Introduction

Overview

Currently, there is a global effort to develop alternative transportation fuels. Dozens of biological feedstock sources have been suggested to meet this effort, and the success of their implementation has varied. In this paper, we discuss algal oil, which has been touted as a potential feedstock for renewable diesel production. Specifically, we present a framework for reporting renewable diesel production from algae. Algae are an extremely diverse group of organisms, and it is not surprising that different species of algae produce different compounds that could be used as alternative fuel feedstock. Five commonly studied algal components or products useful for alternative fuels are: lipids for petroleum fuel substitutes, carbohydrates for ethanol, hydrogen, methane via biomass gasification, and biomass for direct combustion, anaerobic digestion, or thermochemical conversion [1–10]. The conversion pathways that are available (i.e., biochemical conversion, thermochemical conversion, and transesterification) for producing algae-based fuel (e.g., biodiesel, methane, hydrogen, electricity, etc.) have been outlined [11, 12]. The framework presented here is focused on characterizing the production of renewable diesel from algae. The term, “renewable diesel” is defined as a substitute for conventional diesel fuel that is derived from renewable resources (e.g., biodiesel) [13–15]. Renewable diesel fuels are produced by upgrading a renewable oil material, which is referred to as biocrude, and can be produced from a variety of feedstock and production pathways. It is possible that non-diesel fuels could be produced from algal biocrude by alternative refining techniques. Although this study presents a framework for reporting the production of renewable diesel fuel from algae, it is expected that similar methods can be used to report other algal fuels.

The use of algae for alternative fuels has been studied globally by researchers for several decades. The United States’ National Renewable Energy Laboratory conducted an 18-year research effort, the Aquatic Species Program, which investigated the use of algae for biodiesel [1]. Dozens of other research and industry groups have conducted algae-to-biofuel studies during the time of the Aquatic Species Program and since its conclusion in 1996.

There are multiple production pathways that are being researched for renewable diesel production from algae. Categorically, these include (1) transesterification of extracted algal lipids, (2) thermochemical conversion of algal biomass, and (3) conversion of secreted algal oils (i.e., “milking”). Each of these pathways is discussed in greater detail in the following sections. The reporting framework presented here provides a way to compare results, not only from within the same production pathway, but also among different production pathways.

In general, the potential of algae-derived renewable diesel to be a suitable alternative fuel is dependent on the answers to three questions: (1) how much renewable diesel can be produced, (2) what is the financial cost of production, and (3) what is the energy ratio of producing renewable diesel? In this study, the amount of renewable diesel that can be produced is evaluated with respect to the cultivation volume and growth duration and is therefore expressed in units of grams per liter per day. Different metrics could be used for evaluating the amount of renewable diesel that can be produced on a national or global scale, such as the total land, water, and nutrients required to produce enough renewable diesel to satisfy the US liquid fuel demand. In addition, evaluating the amount of renewable diesel that can be produced at the national or global level requires the consideration of other factors, such as environmental impacts, resource availability, and infrastructure renovations, which are not specifically addressed in this study. However, the three questions listed above are critical for evaluating the potential of an alternative fuel to make a significant contribution to energy supply. Many variables influence the evaluation of each of these questions, and it is important to establish a systematic approach to determine the answers.

Although great progress has been made regarding the use of algae for renewable diesel, the field is relatively young. Consequently, some ambiguity remains about the best way to report research results. In turn, it is easy to misinterpret published results because the nomenclature varies. Furthermore, because an explicit reporting method has not been established, researchers are at risk of inaccurately estimating the potential for algae as a renewable diesel feedstock by accidentally omitting important processing inefficiencies. Finally, due to the lack of standardization, it is difficult for researchers to integrate results from multiple sources. The following section illuminates the inconsistencies discussed here.

Reporting Variability and Inconsistency

The advantages of a particular species, growth environment, or conversion technology depend on the impact it has on the entire production pathway. Said differently, the fundamental interest is in the total energy, materials, and cost balances for renewable diesel production. To enable systems-level analyses, when reporting results associated with individual processing steps (e.g., growth, harvesting, refining, etc.), it would be ideal to use metrics that are compatible with the other steps. Using compatible metrics is also important, but more complicated, if renewable diesel production from algae is integrated into a multi-product facility, in which waste streams of one product are used for another. Multi-product industrial facilities like this are not specifically considered in this study.

To illustrate the variability in reporting methods, Tables 1 and 2 list results from several algae-based renewable diesel studies pertaining to how much renewable diesel can be produced and the financial cost of production. The data presented are specific to the production of biodiesel via transesterification of extracted lipids (as opposed to the other production pathways, all of which are discussed below). Each symbol listed in Tables 1 and 2 (e.g., P GM, φ sep, \( {\tilde{C}_T} \), etc.) is defined in Appendix 1 and discussed in more detail in the following sections (cf. Figs. 2 and 7). These tables illustrate the variety of reporting methods used in the field, but do not encompass or represent all algae-based renewable diesel research that has been published. For most categories shown, there is a wide range of results. For instance, the estimated cost of producing algal “oil” (or lipids) varies from $39 to $209 per barrel across the studies. Additionally, there are different metrics used to report results within the same category. For example, biomass productivity is commonly reported in terms of kilograms per square meter per day or kilograms per liter per day (where square meter is for the growth media surface area and liter is for the growth media volume), which can lead to ambiguities [41]. Both metrics can provide valuable data, and both should be reported when possible. In addition, some studies use metrics to report results that do not include enough information to adequately characterize the potential of algae as a renewable diesel feedstock. For example, the lipid fraction is often used to evaluate the potential of different algal species [1, 16, 21, 41]. As discussed below, the lipid fraction lacks the specificity needed to evaluate the impact of that species, growth condition, and type of lipids produced on the entire production pathway [41]. Finally, in most of the studies shown, several steps in the production pathway are omitted entirely.

