A life-cycle inventory for the production of soap in Europe

The major use for soap in Western Europe is in toilet bars, which comprise a blend of short chain (C 12-14 , derived from coconut oil or palm kernel oil) and long chain (C 16-18 , from tallow or palm oil) fatty acids. These blends are produced by direct saponification of a mixture of the parent oleochemical esters, which also produces glycerine as a coproduct. It has been assumed for the purpose of this study that all soap in Europe is processed as a 20:80 blend of short/long chain materials from one of four possible combinations: Coconut/Tallow (CNO/Ta), Coconut/Palm oil (CNO/PO), Palm Kernel oil/Tallow (CNO/Ta). Coconut/Palm oil (CNO/PO), Palm Kernel oil/Tallow (PKO/Ta) and Palm Kernel oil/Palm oil (PKO/PO). On average, production of 1000 kg of each soap blend requires approximately 2300 kg raw materials (coconuts, fresh fruit bunches, tallow). Energy consumption and emissions associated with soap production are heavily dependent on boundary conditions. For the study, it was assumed that the starting point was fresh fruit buches for the palm oil and palm kernel oil, coconuts for the coconut oil and grass for tallow (via beef cattle). This leads to an overall energy of 46 to 54 GJ/1000 kg for soap blend production. The average energy associated with the tallow systems is somewhat greater than that for the palm oil systems (53.4 vs 46.7 GJ/1000 kg) mainly due to differences in process energy (19.3 vs 9.4 GJ/1000 kg). Major emissions for all blends are carbon dioxide (airborne) and BOD/COD (waterborne). Other significant emissions include SO x , NO x , hydrocarbons and particulates (airborne), and dissolved and suspended solids (waterborne). Significantly larger emissions of carbon dioxide, methane, ammonia, nitrous oxide, suspended solids and solid waste were associated with the tallow system. The environmental profile for tallow is determined by its production characteristics, as well as by the boundary and allocation criteria used for the LCI. It is obvious from the study that these are debatable and that different ones to those chosen may also be relevant. A full examination of this issue is beyond the remit of the present study although an assessment of the main implications is presented