Environmental Impacts of Algae-Derived Biodiesel and Bioelectricity

20 Jul 2011 - the processes used to convert algae biomass energy into a usable energy carrier) affect algae's overall sustainability. In Clarens et al...
0 downloads 0 Views 3MB Size
Download PDF
ARTICLE pubs.acs.org/est

Environmental Impacts of Algae-Derived Biodiesel and Bioelectricity for Transportation Andres F. Clarens,† Hagai Nassau,† Eleazer P. Resurreccion,† Mark A. White,‡ and Lisa M. Colosi*,† †

Civil and Environmental Engineering and ‡McIntire School of Commerce, University of Virginia, Charlottesville, Virginia 22904, United States

bS Supporting Information ABSTRACT: Algae are a widely touted source of bioenergy with high yields, appreciable lipid contents, and an ability to be cultivated on marginal land without directly competing with food crops. Nevertheless, recent work has suggested that large-scale deployment of algae bioenergy systems could have unexpectedly high environmental burdens. In this study, a “well-to-wheel” life cycle assessment was undertaken to evaluate algae’s potential use as a transportation energy source for passenger vehicles. Four algae conversion pathways resulting in combinations of bioelectricity and biodiesel were assessed for several relevant nutrient procurement scenarios. Results suggest that algae-to-energy systems can be either net energy positive or negative depending on the specific combination of cultivation and conversion processes used. Conversion pathways involving direct combustion for bioelectricity production generally outperformed systems involving anaerobic digestion and biodiesel production, and they were found to generate four and fifteen times as many vehicle kilometers traveled (VKT) per hectare as switchgrass or canola, respectively. Despite this, algae systems exhibited mixed performance for environmental impacts (energy use, water use, and greenhouse gas emissions) on a “per km” basis relative to the benchmark crops. This suggests that both cultivation and conversion processes must be carefully considered to ensure the environmental viability of algae-to-energy processes.

’ INTRODUCTION Algae-derived bioenergy has been a subject of increasing attention from academic and industrial groups because of its potential to harness sunlight into usable energy carriers.1,2 Although the notion of algae-derived energy is not new,3 increasing concerns about energy independence, climate change, and large-scale oil spills make algae a seemingly attractive alternative to petroleum. There is particular interest in algaederived transportation energy, as noted in a recent commercialization roadmap compiled by the United States Department of Energy (DOE),4 because transportation currently accounts for almost 30% of global energy demand. Still, several recent studies,5 most notably the DOE roadmap, have highlighted significant technical barriers to large-scale deployment of algaeto-energy systems. One especially critical barrier is a lack of understanding about how systems-level conversion choices (i.e., the processes used to convert algae biomass energy into a usable energy carrier) affect algae’s overall sustainability. In Clarens et al. 2010,5 we quantified the environmental impacts associated with large-scale algae cultivation in open ponds. This analysis used so-called “cradle-to-gate” system boundaries, whereby the delivered output was dried biomass corresponding to a functional unit of 317 GJ.5 Environmental impacts were then compared to three other bioenergy feedstocks: canola, corn, and switchgrass. We concluded that cultivation-phase burdens, most r 2011 American Chemical Society

notably upstream impacts for manufacture of fertilizers and other inputs, make algae-derived energy seemingly more burdensome than the selected benchmark crops. For three key impacts in particular (energy use, greenhouse gas emissions (GHG), and water use), algae burdens were found to be much greater than those of terrestrial crops. Still, we also noted that the land requirement to produce one unit of algae-derived energy is much smaller than that of the terrestrial crops. Taken together, these two observations pointed to an important trade-off between algae’s land use efficiency and environmental performance relative to terrestrial alternatives. The cultivation results from our previous work identified several intriguing opportunities for improving large-scale algae production; however, these results cannot tell us whether algae is more or less suitable for production of usable transportation energy than the benchmark crops. This is because some crops are more easily converted into energy carriers than others, and this was not included in the previous paper’s system boundaries.6,7 The present study addresses this shortcoming by adopting “wellto-wheel” system boundaries; i.e., by expanding our original Received: March 6, 2011 Accepted: July 20, 2011 Revised: June 2, 2011 Published: July 20, 2011 7554

dx.doi.org/10.1021/es200760n | Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology analysis to include conversion of each biomass into transportation energy sources. The array of selected conversion technologies reflects several prominent strategies being promoted by the algae community, in order to assess environmental performance of algae-to-energy systems as they are currently envisioned. Results are reported on the basis of two complementary functional units to illuminate expected trade-offs between land use and other environmental impacts. These functional units are 1) usable energy production per unit land area, as expressed using “vehicle kilometers traveled” (VKT per ha) and 2) environmental burdens (net energy use, water use, and GHG) per VKT. Energy return on investment (EROI) (i.e., the amount of energy produced per energy consumed to deliver one functional unit) was also assessed for each system, because this study evaluates many algae conversion options and it was desirable to make direct comparisons among dissimilar conversion systems. The most promising algae systems (as identified using EROI) were then compared with benchmark terrestrial biofuels.