Table 1 Reported research results associated with algae-derived renewable diesel productivity
Full size table
Table 2 Reported research results associated with algae-derived renewable diesel cost
Full size table

Due to these inconsistencies, it is difficult to accurately determine the potential of algae-derived renewable diesel on a systems level. Additional standardization is needed within the field so that communication among researchers is less ambiguous. To address this problem, we outline three production pathways for algae-derived renewable diesel, present a nomenclature system for reporting results, and propose a framework for characterizing the potential of algae as a source for renewable diesel production. The nomenclature system is presented as a top-level analysis that is applicable to all three of the production pathways. The utility of this nomenclature system is illustrated in Appendix 2 by expanding it to include the detailed production steps for one of the production pathways (namely, transesterification of extracted algal lipids). A detailed expansion of the other production pathways (thermochemical conversion of algal biomass and conversion of secreted algal oils) can be conducted similarly.

Top-Level Algae-Derived Renewable Diesel Production Pathway

This section presents the top-level nomenclature system for reporting productivity, cost, and energy requirements for producing algae-derived renewable diesel, all of which are based on the production pathway flow diagram in Fig. 1. The nomenclature system is necessary for reporting results within the analytical framework that is subsequently presented.

Fig. 1
figure 1

Renewable biodiesel production can be represented in simplified form as three phases: growth, processing, and refining

Full size image

Productivity

First, the renewable diesel productivity, P RD, can be written as

$$\begin{array}{*{20}{c}} {{P_{\text{RD}}} = \frac{{{M_{\text{RD}}}}}{{{V_{\text{G}}} \cdot {t_{\text{c}}}}} = {P_{\text{GM}}} \cdot {\varphi_{\text{proc}}} \cdot {\varphi_{\text{ref}}}} & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \\ \end{array} $$
(1)

where M RD is the mass of renewable diesel produced, V G is the algal growth volume, t c is the cultivation period, and P GM is the grown (algal) mass productivity (g of dry algal biomass/(L-day)). In this proposed nomenclature system, “productivity” is a volumetric measure, based on the growth volume (V G). Although areal productivity is important for many studies [9, 12, 42–44], to determine how much renewable diesel can be produced, volumetric productivity is also a critical measure. It is important to specify the dry weight measurement method used, and it is best to measure organic dry weight (i.e., by removing inorganic solids). The processing efficiency, φ proc, can be defined as the mass of biocrude, M BC, that is obtained by a processing method divided by the amount of (dry) grown mass, M GM, that was present in the growth medium prior to processing (g biocrude/g grown mass), as shown in Eq. 2. The dry grown mass is calculated as the algal concentration (g/L) multiplied by the growth volume (L).

$$ \begin{array}{*{20}{c}} {{\varphi_{\text{proc}}} = \frac{{{M_{\text{BC}}}}}{{{M_{\text{GM}}}}}} & {\left[ - \right]} \\ \end{array} $$
(2)

The refining efficiency, φ ref, is defined as the amount of renewable diesel, M RD, that is produced from an associated amount of biocrude, M BC, which can be expressed as

$$ \begin{array}{*{20}{c}} {{\varphi_{\text{ref}}} = \frac{{{M_{\text{RD}}}}}{{{M_{\text{BC}}}}}} & {\left[ - \right]} \\ \end{array} $$
(3)

Cost

A similar top-level cost analysis of algae-derived renewable diesel production can also be created. The cost of renewable diesel, C RD, can be written as

$$ \begin{array}{*{20}{c}} {{C_{\text{RD}}} = {c_{\text{RD}}} \cdot {\rho_{\text{RD}}} = {{\tilde{C}}_{\text{G}}} + {{\tilde{C}}_{\text{P}}} + {{\tilde{C}}_{\text{R}}}} & {\left[ {\frac{{\text{\$ }}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(4)

where \( {\tilde{C}_{\text{G}}} \), \( {\tilde{C}_{\text{P}}} \), and \( {\tilde{C}_{\text{R}}} \) are the cost of growth, processing, and refining, in dollars per liter of renewable diesel. A tilde (~) is used to differentiate the cost of a production step (i.e., growth, processing, or refining), from the cost of a product (i.e., grown mass, biocrude, or renewable diesel). In Eq. 4, c RD is the cost of producing renewable diesel per kilogram of renewable diesel, where the lowercase “c” denotes a cost on a per mass basis. Thus, the product of c RD and the density of renewable diesel, ρ RD, is equal to the cost of producing renewable diesel per liter. The units used in Eq. 4 are adopted because many products are priced by volume, rather than by mass.

Each cost on the right-hand side of Eq. 4 can be expanded. For instance, the cost of growing algal biomass, \( {\tilde{C}_{\text{G}}} \), can be expanded as

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{G}}} = {{\tilde{c}}_{\text{G}}} \cdot {\text{GMCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{{\text{\$ }}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(5)

In Eq. 5, \( {\tilde{c}_{\text{G}}} \) is the cost of growing algal biomass, in dollars per kilogram of (dry) grown mass, and ρ RD is the density of renewable diesel. GMCF is the grown mass conversion factor, which is the amount of dry biomass that must be grown in order to produce an associated mass of renewable diesel. Conversion factors, such as GMCF, are dependent upon the subsequent production efficiencies, and the GMCF is defined as

$$ {\text{GMCF}} = \frac{1}{{{\varphi_{\text{proc}}}}}\left[ {\frac{{{\text{k}}{{\text{g}}_{\text{GM}}}}}{{{\text{k}}{{\text{g}}_{\text{BC}}}}}} \right] \cdot \frac{1}{{{\varphi_{\text{ref}}}}}\left[ {\frac{{{\text{k}}{{\text{g}}_{\text{BC}}}}}{{{\text{k}}{{\text{g}}_{\text{RD}}}}}} \right] $$
(6)

