Review Trends in Plant Science Vol.13 No.8 canthus x giganteus is also a C-4 perennial plant positive NEB can be considered as economically and conferring most of the advantages of switchgrass.Mis- environmentally sustainable.This is particularly import- canthus shows greater cold tolerance and hence might ant when considering which crops and conversion pro- perform better at higher latitudes.The yield of Mis- cesses might be worthy of substantial biotechnology canthus x giganteus has been reported to be between 7 investment.Even though the economics of corn starch- and 38 Mg/ha/yr and potentially has better nitrogen usage based ethanol and biodiesel production is currently com- than switchgrass [22,23]. petitive with gasoline,their NEB is fairly low or even Another group ofdedicated bioenergy feedstocks is woody negative,in contrast to the favorable NEB of lignocellu- plants,including hybrid poplar,willow and pines.Hybrid losics,as shown in Table 1 [10,28,29].If lignocellulosic poplar is considered a model woody biomass feedstock biomass can be efficiently converted into ethanol,a NEB because of its broad adaptation,available genome sequence of up to 600 GJ/ha/yr is a reasonable expectation,which and transformation techniques,and fast growth.The bio- would provide the highest NEB of all first or second mass accumulation of hybrid poplar is reported to bebe- generation platforms.Recent efforts to build biorefineries tween ~7 to 20 Mg/ha/yr depending on the nutrition and for lignocellulosic biomass processing are the first step to environmental conditions [24-26].From the perspective of fulfilling such potential;however,both low recalcitrance biomass production,switchgrass and hybrid Miscanthus feedstocks and new biocatalysts to improve the proces- seem to have the potential to produce more biomass com- sing efficiency are needed to realize this potential.Among pared with that produced by poplar.Given that a short the different bioenergy crops,switchgrass,Miscanthus, rotation for trees is five years,there is a time lag before and sorghum could potentially produce the highest NEB poplars can be harvested,and then,only the wood is har- 28.30.311 vested.Woody biomass does have a storage advantage over herbaceous feedstocks.However,geography,land-use pat- Environmental and ecological benefits of different terns,agronomy,economics and biology are likely to result platforms in multiple feedstock use.Because of the advantages of Different bioenergy platforms have different pros and cons perennial feedstock,efforts have been put into developing from an ecological and agricultural perspective(Table 1) perennial bioenergy feedstocks via breeding [27]. [41.The near-term economic advantages of ethanol pro- duction from maize and biodiesel production from soybean Plants for biodiesel are often counter-balanced by the detrimental effects of In temperate areas,annual oilseeds such as soybean agricultural practices on the environment.By contrast, (Glycine max),canola (Brassica napus),and sunfower perennial feedstocks such as switchgrass can help to (Helianthus annuus)have all been used as biodiesel feed- decrease soil erosion,improve water quality,and protect stocks.Palm oil (Arecaceae)trees have been successfully natural diversity [4,29,32-34]Perennial biomass crops used as biodiesel plants in the tropics.If we consider also complement food-based and feed-based agriculture potential biodiesel feedstocks for temperate use,the trans- instead of competing with it. portation costs for palm oil would be prohibitively expens- ive for export and would have a positive net energy Global climate change and bioenergy choices balance.In the case of soybean,canola and sunflower, Biomass crops and bioenergy production as an offset to the energy output from grain was estimated to be ~10 fossil fuel have the potential to ameliorate global warming. to 40 GJJ/ha,which is considerably lower than the ~200 Not only does the offset mean that less 'old'carbon is 500 GJ/ha energy gain from lignocellulosic biomass [28]. released into the atmosphere,but the underground bio- Hence,we might conclude that lignocellulosic biomass will mass of perennial biomass crops also acts as a carbon sink. have a greater demand than biodiesel feedstocks. For example,the capacity of Miscanthus x giganteus to fix There are other candidates for bioenergy feedstocks that carbon dioxide is estimated to be 5.2 to 7.2 t C/ha/yr,which are too numerous to detail in this review.Alternative results in a negative carbon balance where more carbon bioenergy plants include additional crops(e.g.sweet sor- dioxide is fixed than emitted [27].In a recent study of ghum),Camelina,grasses (e.g.big bluestem),trees (e.g. maize,switchgrass,soybean,alfalfa,hybrid poplar and willow),and even algae.Potentially,green algae could be reed canarygrass (Phalaris arundinacea),only poplar used for hydrogen production,oil production for biodiesel and switchgrass had a negative carbon balance(carbon platforms,and even biomass production for a bioethanol fixation of ~2t C/ha/yr)[35].However,two recent publi- platform,depending on the biotechnology breakthroughs cations indicate that first generation ethanol platforms Hydrogen is believed to be an important component of the actually have far higher carbon dioxide emissions com- third generation of bioenergy and can be adapted as differ pared with that released from fossil fuels [20,21].Accord- ent energy sources.Many factors determine the choice of ing to one of the studies,even the best lignocellulosic bioenergy crops;these are summarized in Table 1 and ethanol is predicted to have a positive carbon balance discussed further in the next section. 