Page 317 - Materials Chemistry, Second Edition
P. 317

L1644_C07.fm  Page 288  Monday, October 20, 2003  12:10 PM










                                    7.7.2  LCI ANALYSIS
                                    In order to obtain the LCI data of this abbreviated waste incinerator process chain,
                                                      3
                                    the volume flow (Nm /a) of the waste incineration plant is multiplied by the unit
                                    emission files for the waste incinerator in the different cases, and the overall transport
                                    (tkm) is multiplied with the unit emission files for the transport. These unit emission
                                                                                     3
                                    files provide the emission of different air pollutants per Nm and per tkm, respec-
                                    tively. The previously mentioned data and the unit emission files for SIRUSA can
                                    be found in the case study sections of previous chapters; the unit emission files for
                                    the transport processes are presented in Table 7.7. The table only lists the pollutants
                                    for which impact indicators are available that are considered crucial for environ-
                                    mental problems related to transport.  The unit emission data for the MSWI for
                                    Scenarios 1 and 2 stem from the existing LCA study (STQ, 1998; see Chapter 2)
                                    and, for Scenario 3, from Hagelueken (2001; see Chapter 1).
                                       The most obvious result is that the  overall tkm increases significantly from
                                    Scenario 1 over Scenarios 2 and 3. Thus, the additional transport processes somehow
                                    compensate for the decrease of emissions due to the improvement of flue gas treatment.
                                    To what extent transport compensates for the damages avoided through more powerful
                                    gas treatment needs to be shown by environmental damage estimations.

                                    7.7.3  LCIA METHOD
                                    The HTP is calculated according to the CML method (Heijungs et al., 1992) and
                                    the EDIP method (Hauschild and Wenzel, 1998; the nomenclature of the latter is
                                    changed). Further explanations of the HTP can be found in Chapter 3. Table 7.8
                                    presents the HTP results for the three alternatives.

                                    7.7.4  DOMINANCE ANALYSIS AND SPATIAL DIFFERENTIATION
                                    No dominance analysis is necessary for this application. The medium is air because
                                    the impact factors have been calculated for that medium. The pollutants are those
                                    for which site-dependent impact factors have been determined in Section 7.6.4. They
                                    correspond largely to the pollutants identified as predominant in  Chapter 5. The


                                                 TABLE 7.7
                                                 Unit Emissions for Transport Processes
                                                       Pollutant         Emission (kg/tkm)

                                                 Benzo[a]pyrene             4.00 ¥ 10 -10
                                                 Cd                         2.10 ¥ 10 -9
                                                 NMVOC                      4.40 ¥ 10 -4
                                                                            3.12 ¥ 10 -3
                                                 NO x
                                                                            2.00 ¥ 10 -4
                                                 PM 10
                                                                            2.08 ¥ 10 -4
                                                 SO 2
                                                 Source: Kern, 2000


                                    © 2004 CRC Press LLC
   312   313   314   315   316   317   318   319   320   321   322