Tackling industrial waste Cement kilns versus Incinerators - An environmental comparison

7. Conclusion - How can the results be interpreted and how reliable are they?

7.1 Are certain environmental impacts more important than others?

The Life Cycle Assessment by TNO states:

For further interpretation, the results are integrated to one indicator using standardized weighting methods to keep the integration step transparent:

  • The results of CML are integrated with shadowprices – more explanation on this method can be found in Annex 4.
  • The Eco-indicator 99 results are integrated using the default weighting set of 40% for damage to Human Health, 40% for damage to Ecosystems and 20% for depletion of resources.

To follow the ISO guidelines for interpretation, the calculation of the results needs to be transparent. Therefore the unweighed characterized results are presented in Annex 5. The Annex also reports the weighing factors per impact category and the methods used to derive these weighing factors. In addition to the results reported in graphs in this report, the weighed results are also presented in tables in Annex 5.

Source & ©: TNO  LCA of thermal treatment of waste streams in cement clinker kilns (2007),
2. Study approach, 2.3 Impact assessment, p. 11-12

 See also Annex 4 Shadow prices, Annex 5 LCA results, Annex 6 Comments of reviewer and panel members, Comments of Neosys, Weighing and the conformity of the study with ISO 14044


7.2 How would the results change if the waste or process were slightly different?

The Life Cycle Assessment by TNO states:

4.3 Sensitivity analysis

4.3.1 Variation in emissions of cement kilns

The emissions at cement kilns differ due to differences in the cement kiln process characteristics, like wet and dry process, etc. The annex reports for all results the minimum and maximum environmental impacts caused by the cement kiln emissions.

Figure 23  shows the minimum and maximum of environmental impact of emissions at cement kilns related to one ton input for the five waste streams, petcokes and coal and for raw meal, and the minimum and maximum environmental impact.

The small variation is caused by variation in emission of toxic elements (like metals, HF and SOx). However the emission of toxic elements contributes little to the environmental impact expressed in shadowprices, as most impact comes from the CO2 emission. The CO2 emission for 1 ton input does not vary between the cement kilns as CO2 emission is directly related to the carbon content of the waste which was the same in the calculation for all kilns. So, from an LCA point of view, the different cement production processes, including wet and dry processes, only differ marginally. Because of the ‘marginal change’ approach, in which the thermal treatment of one ton of waste is the functional unit, any mutual differences in energy efficiency between the cement kilns in Belgium cannot be made visible in this study. In the framework of this study, and also in the framework of the taxation of high caloric waste streams the question of energy-efficiency is not relevant either, because the energy-efficiency does not depend on the kind of fuel (primary or secondary), but on the cement clinker production process.

4.3.2 ±50% heavy metal content, S, CL and F in paint/ink

A theoretical calculation is made to find out if conclusions can change when metal, Sulfur, Chloride and Fluor content is different. As these elements contribute most in the results for paint/ink mixture, this waste stream is selected for sensitivity analysis. The sensitivity analysis is based on a theoretical variation of 50% higher and lower content of metals, sulfur, chloride and fluoride. This is a theoretical calculation as real variation in not known by TNO. However, 50% higher content of these elements would not °Ccur in practice as emissions limits would be exceeded.

Figure 24  shows the results on the environmental impact for 50% higher and lower content of toxic components in paint/ink. This variation has a small influence on the environmental impact of cement kilns, as these elements have a minor contribution to the result (see also paragraph 4.2.3).

However the water emissions in the rotary kiln have a relative large contribution to the environmental impact, and there a variation on toxic components has a large influence on the results. A decrease of 50% in toxic components in paint/ink would result in a similar environmental impact between cement kilns and rotary kiln for the pretreatment, transport and incineration of 1 ton paint/ink. However the environmental bonus for energy substitution is has not changed and is larger for cement kilns. Therefore, the conclusion that cement kilns are preferred from an environmental point of view does not change, even not with a 50% lower toxic component content.

4.3.3 -10% Carbon content in Fluff

Fluff is a mixture of plastics, textiles etc. Differences in composition may lead to differences in carbon content, resulting in a different environmental profile for fluff. In the sensitivity analysis, a theoretical calculation is made: what if the carbon content decreases 10% due to an increased paper fraction. That would result in:

  • A decrease in CO2 emission from 2300 ton CO2 per ton fluff to 2070 ton CO2
  • A decrease in caloric value from 21,6 to 19,1 GJ/ton, resulting in a decreased substitution: for Petcokes from 0,65 T/T to 0,58 T/T, for Coal from 0,64 T/T to 0,56 T/T, for Waste incineration from 21,6 GJ/ton to 19,1

Figure 25  shows the results of the sensitivity analysis on carbon content of fluff. A decrease in carbon content would lead to a decrease of environmental benefit of treatment in cement kilns and a decrease in environmental damage for treatment in waste incinerators. Although the differences between cement kilns and waste incineration become smaller, the conclusion does not change; even with a lower caloric content, the cement kilns have a better environmental performance compared to waste incineration.