20].These contradictory estimates are a product of the different methods and models used to assess carbon Bioenergy:environmental,ecological and economic release and fixation.Despite the differences,there is a considerations consensus that second generation ethanol production Net energy balance of different platforms using lignocellulosic platforms should lead to a lower Net energy balance (NEB)is an important concept in carbon balance as compared with first generation plat- choosing a bioenergy platform because only a high forms.Enhancing the ability of perennial feedstocks to 424canthus giganteus is also a C-4 perennial plant conferring most of the advantages of switchgrass. Mis￾canthus shows greater cold tolerance and hence might perform better at higher latitudes. The yield of Mis￾canthus giganteus has been reported to be between 7 and 38 Mg/ha/yr and potentially has better nitrogen usage than switchgrass [22,23]. Another group of dedicated bioenergy feedstocks is woody plants, including hybrid poplar, willow and pines. Hybrid poplar is considered a model woody biomass feedstock because of its broad adaptation, available genome sequence and transformation techniques, and fast growth. The bio￾mass accumulation of hybrid poplar is reported to be be￾tween 7 to 20 Mg/ha/yr depending on the nutrition and environmental conditions [24–26]. From the perspective of biomass production, switchgrass and hybrid Miscanthus seem to have the potential to produce more biomass com￾pared with that produced by poplar. Given that a short rotation for trees is five years, there is a time lag before poplars can be harvested, and then, only the wood is har￾vested. Woody biomass does have a storage advantage over herbaceous feedstocks. However, geography, land-use pat￾terns, agronomy, economics and biology are likely to result in multiple feedstock use. Because of the advantages of perennial feedstock, efforts have been put into developing perennial bioenergy feedstocks via breeding [27]. Plants for biodiesel In temperate areas, annual oilseeds such as soybean (Glycine max), canola (Brassica napus), and sunflower (Helianthus annuus) have all been used as biodiesel feed￾stocks. Palm oil (Arecaceae) trees have been successfully used as biodiesel plants in the tropics. If we consider potential biodiesel feedstocks for temperate use, the trans￾portation costs for palm oil would be prohibitively expens￾ive for export and would have a positive net energy balance. In the case of soybean, canola and sunflower, the energy output from grain was estimated to be 10 to 40 GJ/ha, which is considerably lower than the 200– 500 GJ/ha energy gain from lignocellulosic biomass [28]. Hence, we might conclude that lignocellulosic biomass will have a greater demand than biodiesel feedstocks. There are other candidates for bioenergy feedstocks that are too numerous to detail in this review. Alternative bioenergy plants include additional crops (e.g. sweet sor￾ghum), Camelina, grasses (e.g. big bluestem), trees (e.g. willow), and even algae. Potentially, green algae could be used for hydrogen production, oil production for biodiesel platforms, and even biomass production for a bioethanol platform, depending on the biotechnology breakthroughs. Hydrogen is believed to be an important component of the third generation of bioenergy and can be adapted as differ￾ent energy sources. Many factors determine the choice of bioenergy crops; these are summarized in Table 1 and discussed further in the next section. Bioenergy: environmental, ecological and economic considerations Net energy balance of different platforms Net energy balance (NEB) is an important concept in choosing a bioenergy platform because only a high positive NEB can be considered as economically and environmentally sustainable. This is particularly import￾ant when considering which crops and conversion pro￾cesses might be worthy of substantial biotechnology investment. Even though the economics of corn starch￾based ethanol and biodiesel production is currently com￾petitive with gasoline, their NEB is fairly low or even negative, in contrast to the favorable NEB of lignocellu￾losics, as shown in Table 1 [10,28,29]. If lignocellulosic biomass can be efficiently converted into ethanol, a NEB of up to 600 GJ/ha/yr is a reasonable expectation, which would provide the highest NEB of all first or second generation platforms. Recent efforts to build biorefineries for lignocellulosic biomass processing are the first step to fulfilling such potential; however, both low recalcitrance feedstocks and new biocatalysts to improve the proces￾sing efficiency are needed to realize this potential. Among the different bioenergy crops, switchgrass, Miscanthus, and sorghum could potentially produce the highest NEB [28,30,31]. Environmental and ecological benefits of different platforms Different bioenergy platforms have different pros and cons from an ecological and agricultural perspective (Table 1) [4]. The near-term economic advantages of ethanol pro￾duction from maize and biodiesel production from soybean are often counter-balanced by the detrimental effects of agricultural practices on the environment. By contrast, perennial feedstocks such as switchgrass can help to decrease soil erosion, improve water quality, and protect natural diversity [4,29,32–34] Perennial biomass crops also complement food-based and feed-based agriculture instead of competing with it. Global climate change and bioenergy choices Biomass crops and bioenergy production as an offset to fossil fuel have the potential to ameliorate global warming. Not only does the offset mean that less ‘old’ carbon is released into the atmosphere, but the underground bio￾mass of perennial biomass crops also acts as a carbon sink. For example, the capacity of Miscanthus giganteus to fix carbon dioxide is estimated to be 5.2 to 7.2 t C/ha/yr, which results in a negative carbon balance where more carbon dioxide is fixed than emitted [27]. In a recent study of maize, switchgrass, soybean, alfalfa, hybrid poplar and reed canarygrass (Phalaris arundinacea), only poplar and switchgrass had a negative carbon balance (carbon fixation of 2 t C/ha/yr) [35]. However, two recent publi￾cations indicate that first generation ethanol platforms actually have far higher carbon dioxide emissions com￾pared with that released from fossil fuels [20,21]. Accord￾ing to one of the studies, even the best lignocellulosic ethanol is predicted to have a positive carbon balance [20]. These contradictory estimates are a product of the different methods and models used to assess carbon release and fixation. Despite the differences, there is a consensus that second generation ethanol production using lignocellulosic platforms should lead to a lower carbon balance as compared with first generation plat￾forms. Enhancing the ability of perennial feedstocks to Review Trends in Plant Science Vol.13 No.8 424