Biofuels is the common name for a large portfolio of fuels produced from biomass. Although some biofuel critics regards biofuels for climate mitigation purposes as a dead end, biofuels are also regarded as contributors to energy security in order for future societies to be less reliant on (import of) fossil resources [1
]. The fossil reserves are unknown, but estimates claim that peak oil will occur somewhere between 2008 and 2040 [2
]. The longevity of the reserves depends on usage rate, technology development and economy. Furthermore, a few oil and gas producing countries possess a large share of the total oil and gas production in the world, with consequences for both access to energy and distribution of wealth. Biofuels are thus also promoted as a means to create rural development and improve trade balance [4
]. In a resource perspective, biofuels cannot be the sole solution to the future transport sector, simply because the world cannot produce enough biomass to both feed and transport mankind without compromising sustainability, but they can represent a part of the solution. According to Smeets et al.
], with improved agricultural technology and geographically optimized use of land, the land area needed for food production can be reduced by 72%, hence releasing more area for other purposes. In a biomass resource assessment it was estimated that biomass could supply between 20%–50% of the transport demand in 2030 in EU27 [7
In an environmental perspective, the promotion of biofuels have trade-offs in terms of land and water use, competition for food resources, reduction of biodiversity and impacts related to other pollutants than greenhouse gas (GHG) emissions. As the biofuel discussion is mostly triggered by climate mitigation goals and reduction of GHG emissions, the expansion of native feedstock production for biofuels have been criticized for not taking into account impacts related to direct and indirect land use changes. Soil carbon emissions may create a carbon debt (i.e.
, GHG emissions which may eventually be accumulated in new biomass), with payback times potentially exceeding the climate mitigation goals [9
]. Therefore, the EU renewable energy directive and EU fuel quality directive [10
] sets environmental requirements for biofuels for the future, namely that biofuels must demonstrate at least a 35% reduction in GHG emissions and must not impose damage to sensitive ecosystems.
Life cycle assessment (LCA), a tool for assessing the environmental impacts from products and services, is under development to integrate the complex issue of soil emissions, and LCAs undertaken up to date demonstrate that biofuels done right may have benefits for the environment, but also that biofuels done wrong, may have negative impacts on the environment. Moreover, one may question whether biomass for biofuel purposes is the best use of the limited biomass resources, in a climate mitigation perspective, as alternative uses of the biomass have greater climate mitigation potential.
Decision-makers responsible for regulations or investments in biofuels need input on the environmental and financial viability of biofuels. The main objective of this article is to provide a literature review in order to rank three kinds of biodiesel fuels, namely transesterified lipids, biomasss-to-liquid (BTL) and hydrotreated vegetable oils (HVO), combined with selected feedstock sources in an environmental and cost perspective. The three chosen fuel types can be made from different feedstocks with large variations in environmental characteristics and costs. The study is limited to a selection of current feedstocks relevant for northern Europe. This implies that many relevant and promising future (non-edible) feedstock options are excluded, such as jatropha and algae. The perspective of the article is to determine, if possible, which of the chosen biodiesel and feedstock combination will be most viable, and not to determine whether biofuel is the most viable use of biomass. Therefore, alternative biomass sources are explored, but alternative uses of the biomass is outside the scope of the article.
In order to make informed choices about which biodiesel technology and feedstock combination to possibly pursue, their performance will be presented parallel to statistics on resource availability to indicate their potential to fulfill fuel needs of the future. A biofuel made from a waste resource may, for instance, show good environmental properties and low costs, but still be present in such low quantities that its presence in the market becomes negligible. Moreover, some biodiesel feedstock sources may have higher GHG emissions compared to fossil fuels. Biodiesel with such characteristics do not meet one of the main criteria for substituting fossil fuels.