The cost of the processing phase, \( {\tilde{C}_{\text{P}}} \), in dollars per liter of renewable diesel can likewise be expanded as

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{P}}} = {{\tilde{c}}_{\text{P}}} \cdot {\text{BCCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(7)

where \( {\tilde{c}_{\text{P}}} \) is the cost of the processing phase in dollars per kilogram of biocrude and BCCF is the biocrude conversion factor, which is defined as

$$ \begin{array}{*{20}{c}} {{\text{BCCF}} = \frac{1}{{{\varphi_{\text{ref}}}}}} & {\left[ {\frac{{{\text{k}}{{\text{g}}_{\text{BC}}}}}{{{\text{k}}{{\text{g}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(8)

Finally, the refining cost can be expressed as

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{R}}} = {{\tilde{c}}_{\text{R}}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\$ }{{{L_{\text{RD}}}}}} \right]} \\ \end{array} $$
(9)

where \( {\tilde{c}_R} \) is the cost of refining in dollars per kilogram of refined product (i.e., renewable diesel).

It may also be useful to report cost results on a mass basis. The products of the production phases are grown mass, biocrude, and renewable diesel. The cost of producing these products can be expressed as shown in Table 3. Two expressions for the cost of producing renewable diesel on a mass basis ($/kg of renewable diesel) are also included in Table 3. The financial return on investment, FROIRD, can be expressed as

$$ \begin{array}{*{20}{c}} {{\text{FRO}}{{\text{I}}_{\text{RD}}} = \frac{{{R_{\text{RD}}} + {R_{\text{CP}}}}}{{{C_{\text{RD}}}}}} & {\left[ - \right]} \\ \end{array} $$
(10)

where R RD is the revenue generated by renewable diesel, R CP is the revenue generated from co-products, and C RD is the cost of producing renewable diesel. Each of these terms is measured in units of dollars per liter per day; however, the units could be adjusted for batch processing as needed. If additional processing is required for co-products (e.g., converting the biomass co-product to fuels or chemicals), the associated processing costs should be included in the denominator of Eq. 10.

Table 3 Financial costs, energy requirements, and return on investment for renewable diesel production
Full size table

Equations 4, 5, 6, 7, 8, 9, and those listed in Table 3 demonstrate the difference between the cost of production steps, which are denoted with a tilde (e.g., cost of growth, \( {\tilde{c}_{\text{G}}} \)), and the cost of products, which do not include a tilde (e.g., the cost of grown mass, c GM).

Energy

It is important to determine the financial cost and the energy ratio associated with producing renewable diesel from algae. In general, the financial and energy costs should be directly related. However, the economics of energy production includes many variables that can skew this relationship [45, 46]. The energy required to produce renewable diesel (direct and indirect), E RD, in joules per liter, can be calculated as

$$ \begin{array}{*{20}{c}} {{E_{\text{RD}}} = {e_{\text{RD}}} \cdot {\rho_{\text{RD}}} = {{\tilde{E}}_{\text{G}}} + {{\tilde{E}}_{\text{P}}} + {{\tilde{E}}_{\text{R}}}} & {\left[ {\frac{\text{J}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(11)

where \( {\tilde{E}_{\text{G}}} \), \( {\tilde{E}_{\text{P}}} \), and \( {\tilde{E}_{\text{R}}} \) are the energy requirements for the growth, processing, and refining production steps of renewable diesel. The energy required to produce a kilogram of renewable diesel is defined as e RD. Allocating direct and indirect energy requirements for energy production systems has been presented in previous studies [47–55]. The energy required to produce renewable diesel, E RD, can be calculated using the methods described by Mulder and Hagens [47].

Each of the terms on the right-hand side of Eq. 11 can be further expanded as

$$ \begin{array}{*{20}{c}} {{{\tilde{E}}_{\text{G}}} = {{\tilde{e}}_{\text{G}}} \cdot {\text{GMCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\text{J}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(12)
$$ \begin{array}{*{20}{c}} {{{\tilde{E}}_{\text{P}}} = {{\tilde{e}}_{\text{P}}} \cdot {\text{BCCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\text{J}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(13)
$$ \begin{array}{*{20}{c}} {{{\tilde{E}}_{\text{R}}} = {{\tilde{e}}_{\text{R}}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\text{J}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(14)

where \( {\tilde{e}_{\text{G}}} \), \( {\tilde{e}_{\text{P}}} \), and \( {\tilde{e}_{\text{R}}} \) are the energy requirements of each production step (growth, processing, and refining) per kilogram of output product (grown mass, biocrude, and renewable diesel, respectively). As done for the financial cost of each product, the energy costs can be evaluated on a per mass basis (J/kg of product) as shown in Table 3.

Based on the framework by Mulder and Hagens, the second-order energy return on investment for renewable diesel, EROIRD, can be defined as

$$ \begin{array}{*{20}{c}} {{\text{ERO}}{{\text{I}}_{{\text{RD}}}} = \frac{{{\text{E}}{{\text{C}}_{{\text{RD}}}} + {\text{E}}{{\text{C}}_{CP}}}}{{{E_{{\text{RD}}}}}} = \frac{{{\text{LHV}} \cdot {\rho _{{\text{RD}}}} + {\text{E}}{{\text{C}}_{{\text{CP}}}}}}{{{E_{{\text{RD}}}}}}} & {[ - ]} \\ \end{array} $$
(15)

where ECRD is the energy content of renewable diesel, ECCP is the energy content of co-products, and E RD is the energy required (direct and indirect) to produce renewable diesel. Also in Eq. 15, LHV is the lower heating value of renewable diesel (J/kg), and ρ RD is the renewable diesel density (kg/L). If additional energy is required to produce the final form co-products, this energy requirement should be included in the denominator of Eq. 15.