’ METHODOLOGY A comprehensive LCA model was developed for both cultivation and conversion processes during production of transportation energy from algae, switchgrass, and canola. Systems boundaries accounted for all life cycle processes, from extraction of raw resources to VKT in a passenger automobile. A detailed summary of the model development and pertinent sources can be found in the Supporting Information (SI). Because algae-to-energy systems are an emerging technology, we used stochastic inputs to capture uncertainty about process systems. This was done using the Crystal Ball predictive modeling suite to facilitate Monte Carlo analyses. Modeled systems were identified based on an extensive literature survey, conversations with industry representatives, and first-principles engineering calculations. The resulting cultivation model (based largely on ref 5) assumes that brackish to saline groundwater is used to grow salt-tolerant algae species (e.g., Phaeodactylum sp., Tetraselmis sp., etc.) on marginal land under weather conditions typical of the southwestern USA. Annual algae yield was roughly 91.1 Mg/ha ash free dry weight (AFDW), assuming ash content is roughly 11.2% and total lipid content is roughly 19.6% (both on AFDW basis). Average heat content was 23,140 MJ/Mg-AFDW. Although there are currently several other proposed cultivation configurations under consideration (e.g., photobioreactors,8 floway systems,9 etc.), there is growing academic consensus that open ponds may be the only practically tenable option for algaederived energy production at industrial scale. This is because other systems are generally energetically and/or economically unsustainable,10 12 and this is reflected in emerging industrial practice.13 The conversion model takes a modular approach to understanding individual unit processes. Data for solids-liquids separation, dewatering/drying processes, and anaerobic digestion were from the wastewater treatment literature14,15 and design handbooks.16,17 Diesel production processes were modeled using LCA data for various plant oils, as taken from NREL and others. Burdens for material and energy inputs were obtained from the ecoinvent database.18 It was assumed that bioelectricity and biodiesel are used in commercially available battery electric vehicles (BEVs) and internal combustion vehicals (ICVs), respectively, and published mileage efficiencies (km/MJ) were

ARTICLE

used to compute VKT.19 Capital infrastructure was excluded from this analysis. Figure 1 summarizes modeled cultivation and conversion processes. We modeled four conversion pathways to enable direct LCA comparison among algae conversion options that are presently of interest in the algae bioenergy community. These pathways are depicted in Figure 1 and summarized as follows: A Anaerobic digestion of bulk algae biomass to produce methane-derived bioelectricity; B Production of biodiesel from algae lipids with anaerobic digestion of residual algae biomass to produce methanederived bioelectricity; C Production of biodiesel from algae lipids with direct combustion of residual algae biomass to produce bioelectricity; D Direct combustion of bulk algae biomass to produce bioelectricity. For each pathway A-D, four scenarios were modeled to capture the importance of nutrient procurement, which has been well documented in other algae LCA studies.5,8 This was done to show the range of fuel-cycle results obtainable using various nutrient sources within a single analysis. Scenario 1 involves procurement of virgin commercial CO2. Burden allocation was performed using a mass basis. Scenario 2 embodies CO2 procurement from carbon capture (via chemical sorption) at a coal-fired power plant.20 Scenario 3 corresponds to direct compression and delivery of flue gas comprising 12.5% CO2. Scenario 4 uses Scenario 3 as baseline (CO2 from compressed flue gas) but also accounts for nitrogen (N) and phosphorus (P) supplementation via use of treated wastewater to balance net evaporation (roughly 11% of total flow). N and P for Scenarios 1 3 were strictly from commercial fertilizer; in Scenario 4, most of the fertilizer is from commercial fertilizer, but roughly 4% of N and 7% of P come from wastewater effluent nutrients. For Scenarios 2 4, LCA impacts associated with recycled CO2 were allocated between power production and algae cultivation on an energy basis. See the Supporting Information for more detail. Conversion systems A D were evaluated for each of the four nutrient procurement scenarios referenced above, for a total of 16 algae cases. Cases exhibiting EROI significantly above 3 (the minimum sustainable EROI according to ref 21) were then compared to two terrestrial crop systems; i.e., 1) canola biodiesel production with combustion of residual biomass to produce bioelectricity; and 2) switchgrass combustion to produce bioelectricity. Annual canola dry yield was roughly 7.1 Mg/ha, 2.4 Mg/ha as seed (41% lipids at 21,000 MJ/Mg) plus 4.7 Mg/ha as “residuals” (stalks, pods, etc. at 18,000 MJ/Mg).5,18 Annual switchgrass dry yield was 21.5 Mg/MJ with energy content of 17,900 MJ/Mg.5,18 Comparisons between algae and the selected benchmark crops were facilitated via computation of VKT and VKT-normalized impacts for three different burdens: net energy use (in MJ/km), water use (in m3/km), and GHG emissions (in kg CO2 equivalents/km).

