1 Introduction

Microalgae have been touted as a potential and promising feedstock to produce a variety of end-products (including biofuels, chemicals, biomaterials and various other useful energy products), due to their high quantities of lipids, carbohydrates, proteins, pigments, vitamins and enzyme contents [1, 2]. In comparison with other bioenergy feedstocks (such as first generation and second generation feedstocks), microalgae offer many potential advantages as a feedstock for the biofuels production and other energy applications that have been discussed widely in the Refs. [3,4,5,6].

Potential microalgae advantages [3,4,5,6] include: (1) high growth rate (typically, microalgae can double their biomass in a time period of less than 24 h), (2) higher photon conversion efficiency compared with terrestrial plants, (3) high CO2 sequestration capacity, (4) waste water-based growth; therefore, reducing the fresh water use, (5) growth on non-arable land; therefore, the cultivation of microalgae does not compete with the arable land-based food/crop production, (6) provide potential benefits associated to waste water treatment by possessing the capability to obtain the nutrients from waste water streams (especially phosphorous and nitrogen), (7) minimize environmental impact, and (8) potential feedstock for the production of valuable co-products such as pigments, proteins, fertilizer, animal feed, etc. Furthermore, many microalgal species are rich in oil [7], thus receiving significant attention during the last decades [8, 9].

Microalgae can utilize CO2 as a carbon source (1 kg of dry microalgal biomass utilizes about 1.83 kg of CO2 [3]), hence, the cultivation of microalgae can contribute to restore the atmospheric carbon balance [10]. The idea of using microalgae as a renewable feedstock to produce biofuels and other valuable products is relatively new, and the involved technologies are in the development phase. Under the biorefinery concept, there exist many options/alternatives for microalgae species, potential processing alternatives, and a wide range of products. Even at this early development stage, it is still imperative to assess the economic potential and identify some of the most promising feedstock (species), products, and the processing routes in the viewpoint of overall economics, energy balance, and CO2 balance [11, 12]. Therefore, there is a need to develop high-level models and assess the economic and environmental potentials for a wide range of potential products from microalgae. Accordingly, the objective of this article is to identify and discuss the key issues, challenges and opportunities for the systematic design and optimization of sustainable and robust biorefinery configurations by using process systems engineering (PSE) tools and methods.

In this paper, we first review the potential processing routes/networks for producing biofuels and various platform chemicals from microalgal biomass and their residues, highlight the technical challenges, and demonstrate that the development of a systematic methodological framework can play a leading role to address these challenges. Additionally, we describe the key components of a superstructure-based modeling framework, and provide a comprehensive overview from the modeling and optimization standpoint by identifying the potential research avenues for the systematic design of sustainable, robust and integrated microalgal biorefinery structures. Key issues on sustainability modeling and uncertainty considerations in the biorefinery design phase are also discussed.

2 Concept of microalgae-based biorefinery

The term biorefinery is devised to represent a processing facility where biomass is converted into a variety of end-products such as biofuels and valuable co-products by integrating physical, chemical and bioprocessing technologies in a cost-effective and environmentally sustainable manner [3, 4, 13]. The concept of biorefinery is analogues to a conventional petroleum refinery where multiple fuels and co-products are obtained from the fossil oil [14]. The biofuels are alternatives to the fossil fuels while producing less greenhouse gases when burned [15]. Therefore, the utilization of the biofuels in the transportation and energy sectors may have less environmental impacts than the fossil fuels [16, 17]. Microalgae-based biorefinery deals with the cultivation of microalgae, and the downstream processing and conversion of microalgal biomass and their residues into a number of products, e.g., biofuels, value-added chemicals, pigments, fertilizers, animal feed, heat/power, etc. [18, 19].

Conceptually, a microalgae-based biorefinery comprises of all the processing stages/steps required for the conversion of its raw material (i.e., microalgae) into end-products. These processing stages include microalgae cultivation, dewatering and harvesting, pre-treatment of the harvested microalgal biomass (if necessary), extraction of lipids, transesterification (for conversion of lipids into biodiesel) or thermochemical processing, handling and processing of microalgal residue into valuable energy products [20]. In this study, microalgal residue (adapted from [20, 21]) represent the residuals which are left over after the lipids extraction, and mainly comprise of proteins and carbohydrates. The microalgal residue can also be utilized to produce numerous useful products such as bioethanol, bio-oil, biohydrogen, biogas, fertilizers, animal feed, nutrients, etc. [21, 22]. The energy obtained from the residue-based products (e.g., biogas) can be used for the operation of various processes within the biorefinery or can be sold [22]. Therefore, the proper exploitation of microalgae residue can improve the cost effectiveness of microalgal biorefinery in an environmentally sustainable manner. A simple illustration of the microalgae-based biorefinery is shown in Fig. 1 (modified from [11, 23, 24]). The idea of microalgal biorefinery can assist the production of microalgae-based biofuels to become an economically and environmentally sustainable option [5, 25].

