Optimization of calciners in the cement industry

Since calciner technology was introduced in the 1960s, there has been no comparable leap in the pyroprocessing technology for the manufacturing of cement. Steinbiss in [1] provides a good overview of earlier developments in calciner technology, as well as a prediction of the economic prospects of burning residual materials. Apart from some innovations in the mechanical technology of grinding systems [2–8], there have been some developments in the field of burning technology as regarding calciner and cooler technology [9–15]. While the new clinker cooler designs primarily involve mechanical concepts, new calciner developments have often brought abaout process-relevant changes. This is mainly due to the fact that attempts to increase the usage of secondary fuels have necessitated the creation of new options for the fuel burnout and retention times in the calciner. Consequently, the calciner is currently often at the focus of process-technological developments as a complex pyroprocessing unit, in which two separate processes have to be coordinated, namely transformation of the fuel into energy and the calcination of the raw meal [16–20].

Since the introduction of calciner technology in the 1960s, a large number of process-technological concepts and new designs have been developed and the large number of proprietary names has led to a confusing variety of terms describing the available technology. Possible categories for type classification would be:

– structural form of a flash tube reactor or as an extended vessel

– process functioning with or without tertiary air

– integration into the pyro-process as a so-called inline calciner in the exhaust gas system of the kiln (with or without the tertiary air being previously mixed in) or as a so-called separate-line calciner that is only supplied with the tertiary air

– the type of air or gas supply splitting

– the type of combustion system and its suitability for burning large-sized solid fuels

The fundamental task of a calciner, namely decarbonization of the raw meal, is common to all kinds of calciners, but is performed differently by the different types of systems. The quality of the calcination process is shown by the uniformity of the raw meal decarbonization and in the achievement of a homogenous temperature distribution throughout the calciner at a minimum possible level. In various ways, the process quality largely depends on the flow characteristics of the media involved. Heat transfer, reaction transformations, material transport and pressure drop are directly dependent on how the media flow, where and how they are blended together and how their heat exchange takes place.

From the energetic point of view, the combustion in the calciner is a heat source and the calcination process is a heat sink. Efficient performance of the process is characterized by the heat source being in close proximity to the heat sink, so that no losses are incurred because of heat radiation and the released fuel energy is applied to the maximum possible extent for the decarbonization of the raw meal. In systems where no large-sized solid fuels are used, these needs of the process can be met by suitable infeeding of the mass flows of finely ground material. The closer to each other the injection points of fuel and raw meal are, the smaller will be the occurring temperature peaks and the more uniform will be the precalcination and consequently the smoother will be the kiln operation. In this sense, a calciner can be optimized to maximize the process efficiency by implementing corresponding design measures.

Due to the cost advantages presented by the use of large-sized secondary fuels, increasing employment is being made of, for example, shredded tyres, production wastes and biogenic residues. In such cases, the fundamental dilemma in the calciner designing  is that the ignition and burnout of the secondary fuels, which are often rather unreactive, have to be ensured by separating the calcination process, which acts as a heat sink, locally and time-related from the combustion process. This means that in systems for burning secondary fuels it is essential for the heat source and the heat sink to be separated. A good example for the nature of this problem is provided by the combustion chambers that have recently often been installed for burning large-sized solid fuels. Figure 1 is a diagrammatic representation of such a combustion chamber, whose design varies only very little from vendor to vendor.

In these combustion chambers a zone is created in the upper section of the cylindrical core area, in which the fuel can ignite with the largest possible amount of oxygen and at the highest possible temperature. In order to allow the use of fuels that are difficult to ignite, meal should be prevented from entering this zone because of the endothermic nature of the decarbonization process. In this zone, the generation of an open flame is required. However, at the same time the walls of the combustion chamber have to be protected against the resulting high temperatures, and for that reason they are screened from the flame radiation by a curtain of meal. This meal curtain is usually created by  several tangentially-introduced streams of tertiary air containing the meal. This design of combustion chambers thus does not implement the principle of closest-possible proximity of heat source and heat sink. It is therefore clear that the decarbonization of the meal is more heterogeneous, because the individual particles are exposed to different radiation and heat influences. In order to assure achievement of the required degree of precalcination, these combustion chambers are often operated at higher temperatures than normal, which can, however, lead to problems with local overheating.