4.3.4 -10% caloric value

Another topic of sensitivity analysis is the caloric value of waste: it may decrease as a result of increased moisture content. If 1 ton of waste contains 10% more water, this would result in:

  • 10% lower emissions (as the emissions are related to the waste and not to water)
  • 10% less energy substitution as a result of 10% less waste in a ton ‘wet waste’
  • Less energy substitution as a result of energy loss for water heating and evaporation

The results are shown in Figure 26 . Comparing these outcomes to the summary results in Figure 6 , it can be seem that the conclusions are still the same. This is of no surprise as the contribution of the energy loss of water evaporation is limited compared to the caloric content of the waste. 10% less waste per ton results in 10% lower emissions and a little more than 10% less energy substitution.

4.3.5 VOC content in paint/ink

The VOC emissions related to storage of impregnated sawdust are estimated based on the assumptions:

  • 10% average VOC content of paint ink mixtures as processed in the pre-treatment facility
  • 10% of these VOCs are emitted to air when stored in open silo’s or during transport.

A sensitivity analysis is made: do the conclusions change if VOC emissions as a result of open storage are doubled or 50% lower. The results are presented in Figure 27 .

When the VOC emissions related to storage of impregnated sawdust double, the environmental impact of pre-treatment and transport expressed in shadow prices increases from € 38 to € 58. When the VOC emissions are only 50% of the assumed emissions, the environmental impact of pre-treatment and transport decreases to € 28. As a result the net environmental result (after fuel substitution) varies. However the conclusions regarding the comparison with waste incineration do not change.

4.3.6 Shadow prices for CO2 and SOx

The shadow prices are based on Dutch policy. For SOx and CO2 Flanders has set shadow prices in previous studies, and these are used in a sensitivity analysis.

TNO uses a shadow price of 50 € /ton CO2 (based on Dutch policy). For CO2 several shadow prices (or marginal costs) can be used, depending on the policy, on future scenarios for economic development and the time frame. For CO2 the shadowprices range from 15 to 18 €/ton for low economic development and from 40 to 48 €/ton for high economic development in the BAU+ study from VITO [10]. In the EU trade market for CO2, the current price for CO2 is around 20 € per ton [11]. This value is used in the sensitivity analysis, as it is the lowest value that is used in Europe in long-term assessments, and it is close to the low range of CO2 shadowprices used in Flanders in other studies. If an even lower value would be taken, also the other shadowprices would have to be changed, because these values, as used now, assume long-term perspectives.

For SOx, TNO has used a shadowprice of 4 €/kg SOx equivalent. A Flemish study on environmental costs in Flanders shows that the Flemish goal on acidification can be met with a marginal cost level of 2,5 €/kg SOx eq [12]. This value is used in the sensitivity analysis.

Figure 28  shows the results when the shadowprices for global warming (CO2) and acidification (SOx) are adapted. When we compare these results to the results calculated with the original unadapted method in Figure 6, we see a smaller contribution of global warming and acidification to the overall result. However the conclusions on the comparison do not change. It can be deduced from figure 25  that, even if a value of € 15,- per ton CO2 would have been used in the calculation, and the contribution of global warming would decrease with 25%, the conclusion would not change.

4.3.7 Transfer coefficients for waste incineration

As some of the transfer coefficients available from the used ecoinvent data [9] on waste incineration are rather old – see Annex 3 – a sensitivity analysis is made using lower transfer coefficients for emissions to air and water. Table 5  shows that factors that are varied in this sensitivity analysis: the factor left of the slash (/) are the factors as reported in Annex 3. The factors right from the slash are applied in this sensitivity analysis for both the rotary kiln (air and water emissions) and fluidized bed (air emissions only). As the contribution of solid residues to the environmental impact of waste incineration is very small (maximum 0.5%), the increased concentration in solid residues is not taken into account in the calculation as this would hardly change the result.

Figure 29  shows the results for calculations with the Eco-invent transfer coefficients for waste incineration and for lower transfer coefficients for emissions to air and water as given in Table 5  (SA TC). Applying lower transfer coefficients for emissions to air and water results in a smaller environmental impact results in 2%, 3% and 5% reduced environmental impact for waste incineration for fluff, paint-ink and solvents respectively. The environmental impact of filter cake in waste incineration decreases 12%, and the environmental benefit of sludge in a fluidized bed doubles as a result of lower emissions to air. However when waste incineration of these five waste streams is compared to the environmental impact of waste in cement kilns, the conclusions do not change: even with lower emissions for waste incineration, the co-combustion in cement kilns has a preference from an environmental point of view.

Source & ©: TNO  LCA of thermal treatment of waste streams in cement clinker kilns (2007),
4. Results, 4.3 Sensitivity analysis, p. 33-39

 See also Annex 3 Waste incineration, Annex 4 Shadow prices

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