Fig. 1
figure 1

(modified from [11, 23, 24])

A typical illustration of microalgal biorefinery

Full size image

2.1 Microalgae-to-biofuels pathway

Microalgae have been receiving wide-attention as a potential renewable raw material for multi-product portfolios [26]. As shown in Fig. 1, from a biorefinery perspective, the lipid contents of microalgae are generally used for biodiesel production whereas the microalgae residue can be utilized simultaneously to produce many useful energy products, through a variety of processing pathways. In this section, we describe the major processing stages for microalgae-based biofuels productions.

2.1.1 Microalgae-to-biodiesel

Biodiesel is a renewable fuel for diesel engines. It is mainly derived from vegetable oil (for example, soybean oil, canola oil, rapeseed oil, sunflower oil, palm oil, etc.) and animal fat by the transesterification or esterification process [27, 28]. However, several concerns have been raised on biodiesel sustainability from vegetable oil and animal fats due to competition with the food market [29]. An alternative production pathway for biodiesel has been touted in recent years, i.e., from microalgae. Microalgae are non-food feedstocks which clearly present many advantages over terrestrial food crops, and can be considered to be a renewable source of biodiesel production with the potential to displace fossil diesel [3, 30]. Unlike to other energy/oil crops, microalgae possess rapid growth, and many species have high oil contents. For biodiesel production, the species with high oil productivities are desired [3]. Biodiesel obtained from microalgae is carbon neutral, i.e., it fixes about as much CO2 during the microalgae growth as it releases upon combustion [31]. The major processing steps for biodiesel production from microalgae are: (1) microalgae cultivation, (2) dewatering and harvesting, (3) pretreatment (e.g., drying and cell disruption), (4) extraction of lipids, (5) transesterification, and (6) post-transesterification purification [32].

In addition, during biodiesel production via transesterification, glycerol is obtained as a by-product which can be used in many industries such as pharmaceuticals, paints, etc. [33, 34]. Furthermore, glycerol can also be utilized in order to produce many useful chemicals, for example, 1,3-propanediol that can be further used as a renewable feedstock for the production of industrial chemicals such as ethylene, propylene, etc., which are generally obtained from petroleum-based feedstocks [35, 36].

2.1.2 Microalgae residue-to-energy products/biofuels

2.1.2.1 Bioethanol

Bioethanol is one of the most promising biofuels and recognized as a potential alternative to gasoline. Generally, it is produced via fermentation route from numerous feedstocks such as sugar, corn, molasses, lignocellulosic biomass, etc. Microalgae have several advantages over sugar and corn-based feedstocks for the production of bioethanol. Most notably, they have no competition with the food market. Microalgae species rich in carbohydrates and proteins are more suitable as a feedstock for bioethanol production through fermentation processes [9]. However, bioethanol can also be produced from the microalgae residue which mainly contains carbohydrates and proteins, thus enhancing the utilization of microalgae residue for the production of valuable energy products in a profitable and environmentally sustainable manner [21, 37].

The major processing steps involved in the production of bioethanol from microalgae are: (1) hydrolysis (or pretreatment) of whole microalgae or microalgae residue, (2) fermentation process, and (3) ethanol separation and purification. Yeast, bacteria, and fungi are the micro-organisms used in the fermentation process for the production of bioethanol. During fermentation, water and CO2 are obtained as by-products. The CO2 can be recycled to be used as a carbon source for microalgae cultivation, thereby, mitigating greenhouse gas emissions as well [9].

2.1.2.2 Biogas

Biogas is produced via anaerobic digestion of organic matters. It can be used for electricity generation and many other energy applications. It can also be upgraded to biomethane to be used as a fuel gas or generate electricity [38]. Biogas mainly consists of a mixture of methane (55–75%) and carbon dioxide (45–25%). Biogas can be produced from many sources such as municipal waste, agricultural waste, manure, food waste, sewage, plant and wood waste, microalgae, etc.

Microalgae usually yield higher biogas production rates than land-based biomass [39]. After the lipids extraction, the microalgae residue can be digested anaerobically to produce biogas as a valuable byproduct. Anaerobic digestion of microalgal residue is actually a biological process that integrates renewable energy production with the residue processing in a sustainable manner. It can also provide nutrients, water and CO2 that can be used/recycled for the cultivation of microalgae [40]. The major processing steps involved in the production of biogas from microalgae residue are: (1) pretreatment of microalgae residue (if needed), and (2) anaerobic digestion of microalgae residue.

2.1.2.3 Bio-oil

Bio-oil, also known as bio-crude or pyrolysis oil, is a synthetic fuel being studied and investigated as an alternative to petroleum. It is obtained through pyrolysis (or fast pyrolysis) of biomass, and needs to be upgraded before its use as liquid fuel given its high oxygen content [28]. Both whole microalgae and microalgae residue can be converted to bio-oil through pyrolysis [21]. In some studies, it is reported that the pyrolysis of microalgal biomass results in superior (in some respects) bio-oil compared with that obtained from lignocellulosic biomass [41,42,43]. The major processing steps for bio-oil production from microalgae or microalgae residue are: (1) pretreatment of microalgal biomass/residue (if needed), and (2) pyrolysis of microalgal biomass/residue.