’ RESULTS AND DISCUSSION EROI for Algae Systems. As a first step in evaluating the viability of various algae-to-energy systems, we computed EROI for sixteen algae cases; i.e. four conversion pathways (A-D)  four nutrient procurement scenarios (1 4). This ratio has been identified as a particularly effective metric for assessing proposed bioenergy pathways. As point of reference, previously reported 7555

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology

ARTICLE

Figure 1. Schematic of a generic algae-to-energy production system (top), as evaluated in this investigation. Arrow widths are proportional to mass flows. Energy flows are indicated using dashed lines. Unit operations pertinent to assessment of four specific conversion pathways are shaded (gray) in panels A-D at bottom. These pathways include A) anaerobic digestion to produce bioelectricity; B) biodiesel production from algae lipids with anaerobic digestion of residuals to make bioelectricity; C) biodiesel production from algae lipids with combustion of residuals to make bioelectricity; and D) combustion of bulk algae biomass to make bioelectricity.

Table 1. Energy Return on Investment (EROI) for Four Algae Conversion Pathways Producing Combinations of Bioelectricity (BioE) and Biodiesel (BioD) via Anaerobic Digestion (AD) or Direct Combustion (DC)a case scenario

1. virgin CO2

2. carbon capture

3. flue gas

4. wastewater supplementation

A. AD to BioE

1.06

1.14

1.69

1.72

B. BioD + AD to BioE

0.65

0.72

1.11

1.13

C. BioD + DC to BioE

0.99

1.36

1.99

1.99

D. DC to BioE

1.53

2.90

4.10

4.09

a

Each pathway is modeled for four nutrient procurement scenarios: 1) virgin commercial CO2; 2) CO2 from carbon capture at a coal-fired power plant; 3) direct compression of flue gas; and 4) flue gas with fertilizer offsets from use of wastewater effluent to balance pond evaporation. Values are medians from 10,000 trials. See the SI for corresponding means, confidence intervals, and discussion of uncertainty. Canola and switchgrass systems exhibit median EROI values of 2.73 and 15.90, respectively.

EROI for corn ethanol has been on the order of 1.25.22 To reiterate from above, it is been suggested that the minimum sustainable EROI is roughly 3 but that values from 5 to 10 will be required to maintain quality of life in the absence of readily abundant fossil energy.21 Table 1 summarizes the range of algae EROI values computed in this study, spanning from 0.65 to 4.10. Table 1 suggests that many combinations of cultivation and conversion approaches will consume more energy than they produce (i.e., EROI < 1). Beyond this, Table 1 also points to several interesting trends related to cultivation and conversion choices. These trends emerge from comparison among nutrient procurement strategies (columns) or conversion pathways (rows). Interestingly, the extent of EROI variability among

conversion pathways is roughly the same as EROI variability among nutrient procurement scenarios. This indicates that both factors have an important effect on overall energy balances. Comparison among rows in Table 1 offers insight into the relative favorability of proposed algae conversion pathways. EROI generally follows the same trend, D > A > C > B, and from these values it is apparent that not all conversion technologies are equally efficient. For example, only Case D exhibits median EROI values that are ever greater than the threshold defined by ref 21. Also, most cases involving biodiesel production exhibit median EROI greater than 1 (and are thus net energy producing), but none of these come close to the desired EROI threshold. Finally, paired comparison of EROI for A vs D and B 7556

