Microalgae to biofuels lifecycle assessment — Multiple pathway evaluation

Abstract

A variety of researchers have constructed and presented lifecycle assessments of the microalgae-to-biofuel process, however, inconsistencies in system boundary definitions and high-level process modeling have led to a wide range of results. This study integrates engineering process models validated through experimental and modeling research to perform an environmental assessment of four microalgae-to-biofuel production scenarios leveraging the Argonne National Laboratory GREET model. The baseline scenario consists of a down flow open pond growth system, three phase de-watering step (settling, dissolved air flotation, and a centrifuge), hexane extraction and nutrient recovery using anaerobic digestion. The net energy ratio (NER), defined as energy consumed over the produced energy, and greenhouse gases (GHG) for the baseline scenario are 0.7 MJ MJ^− 1 and − 41.7 g CO_2-eq MJ^− 1 respectively. Three alternative scenarios are also evaluated: 1) Improved microalgal productivity, 2) supercritical CO₂ extraction, and 3) no nutrient recycle. This research shows that supercritical CO₂ extraction is neither currently energetically- nor environmentally favorable and that nutrient recycle plays an integral role in achieving favorable NER and GHGs. The study highlights on the systems level, two findings related to the NER; 1) the NER is minimally impacted with increased productivity and 2) increasing microalgae lipid content detrimentally affects the NER which is attributed to the reduction in the total energy that can be captured by the anaerobic digester.

Introduction

The next generation of biofuel feedstock processes must be critically analyzed to quantify the potential scalability and corresponding environmental impact. Compared to first-generation biofuel feedstocks, microalgae are characterized by higher solar energy yield, year-round cultivation, the use of lower quality or brackish water, and the use of less- and lower-quality land [1], [2], [3], [4], [5], [6]. The theoretical maximum production of oil from microalgae has been calculated at 350,000 L·ha^− 1·yr^− 1 (38,000 gal·acre^− 1·yr^− 1) and is dramatically larger than that of traditional terrestrial crops [7]. Scalable experimental data have shown a near term realizable production of 46,000 L·ha^− 1·yr^− 1 (5000 gal·acre^− 1·yr^− 1), compared to 2500 L·ha^− 1·yr^− 1 (270 gal·acre^− 1·yr^− 1) of ethanol from corn or 580 L·ha^− 1·yr^− 1 (60 gal·acre^− 1·yr^− 1) of biodiesel from soybeans [8], [9], [10], [11], [12]. Researchers have shown that microalgae feedstock cultivation can be coupled with combustion energy plants or other CO₂ sources and has the potential to utilize nutrients from wastewater treatment plants [3]. These advantages have led to a continuing interest in microalgae as an alternative feedstock for biofuel production.

Lifecycle assessment (LCA) has emerged as the fundamental tool to evaluate the sustainability of next generation biofuels. The LCA literature makes use of the metrics of net energy ratio (NER, defined here as the ratio of energy consumed to fuel energy produced) and greenhouse gas (GHG) emissions per unit of energy produced as the functional units. The results from LCA are highly sensitive to definitions of system boundaries, lifecycle inventories, process efficiencies, and functional units [10], [13], [14]. A variety of researchers have constructed and presented LCAs of the microalgae-to-biofuel process, however, inconsistencies in system boundaries and high-level process modeling with large uncertainties in sub-process modeling have led to a wide range of results [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41]. A survey of the literature shows the NER for microalgae biofuel production at scale ranges from a low value of 0.2 (comparable to petroleum) reported by Luo et al. [31] to 7.8 reported by Brentner et al. [19] with an extreme value of 1086 reported by Beal et al. [18] based on the extrapolation of small-scale laboratory data. The GHG results from the surveyed literature range between − 75.29 g-CO_2-eq MJ^− 1 reported by Batan et al. [17] and 534 g-CO_2-eq MJ^− 1 reported by Brentner et al. [19] with other studies reporting values between these extremes [16], [17], [18], [19], [20], [21], [22], [28], [30], [31], [33], [38], [40]. The large variability in the NER and GHGs in previous LCA studies is due to the wide range of processes investigated as well as the assumptions made with regard system boundaries, key parameter values, sources of fossil energy, and co-product allocation, which all complicate comparison of results among studies [10], [13], [14], [42], [43], [44]. The majority of the studies surveyed now integrates anaerobic digestion as a way to effectively recover nutrients and generate on-site heat and energy from the lipid extracted algae (LEA) [20], [21], [23], [24], [25], [30], [32], [37], [38], [39], [40]. In the meta-analysis of Liu et al. [30] an anaerobic digester was added to studies that excluded it in an effort to harmonize a baseline scenario and compare results from various researchers. The integration of the anaerobic digestion system has been shown to favorably impact the NER and GHGs.

This study focuses on a LCA founded on the integration of a systems engineering model informed by experimental and modeling research together with the Argonne National Laboratory's Greenhouse Gases, Regulated Emissions, and Energy USE in Transportation (GREET) model to directly compare multiple process paths [25], [45]. The boundary for the analysis is such that a comparison to literature and soy-based biofuel can be transparently performed. The work focuses on the evaluation of alternative process scenarios to determine process tradeoffs on a systems level. The discussion focuses on the environmental impact of changes in the process parameters on system results, the environmental implications of the integration of an anaerobic digestion system, and comparison of results from this study to previous environmental impact assessments.