2.1.2.4 Biohydrogen

Microalgae possess a tremendous potential and the necessary characteristics (e.g., genetic, metabolic, enzymatic, etc.) to produce biohydrogen through anaerobic fermentation [44]. Biohydrogen is a renewable fuel and also recognized as a clean energy carrier. It possesses higher energy density compared with other biofuels [45]. Apart from its use as a fuel, hydrogen is also employed as a feedstock to produce many useful chemicals [46]. Under the concept of biorefinery, microalgae residue can be utilized to produce biohydrogen [21, 47]. However, the pretreatment of the substrate is necessary prior to the fermentation in order to enhance the hydrogen yield [48]. The major processing steps involved in the production of biohydrogen from microalgal biomass/residue are: (1) pretreatment of microalgal biomass/residue (if needed), and (2) anaerobic fermentation of microalgal biomass/residue.

2.2 Microalgae-to-specialty chemicals/products pathway

Apart from the use of microalgae for the production of biofuels, they can also be used as a potential source of various platform chemicals, biomaterials for the pharmaceutical industry, food supplements, fertilizers, animal feed, etc. [49]. Some of these products are omega-3 fatty acids, docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA) and chlorophyll, each having pharmaceutical applications [9, 50]. The selection of a particular species, its composition and characteristics, cultivation and harvesting strategies are very important and play a key role for the effective utilization of microalgae for these specific (microalgae-based) applications [50]. A number of studies can be found in the literature on the potential use of microalgae for bio-based products (i.e., platform chemicals, biomaterials, etc.). However, to the best of author’s knowledge, the specific information on the processing pathways/routes required to produce these bio-based products from microalgae is not well-documented, thus limiting the proper exploitation of microalgae for these applications under the biorefinery framework.

2.3 Difficulties in biorefinery pathways

Microalgal biorefinery deals with all the upstream and downstream processing of microalgae, from growing the microalgae-to-obtaining the end products, by integrating a number of processing stages [22]. The major processing stages required for biofuels production from microalgal biomass and their residues are described in Sect. 2.1. To carry out the corresponding task at each processing stage, there exist many potential technological alternatives. For example, open ponds and photobioreactors are the two potential alternatives for the cultivation of microalgae with a number of designs and configurations. Similarly, numerous technological alternatives with diverse combinations are available for the downstream processing stages such as harvesting of microalgal biomass, drying, cell disruption, lipids extraction, transesterification, and the processing of microalgae residues [51, 52]. In addition, there are also a large number of microalgal species such as Chlorella vulgaris, Nannochloropsis sp., Chlorella emersonii and many more that can potentially be considered as feedstock to produce multiple products, depending on their composition and growth culture [50]. Consequently, all these potential alternatives along with a number of microalgal species generate a large number of production routes/networks for the biorefinery development to produce biofuels and many other useful products from microalgae [11]. To evaluate all these alternatives and determine the most promising ones considering (1) technical, (2) economic and (3) environmental constraints represents one of the major hurdles in the design and development of profitable and sustainable biorefinery. Accordingly, there is a need to develop systematic methodologies for assessing the economic and environmental potentials of processing alternatives, and locate the most promising ones [11, 12]. Furthermore, the selection and evaluation of the potential of various microalgal species is another critical issue that requires careful attention in the biorefinery development.

Another major difficulty arises from the preliminary and uncertain nature of the technical data available in the literature on these technological alternatives [53]. The dataset comprises of the process information and/or various parameters that can be used to evaluate the processing alternatives for the biorefinery development. Some of these parameters are the composition of microalgae (especially the lipid contents), lipids productivity, CO2 fixation rate, harvesting efficiency, lipids yield, conversion of lipids into biodiesel, conversion of microalgal residue into value-added products, utilities consumption, etc. In addition to the uncertainties in the process data, the development of microalgal biorefinery is also associated with many economic uncertainties such as the uncertainty in the cost of feed, the cost of utilities, selling price of biofuels and other products obtained from microalgae, etc. [53, 54]. Moreover, further developments and future improvements in the processing technologies might be another major source of techno-economic uncertainties associated with development and design of microalgal biorefinery [53]. Consequently, the developed methodologies should also be able to handle and take the technical and economic uncertainties into account, and therefore will be very useful tools to evaluate the viability of biorefinery pathways in the future [53, 54].