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology vs C highlights the differences between anaerobic digestion and direct combustion as a means to valorize nonlipid algae residuals. Cases A and B rely on anaerobic digestion to convert some or all algae into methane, while Cases C and D rely on direct combustion to convert some or all algae into bioelectricity. From these paired comparisons, direct combustion is seemingly more efficient than anaerobic digestion regardless of whether or not algae lipids are extracted to make biodiesel. This is somewhat surprising since anaerobic digestion has been proposed as a purportedly efficient means to simultaneously generate bioelectricity (via methane) and recycle algae’s nutrient inputs during algae cultivation.11,13,14 Comparison of digestion and direction combustion EROIs for identical algae biomass makes it possible to evaluate the relative efficiency of these two specific algae conversion processes. This is important since the observation that anaerobic digestion may be less energetically favorable than direct combustion is somewhat at odds with the generally positive portrayal of digestion in the academic algae literature. From our models, we assert that there are three factors that may make algae digestion systems less attractive than they have been generally portrayed. First, measured digestibility of algae volatile suspended solids (VSS) is only about 50% (range 40 60%), such that a large recalcitrant fraction of the biomass must be cycled through the digestion system and then purged for use as soil amendment (or simply landfilled). N and P bioavailability in the resulting digestate is quite low,23 yielding modest life cycle offsets on the order of 1 Mg N and 0.1 Mg P fertilizer avoided per 100 Mg digestate solids. Second, reported methane and ammonia yields during algae digestion are significantly lower than their theoretical maxima. For Case A, CH4 and nitrogen production are only 50 70% of what would be computed using empirical digestion relationships; i.e., 0.49 vs 0.67 L CH4/g VSS and 30 vs 55 mg N-NH3/g algae VSS. It has been suggested that further research into algae digestion could improve overall digestibility and methane/ ammonia recovery,14 and given the model’s sensitivity to these parameters, this appears to be an important area of future work. The final reason algae digestion may not be as efficient as previously thought is because a substantial fraction of the methane-derived bioelectricity is required to offset heat and electricity demands associated with digestion operations (homogenization, digestion mixing/heating, and belt-filter press dewatering). For Cases 1A and 1B, combined electricity demand, heat demand, and upstream heat burdens consume 33 45% of the bioelectricity output. This makes it such that energy offsets corresponding to CO2 and nitrogen recycle via anaerobic digestion are on the same order of magnitude as methane-derived bioelectricity; i.e., nutrient offsets are 87% and 31% of bioelectricity production for Scenarios 1 and 3, corresponding to most and least energy-intensive CO2 procurement, respectively. This demonstrates that the full benefits of algae digestion can only be realized when all output streams (recycled nutrients; digestate solids; CO2 from digestion/combustion) are fully utilized. With respect to comparison among columns in Table 1, the four modeled scenarios always exhibit the same EROI trend: Scenario 1 < Scenario 2 < Scenario 3 ≈ Scenario 4. This is unsurprising since nutrient procurement choices should not affect relative favorability among conversion pathways. We also see a pronounced impact of CO2 source on overall cultivation sustainability, whereby use of virgin commercial CO2 (Scenario 1) is much more burdensome than use of CO2 captured from fossil flue gas chemical sorption (Scenario 2), which is, in turn, somewhat more burdensome than compressed flue gas.

