OPTIMIZING THE COMBUSTION IN THE CALCINER OF A CEMENT PLANT

Relatively large quantities of thermal nitrogen oxide are formed in th e primary firing system of a rotary tube kiln during the clinker burni ng process because of the combustion gas temperatures of about 1800-20 00 degrees C required for an effective sintering process. This nitroge n oxide is transported into the calciner with the rotary kiln gases wh ere further nitrogen oxide is produced by the combustion of secondary fuels, with the result that NO concentrations equivalent to about 1.0 to 1.7 g NO2/m(3) are found at the end of the system. It can be assume d that this value will have Co reduced significantly in the future. A reduction can be obtained by of optimization of the primary firing sys tem in the rotary tube kiln and of the secondary firing system in the calciner. This article investigates the optimization of the combustion in the calciner by using multi-stage process control with stepped sup ply of fuel and air - which is state of the art in combustion processe s for power generation or in waste incineration. In a first step the m utual effects of the rotary tube kiln and calciner on each other are i nvestigated with the aid of partial and complete balances. This also s hows how the fuel is divided between the primary and secondary firing systems. The combustion in the calciner is then optimized, for which i t is subdivided into stages which are each considered as ideal stirred vessels. The effect of the stepped supply of fuel, kiln feed and air on the NO emissions is examined. A reduction rate for NO of about 65% is achieved when the calciner is divided into two stages with a substo ichiometric first zone. The temperature in the second stage is relativ ely low, so increased emissions of hydrocarbons and carbon monoxide ar e to be expected. As a result a three stage variant is therefore exami ned which contains in addition a super-stoichiometric zone with high t emperatures for complete burn-out. The temperature in this zone is adj usted through the proportion of raw meal supplied. This has only a ver y slight effect on the NO concentrations measured at the end of the sy stem. The NO reduction drops in comparison with the two-stage variant to about 52%. However there is a gain in the degree of freedom in rela tion to possible ways of influencing the burn-out. The theoretical inv estigations tend to show good agreement with experiments carried out i n parallel in a downpipe reactor.