The next section describes how the study is undertaken with a overview of chosen fuel and feedstock sources. Thereafter, the results of the feedstock and fuel ranking, together with numbers on fuel costs and feedstock availability, are presented in Section 3
, where also the results are discussed, with a special emphasis on the complex issue of climate change. Finally, concluding remarks are presented in Section 4
2. Methodology: Dimensions and Sources for Comparison
The goal of the study is to compare and rank three biodiesel fuel technologies that are available in the short and medium term and combine them with the best feedstock options available in the short and medium term, based on environmental assessments presented in literature. Biodiesel fuels were chosen due to the fact that diesel engines have a more favorable fuel economy than petrol fuels as compression ignition is far superior to spark ignition in a fuel efficiency perspective. This may prove important for a future transport sector with limited resource access. Furthermore, a diesel deficit is expected in Europe by 2015 [12
], and the development of viable alternative diesel fuels may also prove important for a European diesel fuel sector in urgent need for diesel refinery upgrading. However, diesel engines have demonstrated tailpipe emission drawbacks compared to gasoline motors, and local air pollution (due to
emissions) may be a problem connected to the expanded use of diesel engines.
Easily igniting compounds such as straight chain hydrocarbons (paraffins) are preferred in diesel motors [13
]. HVO and BTL are paraffinic diesel fuels with several fuel advantages over transesterified lipids. HVO and BTL have higher cetane number, implying easier ignition and more efficient combustion, lower cloud point, better storage stability, better cold properties, less tailpipe
emissions, have higher renewability fraction of the fuel (97%–98% renewable mass inputs versus 90% renewable mass inputs of transesterified lipids) and can theoretically be used in existing vehicles up to 100%, but in order to fulfill EN590, up to 30% can be blended into diesel [14
], compared to only 7% of transesterified lipids. A drawback for HVO is that lubricity is poorer [5
The three fuels have different properties such as lower heating values, and in production the fuel yield vary considerably, as shown in Table 1
Fuel properties [15,16,17,18].
| ||HVO||Transesterified Lipids||BTL|
|Typical Volumetric yield (%)||88–99||100|| |
|Typical Mass yield (%)||75–85||96||12–22 (wet input-30%)|
|Lower heating value (MJ/kg)||44||37–38||44|
|By product yield (mass)||5–14%||10%||1–13%|
Transesterified lipids are produced in a process where lipids and alcohol (typically methanol, but also ethanol) are blende with a catalyst (acid, base or enzyme), and fatty acid methyl ester (FAME) or fatty acid ethyl ester (FAEE) are formed. Glycerol is a byproduct [19
]. Inputs are oils, fats or fatty acids, steam, electricity, catalyst and methanol [19
]. The fuel quality depend on the parent oil or fat, the fraction of unsaturated fatty acids versus saturated fatty acids and the length of the fatty acid chain [21
The environmental impacts from using so-called transesterified lipids from vegetable oils have been discussed for a while, and second generation biofuels from lignocellulosic biomass have been launched as an even better alternative for the transport sector, as this does not compromise food production, have lower feedstock production impacts and higher fuel quality. BTL is produced via Fischer–Tropsch (FT) synthesis, where lignocellulosic materials such as wood is turned into fuel via gasification. The product gas is then cleaned and processed to form synthesis gas. The synthesis gas is then converted into long-chain hydrocarbons with the FT synthesis. Hydrocracking or hydrotreatment is required as a final refining step to produce the desired fuel. The process yield significant amounts of byproduct energy, and fuel energy yield is therefore relatively low (45%–60%). The mass fuel yield per mass input is low compared to the two other biofuels considered. The process is still in early stages of development and there are no commercial scale production plants established in Europe due to high plant investment costs, but several companies have launched interest in producing BTL from wood.
While profitability of BTL is investigated, HVO has been introduced as a feasible alternative to make high quality (paraffinic), high cetane biofuels from the same lipid feedstocks used for transesterified lipid production. HVO is a range of petrodiesel-like (paraffinic) fuels derived from biological sources, where double bonds and oxygen are converted to hydrocarbons by saturation of the double bonds and removal of oxygen (decarboxylation, decarbonylation, dehydration) [22
]. The reaction may require hydrogen, and saturated fats requires less consumption of hydrogen than unsaturated fats. Byproducts are light petroleum gas, propane and naptha that can be used internally for energy production. The main constituent of such petrodiesel-like fuels are alkanes. At present, HVO is produced in several countries, such as Finland (Neste Oil), Sweden (SunPine), Ireland (Conoco Phillips), Australia (British Petroleum) and Italy (UOP/Eni Ecofining).
