Conclusions
From various for agricultural residues of Thailand, this study selected five kinds that have the most potential as feedstocks to produce bio-DME, including rice straw, palm empty fruit bunches, cassava rhizome, sugarcane tops and leaves, maize stem. The LC-GHG of bio-DMEs are compared with that derived from lignite coal. The results indicated that bio-DME derived from sugarcane tops and leaves has the highest NER of 4.83 and lowest GHG emissions at 0.89 kg CO2e/kg DME. This is caused by its low GHG burden in feedstock acquisition and high heating capacity. When compared with that derived from lignite coal or fossil-based diesel, it was found that the bio-DME is promising to use as transportation fuel with GHG benefit as well as high NER. This is because it has the proper chemical properties and high heating value in addition to adequate supply availability. On the other hand, EFB and cassava rhizome show low possibility of utilization as DME feedstock due to their low intensity of cultivation and high moisture content which makes them not attractive in the view of GHG benefits and a large amount of collected and transported feedstock.
DME derived from low-range coal classified as lignite has NER lower than one (0.58) and emitted 3.90 kg CO2e/kg DME with 78% as direct combustion for generated heat to run the DME process. Moreover, when the GHG emitted from use phase of 2.60 kg CO2e/kg DME is included, the resulting value is higher than diesel on per energy (MJ) basis.
The practical solutions need to be identified for using cassava rhizome and palm EFB as DME feedstock. This study focused on reducing their moisture to decrease the overall GHG emissions in consequence. This could possibly be achieved by recovering waste flue gas from syngas production to dehydrate from both feedstocks or using conventional direct solar drying at drying yard for cassava rhizome, and the last option is to use the dry stripping process for separating EFB prior to delivering to the palm oil mill instead of the existing sterilization process at the palm oil mill itself. The results showed that some options would reduce moisture in feedstock significantly. Therefore, life cycle GHG emissions will be consequently reduced as well as energy efficiency increased i.e. palm EFB from dry stripping process can be reduced from 1.75 to 0.84 kg CO2e/kg DME or calculated as 52%, cassava dried by solar drier can reduce GHG emissions from 2.11 to 1.33 kg CO2e/kg DME or calculated as 37%. These are comparable to other agricultural residues for producing bio-DME.
Other improvement measures for reducing GHG emissions of bio-DME production, for instance, improving the feedstock transportation with increased bulk density by machine use, dehydration after residue is collected, reducing methane from rice fields, utilizing the ash from DME production and increasing yield of DME production with lower temperature and pressure conditions.