Pollutant formation and int ... combustion of heavy liquid fuels

CHAPTER V. EXPERIMENTAL DETERMINATION OF THE AXIAL FLAME TEMPERATURE PROFILES IN THE DROP-TUBE FURNACE

1. Objective

The flame temperature is an important variable when considering the processes occurring in a flame. The conversion of fuel-N and especially the formation of thermal-NO_X are processes that are strongly influenced by the flame temperature, as is the interaction of nitrogen and sulphur species in the flame. Conversely, the flame temperature may also be affected by the interaction between these species.

In order to simulate numerically the formation of NO_X from the fuel, and also the interaction of oxides of nitrogen with sulphur compounds it is necessary to have accurate measurements of the flame temperature in the path of the gas species.

Initial attempts were carried out to estimate the temperature profile by measuring the output power into the furnace zones. Since the flame radiates energy towards the furnace wall, less current would be needed in zones where the flame is located, thus indicating its approximate length. However, the information so obtained was purely qualitative and, although the length of the radiant part of the flame could be estimated with some accuracy, it did not provide absolute values of the flame temperature.

2. Experimental procedure and conditions

The flame temperature was measured by means of an Inconel sheath, K-type thermocouple inserted vertically in the sampling probe. The diameter of the thermocouple was 3 mm. Although the large diameter was disadvantageous with respect to radiation heat losses, a robust thermocouple was needed in order to endure the aerodynamic conditions in the furnace. Heat losses in radiation from the thermocouple were compensated for at a later stage.

Another source of heat losses from thermocouples is conduction to cooler surfaces. Great care was exercised to place the tip of the thermocouple at a minimum distance of 90 mm from the top of the water-cooled sampling probe so that conductive losses onto the probe were minimised. Thus it was also possible to measure the flame temperature at very short distances from the atomiser, which are not reached by the sampling probe.

The fuels used were Orimulsion and the heavy coker gas oil M1. It was deemed that the latter fuel would represent the group of heavy gas oils used in this thesis. No large differences in flame temperature are expected to be caused by the different fuel-N and fuel-S contents in other fuels of similar physical and chemical characteristics.

The experimental conditions in the drop-tube furnace were similar to those in experiments reported in other sections of this thesis. Three stoichiometric ratios were investigated by adjusting the amount of secondary air: Fuel-lean ( = 0.833), stoichiometric ( = 1.000) and fuel-rich ( = 1.200). The amount of atomisation air was maintained at 23.2 g/min. In the case of Orimulsion, only combustion in fuel-lean conditions was investigated. As in other experiments the furnace wall temperature was 900 °C.

The objective of both the numerical modelling and experimental runs was to measure the emissions at a distance of 500 mm from the atomiser nozzle. The temperature profiles were obtained by performing vertical traverses from approximately 500 mm up to the closest achievable distance to the atomiser nozzle. The signal from the thermocouple was collected by the Orion data-logger and plotted versus the thermocouple tip position.

Table of Contents

Pollutant formation and interaction in the combustion of heavy liquid fuels
Luis Javier Molero de Blas, PhD thesis, University of London, 1998
© Luis Javier Molero de Blas