles through the flame front have to be measured or calculated employing reasonable assumptions
* Corresponding author.
on heat losses. These requirements can be met using configurations of counterflow premixed flames [11] or a method of the adiabatic flame stabilization on the flat burners introduced by de Goey et al. [12]. This method is based on balancing of the heat loss required for the flame stabilization by the convective heat flux from the burner surface to the flame front. It was demonstrated that the Heat Flux method is suitable for the determination of the adiabatic flame temperature and flame burning velocity [13-16]. This method also allows for sampling measurements of the NOX concentrations in the post-flame region of the premixed adiabatic flames [17-20]. The aim of this work was to study nitric oxide formation in methane flames with admixture of ammonia to simulate nitrogen containing fuels. Premixed adiabatic flames were studied employing a gas mixture of methane with (5000 ± 500) ppm of NH3. Carbon dioxide was used as an inert gas to avoid NOX formation from atmospheric nitrogen.
EXPERIMENTAL DETAILS
The experimental data were obtained in flames stabilized by means of the Heat Flux method on the perforated plate burner. The detailed descriptions of the method, burner and experimental set-up can be found elsewhere [12-16]. Only the most essential characteristics are described in this paper. Figure 1 represents the scheme of the burner. It consists of a plenum chamber and the burner head. The outlet of the burner head is the plate perforated with small holes, with six thermocouples soldered into the plate surface at the upstream side. The burner head has a heating jacked with thermostatic
19
2*
water supply. In these experiments it was kept at 368 K to avoid the cellular instability of the flames taking place at lower temperatures [21]. The plenum chamber also has an independent thermostatic jacket and its temperature was kept at 298 K. Thus, during the measurements the burner plate temperature is higher than the initial gas temperature and the unburned gases passing through it are heated up. If the flame is stabilized under sub-adiabatic conditions, the gas velocity is lower than the adiabatic flame burning velocity and the sum of the heat loss and heat gain is higher than zero, there is a net outgoing heat flux from the plate. Then the center of the burner plate is hotter than the heating jacket. If the un-burned gas velocity is higher than the adiabatic burning velocity (super-adiabatic conditions), the net heat flux is lower than zero and the center of the burner plate is cooler than the heating jacket. By changing the flow rate of the gas mixture an appropriate value of the gas velocity can be found to nullify the net heat flux. In this case the radial temperature distribution in the burner plate is uniform and equal to the temperature of the heating jacket. Experimentally the series of thermocouples attached to the burner plate allow for measuring the temperature distribution in it. The flow rate at which the net heat flux is zero is shown to be an adia-batic flame burning velocity [13, 15].
The gas supply system for the burner consists of three ducts for the fuel, oxygen and diluent gas, assembled on one panel. Each duct is connected to the appropriate gas cylinder and has a buffer vessel and a mass flow controller (MFC). Two water thermostats were used to provide the water supply to the thermostatic jackets of the burner. The coils of the gas ducts were immersed to the thermostatic bath at 298 K before the MFCs to uniform and control the gas temperature.
The experiments were performed with the mixtures with dilution ratios of oxygen, O2/(O2 + CO2), varied from 0.26 to 0.35.
Concentration measurements were made using a non-cooled quartz probe. It has a 0.9 mm inlet diameter, 6 mm external diameter and 1 mm wall thickness. The probe was introduced to the flame by means of the XFZ-position stage with the nozzle directed upstream along the burner axis. The sample gas was directed to the gas analyzers through the gas line via a conditioning unit, a membrane pump and a filter. Moisture was separated in the conditioning unit by rapid chilling at the dew point of water 5°C, without dissolution of gases in the liquid phase. Consequently, the measured concentrations were compared to the results of modeling recalculated to the dry basis.
The Fisher Rosemount Model 951A NO/NO2 chemi-luminescence analyzer was used for measuring the concentrations of NOX as a sum of NO and NO2, because in the present study the measurements of NO or NOX were indistinguishable within the experimental accuracy. Therefore the NOX concentrations measured in the post-flame zone presented here were attributed to the concentrations of NO. An overall accuracy in these measurements was better than 10%. Ranges of measurements, instrumental errors of the analyzer, properties of calibration mixtures, calibration procedure, probe sampling errors and possibility of conversion of NO to NO2 inside the probe were discussed elsewhere [16, 17].
MODELING DETAILS
A detailed C/H/N/O reaction mechanism for the combustion of small hydrocarbons was used for the modeling [22, 23]. The current version of the mechanism (Release 0.5) consists of 1200 reactions among 127 species. This mechanism has been validated with experimental data available for oxidation, ignition, and flame structure of hydrogen, carbon monoxide, formaldehyde, methanol, methane, ethane, propane, and some of their mixtures.
The CHEMKIN-II collection of codes [24-26], including transport properties [27] from Sandia National Laboratories, were used. Multi-component diffusion and thermal diffusion options were taken into account. The adaptive mesh parameters GRAD and CURV were respectively 0.05 and 0.5. Modeling was carried out for the free propagating adiabatic flame and for the flame with downstream heat losses. For the second case the calculated adiabatic temperature profile was modified downstream the flame front assuming a temperature gradient of 100 K/cm [28, 29] as:
T = Tad - 100X,
where X is the axial distance from the burner surface in cm. The modeling was performed with the given temperature profile using a "burner-stabilized flame" option of the CHEMKIN code.
RESULTS AND DISCUSSION
The adiabatic burning velocities measured for the studied gas mixtures agree well with the results of the modeling and those obtained for the mixtures of methane-oxygen-carbon dioxide [30] with the corresponding dilution ratios (Fig. 2). No influence of the admixture of ammonia on the flame velocities is experimentally observed. An accuracy of the burning velocity measurements (±0.8 cm/s, double standard deviation with 95% confidence level) and relative accuracy of the equivalence ratio (±0.54%) were estimated from calibration data for MFC's [16].
The nitric oxide concentrations were measured at fixed distances from the burner as a function of the equivalence ratio. The results of these measurements for the adiabatic flames with the oxygen di
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.