Production Pathways

The three production pathways listed above are presented in more detail in this section and displayed in Figs. 2, 3, and 4. Within each pathway diagram, several different technology options are listed for each conversion process to advance from one level of a production pathway to the next. More research is needed to identify the most advantageous technology for each conversion step. Also, different algal species may require different conversion technology choices. Therefore, the selection of algal species and efficient processing technologies for renewable diesel production are inter-related.

Fig. 2
figure 2

The production of algal biodiesel via transesterification of algal lipids contains several steps, and each step can be accomplished with various technologies. †Mechanical damage includes French press, bead beater, mortar and pestle, etc. *At this resolution, it is not clear which lipids will be useful for biodiesel production; therefore, these lipids could be an assortment or of a single type. **Biodiesel is a type of renewable diesel. a [20, 23–28], b [11, 29–34], c [27, 31, 35], d [27, 33, 36–40]

Full size image
Fig. 3
figure 3

The production of renewable diesel by thermochemical conversion of algal biomass contains several production steps, and many thermochemical conversion processes exist. †Harvesting may or may not include complete drying, depending on the thermochemical conversion process. *Co-products composition depends on the conversion process used. a [20, 23–28], b [11, 33, 37, 56–67], c [11, 33, 56, 59, 64], d [27, 33, 36–40]

Full size image
Fig. 4
figure 4

The production of renewable diesel from algal oil that is secreted into the growth medium is a relatively new approach. *In this process, the algal culture often consists of genetically modified organisms [68–76]

Full size image

Each pathway is segmented into the three production phases listed in Fig. 1 (i.e., growth, processing, and refining). All of the production steps within each phase impact the efficiency, cost, and energy requirement for that phase. In this regard, the pathways shown here can be further expanded to include additional sub-level production steps and can be tailored to accommodate other production methods. The degree to which a pathway is segmented into discrete steps and the categories that are used to group the steps are somewhat arbitrary and left to the discretion of the practitioner.

Transesterification of Extracted Lipids

Figure 2 is a flow chart showing the main processes that are required to produce biodiesel, a form of renewable diesel, from extracted algal lipids, which is the most commonly investigated of the three pathways presented here. After algae are grown in an open pond, photobioreactor, or fermentor (i.e., heterotrophic growth), the algae are harvested from the growth medium. After harvesting, the lipids are extracted from the algal cells. This extraction generally consists of a lysing process to rupture the cells followed by a separation of the lipids from the other biomass. Alternatively, direct solvent extraction can be conducted to extract lipids, although it may be infeasible on the industrial scale. Additional separations may be required to obtain only the lipids that are specifically useful for biodiesel production. These lipids are referred to as useful lipids and are discussed in more detail below. Once the useful lipids have been separated, they are converted to fatty acid methyl esters (FAME) via transesterification. Biodiesel is a term that we use to refer to a composition consisting mainly of fatty acid methyl esters that complies with standard fuel specifications. There is also potential for producing valuable co-products, such as glycerol, protein, or biomass. The biomass co-product could be converted to fuels (via anaerobic digestion or gasification) or used to produce electricity (via direct combustion). A more detailed discussion of the intermediate products in this production pathway is provided in Appendix 2.

Thermochemical Biomass Conversion

Thermochemical conversion of algal biomass is another processing method that can be used to produce renewable diesel, and the associated production pathway is shown in Fig. 3. After the algae are grown, they are harvested and, in some cases, dried (depending on the thermochemical conversion process applied). Biomass is the substrate for thermochemical processes, and therefore, the entire algal cell undergoes conversion. The most common thermochemical processes are liquefaction, pyrolysis, and gasification with subsequent Fischer–Tropsch conversion.

Each of these processes converts algal biomass to products that can potentially be upgraded to renewable diesel. Liquefaction converts high molecular weight organic compounds to low molecular weight oils at temperatures around 250–350°C, high pressure (0.5–20 MPa), and often with the aid of a catalyst [11, 33, 37, 56–58, 77, 78]. Pyrolysis is defined as the conversion of high molecular weight organic compounds to oil under high temperature (~480–700°C), in the absence of oxygen, and under operating pressures of ~0.1–0.5 MPa [37, 59, 77–79]. Algae may also be converted to syngas by gasification, which could then be converted to biocrude by the Fischer–Tropsch process [9, 60–63, 80–84]. The co-products of each thermochemical conversion process vary and can include gases, aqueous liquids, and solid char.

Following thermochemical conversion, the oils (i.e., biocrude) are separated (usually by solvent extraction with chloroform or dichloromethane) [11, 33, 56, 59, 64] and may be refined into renewable diesel. There is also potential for the production of valuable co-products with thermochemical algal biomass conversion. For instance, the gaseous product contains methane, and the solid char residue could be used as a combustion fuel or soil additive.

Conversion of Secreted Algal Oils

Another method that has been explored for renewable diesel production is the collection and conversion of secreted algal oils (sometimes called milking). The aim of this method is to use genetically modified organisms that secrete oils into the growth medium. The production of renewable diesel from secreted algal oils is the least mature of the three production pathways presented here. Much of the work in this area to date is proprietary. As a result, the feasibility of this production pathway is unclear. The increased cost of engineering a suitable organism and maintaining a monoculture may be offset by a reduction in processing cost required to produce renewable diesel, as compared to the lipid extraction and thermochemical conversion production pathways shown above. The most general steps required for producing renewable diesel from secreted algal lipids are shown in Fig. 4.

