3.1.1 Antioxidant Activity

Interest in natural antioxidants for both health and improved food stabilization has intensified dramatically since the last decade of the twentieth century. Health appli­cations have been stimulated by observations that free radicals and oxidation are involved in many physiological functions and cause pathological conditions. Natural antioxidants offer food, pharmaceutical, nutraceutical, and cosmetic manufacturers a “green” label, minimal regulatory interference with use, and the possibility of multiple actions that improve and extend food and pharmaceutical stabilization [168]. Determining antioxidant capacity has become a very active research topic, and a plethora of antioxidant assay methods are currently in use. Despite of it, there are no standard methods due to the sheer volume of claims and the frequent contra­dictory results of “antioxidant activities” of several products.

Reactive oxygen species which include superoxide anion (O2- . a free radical), the hydroxyl radical COH) and hydrogen peroxide (H2O2) are produced by ultravio­let light, ionizing radiation, chemical reaction, and metabolic processes. These reac­tive species may contribute to cytotoxicity and metabolic changes, such as chromosome aberrations, protein oxidation, muscle injury, and morphologic and central nervous system changes in animals and humans [34]. Effective antioxidants must be able to react with all these radicals in addition to lipids, so, consideration of multiple radical reactivity, in antioxidant testing, is critical.

In general terms, three big groups can be distinguished: chain reaction methods, direct competition methods, and indirect methods [154].

1. Among the chain reaction methods two approaches have been used: measuring the lipid peroxidation reactions or the kinetics of substrate oxidation.

There are two modes of lipid peroxidation that may be used for testing. The first one is autoxidation, in which the process is progressing spontaneously, with self-acceleration due to accumulation of lipid hydroperoxide (LOOH). The kinet­ics of autoxidation is highly sensitive to admixtures of transition metals and to the initial concentration of LOOH. As a result, the repeatability of experiments based on the autoxidation is still a problem. The second, much more promising approach, is based on the use of the kinetic model of the controlled chain reaction. This mode offers to obtain reliable, easily interpretable, and repeatable data. This approach has been applied, among others, to test natural water-soluble antioxi­dants, microheterogeneous systems, micelles, liposomes, lipoproteins (basically low-density lipoprotein [LDL]), biological membranes, and blood plasma [154].

When choosing a substrate of oxidation, preference should be given to individual compounds. Among individual lipids, methyl linoleate, and linoleic acid seem to be the most convenient. These compounds are relatively cheap and their oxidation is quite representative of the most essential features of biologi­cally relevant lipid peroxidation. The main disadvantage, when using them in biological materials, is that the extract must be free of the elected compound, as it is impossible to provide the identity of substrate. Besides, biologically originated substrates usually contain endogenous chain-breaking antioxidants (vitamin E, etc.), which can intervene in the testing procedure.

2. The direct competition methods are kinetic models, where natural antioxidants compete for the peroxyl radical with a reference-free radical scavenger:

• b-Carotene bleaching: competitive bleaching b-carotene during the autoxidation of linoleic acid in aqueous emulsion monitored as decay of absorbance in the visible region. The addition of an antioxidant results in retarding b-carotene decay [114].

• Free-radical induced decay of fluorescence of R-phycoerythrin: The intensity of fluorescence of phycoerythrin decreases with time under the flux of the peroxyl radical formed at the thermolysis of APPH (2,2′ — azobis-2-methyl — propanimidamide) in aqueous buffer. In the presence of a tested sample containing chain-breaking antioxidants, the decay of PE fluorescence is retarded [147].

• Crocin bleaching test: Crocin (strongly absorbent in the visible range) under­goes bleaching under attack of the peroxyl radical. The addition of a sample containing chain-breaking antioxidants results in the decrease in the rate of crocin decay [12].

• Potassium iodide test: KI reacts with the AAPH-derived peroxyl radical with the formation of molecular iodine. The latter is determined using an auto­matic potentiometric titrator with sodium thiosulfate. In the presence of antioxidant-containing samples, the rate of iodine release decreases [154].

3. When the indirect approach method is applied, the ability of an antioxidant to scavenge some free radicals is tested, which is not associated to the real oxidative degradation, or effects of transient metals. For instance, some stable colored free radicals are popular due to their intensive absorbance in the visible region [154]. There are two ways for presenting results, as equivalents of a known antioxidant compound (i. e., Trolox Equivalent Antioxidant Capacity, TEAC) or as the con­centration needed to reduce concentration of free radicals by 50% (EC50).

