Page 153 - Materials Chemistry, Second Edition
P. 153

140                                                    S. H. Gheewala

            Fig. 6 Proportion of energy
            inputs in the production of
            molasses ethanol (cradle-to-
            gate). a Total energy. b Fossil
            energy





























            even more modest than that of the cassava ethanol. The NEB for 1,000 L of
            ethanol is 3,350 MJ: once again positive but lower than cassava ethanol. Until this
            stage, the molasses ethanol seems to be doing slightly worse than that produced
            from cassava.
              However, the calculation of renewability shows a significantly different picture.
            The results of only fossil energy input are presented in Fig. 6b. Here, it can be seen
            that the major contributor to energy use, ethanol conversion, is absent because all
            the energy in this step is provided by rice husk and recovered biogas (from
            wastewater treatment). This has a significant effect on the renewability which
            amounts to 3.02, substantially higher than the NER of molasses ethanol and even
            much higher than the renewability of cassava ethanol. This in fact goes on to show
            the importance of the use of renewable energy sources in the life cycle, particularly
            in an energy-intensive step like ethanol conversion.


            5.2.2 LCA of Molasses Ethanol

            Figure 7 shows the contributions of the various life cycle stages to the potential
            environmental impacts of ethanol production from sugarcane molasses. Global
            warming, acidification, eutrophication, and human toxicity are 685 kg CO 2 eq,
                                     3-
            12.5 kg SO 2 eq, 19.55 kg PO 4 eq, and 19.11 kg 1,4 DCBeq, respectively. The
   148   149   150   151   152   153   154   155   156   157   158