By Jason MacDonald
Biofuels have been touted by many politicians and scientists as the antedote to our fossil fuel dependence in the transportation sector. Energy from the sun will be soaked up by verdant fields and forests that can be harvested and converted to liquid fuels that will reduce our dependence on foreign oil, cut greenhouse gas emissions, and prevent more energy-related international conflict. And yet debate rages as to the effectiveness of these fuels to actually reduce fossil fuel consumption. In the present analysis, I attempt to look at some first-gen ethanol biofuels and identify where the discrepancies in the data arise.
Biofuels have been touted by many politicians and scientists as the antedote to our fossil fuel dependence in the transportation sector. Energy from the sun will be soaked up by verdant fields and forests that can be harvested and converted to liquid fuels that will reduce our dependence on foreign oil, cut greenhouse gas emissions, and prevent more energy-related international conflict. These renewable fuels can utilize the existing infrastructure and distribution network of liquid fuels and require little to no invention in transportation technology for their use. However, not all biofuels are created equal. This simple solution becomes more and more complicated as we consider the required land area to cultivate these energy crops, the effect it may have on food prices, and the fossil fuel required for their cultivation. The present discussion examines one such issue by focusing on the net energy yield of bioethanol production using a full fuel life cycle perspective to attempt to understand the lack of consistency in previously published fossil fuel energy consumption results (Hammerschlag 2006; Balat, Balat et al. 2008; Davis, Anderson-Teixeira et al. 2009).
In order to adequately examine the fossil fuel energy inputs, a focus on previous research investigating biofuels from a full fuel life cycle framework were examined. Figure 1 shows a generalized biofuel production system with energy inputs adapted from Davis, et al, 2009. The dotted line indicates what steps were included in the studies examined. The energy inputs for each phase are primary fossil fuel energy. The exception is the fuel processing phase, which does not specify fossil fuel as the input energy because some processes use left over biomass to power the fuel processing plant operations.
Bioethanol life cycle energy consumption varies widely among the literature. As expected it varies among feedstocks, but it also varies widely among studies examining the same crop. We will examine life cycle energy inputs of first generation bioethanol fuel processing from the two main feedstocks, sugarcane and corn, to determine the sources of this variation. To facilitate this comparison, energy values given in the literature will be converted to the commonly used energy ratio, re, metric. The energy ratio is defined as the ratio of the heat content of the fuel (in MJ/kg) to the non-renewable primary energy consumed to produce one kg of that fuel over the entire life cycle of the fuel (Gnansounou and Dauriat 2005).
We will first examine the most common form of bioethanol in the United States, ethanol derived from corn starch. There are two common processes that are used to synthesize ethanol from corn starch, wet milling and dry milling. Wet milling is a process designed to separate different products from the corn, including starch, oil, and specialty feed ingredients. Dry milling is a process in which shelled corn is hammer-milled. It yields starch for fermentation into ethanol and a mixture called distillers dried grains with solubles, which is often used as a feed supplement because of its high nutritional content (Graboski 2002). Examining the energy ratio results of the five articles that were found with sufficient data on process energy inputs, it is clear that there is a wide variation in calculated energy ratios for corn starch-based ethanol (see table 1), with the mean and median right in agreement near 1.3.The variation in results is more easily understood when examining process specific data. In order to make each dataset comparable, the specific process components required aggregation according to the life cycle frame work in Figure 1 and then converted to a unitless value that shows the relative strength of energy inputs. To create this value, each process energy data point (all of which were in different units), were divided by the available energy in an appropriate unit of ethanol production (see Table 2). This yields a unitless number that represents process fossil fuel input energy over ethanol energy.