The first processing step required in this production pathway is the separation of secreted oil from the growth medium. The term “oils” is used here rather than “lipids” because the exact composition of the secreted products is not yet known. As a result, secondary separations may be required to recover oils that are specifically useful for refining into renewable diesel

Framework Principles

In this section, a proposed framework is presented that uses the nomenclature developed in this work (specifically, for transesterification of algal lipids as detailed in Appendix 2). The framework is based on three principles: using strong reporting metrics, using symbolic notation to include unknown values, and ensuring that results are presented consistently. Each of these principles is discussed in detail below.

Strong Metrics

First, results should be reported with the strongest metric possible. The strength of a metric refers to the amount of information relevant to renewable diesel production that it contains. For instance, the metric “triacylglycerol per dry weight” is stronger than “lipid per dry weight” because it includes additional information about the composition of the lipids. Similarly, the renewable diesel productivity, P RD, is a stronger metric than triacylglycerol productivity, P TAG, and this concept is illustrated in Fig. 5.

Fig. 5
figure 5

Metric strength increases as the amount of information that the metric contains increases. It is important that results are consistent in algal species and growth conditions and include all relevant inputs and processing steps. It is assumed for this figure that renewable diesel (specifically, biodiesel) can only be made from triacylglycerol, rather than all lipids

Full size image

The scope of a particular study determines the amount of information that is obtained and the associated reporting metric. Figure 5 illustrates how metric strength increases as the breadth of information conferred by the metric increases. It does not include all relevant metrics for renewable diesel production from algae. For example, a primary study may focus on determining the amount of CO2 required for large-scale algal cultivation, a subset of the “Materials Consumed for Growth” metric.

Figure 6 lists the productivity, cost, and energy requirements associated with each intermediate product in the production pathway (specifically, transesterification of algal lipids) in order of metric strength.

Fig. 6
figure 6

The productivity, cost, and energy requirements associated with producing each intermediate product can be written as an inverted pyramid, with the strongest metrics representing the entire production pathway. The subscripts on the left-hand side of the equations are renewable diesel (RD), fatty acid methyl esters (FAME), separated useful lipids (SUL), lysed mass (LM), harvested mass (HM), and grown mass (GM). The overall units for these equations are grams per liter of growth volume per day, dollars per liter of renewable diesel, and joules per liter of renewable diesel for the productivity, cost, and energy equations, respectively (recall that the tilde denotes a processing step cost or energy requirement and that this nomenclature is specific to transesterification of algal lipids)

Full size image

Use of Symbolic Notation

The second principle for the proposed characterization framework is that results from studies with limited scope can be reported with strong metrics by including unknown information in symbolic notation. There are two main advantages of presenting information in this manner: (1) it ensures that results are not taken out of context, thus helping to avoid incomplete estimates for the potential of renewable diesel from algae, and (2) it explicitly identifies the areas where additional data are needed to complete the production pathway analysis. In addition, using symbolic notation enables results to be incorporated into systems-level analyses more directly. The nomenclature used in this section is described in detail in Appendix 2.

To demonstrate reporting results with unknowns in symbolic notation in an example, the triacylglycerol productivity, P TAG, of a culture can be expressed as

$$ \begin{array}{*{20}{c}} {{P_{\text{TAG}}} = \frac{{{M_{\text{TAG}}}}}{{{V_{\text{G}}} \cdot t}} = {P_{\text{GM}}} \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{cellys}}} \cdot {\varphi_{\text{sep}}}} & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \\ \end{array} $$
(16)

where P GM is the grown mass productivity and the efficiencies are defined in Appendix 2. Note that triacylglycerol is a subset of useful lipids, and therefore, the separation efficiency (cf. Eq. 33), φ sep, can be evaluated as

$$ \begin{array}{*{20}{c}} {{\varphi_{\text{sep}}} = {\text{LF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{L}}}}} \cdot {\text{TAGF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{TAG}}}}}} & {[ - ]} \\ \end{array} $$
(17)

where TAGF is the triacylglycerol fraction, which is the fraction of lipids that are triacylglycerol (g triacylglycerol/g lipid), and \( {\varphi_{{\text{se}}{{\text{p}}_{\text{TAG}}}}} \) is the efficiency with which the triacylglycerol can be separated from the other lipids. These terms have been substituted for the useful lipid fraction, ULF, and useful lipid separations efficiency, \( {\varphi_{{\text{se}}{{\text{p}}_{\text{UL}}}}} \), of Eq. 33. In a situation where a researcher may not have all of the above information, the results could be reported in symbolic notation. For instance, in a study by Richmond et al., Nannochloropsis salina was produced at a rate of 24.5 g/(m2-day) with a lipid fraction of about 16% in a pond with 0.12 m depth (cf. Sheehan et al. 1998, p. 191). This information translates to a triacylglycerol productivity that can be reported as

$$ \begin{array}{*{20}{c}} {{P_{\text{TAG}}} = \frac{{24.5}}{{0.12 \cdot 1,000}} \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{cellys}}} \cdot 0.16 \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{L}}}}} \cdot {\text{TAGF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{TAG}}}}}} & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \\ {{P_{\text{TAG}}} = 0.033 \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{cellys}}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{L}}}}} \cdot {\text{TAGF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{TAG}}}}}} & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \\ \end{array} $$
(18)

Leaving the triacylglycerol fraction, TAGF, in symbolic form helps to avoid using the terms “lipid” and “triacylglycerol” synonymously. Furthermore, leaving the processing efficiencies in symbolic notation clarifies that the triacylglycerol productivity, P TAG, is dependent on the processing methods.