ARTICLE

This trend reflects underlying energy “costs” per Mg CO2, as adjusted to account for pertinent allocations: 4130 MJ/Mg for Scenario 1, 530 MJ/Mg for Scenario 2, and 180 MJ/Mg for Scenario 3. The benefit of using reclaimed flue gas, as clearly articulated in Table 1, is consistent with other algae LCAs in which authors have advocated for use of “free” CO2 during cultivation.8,24,25 Although this is an obvious conclusion arising from LCA, we wish to reiterate that flue gas capture is not presently viable at industrial scale, and it remains unclear how various flue gas constituents might impact algae yields.26 These issues will need to be addressed before the potential benefits of flue-fed algae cultivation can be achieved. Also, even reclaimed flue gas is not completely “free” from an LCA perspective, such that it is important to account for CO2 capture and transport burdens as we have done here. Finally, the use of wastewater effluent as an N and P source has almost no effect on EROI values in Table 1 (Scenario 3 ≈ Scenario 4). This is because effluent accounts for only 11% of the total flow in the modeled system when it is used to balance net evaporation, and because conventional effluents possess such low N and P concentrations.27 The effect of wastewater supplementation on nutrient burden reduction would undoubtedly increase for freshwater systems, since a much larger percentage of the flow could be delivered as effluent; however, our preliminary analyses suggest that the increased productivity of marine algae relative to freshwater algae more than compensates for the reduction in fertilizer offsets afforded by use of larger wastewater flows in a freshwater system (data not shown). EROI values in Table 1 mask several subtleties related to the relative efficiency of electricity production via different mechanisms. Components of the denominator (EIN) included direct electricity use, direct heat use, and upstream energy use for materials and energy inputs. Since it was assumed that bioelectricity produced from each algae system would be used to offset direct electricity use within that system, upstream burdens were only assessed for the portion of electricity demand in excess of the amount generated. This is an important differentiation because the upstream energy impact factor for US electricity is quite high, roughly 3.5 MJ/MJ. For the case in which an algae system produced a surplus of bioelectricity, this same impact factor was used in the numerator (EOUT) to account for the “virtual” upstream burden that would have otherwise accrued on electricity from the US grid. Although these calculations are consistent with the industry-standard GREET model,28 they do impart some bias toward bioelectricity systems by doubleemphasizing (denominator and numerator) creation of an electricity surplus and also charging an “opportunity cost” for algae which does not become bioelectricity. For Case A, we see that algae anaerobic digestion is energetically favorable (all EROIs > 1); however, the definition of EROI imparts an indirect penalty for both the fraction of algae biomass that becomes digestate and the fraction of methane bioelectricity which is used to operate the digestion system. These could be converted into energy more efficiently via direct combustion, generating a larger electricity surplus. This partly explains why Case D is always preferable to all others, and it is one of several reasons why algae bioelectricity is a seemingly appealing transportation energy source. Comparison of Vehicle Kilometers Traveled from Algae vs Other Crops. Having determined which combinations of algae conversion processes result in favorable EROIs, we wished to evaluate the amount of usable transportation energy that could be derived from each system and also assess their relative environmental 7557

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology

Figure 2. Annual vehicle kilometers traveled (VKT) per hectare-year for four bioenergy systems (error bars are median (1 standard deviation). Section 7.2 of the SI summarizes VKT calculations. Media values are (from left to right): 584,600 km/ha-yr; 124,700 km/ha-yr; 462,700 km/ha-yr; and 33,400 km/ha-yr.

favorability. Taken together, these metrics enable more meaningful comparison between algae and terrestrial alternatives and also articulate trade-offs between maximizing usable outputs (VKT) and minimizing system burdens (energy, water, GHG). Figure 2 summarizes the VKT that could be derived annually from one hectare of biomass cultivation for four bioenergy systems: algae biodiesel production with conversion of residuals into bioelectricity (Case 4C); canola biodiesel production with conversion of residuals into bioelectricity; algae bioelectricity production (Case 4D); and switchgrass bioelectricity production. Recall that Scenario 4 assumes that compressed flue gas is used for CO2 and that wastewater nutrients offset some fertilizer and water demands. Cases 4C and 4D were selected for comparison with the benchmarks because they produce the only EROIs for which the entire 90% confidence intervals are greater than 1 and the only median values which are greater than 3. From Figure 2, both selected algae systems dramatically outperform the terrestrial crop systems in terms of VKT production per hectare. Algae generates, on average, 4.2 times and 15.7 times more VKT than the switchgrass and canola systems, respectively. Misalignment of system boundaries precludes direct comparison with corn ethanol, but we estimate that the average algae VKT from Figure 2 is roughly nineteen times greater than could be derived from corn ethanol (27,000 km/ha-yr) even when accounting for ethanol coproducts.29 The results in Figure 2, whereby algae outperforms terrestrial crops in VKT production per hecature, are not unexpected. It was previously demonstrated that land use efficiency was algae’s most notable advantage relative to corn, canola, and switchgrass when these were compared on a raw biomass energy basis.5 The VKT values in Figure 2 demonstrate that accounting for conversion of each crop into usable transportation energy does not erode algae’s superior cultivation-phase land use efficiency. Figure 2 highlights the economic incentive associated with algae-derived transportation energy, since only a finite amount of land is available to produce bioenergy feedstocks, and algae cultivation leverages land area most efficiently. Finally, Figure 2 most likely offers a conservative estimate of the differences in land use among crops since it does not explicitly consider land quality, and it is expected that algae could be grown on marginal lands that are unsuitable for terrestrial agriculture.30 Comparison of VKT from Two Algae Conversion Systems. From Figure 2, VKT output for Case 4D is greater than for Case

ARTICLE

Figure 3. Environmental impacts per vehicle kilometer traveled (VKT) for four transportation bioenergy systems (error bars depict medians (1 standard deviation). “Algae biodiesel + bioelectricity” is Case 4C in Table 1; “Algae bioelectricity” is Case 4D in Table 1.