The variation seen is largely due to the inclusion of sub processes that are not uniformly seen, inclusion of co-product credits, allocation of co-product credits, differences in crop yield due to location and size of sample data. It is also important to note that each of these datasets do not represent the same processes. Pementel and Hill are using Dry Milling data, Graboski and Shapouri are using weighted averages of wet and dry milling, and Dias de Oliviera does not specify the process that corn undergoes. A visual representation of the process energy data is shown in Figure 2, below. Both Graboski and Pementel do account for seeding and soil management (Fertilizer), but it is part of corn production. It is unclear if Hill and Pementel are lumping transportation of the feedstock to the biorefinery (Corn Transport) and distribution of Ethanol together. It is interesting that neither Pementel or Dias de Oliviera take into account any co-product credits, and also that Shapouri’s co-product valuation is so large.
The second feedstock that we consider for ethanol production is sugarcane. Producing ethanol from sugarcane does not require breaking down starches to create glucose to ferment. Instead, cane juice is separate from bagasse through pulverization, and then sent directly to ferment into hydrous ethanol. The bagasse is burned to provide power for the entire facility and usually is considered a co-product because more bagasse is created than can be used at a single site (Macedo, Seabra et al. 2008). Sugarcane is primarily used as a feedstock crop in Brazil and there is limited life cycle data available that discusses process level fossil energy inputs. We discuss two studies with a range of energy ratios from 9.3(Macedo, Seabra et al. 2008) to 3.7(Dias De Oliveira, Vaughan et al. 2005).
Comparing the datasets in Table 3 and Figure 3, there are two obvious reasons why Macedo and Dias de Oliviera disagree. First, the combined soil management and crop production of the the Dias de Oliviera study is almost triple that of Macedo, which Dias de Oliviera contends is a failing of Macedo’s. Second, Macedo does not include any distribution energy allocation, his study stops when the ethanol is ready to be shipped. Neither of the two overtly discuss co-products, despite the fact that sugar is refined alongside ethanol in Brazilian factories. Macedo says that the energy has already been allocated appropriately and does not have a transparent analysis available for perusal.
The figure below compares both sugarcane and corn ethanol sources side by side. It is very clear that is holding corn ethanol back is the energy inputs in the production phase. This is likely because there is no biomass burned to produce the energy for the plant processes and also because there is the additional energy intensive processing step of breaking up the starches into glucose. However, even in the worst case presented, ethanol still has a better energy ratio than, which suffers from an re of 0.76 because of all the upstream fuel costs associated with its production (Hammerschlag 2006).
Balat, M., H. Balat, et al. (2008). “Progress in bioethanol processing.” Progress in Energy and Combustion Science 34(5): 551-573.
Davis, S. C., K. J. Anderson-Teixeira, et al. (2009). “Life-cycle analysis and the ecology of biofuels.” Trends in Plant Science 14(3): 140-146.
Dias De Oliveira, M. E., B. E. Vaughan, et al. (2005). Ethanol as fuel: energy, carbon dioxide balances, and ecological footprint, Univ California Press. 55: 593-602.
Gnansounou, E. and A. Dauriat (2005). Energy balance of bioethanol: A synthesis.
Graboski, M. S. (2002). Fossil energy use in the manufacture of corn ethanol.
Hammerschlag, R. (2006). Ethanol’s Energy Return on Investment: A Survey of the Literature 1990−Present. 40: 1744-1750.
Hill, J., E. Nelson, et al. (2006). Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels, National Acad Sciences. 103: 11206.
Macedo, I. C., J. E. A. Seabra, et al. (2008). “Green house gases emissions in the production and use of ethanol from sugarcane in Brazil: The 2005/2006 averages and a prediction for 2020.” Biomass and Bioenergy 32(7): 582-595.
Pimentel, D. and T. W. Patzek (2005). Ethanol production using corn, switchgrass, and wood; biodiesel production using soybean and sunflower, Springer. 14: 65-76.
Shapouri, H., J. A. Duffield, et al. (2002). The energy balance of corn ethanol: An update, US Dept. of Agriculture, Office of the Chief Economist: Office of Energy Policy and New Uses.