• DPPH test: It is based on the capability of stable-free radical 2,2-diphenyl-1- picrylhydrazyl (DPPH) to react with H-donors. As DPPH’ shows a very intensive absorption in the visible region (514-518 nm), it can be easily deter­mined by the UV-Vis spectroscopy [13]. This method has been applied online with TLC [65] and HPLC [5] to determine antioxidant activity in different algae extracts.

• ABTS test: The decay of the radical cation ABTS+^ (2,2′-azinobis(3-ethylben — zothiaziline-6-sulfonate) radical cation) produced by the oxidation of ABTS+^ caused by the addition of an antioxidant-containing sample is measured. ABTS+^ has a strong absorption in the range of 600-750 nm and can be easily determined spectrophotometrically. In the absence of antioxidants, ABTS+^ is rather stable, but it reacts energetically with an H-atom donor, such as pheno — lics, being converted into a noncolored form of ABTS [115].

• Ferric reducing antioxidant power (FRAP): The FRAP assay is based on the ability of antioxidants to reduce Fe3+ to Fe2+ [154]; if the reaction is coupled to the presence of some colored Fe2+ chelating compound like 2,4,6-trypyridyl — s-triazine, it can be measured spectrophotometrically.

• Cyclic voltammetry: The general principle of this method is as follows: the electrochemical oxidation of a certain compound on an inert carbon glassy electrode is accompanied by the appearance of the current at a certain poten­tial; while the potential at which a cyclic voltammetry peak appears is deter­mined by the redox properties of the tested compound, the value of the current is proportional to the quantity of this compound, in the presence of an antioxi­dant compound the signal will be lower [155].

1.1.3 Biogas Composition

Biogas is formed during AD and has two main constituents: methane (about 55-70% by volume) and carbon dioxide (30-40%). Depending on the source of the biogas, other minor components include nitrogen (<2%), hydrogen, oxygen (<1%), hydro­gen sulfide (0-50 ppm), and other sulfide compounds, volatile organic compounds (VOC) 10-270 mg/m3, and siloxanes with concentration ranging from 80 to

2,500 mg/m3 [98, 99]. The VOC comprise aromatic and halogenated compounds. Large amounts of noxious VOC can be produced during digestion of household wastes [100, 101].

Carbon dioxide is not a harmful inert gas but the presence of carbon dioxide in biogas reduces its calorific value. The removal of carbon dioxide is an expensive process and power generation equipment commonly operates with carbon dioxide concentrations up to 40-50%.

The most abundant sulfur compound in biogas is hydrogen sulfide but other reduced sulfur chemicals (e. g., sulfides, thiols) are present as well. The main source of sulfur in biogas is degradation of sulfur containing amino-acids—cysteine and methionine. Hydrogen sulfide at concentrations higher than 300-500 ppm can form unhealthy and hazardous sulfur dioxide (SO2) and sulfuric acid (H2 SO4) which corrodes pipeline metal parts, storage tanks, compressors, and engines [102]. Frequently, it is necessary to install sulfur removing facilities before the biogas application. Another corrosive contaminant is ammonia (NH.) . The burning of biogas with high ammonia concentration increases the emission of nitrogen oxides to the atmosphere. Both hydrogen sulfide and ammonia are contaminants that pose a health risk.

Other compounds of concern in biogas are siloxanes—the organic polymers of silicon coming from a wide range of industrial, personal care, pharmaceutical, and other products. These organic compounds can be oxidized to silicon dioxide and accumulate on valves, gas turbines, and engines causing erosion and decreasing the operating efficiency [103].

Co-digestion of different substrates has been recognized recently as an attractive approach that has many economical, logistic, and sanitary benefits [404]. The main advantage of co-digestion is the improvement of the nutrient balance that leads to greater reduction of VS and methane yield [405] . The C:N:P ratio is an important factor for AD as discussed earlier in the chapter. Co-digestion of low nitrogen bio­mass (municipal solid waste, paper, sisal pulp, straw, grasses, wood wastes) with higher nitrogen wastes (sewage sludge, chicken or livestock manure, slaughter­house, meat or fish processing wastes) can increase the loading rate and methane yield up to 60-100% [399, 406-409].

