Tackling industrial waste Cement kilns versus Incinerators - An environmental comparison

3. How is waste used in cement production?

3.1 How is cement produced?

The Life Cycle Assessment by TNO states:

3.3 Thermal treatment in cement kilns

Cement manufacturing consists of raw meal grinding, blending, pre-calcining, clinker burning and cement grinding. This is visualised in figure 3 . In short, limestone and other primary and secondary materials, containing calcium, silicon, aluminium and iron oxides are crushed and milled into a raw meal. This raw meal is heated in the pre-heating system, to initiate the dissociation of calcium carbonate to calcium oxide and carbon dioxide. The preheating system consists of one or more cyclones (mostly 3 or 4), and a pre-calciner. Part of the fuel is fed into the pre-calciner to keep the temperature sufficiently high. The temperature at this place is at least 600°C (up to 1000°C). After pre-heating and pre-calcining the meal is fed into the kiln, for further heating and reaction between calcium oxide and other elements to form calcium silicates and aluminates, at a temperature up to 1450°C. Most of the fuel is used to keep the temperature high enough in the burning zone for the chemical reactions to take place. The reaction products leave the kiln as a nodular material, called clinker. The clinker is than inter-ground with gypsum, limestone and/or ashes to the final cement product. A more extended description of the cement production process is given in Annex 2.

Source & ©: TNO  LCA of thermal treatment of waste streams in cement clinker kilns (2007),
3. Model description, 3.3 Thermal treatment in cement kilns, p. 16

 See also Annex 2 Cement clinker production


3.2 How can waste replace part of the fuel and raw materials used in cement kilns?

The Life Cycle Assessment by TNO states:

2.2 Substitution scenarios

The substitution scenario is based on the caloric value of the specific waste stream. The input of a certain amount of energy in the form of waste in a clinker kiln avoids the same amount of energy in the form of petcokes or coal. The substitution by coal is relatively low, compared to that of petcokes. In practice approximately 20% of the total fuel requirement is fulfilled by means of coal. As most wastes contain more minerals compared to petcokes or coal, the difference is compensated by a decrease in raw meal. Solvents and waste oils are an exception: they contain less minerals compared to petcokes and coal, and therefore a little more raw meal is needed.

Table 2. Substitution scenarios for each of the five wastes in clinker kilns 

Source & ©: TNO  LCA of thermal treatment of waste streams in cement clinker kilns (2007),
3. Model description, 3.3 Thermal treatment in cement kilns, p. 16-18


3.3 How does the use of waste change the outputs of cement production?

The Life Cycle Assessment by TNO states:

The production of cement clinker is energy intensive. The energy requirement for the production of clinker is approximately 3.5 GJ/tonne clinker produced in the dry process (and approximately 5 GJ/tonne in the wet process). Regular fuels are fuel oil, petcokes and coal. In order to meet financial and environmental standards, conventional raw materials and fuels are substituted by secondary streams. This study focuses on that waste streams or secondary streams that are used for energy supply as such. All these secondary fuels are fed into the cement kiln on the hot side of the kiln.

For the study, a linear plant model is used, which means that linear transfer functions between inputs and outputs of predefined species are defined. Knowing these functions enable us to relate changing inputs to resulting outputs. For example, if we know the transfer function of sulphur, we can calculate the change of sulphur emission at the stack as a result of the change of sulphur input, caused by a change in the fuel package. Although the process of clinker production is by far linear, in contrast to for instance incineration of waste in grate incinerators, the use of a linear model can be justified by the following reasons:

  • The used approach is the method of marginal changes (see 3.5). By application of this method, based on the extra input of 1 ton of specified waste, defined changes to the system are small and can thus be regarded linear.
  • The study focuses on species that, to some extent, have linear behaviour in the system. Exceptions are for instance sulphur and NOx.

For each kiln in the Belgian cement industry, inputs of chemical species are defined from raw materials and fuels. In addition, data of emissions to water, air, soil and product were retrieved. For each kiln, a set of transfer coefficients is determined based on these data. The transfer functions are not published in this report, as agreed with the expert panel, because of confidentiality reasons6

6 Not all expert panel members were happy with this (see also Annex 6). However, from confidentiality viewpoint this was the only solution. The confidential data and calculations were reviewed by Jürg Liechti of Neosys.

In the study, the main assumptions with regard to the treatment in cement kilns are:

  • As a result of the application of waste streams as energy supply, no additional changes to maintenance of the cement kilns are required. This means that from the base-case no change to the usage of materials is applied, for instance insulation and lubrification.
  • For NOx, no transfer model can be obtained due to the nature of the formation of this component. In fact 90% or more of the NOx that is formed in a cement kiln is not caused by the nitrogen of the raw materials and fuels, but by the high temperature conversion of nitrogen from the air. This 90% of the NOx emission is therefore process intrinsic, and does not depend on the usage of more or less waste as a fuel. This part is out of the scope of the study because of the approach of marginal change. As far as the 10% originating from the fuel is concerned, from measurements in practical situations, it is known that the use of waste streams leads to a decrease in the formation of NOx. These measurements were done, for example, at the ENCI plant in Maastricht. The actual data however are confidential. This means that in this study the assumption that the NOx emission will not increase, as a result of the use of more waste as a fuel, in fact is the worst case. So, in this study 90% of the NOx from the clinker kilns is assumed to be process intrinsic, and in not due to marginal changes. With regard to the remaining 10% it is assumed that the NOx emission caused by one ton of waste is the same as that caused by the same equivalence, in terms of caloric value, of petcoke or coal,.
  • Streams that are not leaving the plant, but are rather recycled or mixed with the final product, are not regarded as ‘waste streams’ or emissions. This concerns especially the fines from the E-filter. These fines are recycled into the product. Due to the choice of the system boundaries, only emissions outside the plant are taken into account.
  • All transfer functions are linear. For species streams, driven by temperature, this assumption holds. However, for species that are formed from chemical transitions, this approach fails. Due to the small changes to the base-case, linearity around the base-case situation is justified, and linear models for the transfer coefficients can be applied.

Source & ©: TNO  LCA of thermal treatment of waste streams in cement clinker kilns (2007),
3. Model description, 3.3 Thermal treatment in cement kilns, p. 16-18

 See also Annex 6 Comments of reviewer and panel members

The Three-Level Structure used to communicate this Life Cycle Assessment is copyrighted by GreenFacts asbl/vzw