4C, consistent with Table 1. Although this is somewhat at odds with conventional wisdom, which holds that algae are particularly promising for production of liquid fuels,2,31,32 our model points to two reasons why bioelectricity systems outperform combined biodiesel/bioelectricity systems. First, the amount of energy comprising nonlipid algae residuals is much larger than the amount of energy comprising lipid-derived biodiesel. Using average values from the model and ignoring minor losses, nonlipids constitute roughly 1,554,100 MJ/ha-yr: 0.84 g residuals/g algae  91 Mg algae/ha-yr  20,200 MJ/Mg residuals. This is almost three times as much energy as is available as biodiesel: 0.16 g neutral lipids/g algae  91 Mg algae/ha-yr  0.96 g biodeisel/g neutral lipids  37,700 MJ/Mg biodiesel = 527,000 MJ/ha-yr. To compute VKT from these quantities, it is necessary to multiply by efficiency factors. For the nonlipids, these are as follows: boiler efficiency (0.54)  transmission efficiency (0.92)  charging efficiency (0.90)  BEV mileage efficiency (1.3 km/MJ) = 0.58 km/MJ. For the lipid fraction, it is necessary to multiply by ICV mileage efficiency, 0.39 km/MJ. Since the bioelectricity factor is greater than the biodiesel factor, multiplying them by different amounts of energy yields significantly more VKT from the nonlipid algae residuals. Second, the amount of energy expended to convert lipids into 1 MJ biodiesel energy is much larger than the amount of energy expended to convert nonlipids into 1 MJ bioelectricity: 680 kJ/MJ (for homogenization, lipid extraction, transesterification, and biodiesel processing) versus 5 kJ/MJ (for drying and ash transport). The combined effects of more available energy and lower conversion-phase energy inputs leads to larger VKT for Case D (direct combustion) versus Case C (direct combustion following biodiesel production). Finding uses for the nonlipid fraction of algae biomass is a major goal for the algae industry. For Case C in Figure 3, biodiesel and bioelectricity account for 40% and 60% of total VKT, respectively. Since more than half of the usable transportation energy output constitutes bioelectricity, it is no surprise that many algae LCA studies to date have emphasized the importance of algae “co-products” in driving overall environmental or economic sustainability. We have considered only energy coproducts for this analysis; however, a wide variety of other purported uses do exist for the nonlipid fraction of algae biomass (e.g., animal feed). LCA analysis of these proposed coproduct systems 7558

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology to validate their viability would be an important contribution. Finally, we should emphasize that our current analysis excludes manufacture of BEVs because we were focused on understanding fuel cycle burdens. This affects comparison between Cases C and D, such that better LCA data for manufacture of BEVs will be needed as the market grows in the coming years. Life Cycle Burdens for Transportation Energy from Algae vs Other Crops. Figure 2 highlights the promising potential of algae-derived transportation energy insofar as it demonstrates algae’s superior land use efficiency relative to the benchmarks. Although land use is an important life cycle consideration, other environmental burdens must also be evaluated. Figure 3 depicts VKT-normalized net energy use, water use, and GHG emissions for the bioenergy systems in Figure 2. From Figure 3, algae-derived transportation energy exhibits mixed performance relative to the canola and switchgrass systems. Algae biodiesel and bioelectricity systems exhibit higher net energy use but lower water use and GHG emissions per km than their respective terrestrial benchmarks. These results cast algae-derived transportation energy in a better light than our previous results, which accounted for only cultivation-phase burdens.5 Algae’s improved performance relative to the other crops reflects use of VKT-normalized burden comparison, because this metric captures algae’s significantly higher energy yield per unit land area compared to switchgrass and canola. We should note that algae’s superior land use efficiency is more pronounced in this analysis because we have modeled marine algae species exhibiting higher yields than the freshwater species used in our previous analysis. Canola’s very large GHG burden reflects underlying differences among crops in cultivation energy intensiveness. From ecoinvent, GHG emissions for canola, corn, and switchgrass are as follows: 1.9 kg CO2 eq/kg canola seed, 0.4 kg CO2 eq/kg corn kernels, and 0.2 kg CO2 eq/kg switchgrass.18 Finally, the seeming efficiency of algae-derived bioelectricity, as revealed in Figure 3, suggests that it may be worthwhile to compare this system to photovoltaic power production in subsequent analyses. Implications and Summarizing Remarks. The results of this study represent significant progress toward better understanding algae’s potential use as a sustainable transportation energy source in three ways. First, they offer a more complete picture of algae’s performance relative to other bioenergy feedstocks using a comprehensive well-to-wheel system boundary. Detailed assessment of both cultivation and conversion processes for multiple crops within the same system boundaries makes it possible to evaluate algae’s potential within the context of the current bioenergy landscape. Second, multiple algae conversion scenarios are compared within the same study. This rectifies a shortcoming of previously published LCAs. Third, these results emphasize a key trade-off, namely: Algae are capable of producing many more VKT per hectare than the selected terrestrial crops, but in doing so they create larger environmental burdens on a per-km basis. This is consistent with the qualitative conclusions of several recent reports, including the DOE Algae roadmap and Lundquist et al.13 Unfortunately, even the most rigorous LCA analyses fall short when it comes to making normative judgments. Thus, our results cannot tell us whether the US and other countries should pursue algae-derived transportation fuels. Economic considerations, which we have not addressed, will undoubtedly have a large impact on whether algae-derived biodiesel and bioelectricity become widely commercialized. On one hand, direct biomass energy technologies have a favorable levelized cost of energy vs