The C to N ratio varies for macro — and microalgae due to significant differences in biochemical composition. Carbohydrates are the main components of macroalgal

biomass with concentrations between 50 and 70% of dry weight. Microalgae usu­ally have a carbohydrate content in the range 10-20% while the protein content can reach 30-50% in some species. These differences in biochemical composition influence the C:N ratios and the strategies needed for co-digestion processes with macro — or microalgae.

The following review focuses on those studies that are most relevant to the assess­ment of gas hydrate resource potential.

1.1.1 Synopsis of Global Research Activities

Japan took a leading role in the effort to explore the potential of geologic hydrate deposits as an energy source by establishing a research program in 1995, which led to the drilling and installation of the first well in marine gas hydrate deposits in the Nankai Trough offshore Japan at a water depth of 945 m [114, 190, 199, 203]. This was succeeded by a larger multi-well exploration program [114], and is probably the most advanced program in the world in terms of proximity to commercial produc­tion. As part of this program, 36 wells were recently drilled in gas hydrate-bearing sand reservoirs at the same location [189]. Fujii et al. [47,48] and Saeki et al. [172] described the variety of gas hydrate occurrence found in the Nankai region, and Kurihara et al. [94, 96] discussed the technical challenges and the relative economic favorability of GH in different geologic settings. Production is expected to begin around 2016. Japan has also collaborated with Canada and other nations to conduct scientific studies and production tests from GH in the Canadian Arctic. Canada has
established a large gas hydrate research and development program that resulted in the Mallik production field test [35], the most significant to-date development in the quest for gas production from hydrates (see later discussion).

In the United States, studies on GH as a resource began in 1980s, and Collett [19] conducted the first systematic assessment. He estimated the 50% probability (mean) estimate of hydrate resources within the United States at 9 x 1015 m3 of CH4 (with the 95% probability estimate at 3 x 1015 m3 and the 5% probability estimate at

1.9 x 1016 m3), i. e., the mean value indicates 300 times more hydrated gas than the gas in the total remaining recoverable conventional resources. The Methane Hydrate Research and Development Act (MHR&D Act) of Congress in 2000 authorized funding to uncover the physical nature, economic potential, and environmental role of naturally occurring GHs. Over the first 10 years of the MHR&D Act, hydrate sci­ence advanced significantly, both in terms of knowledge of natural hydrate occur­rences, hydrate physical/chemical properties, and in the tools available to researchers. Researchers gained a greater understanding of the complexity of hydrate accumula­tions through laboratory work [40, 56, 86, 204,215], numerical simulation analyses [88, 129, 131-134, 136, 137], and national and international collaborative field experiments [35], and began the development of the precursors to tomorrow’s hydrate exploration and evaluation technologies. By 2005, it was clear that, given certain reservoir conditions, production of methane from hydrate was technically feasible and potentially commercially viable through specially tailored application of existing technologies [8]. Current research activities in the United States include laboratory experiments and simulation studies [8], in addition to field studies that focus on onshore Alaska and the offshore Gulf of Mexico (GOM)-i. e., sites of proven exploration targets in the United States [19-21]. Major federal-industry part­nerships have been formed in both the GOM and on the North Slope of Alaska [20]. It is likely that the first US domestic production from hydrates may occur in Alaska because of easier access, although the possibility of first production from GHs in the GOM cannot be discounted because of pipeline capacity and easier access to markets.

The government of India is funding a large national GH program to meet its growing gas requirements. Earlier seismic data from the Indian continental margin and GH occurrences that had been accidentally discovered during drilling for con­ventional oil and gas resources [20] provided the impetus for a hydrate-focused scientific expedition in the summer of 2006. This expedition confirmed large GH deposits at four offshore locations, from which many hydrate-bearing cores were obtained. Most notable was the 130-m thick fractured shale occurrence in the Krishna-Godowari basin that contained GH saturations SH previously unseen in shale-dominated reservoirs [ 28, 29 ] .

China has pursued gas hydrates R&D for more than a decade [44] , and con­ducted its initial drilling and coring program in the South China Sea in early 2007. That expedition found GH occurrences with SH up to 40% in clay-dominated sedi­ments at several sites [224]. As in the 2006 India expedition, these results were unexpected, and indicated that, given adequate sources of gas, hydrates are remark­ably effective at filling any available pore space.

Korea has established a significant research program that aims to assess the potential hydrate resources in the Korean East Sea. Preliminary surveys conducted by the Korea Institute of Geoscience and Mineral Resources (KIGAM) between 2000 and 2004 suggest a significant potential for gas hydrate occurrence in the Ulleung Basin [146], and numerical simulation studies have raised intriguing possibilities about the production potential of these deposits [136] . In late 2007, a drilling and coring program in Korea’s East Sea reported several 100-m thick occurrences [105].