3 Application of superstructure-based modeling and optimization to microalgae-based biorefinery

3.1 Overview

PSE concepts and tools can be employed to address and overcome the difficulties and challenges that microalgae-based biorefinery is currently facing. Under the domain of PSE, superstructure-based modeling and optimization can be a very useful tool to identify the optimal or promising biorefinery structures in a cost effective and environmentally sustainable manner [55, 56]. Superstructure-based modeling and optimization is a systematic approach that employs mathematical programming to identify the optimal processing network or optimal configuration of a chemical process for a defined objective function [57]. It generally consists of four main steps [58,59,60]. First, we have to define the problem by identifying its scope and the selection of performance metrics for the analysis. In the second step, the data about the raw materials, products and all the potential technological alternatives is collected. After collecting the data, the connections between the processing alternatives are established to process and convert raw materials into desired products through all possible combinations and configurations, in order to develop a superstructure. The next step deals with formulating and developing the models for the processing alternatives included in the superstructure. This step results in the formulation of a mixed integer optimization model, e.g., mixed integer linear programming (MILP), mixed integer nonlinear programming (MINLP), etc. The optimization models usually include an objection function to be optimized, and a set of equality and inequality constraints. The mixed integer optimization models comprise of both binary and continuous variables. However, the binary variables are the main decision variables that determine the optimal processing network. If a technological alternative in the superstructure is selected, the value of the corresponding binary variable will be equal to one; otherwise, it will be equal to zero. In the final step, the optimization problem is solved to determine the optimal solution (i.e., optimal processing pathway). A schematic representation of the superstructure-based optimization framework (extracted from [32, 61]) is given in Fig. 2.

Fig. 2
figure 2

Schematic representation of the superstructure-based optimization framework

Full size image

Depending on the type of optimization problems, a number of algorithms and solution strategies are available to solve them. For example, Branch and Bound algorithm [62] is a well-known method for solving MILP problems. Whereas, for the MINLP problems, Generalized Benders Decomposition [63] and Outer Approximation [64, 65] are the two most commonly used methods. These algorithms to solve mixed integer optimization problems are successfully implemented and executed by a number of commercially available solvers [66].

The application of process synthesis (i.e., superstructure-based modeling) for the design of optimal processing networks is presented in various studies [57, 58, 67]. These studies provide the basis to employ mathematical programming in order to solve the process synthesis problems, thus determining the optimal processing pathways/routes. Yuan et al. [68], highlighted that the process synthesis can play a major role in the development, design and operation of biorefinery processes. Thus, the process synthesis can be recognized as a powerful tool to identify the optimal/promising production routes for microalgae-based biofuels and value-added products. In another study, Yue et al. [56] emphasized the need to use the multiscale modeling and optimization approach for biomass to biofuels and bioenergy production. The presented approach allows the integration across various scales such as from the unit operations to biorefinery processes and to biofuel supply chains, as well as from the operational level to strategic level. However, the focus of the current article is on the modeling and optimization of microalgae-based biorefinery networks. The details on the modeling and optimization of other biomass feedstocks-to-bioenergy/biofuels can be found in Yue et al. [56], Yuan et al. [68], Morales et al. [69], Bertran et al. [70], Zondervan et al. [71], Quaglia et al. [61], Yuan and Chen. [72], Yilmaz and Selim [73] and Daoutidis et al. [74].

3.2 Development of superstructure based on microalgae

To incorporate all the potential alternatives/processing routes for producing biofuels and other valuable products from microalgae, a biorefinery superstructure can be developed. The superstructure contains all the potential candidates for feasible and optimal solution [67]. A general representation of the superstructure (adapted from [11]) is given in Fig. 3.

Fig. 3
figure 3

(adapted from [11])

A general representation of the superstructure

Full size image

To develop such a biorefinery superstructure, first there is a need to identify all the processing stages/steps, potential microalgae species, and targeted end-products. Following the conventional microalgae-to-biofuels pathway, the processing stages are given as follows: (1) microalgae cultivation, (2) dewatering and harvesting (3) pre-treatment of harvested biomass (such as drying, cell disruption, etc.), (4) extraction of lipids, (5) transesterification, (6) post-transesterification purification, (7) pre-treatment of microalgal residue, and (8) processing of microalgal residue to produce value-added products [20]. For each processing step, all the potential technological options/alternatives should then be identified, and the appropriate connections should be made to construct the superstructure. In Fig. 3, each box represents a technological alternative at the corresponding processing stage. Depending on the microalgal species and the targeted end-products, the processing stages/pathways may vary, and therefore, must be added in the superstructure accordingly.

3.3 Establishment of optimal biorefinery configurations/networks

The superstructure-based modeling framework can be developed to identify the optimal design of microalgal biorefinery in order to produce biofuels and value-added products from microalgae. In general, it can be stated as: given a biorefinery superstructure (explained in Sect. 3.2) encompassing raw material (microalgae), potential technological options/alternatives, and products; the objective is to develop a systematic methodology to identify the optimal (or promising) biorefinery configuration for a defined objective function. Various objective functions (such as maximization of yield, minimization of cost, maximization of profit, minimization of environmental impacts, etc.) can be chosen for the optimization formulations, i.e., mixed integer formulations [12, 32, 75,76,77,78]. In this section, the relevant studies are reviewed and reported to elucidate the various opportunities in the superstructure-based modeling and optimization of microalgal biorefinery.