To further illustrate the use of symbolic notation, renewable diesel productivity, P RD, can be used to present the results for heterotrophic growth of Chlorella protothecoides presented by Li et al. [18]. The 8,000-L growth volume used in that study produced an algal density of 14.2 g/L with a lipid fraction, LF, of 44.3%, and 98% of the lipids were converted to FAME via transesterification. The cultivation period was 8.33 days, yielding a grown mass productivity, P GM, of 1.70 g/(L-day). However, the transesterification efficiency was not reported on a mass basis, the harvesting efficiency was not reported, and post-processing was not conducted. Therefore, these values can be best represented in symbolic notation, and the renewable diesel productivity in the study by Li et al. can be reported as

$$ \begin{array}{*{20}{c}} {{P_{\text{RD}}} = {P_{\text{GM}}} \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{cellys}}} \cdot {\varphi_{\text{sep}}} \cdot {\varphi_{\text{trans}}} \cdot {\varphi_{\text{post}}}} \hfill & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \hfill \\ {{P_{\text{RD}}} = {P_{\text{GM}}} \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{cellys}}} \cdot {\text{LF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{L}}}}} \cdot {\text{ULF}} \cdot {\varphi_{{\text{se}}{{\text{p}}_{\text{UL}}}}} \cdot {\varphi_{\text{trans}}} \cdot {\varphi_{\text{post}}}} \hfill & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \hfill \\ { = 1.70 \cdot {\varphi_{\text{harv}}} \cdot 1 \cdot 0.443 \cdot 1 \cdot 1 \cdot 1 \cdot {\varphi_{\text{trans}}} \cdot {\varphi_{\text{post}}}} \hfill & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \hfill \\ \end{array} $$
(19)

or

$$ \begin{array}{*{20}{c}} {{P_{\text{RD}}} = 0.75 \cdot {\varphi_{\text{harv}}} \cdot {\varphi_{\text{trans}}} \cdot {\varphi_{\text{post}}}} & {\left[ {\frac{\text{g}}{{{{\text{L}}_{\text{G}}}\;{\text{day}}}}} \right]} \\ \end{array} $$

In Eq. 19, the cell lysing efficiency, φ cellys, lipid separation efficiency, \( {\varphi_{{\text{se}}{{\text{p}}_{\text{L}}}}} \), and the useful lipid separation efficiency, \( {\varphi_{{\text{se}}{{\text{p}}_{\text{UL}}}}} \), were assumed to be unity because the lipid fraction was determined from the amount of lipid separated, thus already containing the lysing and lipid separations efficiency, and all extracted lipids were used for the production of FAME. For this calculation, it is also assumed that the lipid fraction remains constant throughout harvesting and lysing. The validity of that assumption is not known.

Similarly, symbolic notation can be used to report financial and energy costs. For example, Schenk et al. [16] cited the cost of producing algal oil (between $126 and $209 US(2008)/bbl or between $0.79 and $1.31 US(2008)/L) based on Seambiotic Inc. biomass growth and harvesting costs of $0.34/kg of dry algae (i.e., harvested dry mass). The algae in that study were reported to have a lipid fraction, LF, of 24%. Schenk et al. assumed no additional processing cost [16]. We suggest that the cost of producing renewable diesel from this biomass is aptly characterized by Eq. 20, which is

$$ \begin{array}{*{20}{c}} {{C_{\text{RD}}} = {c_{\text{RD}}} \cdot {\rho_{\text{RD}}} = {{\tilde{C}}_{\text{G}}} + {{\tilde{C}}_{\text{H}}} + {{\tilde{C}}_{\text{CL}}} + {{\tilde{C}}_{\text{S}}} + {{\tilde{C}}_{\text{T}}} + {{\tilde{C}}_{\text{PP}}}} & {\left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(20)

To evaluate Eq. 20, the growth and harvesting processing costs can be combined to obtain an expression for the cost of harvested (algal) mass. Combining Eqs. 5 and 41, the processing costs of growing and harvesting algal biomass in dollars per liter of renewable diesel, \( {\tilde{C}_{\text{G}}} + {\tilde{C}_{\text{H}}} \), can be expanded as

$$ {\tilde{C}_{\text{G}}} + {\tilde{C}_{\text{H}}} = {C_{\text{HM}}} = {\tilde{c}_{\text{G}}} \cdot {\text{GMCF}} \cdot {\rho_{\text{RD}}} + {\tilde{c}_{\text{H}}} \cdot {\text{HMCF}} \cdot {\rho_{\text{RD}}} \left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right] $$
(21)

where C HM represents the cost of harvested (algal) mass (i.e., grown and harvested) per liter of renewable diesel. Equation 21 reduces to

$$ {\tilde{C}_{\text{G}}} + {\tilde{C}_{\text{H}}} = \left( {{{\tilde{c}}_{\text{G}}} \cdot \frac{1}{{{\varphi_{\text{harv}}}}} + {{\tilde{c}}_{\text{H}}}} \right) \cdot {\text{HMCF}} \cdot {\rho_{\text{RD}}} \left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right] $$
(22)

and the cost of harvested (algal) mass per kilogram, c HM, is

$$ \begin{array}{*{20}{c}} {{c_{\text{HM}}} = \left( {{{\tilde{c}}_{\text{G}}} \cdot \frac{1}{{{\varphi_{\text{harv}}}}} + {{\tilde{c}}_{\text{H}}}} \right) } & {\left[ {\frac{{\text{\$ }}}{{{\text{k}}{{\text{g}}_{\text{HM}}}}}} \right]} \\ \end{array} $$
(23)

Equation 23 is useful to illustrate the distinction between \( {\tilde{c}_{\text{H}}} \), which is the cost of the harvesting process per kilogram of harvested mass, and c HM, which is the total cost of producing harvested mass (which includes growth costs). Equation 21 can therefore be written more concisely as

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{G}}} + {{\tilde{C}}_{\text{H}}} = {C_{\text{HM}}} = {c_{\text{HM}}} \cdot {\text{HMCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(24)