Other countries (e. g., Norway, Russia, Mexico, Taiwan, Vietnam, Malaysia) have either embarked on, or are investigating the viability of, government-sponsored research programs to investigate the potential of gas production from national hydrate deposits. This list is only expected to grow. In Europe, research programs like Hydratech and Hydramed have focused primarily on scientific and environ­mental issues.

Recently, a growing number of deep sea drilling expeditions have been dedicated to locating marine GHs and obtaining a greater understanding of the geologic con­trols on their occurrence. The earliest projects were those of the Ocean Drilling Program (ODP) and the Integrated Ocean Drilling Program (IODP), including ODP Legs 164 [149] and 204 [194] and IODP Expedition 311 [162], as well as the 1998 and 2005 drilling programs conducted in the Nankai Trough by the MH21 consor­tium [47, 181]. More recently, the Gumusut-Kakap project offshore Malaysia [57], the Department of Energy (DOE)-sponsored drilling Legs I and II under the Joint- Industry Project in the GOM [9, 13, 166], and the India NGHP Expedition 01 [23, 28,29], as well as those in the offshore of China [218] and South Korea [147], have continued to expand the GH knowledge base.

Given the difficulty and the large costs of conducting field studies on hydrates, significant effort is invested in international collaborative projects. The most well known (and probably the most important, in terms of knowledge generated) was the 2002 Mallik project, conducted at that site in Canada’s Mackenzie Delta (Northwest Territories) by an international consortium that included seven organizations from five countries, as well as the International Continental Scientific Drilling Program. Current international collaborative projects include the Mallik 2007-2008 project [33, 37] (Japan and Canada), as well as other bilateral collaborations, e. g., US-India [27] and US-China [218].

Laboratory investigations are performed on natural samples and laboratory-synthe­sized samples to understand the thermodynamic properties, gas composition, mechanical, electrical, thermal, and hydrologic properties of HBS. Understanding these properties is important in predicting (a) the amount of gas that can be pro­duced from a reservoir, (b) providing “ground-truth” for geophysical and well-log measurements [23, 28, 29], (c) understanding the mechanical strength of the reser­voir medium, and (d) understanding how gas will be produced.

As archetypes of O2 tolerant hydrogenases, the R. eutropha enzymes have been extensively studied using various spectroscopic techniques and site directed muta­genesis (reviewed in [ 17, 36 ] ) . Although these studies have proven that direct manipulation of the hydrogenases is possible, it has also shown the difficulty of improving activity of the enzymes. Therefore, direct manipulation of the hydroge — nases to optimize the generation of energy and reducing equivalents for IBT produc­tion would be a significant challenge, and likely not a viable approach. In contrast, hydrogenase gene expression will be enhanced to assure that sufficient reducing equivalents will be available for optimal IBT yield.

To optimize IBT yield, it is important to balance the reducing equivalents gener­ated by the SH with the ATP synthesized by the MBH. Since carbon fixation will be maximized and the IBT synthesis pathway consumes two NAD(P)H molecules for every two molecules of pyruvate reduced to IBT, the hydrogenase activity balance will be shifted towards the SH.

3.2 Enhancement of Carbonic Anhydrase

Because of the inefficiency of the enzyme, the CO2 fixation by RuBisCO is likely the limiting step for an efficient production of IBT by R. eutropha. Enhancing expression and/or activity of CAs could help to increase the CO2 concentration in the cytosol, thereby limiting the competing oxygenation reaction by RuBisCO and increasing the CO2 flux through the CBB cycle and subsequently to IBT.

4 Outlook

In addition to the aforementioned efforts to ensure deep penetration of light into the culture, it must be constituted that the presence of dark volume elements in a pho­tobioreactor does not necessarily decrease volumetric productivity of the system. On the contrary, the circulation of algal cells between sufficiently illuminated and
dark volume elements can increase the overall volumetric productivity in reactors at saturating light intensities. This phenomenon is referred to as intermittent (or “flashing”) light effect [22]. As long as microalgae are located in illuminated areas, photons are captured and the photosynthetic apparatus generates ATP and NADPH (light reactions). Light reactions stop when cells are located in a dark volume ele­ment. Nevertheless, dark reactions, that are driven by ribulose bisphosphate car­boxylase (Rubisco) and that are kinetically limiting for the overall CO2 fixation process, can proceed. The overall productivity can eventually be increased in a reac­tor if light intensities exceed saturation levels and frequencies of circulation between dark and illuminated volume elements are beneficial for the algae cells.