This situation prompts the question of how the two contradictory demands – reliable burnout of the fuel on the one hand and uniform calcination on the other hand – can be efficiently combined in the calciner. The task arising is to transport the heat from the initially occurring fuel oxidation to the meal that has to be calcined. Naturally, flow and mixing processes play the decisive role here. For this reason, process-technological optimization is aimed at achieving the most suitable concentration and temperature fields within the calciner.

It is not necessary for the fuel to be immediately converted as a complete oxidation. Processes are also possible – and have been industrially implemented – in which the fuel is initially converted substoichiometrically and then fully oxidized, either subsequently or in a different process component.

On the basis of the great deal of experience that has been gained in recent years with a wide range of calcination processes, it has become increasingly clear that the processes taking place inside different calciners, which often look externally similar, rarely actually are identical. It is now a widespread perception that system and application-specific influences affect the process operation mode more strongly than previously supposed. This is due to the fact that the flow and fuel conversion conditions differ unexpectedly strongly, even in calciners that appear to be externally similar. If, for example, the momentum of the entering secondary fuel flow differs because it is fed into the calciner through a different design of injection system, or possibly a flap valve is installed several metres higher, then the suspension of the fuel in the gas stream also changes, so that the entire conversion conditions are different. As a consequence, there may be a shift in the local temperature fields that negatively influences the calcination. Such deviations from the optimum design point of a calciner can be proved with the aid of mathematical-physical modelling, such as CFD (Computational Fluid Dynamics). An analysis of the flow and concentration fields reveals suitable optimization measures that can be implemented as purposive system modifications.

The burning of large-sized solid fuels in the calciner is currently one of the greatest challenges for the engineers. A CFD-aided process analysis has therefore been conducted in order to optimize the combustion of shredded tyres in a flash tube calciner. Its objectives were to calculate the most suitable injection point, as well as the maximum possible feed rate and the corresponding optimum operating parameters.

This computation involves a considerable amount of complex modelling and simulation work. For calculating calcination processes in the pyroprocessing line of a cement factory, the options of commercial CFD codes have to be extended in order to enable the calculation of, for instance, calcination reactions and heat transfer by radiation. Moreover, further model extensions have to be implemented to incorporate the combustion of large-sized, non-spherical fuel particles.

Figure 2 shows the schematic sequence of the model for the computation of shredded tyre combustion. As usual, the oxidation takes place in two steps. As the particle temperature increases, the pyrolysis first takes place and the various hydrocarbons volitilize and are released from the particle. Their oxidation at the particle surface provides the source of heat that drives the pyrolysis process. In the second step the coke combustion takes place, in which the constituents remaining in the pyrolysis coke are oxidized more slowly. The dynamics of these two steps are characteristic for the respective fuels, or for their constituents, and naturally also dependent on the specific surface that is available for the reaction, meaning that it is ultimately also dependent on the particle size.

The pneumatic behaviour of the shredded tyres in the gas stream is characterized by the fact that individual particles can take up different orientations in the flow, which has consequences for their gas stream environment and thus ultimately for the transport of heat into the particles and for their pyrolysis behaviour. It is also known from the literature that the fuel particles break up at a certain stage of combustion, a fact that also has to be taken into account in the modelling.

The existing process is performed in an inline flash tube calciner (Fig. 3) with split feeding of fuel, meal and tertiary air.
The fuels employed in both the kiln inlet burner and in
the calciner are coal, fluff and MBM (Meat and Bone-Meal). Table 1 presents the most important volume and mass flows of the process.

The aim of the study was to identify the most suitable injec­-tion point for the shredded tyres and the maximum possible rate of shredded tyre use at the specified particle size, and to calculate the process parameters to be set for this purpose. For the computation, the shredded tyres were divided into 8 size categories from 5 to 50 mm edge length (in 2 dimensions) and each with a height of 10 mm. The applied net calorific value was 7450 kcal/kg. Table 2 shows the size distribution of the shredded tyres.

The criterion of maximum applicable feed rate was defined by three preconditions:

1. No shreds may fall through into the kiln inlet chamber.

2. The CO concentration at the calciner outlet must be under 100 ppm and the oxygen content must be in the usual range of approx. 4  %.

3. The shreds of the largest size fraction have to be 75  % burnt upon leaving the calciner.

In view of the fact that no uniform stipulations exist up to now regarding a generally tolerable degree of residual burnout, the third criterion is certainly open to discussion. The limit selected for the described calculation is derived from the assumption that the coke no longer has a sticky surface because of the advanced degree of burnout.