ARTICLE

other alternative energy technologies and are cost-competitive with fossil power plants.33 On the other hand, establishing and maintaining the infrastructure for algae cultivation and conversion will be quite expensive. Regardless, the tremendous demand for transportation energy, increasing fuel prices, and a lack of mechanisms for monetizing environmental performance in the US make it reasonable to expect that algae’s excellent land use efficiency could render it financially attractive over the next several decades. For this reason, environmental and economic LCA studies will be key tools for improving the overall sustainability of algae-derived transportation energy systems.

’ ASSOCIATED CONTENT

bS

Supporting Information. Detailed model documentation. This material is available free of charge via the Internet at http:// pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 434-924-7961. E-mail: [email protected].

’ ACKNOWLEDGMENT The authors gratefully acknowledge funding for this study from a University of Virginia (UVA) Energy Seed Grant and a UVA Fund for Excellence in Science & Technology (FEST) Grant. ’ REFERENCES (1) Brune, D. E.; Lundquist, T. J.; Benemann, J. R. Microalgal Biomass for Greenhouse Gas Reductions: Potential for Replacement of Fossil Fuels and Animal Feeds. J. Environ. Eng. 2009, 135 (11), 1136–1144. (2) Chisti, Y. Biodiesel from microalgae. Biotechnol. Adv. 2007, 25 (3), 294–306. (3) Benemann, J.; Oswald, W. Systems and Economic Analysis of Microalgae Ponds for Conversion of CO2 to Biomass - Final Report; Department of Energy: Pittsburgh, 1996; p 201. (4) USDOE. National Algal Biofuels Technology Roadmap; U.S. Department of Energy: College Park, MD, 2010. (5) Clarens, A. F.; Resurreccion, E. P.; White, M. A.; Colosi, L. M. Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks. Environ. Sci. Technol. 2010, 44 (5), 1813–1819. (6) Clarens, A. F.; Resurreccion, E. P.; White, M. A.; Colosi, L. M. Response to Comment on “Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks”. Environ. Sci. Technol. 2010, 45 (2), 834. (7) Starbuck, C. M. Comment on “Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks”. Environ. Sci. Technol. 2010. (8) Batan, L.; Quinn, J.; Willson, B.; Bradley, T. Net Energy and Greenhouse Gas Emission Evaluation of Biodiesel Derived from Microalgae. Environ. Sci. Technol. 2010, 44 (20), 7975–7980. (9) Mulbry, W.; Kondrad, S.; Pizarro, C.; Kebede-Westhead, E. Treatment of dairy manure effluent using freshwater algae: Algal productivity and recovery of manure nutrients using pilot-scale algal turf scrubbers. Bioresour. Technol. 2008, 99 (17), 8137–8142. (10) Jorquera, O.; Kiperstok, A.; Sales, E. A.; Embiruau, M.; Ghirardi, M. L. Comparative energy life-cycle analyses of microalgal biomass production in open ponds and photobioreactors. Bioresour. Technol. 2010, 101 (4), 1406–1413. (11) Stephenson, A. L.; Kazamia, E.; Dennis, J. S.; Howe, C. J.; Scott, S. A.; Smith, A. G. Life-Cycle Assessment of Potential Algal Biodiesel 7559