Aforementioned mixed integer formulations (MILP/MINLP) have been implemented for the optimal synthesis and design of microalgal biorefineries. Table 1 summarizes the relevant studies on the application of the superstructure-based approach to identify the optimal biorefinery configurations based on microalgae. Inspired by the superstructure-based formulations, Rizwan et al. [32] proposed a superstructure-based optimization framework for biodiesel production from microalgal biomass. An MINLP problem was formulated to identify the optimal processing route for biodiesel production from microalgal biomass under three optimization scenarios with different objective functions; (1) maximization of product yield, (2) maximization of gross operating margin (GOM), and (3) maximization of GOM and minimization of waste (multi-objective). In their follow-up study [20], based on the concept of biorefinery, they extended the previously developed framework for the production of multi-products from microalgae, in order to improve the overall economics of biofuels production from microalgae in an environmentally sustainable manner. An extended biorefinery superstructure was developed based on Chlorella vulgaris, and then optimized to identify optimal biorefinery pathways for (1) biodiesel production from microalgal lipids, and (2) simultaneous conversion of microalgal residue into value-added energy products. In their studies [20, 32], the focus was on developing a generic optimization framework while making the framework computationally efficient with the capability to evaluate all potential conversion alternatives. Gupta et al. [78] also presented a superstructure-based approach for biodiesel production from microalgae. Based on the superstructure, an MILP model was formulated to optimize the biodiesel production flowsheet from Chlorella with the objective to minimize the net annualized life cycle cost (ALCC). They also optimized the operational parameters as well as the equipment specifications at each processing stage of the microalgae-to-biodiesel process. In their follow-up work [79], they used the same approach to address the optimal design of integrated biorefinery producing biodiesel (as main product) and various co-products.

Table 1 Summary of the relevant studies on the superstructure-based optimization approach to microalgae-based biorefinery
Full size table

Some other studies also focused on the application of superstructure-based optimization to microalgal biorefinery. Martin and Grossmann [80] optimized a superstructure consisting of five technological alternatives in the transesterification stage of biodiesel production process from algae and waste cooking oil, by formulating an MINLP model. Heat and water integration was also simultaneously optimized. Slegers et al. [81] presented a model based combinatorial approach (based on the superstructure) to determine the design configuration for the energy efficient processing of microalgae to produce biodiesel. The objective function was to maximize the net energy ratios. Gebreslassie et al. [82] developed an optimization model based on algal biorefinery superstructure for simultaneously producing hydrocarbon biofuels from microalgae and the sequestration of CO2 from the power plant flue gases. Based on the developed superstructure, a multi-objective optimization model (MINLP) was formulated to simultaneously maximize the net present value (NPV) and minimize the global warming potential (GWP). In another study, Gong and You [76] also optimized a biorefinery superstructure of algae processes, by formulating and solving the optimization model, in order to determine the optimal biorefinery processes. The objective function was the minimization of the unit carbon sequestration and utilization cost. In their follow-up work [12], they developed a generic optimization framework for the global superstructure-based optimization of algae-based processing networks. A multi-objective optimization model (MINLP) with the objective of simultaneously optimizing the unit annualized cost and GWP was formulated. The developed modeling framework was successfully implemented for the optimal design of sustainable algae-based processing network for the production of biofuels as well as CO2 mitigation, considering the entire lifecycle. Nodooshan et al. [83] focused on algae-based biofuel supply chain optimization. A multi-objective MILP model was formulated for the sustainable production of biodiesel.

Yu et al. [77] developed a mathematical model to determine an economical design of biodiesel production process from microalgae by systematically integrating microalgae strain selection with the downstream processing stages. Based on the hypothetical case study, they identified that the strain properties such as lipid contents, productivity and effective diameter have major effects on the system configuration and the production cost. The cheapest processing route involves the production cost of S$ 2.66/kg that is higher than the current price of diesel (S$1.05/kg). However, the price of biodiesel production can be reduced by co-production of other value-added products under the concept of integrated biorefinery [84].

3.4 Technical insights

Despite the preliminary nature of the available dataset used to solve the optimization formulations, the obtained optimization results can guide the researchers to focus on the shortlisted candidates (i.e., processing pathways) with high economic potential and low environmental impact [20, 32]. In this way, a more reliable and consistent dataset can be generated which will ultimately help to determine the most promising pathways under both economic and environmental constraints [81].

The analysis of the results (e.g., by performing the sensitivity analysis) provides insights on how the optimal solutions and their performance vary as a function of key model parameters changes. Thus, the parameters with high (positive) influence can be targeted further for the potential improvements. In the findings of Rizwan et al. [20], three parameters are enlisted as the influential ones including the cost of feed, residue conversion and lipid contents, in viewpoint of overall economics of biofuels production from microalgae. Yu et al. [77] modeled and analyzed the strain properties, and identified that lipid contents, productivity and effective diameter significantly influence the system configuration as well the production cost for the case of biodiesel production from microalgae. Overall, the microalgae-based biorefinery demads significant development/improvement from both engineering and biological realms to become economically viable. Moreover, the superstructure-based modeling frameworks can play a key role for identifying the potential technologies for such developments in the future.