Transesterified lipids and HVO may employ the same feedstock sources. There are several feedstock candidates, but here restricted to rapeseed, palm oil, tallow, used cooking oil (waste oil) and tall oil. Rapeseed can be cultivated in Northern Europe and is the most popular crop for transesterified lipids production in Europe. In addition, palm oil from Asia (Malaysia or Indonesia) is included as a possible feedstock. Tall oil, used cooking oil and tallow are wastes or byproducts that can be used for transesterified lipids and HVO production instead of higher priced refined vegetable oils. These feedstock sources have higher amounts of so-called free fatty acids and water, and may need pre-treatment before processing. For BTL production, different kinds of wood material can be used, hence BTL is different from the two other fuels with respect to feedstock. It is important to emphasize that results for BTL production are preliminary since the technology is still in early stages.
There are few studies comparing all three biofuels scrutinized here, and most studies compare only two of the fuels. No studies included all the chosen fuels and feedstock combinations. The ranking of the biodiesel fuels is mainly qualitative, based on individual studies where either different biofuel technologies or feedstock options are compared in several dimensions. Some studies compare only one fuel technology with different feedstocks, and from these studies a feedstock ranking can be attempted. Other studies compare technologies, and from these studies a technology ranking can be attempted. Combining information from these two kind of studies can give an indication on the best feedstock and fuel technology combination. As the feedstock has the greatest influence on the environmental feasibility, the studies that compare technologies cannot be solely used for the benchmarking. Furthermore, the availability and economic feasibility of a specific feedstock will be an important criterion when opting for a specific biofuel, and may limit the attractiveness of a specific biodiesel and feedstock combination. Even if individual studies differ in important methodological aspects such as goal of the study and choice of system boundaries, which may reduce the reliability of inter-study comparisons, a numerical comparison of GHG emissions across studies has been attempted.
Many LCA studies report GHG emissions, energy use and possible savings in total and non-renewable energy, and it is demonstrated that different energy indicators correlates well with most environmental life cycle impact categories [23
]. Where studies lack a full range of environmental indicators, the GHG and energy balances will have to serve as proxy measures for the environmental performance of the different fuels in order to perform a ranking. The fossil energy use does not correlate well with impacts due to land use, land use changes and toxicity impacts that may be important in biological systems, and detailed LCA studies are preferable. In the results section, GHG emissions are in focus, but other impacts are also reported. There are, however, reasons to be concerned whether GHG emission savings from substituting biofuels for fossil fuels come at the expense of other environmental impacts such as acidification and eutrophication. Moreover, no matter how environmentally friendly a product is, the market will not respond if the retail price is too high compared to alternatives. Hence, costs are important to reveal in order to secure that a specific biofuel and feedstock combination is a viable alternative.
Production systems often lead to generation of co-products, waste and by-products. This creates difficulties in allocating and assigning environmental loads to the outputs of the system and introduces complications in life cycle assessments. The definitions of byproduct, co-product and waste may differ from study to study. A co-product is generally considered a product produced along with the main product in comparable value as the main product. A byproduct is a product with lower economic value than the main product. Waste is defined as an output with little or no economic value and may turn into a byproduct if its economic value increases. Regarding life cycle assessment, the definitions are very important as a waste is often considered free of environmental charges, while for byproducts and co-products it is common to allocate the environmental burdens between different non-waste output streams. Byproducts and wastes may be recycled within the same production system, often referred to as close-loop recycling, or they may be recycled in alternative production systems, referred to as open-loop recycling. The latter alternative introduces difficulties in determining the quality of the raw material and whether it is able to replace virgin materials.