Experiments with Dunaliella have shown that photosynthetic efficiency can be increased in comparison to continuous illumination for light/dark cycles of 5.32 Hz but efficiency is lower under slow cycles of 0.17 Hz [21]. Reasons for reduced efficiency when slow light/dark cycles are prevalent are not clear, yet. One possible reason might be interference with intracellular control loops on the epigenetic level [30]. Cycle frequencies of >1 Hz are recommended for P. tricornutum [24]. Optimal cycle frequencies are certainly strain-dependent and also strongly influenced by photon flux densities and spatial distribution of light in the liquid volume.

Usually, favorable light/dark cycle frequencies can be attained when prevalent flow regimes in reactors are turbulent so that sufficient radial mixing along the light path is guaranteed. However, this demand imposes two restrictions. Firstly, some algae species are sensitive to shear stress. Microeddies with dimensions comparable to cell size should be avoided. This restriction can be crucial for scale-up when specific light/dark cycles should be obtained but the high energy input required gener­ates cell damaging flow regimes [25]. Secondly, high levels of auxiliary energy input are usually required to attain light/dark frequencies in a desirable order of magnitude. These energy inputs significantly exceed values necessary for sufficient mass transfer. Especially for energetic utilization of biomass the energy balance requires that energy input is minimal. Energy content of algae can range from 20 to 30 MJ/kg for oil rich algae [30]. Unfortunately, little information is available about the correlation between energy input and frequencies of light/dark cycles in different reactor geometries.

Computational flow dynamics (CFD) simulations can be implemented to estimate radial mixing velocities and therewith residence times in dark or illuminated volume elements. Biomass concentration, pigment composition, and light intensities at the surface of a reactor strongly influence the spatial light distribution and the associated ratio of dark and illuminated volume elements. Exemplary simulations for tubular reactors show that static helical mixers could increase light/dark cycle frequencies in a tubular reactor with a factor higher than 20 compared to plain tubes [27].

Razif Harun, Mark Doyle, Rajprathab Gopiraj, Michael Davidson, Gareth M. Forde, and Michael K. Danquah

Abstract With the current global drive towards a low-emission economy, countries need to take a stance. For example, Australia, which is one of the world’s largest polluters, has made a commitment that before 2020 its overall emissions would be reduced by 5-15% below the levels registered in the year 2000. To realise these targets, processes which capture carbon dioxide will prove critically important. One of such emerging processes is carbon dioxide capture for microalgae cultivation and subsequent downstream biomass processing for biodiesel production. This chapter will entail engineering scale-up, economic analysis and carbon audit to ascertain the viability of an industrial scale microalgal biodiesel production plant. This will involve the development of an industrial scale model to determine the feasibility of a real large-scale plant. Data from each process step (cultivation, dewatering, lipid extraction and biodiesel synthesis) will be presented individually and integrated into the overall process framework.

1 Introduction

Lack of sustainable energy resources currently threatens the survival of an increas­ingly globalised world economy. Due to the heavy dependence on limited fossil fuel resources, renewable alternatives which are able to compete with conventional energy options must soon be developed. Steady increases in the price of crude oil, for instance, are being observed, due to rising demands and the escalating scarcity of reserves. In addition, the increasing accumulation of CO2 in the atmosphere and

R. Harun • M. Doyle • R. Gopiraj • M. Davidson • G. M. Forde • M. K. Danquah (H) Bioengineering Laboratory, Department of Chemical Engineering,

Monash University, 3800 Clayton, VIC, Australia e-mail: michael. danquah@eng. monash. edu. au

J. W. Lee (ed.), Advanced Biofuels and Bioproducts, DOI 10.1007/978-1-4614-3348-4_30, 709

© Springer Science+Business Media New York 2013

its impact on climate change have provided a significant driver for change. Amid growing concerns, global agreements to limit greenhouse gas (GHG) emissions, such as the Kyoto Protocol, have been formed. Directly resulting from such agreements, many developed countries have adopted the “cap-and-trade” carbon trading schemes in efforts to curb emissions. Thus projects which capture CO2 to prevent release into the atmosphere will play a significant role in combating climate change.