The following specifications were employed for the calculation model:

– Stationary flow field

– Turbulence model: realizable k-e model

– Transportation of heat by convection and radiation

– Mass transport for N₂, O₂, CO₂, CO, H₂O, volatiles from the coal and shredded tyres

– Finite rate/eddy-dissipation model for volumetric gas phase reactions

– Lagrangian particle tracking (separately for raw meal, coal and shredded tyres, as well as for their coke products)

– Momentum, energy and mass transport coupling between all particles and the gas phase

– Proprietary model for drag forces and heat transport between shredded tyres and gas phase

– Standard combustion model with additional consideration of the CO-formation in the gas and coke combustion

The decisive factor of the selection on the shredded tyre injection point is that the shreds are either immediately lifted by a sufficiently high gas velocity or, if they cannot be reliably lifted, that they are suspended in the gas stream at a point further upstream without falling into the kiln inlet chamber.

Figure 4 presents the velocity and temperature profile in the calciner. It shows the local zones of high velocity that are caused by injection of the fuels and their ignition. The temperature reduction downstream of the meal injection point is also ­visible, as is the compression of the flow on the opposite side. Consequently, the designated point is a favourable position for
feeding in the shredded tyres, as this point fulfils the requirements of high flow velocity of up to 60 m/s for preventing the shreds from falling into the kiln inlet chamber and high temperatures for supporting immediate spontaneous ignition of the shreds.

After the suitable injection point for the shredded tyres was identified, 20  % of the fuel energy requirement of the entire kiln line was substituted by shredded tyres as a first step. In this case that corresponded to a mass flow of 1050 kg/h. Figure 5 shows the resulting particle trajectories, which prove that the shredded tyres do not fall through into the kiln inlet chamber. Although some particles do not flow uniformly through the calciner immediately, but remain in wider sections of the calciner due to their current weight and the local flow conditions, these are finally carried to the outlet because of the decrease in mass that is associated with the progressing oxidation. The figure also shows how the flow pattern is particularly affected by the upper inflow point of tertiary air, although the additional intake of oxygen fosters the coke burnout there.

The fact that the streams of combustion air remain unchanged during the injection of shredded tyres into the process gives
rise to the question of whether the complete combustion can take place and whether the CO concentration is low enough to assure product quality and operational reliability. Figure 6
­illustrates the CO distribution inside the calciner when ­shredded tyres are being burnt. The zones of high CO concentration before the upper tertiary air inflow are conspicuous. At the calciner outlet the mean CO value over the cross-section is 80 ppm. Figure 7 shows the resulting oxygen distribution in the calciner.

The mean oxygen value of 1.3  % over the cross-section is so low that this process cannot be expected to remain stable and unproblematic when normal operating fluctuations are taken into consideration. It can also be assumed that with such an oxygen excess so low the larger tyre shreds in particular will not burn out adequately, so that further operating problems may arise. Table 3 presents the burnout rates as a function of the original particle size for the smallest, medium and largest particle fractions. These data show that although the volatile constituents of all the size fractions have completely burnt out before leaving the calciner, combustion of the residual coke of the largest particle fraction has not been completed.

To improve the process operation it was therefore proposed that the quantity of tertiary air should be increased up to a point where the coke burnout and the residual oxygen reach favourable values. Figure 8 shows the oxygen and CO concentrations after the tertiary air volume was raised by 40  %.

As demonstrated by Table 3, the increase in tertiary air volume produced the desired improvement with regard to coke burnout rates. The particles of the largest size fraction now have a burnout rate of 73  %.

In the cement industry, the increasing use of fuels with a low calorific value or a relatively large particle size has caused engineers to focus on the optimization of calciner processes. The increasing application of low-priced secondary fuels demands careful harmonization of the two coupled processes of combustion and decarbonization. One effective tool for computation and designing favourable process conditions is the mathematical-physical modelling method CFD. When employing CFD, the special features of both the cement manufacturing process and the special fuel properties have to be taken into account. Especially at the typical high loading rates, the chemical and physical properties of the solid fuels have a great influence on the process. Progress made in recent years on the field of flow calculation permits effective application of this method and therefore enables faster and cheaper process optimization, because the cost-intensive conversion measures in the production plant can be reduced by previous optimization of the process on the computer.