The assessment was undertaken as follows: First, the chosen feedstocks (palm oil, rapeseed oil, tallow, used cooking oil/waste oils and tall oil) for transesterified lipids and HVO were qualitatively ranked. Then transesterified lipids and HVO technologies were qualitatively compared based on the few LCA studies that compare these fuels. Next, studies that include woody BTL were compared to results for HVO and transesterified lipids. Lastly, an overview of GHG emissions per MJ of reported fuel and feedstock combinations is presented, as well as fuel costs.
4. Concluding Remarks
Based on the dimensions presented above, the most feasible biodiesel technology and feedstock combination, in an environmental and cost perspective, is likely to exhaust wastes and byproducts for HVO production first. These are the best ranked in all dimensions scrutinized here. Unfortunately, they have limited availability. To expand biofuel production and use, the next preferable feedstock is likely residual woody biomass for BTL. Its relatively poor performance regarding costs can probably be remedied through technological advancements. Both the ability to not compete with food production and its environmental performance makes it a likely better option than most agricultural feedstocks. Thus, finally one should revert to agricultural feedstock options such as rape and palm oil.
Looking at the lifecycle environmental impacts of the fuels, HVO do in most cases outperform transesterified lipids when assessing the same feedstock. Comparing HVO from different lipid feedstocks with woody BTL, the largest differences appear due to the fuel yield of the processes and the feedstock used. If HVO is produced from waste feedstocks with low upstream impacts, the high conversion efficiency of the HVO production process will make HVO the preferred fuel to BTL from any woody feedstock, even woody waste. Fuel yield and energy efficiency is low in BTL production. When native feedstocks as palm oil and rapeseed oil is used for HVO production, BTL is mostly superior due to low feedstock impacts and high energy yields per hectare of most non-waste woody feedstocks. On the other hand, using short rotation forestry for BTL production may in some cases make HVO from native feedstock with high yields, such as palm oil, more feasible. This is due to a combination of low energy efficiency and higher feedstock impacts of short rotation forestry than wood from waste and forestry. In a resource perspective, woody biomass is the most abundant biomass resource in Europe and in particular in Northern Europe, where large forest resources are located. Waste feedstocks available for HVO are sparse, but the availability and potential of agricultural feedstocks are higher. For HVO and transesterified lipids, improvements in agricultural practices may improve the overall environmental impacts for native feedstocks, and will be key to successful implementation of these.
All the challenges and uncertainties brought up in the discussion reduce the confidence in presented results. Many of them are often presented as methodological challenges that can be overcome by using standardized methods for, for instance, carbon releases associated with land use changes. There are, however, reasons to advocate precautions in a belief that any assessment tool will be able to present a correct figure for GHG emissions. Every forest or cropland, or even tree or plant, is specific and categorization can only give rough estimates. Impacts on biodiversity or land use are even more difficult to standardize and quantify for the very same reason. Together with the challenges related to how to allocate environmental burdens between products, byproducts and wastes, how to set system boundaries and how to choose the right inventory data, one can lose faith in ever getting trustworthy answers to what can be a sustainable path.
When including as many aspects as possible in a study and using conservative estimates for the benefits, such as many of the studies presented here are examples of, conclusive results will give a strong indication of the right choices. Hence, using HVO produced from used cooking oil, tall oil and tallow or BTL from forest residues are wise environmental measures for the transport system in a short term. HVO or transesterified lipids from agricultural crops, as well as BTL from short rotation coppice, must be investigated in each specific case to ensure that food production is not compromised and that GHG emission savings are real even when land use change is included.
In order to fully evaluate the sustainability or environmental feasibility of a biofuel feedstock, possible reference or alternative uses for the feedstock should also be evaluated. The difference between impacts in the use for biofuel purposes and alternative purposes should ideally be added or subtracted to the overall environmental impacts. Often there are several alternative routes for a waste or byproduct, and determining the most likely alternative use can be challenging and introduce large uncertainties. Furthermore, LCAs will have to adapt to the three-dimensional structure of sustainability evaluation to include social and economic aspects, and to address the challenge of “how can we in the future produce more with less” [143