Furthermore, the superstructure-based optimization formulations and the results obtained from them can also help to assess the potential of a number of microalgae species in a systematic way for the quick and efficient screening of microalgal species [77], that can be used further for the biorefinery development. Screening and selection of microalgae strains is also a major hurdle in the development of cost-effective and sustainable biorefinery. To this end, the superstructure-based modeling framework developed by Rizwan et al. [20] (for Chlorella vulgaris) can easily be extended to evaluate the potential of other microalgal species given that the required data can be collected. Subsequently, the strain properties and potential of these species can be compared (e.g., freshwater strains vs. marine water strains) to choose the most adequate ones for the biorefinery development. In addition, this analysis can also provide useful information on the product-wise selection of microalgal species by evaluating their potential for multi-product portfolios [11]. As a result, the corresponding strains for the production of fuel-based products (e.g., biodiesel, biogas, bioethanol, bio-oil, etc.) and for non-fuel products (e.g., value-added chemicals, biomaterials, fertilizers, animal feed, etc.) can be shortlisted for further improvement and advancement.

4 Perspectives on integrated design of microalgal biorefinery configurations

4.1 Integration with carbon mitigation

Carbon capture and sequestration (CCS) is a major technological option to capture/reduce CO2 emissions from large point sources such as power plants as well as many other industrial processes [85,86,87,88]. The CO2 is first captured, compressed to a liquid/supercritical fluid, and then stored in a reservoir such as soil, ocean, etc. [10, 89, 90]. However, the recent technological developments in microalgae-based biofuels/products create new options/alternatives for CO2 sequestration by utilizing CO2 from flue gases for growing microalgae during the cultivation stage, thus creating potential opportunities for restoring the carbon balance in the atmosphere. During the production of biofuels from microalgae, the CO2 fixed by microalgae will essentially be recycled. For instance, when the microalgae-based biofuels will be combusted, CO2 will be released to the atmosphere, and then it will be consumed again by the various plants and microalgae during the process of photosynthesis for biomass production. The resulting biomass will again be converted into biofuels, thus creating a balance between the carbon cycle and energy in a sustainable manner [91].

Microalgae have the potential to fix CO2 using sunlight with 10–50 times higher efficiency compared with the terrestrial plants [4, 10]. Generally, for the cultivation of microalgae, high purity CO2 gas is not required; flue gas containing varying amount of CO2 can be fed directly to the microalgal growth culture [92]. This idea of carbon mitigation via microalgae cultivation has been proposed and investigated in many studies [10, 87, 93,94,95,96,97,98,99]. However, the research on how to integrate the carbon source with microalgae growth facilities and downstream processing is relatively limited [56]. Gutierrez-Arriaga et al. [91] investigated and optimized the integration of CO2 fixation (from steam power plant) with algal biodiesel production only, whereas few other studies [12, 76, 82, 100] focused on the integration of carbon mitigation with algae biorefinery processes from the PSE standpoint. Therefore, there is a need to develop an expanded biorefinery modeling framework, e.g., in the preliminary phase, the capture of CO2 from the various sources can be incorporated in the biorefinery superstructure in order to: (1) model/optimize the utilization of CO2 from flue gases for the growth of microalgae and then integrate it with the microalgal biorefinery downstream processing, and (2) assess the competitiveness of this integrated biorefinery system under technical, economic and environmental constraints, considering carbon mitigation. Furthermore, it will also simplify CO2 separation from flue gases significantly [92]. In a techno-economic study [101], it is highlighted that the large scale cultivations of microalgae for CO2 fixation are very capital intensive and require a large amount of land. However, it may become competitive if (1) key parameters such as photosynthetic efficiencies, exposure to sunlight, cultivation system and other limits to the growth of microalgae are targeted and improved, and (2) a stable market is created for microalgae for their use as a valuable resource for biofuels, energy, chemicals, food and pharmaceutical industries, etc.

4.2 Integrated design considering heat/power integration

The production of biofuels from microalgae is economically not viable. It is mainly due to the high operating cost of the processing technologies involved in the design of microalgal biorefinery [20, 102]. The major contribution in this high operating cost comes from the cost of utilities [20]. Nonetheless, the efficient and integrated design of utility system together with biorefinery could be a good idea to cut down the net energy consumption in order to improve the overall economics. Therefore, an integrated design of microalgal biorefinery needs to be developed. The energy consumption within such integrated biorefinery should be self-sustained. Process synthesis can play a key role in establishing the sustainable integrated biorefinery configurations considering heat and power integration [103, 104]. This process integration can improve the overall performance of microalgal biorefinery by optimizing energy conversion and utility integration/consumption. Therefore, in developing biorefinery modeling frameworks based on the superstructure (discussed earlier in Sects. 3.3, 4.1), heat/power integration must be taken into account. The simultaneous approach will ultimately help to identify the most promising biorefinery configuration for the production of biofuels and other valuable products in a potential self-sustained manner.

