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

OBJECTIVE OF THE PRESENT WORK

The quality of heavy fuel oils continues to decrease world-wide. Although the total amount available is large, not only from refining operations but also as natural occurrences, difficulties for their use arise not only from the pitfalls of their manipulation but also as a result of their noxious combustion emissions, predominantly NO_{_X}, SO_X and particulates. These fuels become major contributors to the processes of acidification and eutrophication associated to transboundary pollution by virtue of the large amounts of nitrogen and sulphur compounds that they contain. As a result, strict legislation is being implemented that specifically limits the amount of sulphur contained in some heavy liquid fuel oils.

In this thesis, a number of heavy distillate fuels intended for gas turbine use are studied with respect to their NO_X and SO_X formation capacity. The fuels investigated are by-products of refining operations, and can be differentiated in two groups, namely heavy vacuum gas oils and heavy coker gas oils. The former are distillate fuels produced during the vacuum distillation of crude oil. Distillation at 30-40 mm Hg allows heavier hydrocarbons to vaporise at much lower temperatures than their boiling points at atmospheric pressure, thus avoiding thermal cracking. On the contrary, heavy coker gas oils are produced as a distillate stream during the delayed coking stage. This is a process of thermal cracking, in which heavy hydrocarbons are heated up to 520 °C and passed through a coke drum at 4 - 5 atm, where they break to produce lighter hydrocarbons at a low rate.

In addition, Orimulsion was another fuel studied in this thesis. This oil:water emulsion is a natural occurrence in the Venezuelan Orinoco basin. In spite of its controversial reputation, this 30 %-water emulsion has several advantages over other conventional fuels, such as ease of handling, transportation and processing, and environmental emissions comparable to those of other heavy fuels.

All of the fuels studied contain large amounts of sulphur and nitrogen (details on their composition can be seen in "Appendix II"). Nitrogen and sulphur in the fuel form NO_X and SO_X, which can interact during combustion, sometimes in unpredictable ways. Limited research has been reported in the past about the interactions between NO_X and SO_X. Furthermore, a large part of the work has been focused on the interaction of thermal-NO_X and oxides of sulphur. However, few and contradictory results have been reported on the effect of sulphur oxides on fuel-NO_X, ie nitrogen oxides originating from nitrogen compounds in the fuel.

As well as providing information about the fundamental NO_X formation processes, a knowledge of their interaction with sulphur compounds in combustion can assist devising pollution abatement strategies, similar to the way in which the design of low-NO_X burners was aided by the detailed knowledge of the NO_X formation processes.

It is the object of this thesis to provide experimental information and analysis of the processes whereby fuel-NO_X (NO and NO₂) and -SO_X interact in the combustion of heavy hydrocarbon liquid fuels.

Most of the experimental work was carried out in a semi-industrial scale drop-tube furnace. Although the conditions of combustion in the drop-tube furnace differ from those in a gas turbine (higher combustion intensities, higher pressures and shorter residence times would be encountered in the latter), considerable gains can be obtained from the understanding of the fundamental NO_X-SO_X interaction processes at atmospheric pressure.

In addition, numerical modelling is a tool that provided further information about the interactions sought. The Sandia CHEMKIN suite of codes can simulate different flow regimes, and be customised to suit the experimental combustion conditions, fuel input and reaction system. The effect of varying amounts of fuel-S and -N can be simulated by modifying the input to the code.

Five variables were important when determining the interaction effects between NO_X and SO_X:

The stoichiometry of the combustion environment, which was varied by modifying the amounts of combustion air present in the furnace
The flame temperature, which was varied by setting different furnace wall temperatures and equivalence ratios
The residence time in the combustor, by sampling at various locations along the furnace tube
The amount of total sulphur present in the combustion system: fuels with different S contents were studied, and increased amounts of fuel-sulphur were simulated with the addition of SO_2-gas
The amount of nitrogen present in the fuel, by studying fuels with different N contents

A series of complementary experimental techniques were used to obtain further information about the fuels studied. For instance, experiments with the single suspended droplet technique enabled the determination of their basic ignition characteristics. In addition, the size variation of a fuel droplet prior to ignition, the formation of soot or coke particulates from isolated single suspended droplet experiments were observed by means of video recordings.

Other information, such as the potential for NO_X and SO_X formation of the range of fuels studied, the axial flame temperature profile, the formation and emission of other combustion species and of thermal-NO_X also assisted to establish and validate the numerical model and provide further information about the interactions studied.

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