Equation 24 can be populated with the data reported by Schenk et al. (\( \left( {{c_{{{\text{HM}}}} = \$ 0.34} \mathord{\left/ {\vphantom {{c_{{{\text{HM}}}} = \$ 0.34} {{\text{kg}}}}} \right. \kern-\nulldelimiterspace} {{\text{kg}}}} \right. \) and \( \left. {\rho _{{{\text{RD}}}} = {0.92{\text{kg}}} \mathord{\left/ {\vphantom {{0.92{\text{kg}}} {\text{L}}}} \right. \kern-\nulldelimiterspace} {\text{L}}} \right) \)), and the cost of producing harvested (algal) dry mass per liter of renewable diesel can be calculated as

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{G}}} + {{\tilde{C}}_{\text{H}}} = {C_{\text{HM}}} = 0.34 \cdot {\text{HMCF}} \cdot 0.92} & {\left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(25)

Evaluating the lipid fraction, LF, as 24% in HMCF and reducing yields

$$ \begin{array}{*{20}{c}} {{{\tilde{C}}_{\text{G}}} + {{\tilde{C}}_{\text{H}}} = 1.31 \cdot \frac{1}{{{\varphi_{\text{cellys}}}}} \cdot \frac{1}{{{\varphi_{\text{sep}}}}} \cdot \frac{1}{{{\varphi_{\text{trans}}}}} \cdot \frac{1}{{{\varphi_{\text{post}}}}}} & {\left[ {\frac{\$ }{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(26)

Then, using the cost of the transesterification process per liter of FAME produced (i.e., \( {\tilde{c}_{\text{T}}} \cdot {\rho_{\text{RD}}} \)) to be about $0.13 US(2008)/L [22], the total cost of producing algae-derived renewable diesel (combining Eq. 26 with Eq. 20) can be expressed as

$$ \begin{array}{*{20}{c}} {{C_{\text{RD}}} = 1.31 \cdot \frac{1}{{{\varphi_{\text{cellys}}}}} \cdot \frac{1}{{{\varphi_{\text{sep}}}}} \cdot \frac{1}{{{\varphi_{\text{trans}}}}} \cdot \frac{1}{{{\varphi_{\text{post}}}}} + {{\tilde{C}}_{\text{CL}}} + {{\tilde{C}}_{\text{S}}} + 0.13 \cdot \frac{1}{{{\varphi_{\text{post }}}}} + {{\tilde{C}}_{\text{PP}}}} & {\left[ {\frac{{\text{\$ }}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(27)

Using symbolic notation can also improve the consistency of reporting results associated with the net energy ratio for producing algal renewable diesel. For example, Benemann and Oswald present an energy analysis for fuel inputs required for growing and harvesting algae and report the energy requirement for producing harvested algal biomass, e HM, of between 0.924 and 1.202 kJ/kg (note, \( {\tilde{e}_{\text{GM}}} \cdot \frac{1}{{{\varphi_{\text{harv}}}}} + {\tilde{e}_{\text{H}}} = {e_{\text{HM}}} \), where φ harv is assumed to be 1, cf. Appendix 2) [22]. Therefore, as described in detail in Appendix 2, using e HM = 1.202 kJ/kg, the energy required for growing and harvesting per liter of renewable diesel, \( {\tilde{E}_{\text{G}}} + {\tilde{E}_{\text{H}}} \), can be approximated as

$$ \begin{array}{*{20}{c}} {{{\tilde{E}}_{\text{G}}} + {{\tilde{E}}_{\text{H}}} = 1.202 \cdot {\text{HMCF}} \cdot {\rho_{\text{RD}}}} & {\left[ {\frac{\text{kJ}}{{{{\text{L}}_{\text{RD}}}}}} \right]} \\ \end{array} $$
(28)

Using this value, a LHV of renewable diesel to be 41 MJ/kg [19], and the density of renewable diesel, ρ RD, as 0.92 kg/L in the energy return on energy investment (Eq. 15) yields

$$ \begin{array}{*{20}{c}} {{\text{ERO}}{{\text{I}}_{\text{RD}}}{ = }\frac{{{41,000} \cdot {0}{.92} + {\text{E}}{{\text{C}}_{\text{CP}}}}}{{{1}{.202} \cdot {\text{HMCF}} \cdot {0}{.92} + {{\tilde{E}}_{\text{CL}}} + {{\tilde{E}}_{\text{S}}} + {{\tilde{E}}_{\text{T}}} + {{\tilde{E}}_{\text{PP}}}}}} & {[ - ]} \\ \end{array} $$
(29)

Combining Eqs. 28 and 53 produces the denominator of Eq. 29. One can see that the EROI is dependent upon the energy requirements of all processing steps, and additional data are needed to accurately assess these terms. The energy ratio is also dependent on the allocation of indirect energy requirements. For example, Clarens et al. [51] include the energy embedded in nutrients (including CO2) in their life cycle analysis, which yields a growing and harvesting energy requirement of 22,710 kJ/kg of harvested algae. This result is four orders of magnitude greater than the estimate provided by Benemann and Oswald that is used in Eq. 29. Lardon et al. and Beal et al. have also conducted a net energy balance associated with renewable diesel and have also used unique system boundaries [85, 113].

Reporting Consistency

The third principle for the characterization framework is that results associated with renewable diesel production from algae should be reported consistently. Consistent refers to presenting results that are specific (to algal species, growth conditions, and product composition) and inclusive (of all inputs and processing steps) and that consider the energy, materials, and cost associated with all relevant production pathway steps.

Specific Results

Due to the scope of a particular study, some reported results are not explicit with respect to algal species, growth conditions, or product composition [11, 17, 38, 59, 86–91]. As a result of the variety among algal species, even within a single genus, it is important that the characteristics of one alga are not mixed with those of another for analytical calculations. The energy, cost, and materials required for production typically vary with species. This lack of specificity in the results of many studies limits their utility.