4.3 Modeling sustainability issues

Considering microalgae as a renewable alternative to fossil fuels, the sustainability issues associated with the microalgae-based biorefinery must be addressed systematically. To improve the overall economics as well as assess the environmental benefits of microalgal biorefinery, systematic modeling and optimization frameworks need to be developed, with the ability to guide towards the sustainable development of microalgal biorefinery. In this section, we will suggest how the modeling and optimization tools (based on a biorefinery superstructure) can be used to address the sustainability issues. Only economic and environmental aspects of sustainability are discussed in this work.

4.3.1 Economic sustainability

The concept of economic sustainability is to improve the overall economics of microalgal biorefinery by utilizing and consuming all the components of microalgae (lipids, protein, carbohydrates, etc.) for the production of biofuels and/or other valuable products in an efficient and promising manner under the broad concept of biorefinery [3, 19, 49, 105]. Various economic metrics can be used as objective functions in the optimization models to evaluate the economic potential of microalgal biorefinery. Some of the commonly used metrics are GOM, net present value, annualized total cost, etc. GOM deals with the operation, and identifies the profit obtained by operating a certain biorefinery configuration [20, 32]. Whereas, the net present value and annualized total cost represent the time value of money over the project lifetime (usually 15–20 years) with the use of discounted cash flow analysis [56]. Other objective functions such as minimization of production cost associated with microalgae-based products, minimization of utilities consumption, etc., can also be used as objective functions in the optimization models to assess the competitiveness of microalgal biorefinery. The superstructure-based modeling and optimization approach (discussed in Sect. 3.3) is a very useful tool that can use these economic metrics in order to evaluate the cost competitiveness of microalgal biorefinery. The relevant studies are reviewed in Sect. 3.3.

4.3.2 Environmental impacts

Production of microalgae-based biofuels and other energy products may offer potential environmental advantages over the fossil fuels. Therefore, the use of these biofuels in the transport, energy, and other sectors could possibly alleviate the environmental burdens [16, 17]. In order to evaluate and assess the environmental impact/benefits of microalgae-based biorefinery, a comprehensive life cycle assessment (LCA) can be performed.

LCA is a system-level analysis and a widely used tool to systematically investigate and evaluate the environmental impacts (both direct and indirect impacts) of a process and/or a product/service through all stages of the life cycle [17, 106,107,108]. As shown in Fig. 4, the LCA framework generally consists of four main steps: (1) defining the goals and scope of the analysis, (2) identifying life cycle inventory, (3) assessment of life cycle impacts, and (4) interpretation of the results [108]. The first step deals with defining the goals and scope of the LCA study as well as defining the functional units while identifying the system boundaries for which the LCA analysis has to be performed. The second step comprises of collecting the inventory data such as all the inputs (e.g., flow of raw material, energy resources, etc.) and the outputs (e.g., wastes, and emissions to the environment) to and from the system under investigation. The third step includes determining the environmental impacts of inventories that are identified in the second step. A number of metrics can be used for the impact assessment purpose such as net energy ratio, GWP, energy return on investment, water footprints, ozone depletion, etc. In the final step of LCA, the obtained results are analyzed to evaluate the system under investigation, and then the recommendations and conclusions are drawn based on the environmental performances.

Fig. 4
figure 4

Framework for life cycle assessment by employing superstructure-based approach

Full size image

The environmental impacts of microalgae-based biofuels obtained from a particular processing route have been evaluated in many studies. The details on these LCA analyses can be found in these review articles: Collet et al. [17], Zaimes and Khanna [108], Singh and Olsen [109], Quinn and Davis [110] and Thomassen et al. [111]. However, for the LCA of a number of technological alternatives for the development of optimal microalgal biorefinery structure to produce biofuels and other valuable products, the work is relatively limited, and we suggest to develop a systematic and integrated modeling framework. Gong and You [112] integrated the techno-economic analysis based on the biorefinery superstructure with life cycle analysis for the sustainable design of manufacturing processes to produce biodiesel, hydrogen, and various valuable chemicals from microalgae. A significant reduction (5–63%) in the greenhouse gas emissions was observed under the most environmentally sustainable process for microalgae-based products compared with their petroleum-based counterparts.

The superstructure-based optimization models can be coupled with LCA methodologies to identify the sustainable biorefinery pathways for producing microalgae-based biofuels and other value-added products that may have less environmental impacts [11, 12, 76, 112]. In such integrated framework (as shown in Fig. 4), the optimization models (developed based on the biorefinery superstructure) can be extended first for the life cycle inventory analysis, and then these inventories can be translated into various sustainability metrics for the impacts assessment. The integrated framework will also play a key role in the development and design of economically and environmentally sustainable microalgal biorefinery by setting the future targets for the possible improvements and developments in the processing technologies [11]. Furthermore, multi-objective optimization can be employed to simultaneously optimize the conflicting objectives such as maximizing profitability while minimizing the environmental impacts [56].