Biofuels such as biodiesel and bioethanol are possible alternative fuels. Current biofuel production requires increasing amounts of arable land for biomass cultiva­tion, which compete with terrestrial fuel crops, thus heightening concerns over food affordability. Microalgae offer a unique alternative as it does not compete for culti­vation logistics with agricultural food crops. Microalgae could harbour a substantial amount of lipids for biodiesel production and carbohydrates for bioethanol production.

This chapter entails the engineering scale-up, economic analysis and carbon audit to ascertain the viability of an industrial scale microalgal biodiesel production plant. This will involve the creation of an industrial scale model that can be used to determine the feasibility of a large-scale plant. Four major stages are considered here: microalgal cultivation, dewatering, lipid extraction and transesterification (biodiesel production).

Biodiesel is generally composed of several fatty acid esters including C12:0, C14:0, C16:0, C16:1, C18:0, C18:1, C18:2, C18:3, C20 and C22:1. After oil extraction and purification, the exploitable fatty acid chains from the lipid undergo transesterification to produce fatty acid methyl esters (FAME). The transesterification reaction involves the use of an acidic or alkali catalyst mixed with methanol [36]. The methanol group attaches itself to the fatty acid chains via the bond cleaving activity of the catalyst to produce FAME and glycerol. The methyl ester (biodiesel) produced from this reaction after glycerol separation is crude, hence must be washed, dried and decon­taminated so that all water and particulates within the biodiesel are removed. The purified biodiesel must comply with the regulatory standards set by the Fuel Quality Standards Act 2000.

1.1 Framework for Defining Sustainability Indicators

The commonly understood three dimensions of sustainability as shown in Fig. 1 are largely interrelated, i. e. what is good for the environmental is also good for society, or what is good for the economy can be usually good for society, and so on.

As shown, sustainability exists at the intersection of the three domains represent­ing the economy, environment, and society. We can state that a system becomes more sustainable when all three domains, as represented by the intersection of the three domains, show improvement as a result of a human intervention. This Venn diagram also facilitates identification of the dimensions of the metrics to be used to evaluate relative sustainability of a selected system. For instance, any indicator or

Fig. 1 Generally accepted model for sustainability

metric that represents all three dimensions, such as energy use, will be a 3D or 3D metric. Similarly other metrics could be 2D or 1D, depending on how many domains are represented.

When the task is to compare the relative sustainability of a system against alter­natives, we need to consider the following actions in sequence: define the system, identify the metrics to be used and their dimensionality, prioritize them in terms of their significance, obtain values of those metrics for the competing alternatives, and compare them to arrive at a decision. A small set of indicators is ideally preferable because it simplifies analysis, and sometimes allows a decision by visual inspection of the values of the metrics. Though not always possible, it is advisable that the metrics are deemed to be necessary and sufficient, independent of each other, and are quanti fi able.

In this work the framework previously used by Martins et al. [10] is applied for the sustainability evaluation of microalgae biofuels, taking into account the biofuel supply chain stages that include: microalgae cultivation and harvest, biomass pro­cessing, oil extraction and pre-treatment, biodiesel production and blending, distri­bution, and final use. In practice, one needs to first clearly define the system boundaries with identified supply chain. This process allows identification of the indicators to be used and the kinds of data to be collected for calculating their values. Then, all the relevant inputs and outputs (energy, water, materials, product, by-products, wastewa­ter, gas emissions, solid wastes, etc.) should be identified and quantified, in order to be able to calculate the values of the selected metrics. Finally, when the values of the metrics for all possible alternatives are available, a decision on relative sustainability can be made either by inspection or by use of computational tools.

For deriving sustainability indicators and to evaluate microalgae biodiesel through its supply chain, the following sequential procedure (Fig. 2) can be applied:

1. System boundary definition, including inputs and outputs (energy and mass fluxes) through the supply chain.

2. Identi fi cation of the most relevant environmental, economic and societal impacts that ought to be considered and explicitly included in the indicators to be selected, as well as the data required for their calculation.

3. Selection and prioritization of an adequate set of sustainability indicators based on technical input and data availability. All the relevant inputs and outputs should be identified and quantified, in order to be able to calculate the indicators values.

4. Calculation of the chosen indicator values for sustainability evaluation. The 3D, 2D and 1D metrics are calculated based on the inventory analysis of the process.

5. Interpretation and decision making. Decisions for improving the process are made based on the results of the indicators calculation and on the consideration of other issues, for example cost analysis.