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560

Environmental Science & Technology

ARTICLE

Production in the United Kingdom: A Comparison of Raceways and AirLift Tubular Bioreactors. Energy Fuels 2010, 24 (7), 4062–4077. (12) Zhao, F.; Clarens, A.; Murphree, A.; Hayes, K.; Skerlos, S. J., Structural Aspects of Surfactant Selection for the Design of Vegetable Oil Semi-Synthetic Metalworking Fluids. Environ. Sci. Technol. 2006. (13) Lundquist, T. J.; Woertz, I. C.; Quinn, N. W. T.; Benemann, J. R. A Realistic Technology and Engineering Assessment of Algae Biofuel Production; California Polytechnic State University: San Luis Obispo, 2010; p 178. (14) Sialve, B.; Bernet, N.; Bernard, O. Anaerobic digestion of microalgae as a necessary step to make microalgal biodiesel sustainable. Biotechnol. Adv. 2009, 27 (4), 409–416. (15) Soda, S.; Iwai, Y.; Sei, K.; Shimod, Y.; Ike, M. Model analysis of energy consumption and greenhouse gas emissions of sewage sludge treatment systems with different processes and scales. Water Sci. Technol. 2010, 61 (2), 365–73. (16) Metcalf; Eddy; Tchobanoglous, G.; Burton, F. L.; Stensel, H. D. Wastewater engineering: treatment and reuse; McGraw-Hill: Boston, 2003; p 1819. (17) Poirier, D. Conveyor Dryers. In Handbook of Industrial Drying, 3rd ed.; CRC Press: 2006. (18) Weidema, B. ecoinvent data v2.0. http://www.ecoinvent.org/ (accessed July 27, 2011). (19) Campbell, J. E.; Lobell, D. B.; Field, C. B. Greater Transportation Energy and GHG Offsets from Bioelectricity Than Ethanol. Science 2009, 324 (5930), 1055–1057. (20) Khoo, H. H.; Tan, R. B. H. Life cycle investigation of CO2 recovery and sequestration. Environ. Sci. Technol. 2006, 40 (12), 4016–4024. (21) Hall, C.; Balogh, S.; Murphy, D. What is the Minimum EROI that a Sustainable Society Must Have? Energies 2009, 2 (1), 25–47. (22) Hill, J.; Nelson, E.; Tilman, D.; Polasky, S.; Tiffany, D. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (30), 11206– 11210. (23) Warman, P. R.; Termeer, W. C. Evaluation of sewage sludge, septic waste and sludge compost applications to corn and forage: yields and N, P and K content of crops and soils. Bioresour. Technol. 2005, 96 (8), 955–961. (24) Clarens, A. F.; Resurreccion, E. P.; White, M. A.; Colosi, L. M. Response to Comment on “Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks. Environ. Sci. Technol. 2010, 44 (9), 3643–3643. (25) Subhadra, B. G. Comment on “Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks”. Environ. Sci. Technol. 2010, 44 (9), 3641–3642. (26) Doucha, J.; Straka, F.; Lívansky , K. Utilization of flue gas for cultivation microalgae (Chlorella sp.) in an outdoor open thin-layer photobioreactor. J. Appl. Phycol. 2005, 17, 403–412. (27) Dodd, M. C.; Zuleeg, S.; Von Gunten, U.; Pronk, W. Ozonation of source-separated urine for resource recovery and waste minimization: Process modeling, reaction chemistry, and operational considerations. Environ. Sci. Technol. 2008, 42 (24), 9329–9337. (28) Wang, M. The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) Model; UChicago Argonne, LLC: Chicago, IL, 2010. (29) Farrell, A. E.; Plevin, R. J.; Turner, B. T.; Jones, A. D.; O’Hare, M.; Kammen, D. M. Ethanol Can Contribute to Energy and Environmental Goals. Science 2006, 311 (5760), 506–508. (30) Dale, B. E.; Bals, B. D.; Kim, S.; Eranki, P. Biofuels Done Right: Land Efficient Animal Feeds Enable Large Environmental and Energy Benefits. Environ. Sci. Technol. 2010, 44 (22), 8385–8389. (31) Lardon, L.; Helias, A.; Sialve, B.; Steyer, J.-P.; Bernard, O. Life-Cycle Assessment of Biodiesel Production from Microalgae. Environ. Sci. Technol. 2009, 43 (17), 6475–6481. (32) Sander, K.; Murthy, G. Life cycle analysis of algae biodiesel. Int. J. Life Cycle Assess. 2010, 15 (7), 704–714. (33) Roth, I. F.; Ambs, L. L. Incorporating externalities into a full cost approach to electric power generation life-cycle costing. Energy 2004, 29 (12 15), 2125–2144. 7560

dx.doi.org/10.1021/es200760n |Environ. Sci. Technol. 2011, 45, 7554–7560