4.4 Modeling uncertainties in the dataset

The processing technologies for the design and development of microalgae-based biorefinery are still in the development phase. Therefore, a wide range of uncertainties may arise given the inconsistencies and shortage of process information [53, 113]. Specifically, these uncertainties are associated with the various process parameters at each processing stage such as lipid contents, lipids productivity, composition of microalgal biomass, CO2 fixation rate, harvesting efficiency, lipids yield, conversion of lipids into biodiesel, conversion of microalgal residue into value-added products, consumption of utilities, etc. In addition to these uncertainties, there are also uncertainties associated with the economic parameters such as the uncertainty in the feed cost, utilities cost, selling price of the biofuels and other products from microalgae, etc. [53, 54]. Consequently, the development and design of microalgal biorefinery is associated with a number of techno-economic uncertainties. To ensure robust decision making, these uncertainties must be taken into account while developing the biorefinery models, and these models can be very useful to determine the robust biorefinery structures considering techno-economic uncertainties [53, 54, 113].

Stochastic programming/optimization, under the domain of PSE, is a widely used approach for the synthesis and optimal design of chemical processes under uncertainty [114]. Many studies have elaborated this approach with different scope and applications [115,116,117,118,119,120,121,122], providing the basis to use the stochastic programming for determining the robust biorefinery structures under modeled uncertainties.

The impact of the uncertainties present in the dataset on the optimal pathways of microalgal biorefinery has not been thoroughly investigated yet. Brownbridge et al. [54] presented a global sensitivity analysis for the assessment of economic viability of biodiesel production process from microalgae under techno-economic uncertainties. The biodiesel production cost and the return on investment (ROI) were found to be highly sensitive to algae oil contents, and less sensitive to other parameters such as annual productivity, production capacity, and carbon price increase rate. However, their analysis was limited to only one specific processing route for the case of biodiesel production. The uncertainties in the key and/or influential parameters may significantly affect the biorefinery design as well as the overall performance, therefore, they need to be handled and addressed systematically to ensure robust decision-making [53]. In such an attempt, Rizwan et al. [53] developed a systematic framework to identify the optimal biorefinery configurations under techno-economic uncertainties. The major uncertainties were incorporated into the microalgal biorefinery superstructure constructed in an earlier study [20], and a stochastic MINLP problem with the objective of maximizing the expected value of GOM was formulated. The uncertainties in the model parameters were sampled to generate future scenarios. The resulting stochastic optimization problem was solved under sampled scenarios to identify the optimal biorefinery configuration under uncertainty. In another study, Gong and You [123] formulated a two-stage adaptive mixed integer fractional programming model to handle the uncertainties in the parameters. The developed model was implemented to identify the optimal design of processing networks under uncertainty to produce biodiesel, renewable diesel, hydrogen and other useful bio-products from microalgae with the objective function to maximize ROI. Sy et al. [124] developed a multi-objective target oriented robust optimization framework to address the uncertainties in the model parameters. The presented approach was implemented on a case study to determine the robust optimal configuration of integrated algal biorefinery under uncertainty, with economic and environmental considerations.

An extended biorefinery framework can be developed where based on the biorefinery superstructure, stochastic programming approach (e.g., formulation of two-stage stochastic model [114]) can be applied to evaluate the impact of uncertainties on the optimal design as well as the overall performance of microalgal biorefinery that might turn to be a very promising approach. A simple schematic representation of uncertainty analysis (extracted from [53, 120]) is highlighted in Fig. 5, where the deterministic optimization model (formulated based on the superstructure) can be extended further for the uncertainty analysis. It results in the development of a stochastic optimization problem that can be solved to identify the optimal solution under uncertainty. Furthermore, robust optimization [125, 126], e.g., worst-case analysis, can also be employed to identify the “risk-averse” design to handle potentially bad outcomes [53]. Different risk aversion strategies such as conditional value at risk [127] and downside risk [128] can be formulated to manage and control the risk at the optimal design and operation of microalgal biorefinery.

Fig. 5
figure 5

Uncertainty analysis framework by employing stochastic programming approach

Full size image

5 Conclusions

The development of a systematic modeling framework can play a key role in addressing the current challenges and problems of microalgae-based biorefinery for the production of biofuels and value-added products from microalgae in a cost-effective and environmentally sustainable manner. Some of the major challenges are highlighted and discussed including (1) a number of technological alternatives for the development and design of microalgal biorefinery, (2) economic feasibility, (3) assessment of environmental sustainability, and (4) robust biorefinery design under uncertain dataset. As reviewed in this paper, the superstructure-based modeling approach is a useful tool to address these challenges, and is receiving significant attention. Based on the various research opportunities and potential perspectives discussed in this paper, we suggest the development of an integrated framework where the superstructure-based optimization models can be integrated with LCA methodologies, uncertainty analysis framework and multi-objective optimization techniques. This integrated approach will provide the basis to develop a robust framework/tool to determine the integrated and sustainable biorefinery structures by simultaneously optimizing on technical, economic and environmental criteria as well as taking the uncertainties into consideration. This represents a potential and promising track for further research.