It is also important to be specific when citing results that are dependent upon growth conditions. Combining results for the maximum lipid fraction (g lipid/g of algal mass) of a particular alga (often obtained under nutrient deficient conditions) with the maximum growth rate of that alga (generally obtained under nutrient replete conditions) introduces an inconsistency in the resulting lipid productivity. This practice is misleading because growth rate and lipid production are generally inversely related [1].

There are several different metrics that are often used to evaluate algal biofuel potential including lipid content, neutral lipid content, triacylglycerol content, etc. If these terms are improperly used as synonyms, comparisons among various results are not direct comparisons. For instance, the terms “oil” or “algae oil” are frequently used for reporting results without defining the chemical composition of these substances. These terms have been used to refer to algal lipids, biodiesel, and even ethanol. The lack of specificity regarding these metrics needlessly limits the value of the published results.

Finally, neglecting to distinguish among types of biofuels can also introduce ambiguity in analyses. For example, ethanol contains about 70% of the energy content per volume of biodiesel [19, 92].

Inclusive Results

Many studies specifically addressing the production of renewable diesel from algae could be more widely useful if they included information encompassing more of the production pathway [16, 17, 19, 64, 91, 93–96]. While the scope of a study determines the breadth of information available, it is useful to rigorously place the work in the context of the entire production pathway. As suggested by Griffiths and Harrison [41], this benefit is particularly true for studies that include information regarding the products of interest in algal cultures for biodiesel production (i.e., lipids or triacylglycerol). For example, if a researcher evaluates the impact of different nutrients on lipid production, the results for that study are most useful if they provide more information than simply the lipid fraction, LF, of the cultures.

It is also important to include information about all relevant parts of the production pathway due to the variability of algal cultures. Downstream processing studies have been conducted to evaluate the efficiency of processing algae or “algae oil,” without including relevant information about the algal species, lipid content, or oil composition [93, 94]. Since growth rate, lipid content, and lipid composition can vary widely depending on species or growth conditions [1, 21, 41, 97–105], the processing efficiencies determined for one alga or one composition of “algae oil” may differ from those associated with different algae or oil compositions. The resources required for production may also vary depending on species [58]. Including the algal species and/or the composition of the tested algal oil would reduce these inconsistencies.

Also, there is a significant discrepancy among costs on a lab scale, pilot scale, and commercial scale and among results obtained for short-term versus long-term experiments. Correlating data among these scales and time frames is challenging, and the scalability of algal production for biofuels is an ongoing area of research. To advance this research area, which is critical for producing accurate estimates of the potential of algae for renewable diesel, it is valuable to be as specific about the growth volume and time period as possible because the cultivation scale can impact growth characteristics [1, 16].

Finally, several systems-level analyses have calculated estimates for the total land area (for open ponds) or growth volume (for bioreactors) required to cultivate enough microalgae needed to produce a specified amount of diesel fuel substitute (such as to satisfy the US diesel consumption) [16, 17, 86, 87, 106]. Other studies present estimates for the total financial cost of producing “algae oil” in terms such as dollars per barrel [1, 16, 17, 20, 22, 88, 95, 106]. However, some of these analyses, including some analyses listed in Table 1, omit important pieces of the production pathway, leading to inconsistent results.

Energy, Materials, and Cost Balances

The third way to improve reporting consistency is by considering the energy, materials, and cost requirements for each production process. These requirements are relevant for specific studies conducted by primary researchers (e.g., energy required to grow a culture) and especially for systems-level researchers who characterize the potential for algae-based biofuel. In fact, the two are directly linked. Without energy, materials, and cost information associated with primary studies of individual processes, it is impossible to compile the energy balance of the entire production pathway.

There are numerous primary studies on algal growth, lipid composition, and processing methods [1, 21, 41–43, 64, 98, 100–102, 104, 107–110]. The impact of many primary studies could potentially be increased with the inclusion of energy, cost, and material requirements. For instance, if a study on the lipid fraction of different algal species includes the amount of energy, materials, and cost required to produce the lipid, its results could be more broadly interpreted. Information that is not relevant or not known can be presented in symbolic notation, as suggested above, to enable the use of strong reporting metrics.

It is also important that top-level analyses of renewable diesel produced from algae address the energy, materials, and cost requirements for renewable diesel production consistently. Significant inconsistencies can arise when the requirements for entire production steps are omitted or oversimplified. Several systems-level analyses have been published for renewable diesel from algae (primarily biodiesel), each with a different amount of information regarding energy, cost, and material requirements [1, 16, 17, 20, 22, 37, 58, 88, 95, 106, 111]. These systems-level cost analyses provide good outlines for conducting cost estimates, but sometimes lack specificity to algal species or growth conditions, and may omit some required processing steps. Processing efficiencies and resource requirements may depend on the algal species and composition [16, 20, 23, 58, 93].

Conclusion

For the field of algae-derived renewable diesel to progress, the community of researchers needs to provide accurate answers to the three questions: (1) how much renewable diesel can be produced, (2) how much will this renewable diesel cost, and (3) what are the energy requirements for production? We have proposed a framework and associated nomenclature system for characterizing the potential of algae for renewable diesel that outlines a method for presenting consistent, widely interpretable results. This framework consists of three principles: using strong metrics, using symbolic representation for unknown information, and presenting results that are consistent and include all relevant information. Widespread use of common nomenclature and a consistent reporting framework by primary researchers would allow systems-level analysts to integrate the results of primary research into estimates for the potential of algae for renewable diesel. In turn, widespread use of a framework by systems-level analysts would lead to improved estimates, which are valuable for researchers and policy makers. Accurate and informative estimates of the potential of renewable diesel will help researchers focus their efforts on the most pressing problems and help policy makers make appropriate decisions about funding and resource allocation related to algal biofuel development.