Finally, biomass can be converted into biogas, either directly from the microalgal biomass or indirectly by the anaerobic digestion of the oilcakes. Various kinds of energies can thus be produced, and the perimeters of the study could be different. Methane potential strongly varies depending on the species composition and degradability (Sialve et al., 2009). In the con­sidered studies, it ranges from 0.262 (Collet et al., 2011) to 0.800 m3CH4 kgDM-1 (Brentner et al., 2011). It should be noted that this last value is higher than the theoretical maximum value for Scenedesmus (Sialve et al., 2009). Energy consumption of anaerobic digesters is usu­ally ignored. It is regrettable, since long hydraulic retention times required to digest low bio­degradable materials (from 10 to 40 days) represent a significant energy effort for mixing and heating. Heat consumption is estimated at 2.45 MJ kgDM in Collet et al. (2011), and electricity consumption is estimated at 0..47 MJ kgDM in Brentner et al. (2011) and 0.39 MJ kgDM in Collet et al. (2011).

The solid-bowl decanter centrifuge is characterized by a horizontal conical bowl containing a screw conveyor that rotates in the same direction. Feed slurry enters at the center and is spun against the bowl wall. Settled solids are moved by the screw conveyor to one end of the bowl and out of the liquid for drainage before discharge, while separated water forms a concentric inner layer and overflows an adjustable dam plate. The helical screw conveyor pushing the deposited slurry operates at a higher rotational speed than the bowl.

A solid-bowl screw centrifuge was used to separate various types of algae (Mohn, 1980). Fed with an algal suspension of 2% solids, the separated algal slurry was able to attain 22% solids concentrations. Although the reliability of this centrifuge seems to be excellent, the energy con­sumption is too high. An attempt to concentrate an algae feed of 5.5% solids derived from a flotation process by a co-current solid-bowl decanter centrifuge was not successful (Shelef et al., 1984). Subsequently, algae slurry concentration was improved to 21% TSS by reducing the scroll speed to 5 rpm (Shelef et al., 1984). The solid-bowl decanter centrifuge was recommended for use concurrently with polyelectrolyte coagulant to increase the efficiency.

The most common procedure for cultivation of microalgae is autotrophic mode. Microalgae in photoautotrophic nutrition mode use sunlight as the energy source and inor­ganic carbon (CO2) as the carbon source to form biochemical energy through photosynthesis (Huang et al., 2010). This is one of the most prevailing environmental conditions for the usual growth of microalgae (Chen et al., 2011). In photoautotrophic nutritional mode, photosynthet­ically fixed CO2 in the form of glucose serves as a sole energy source for all metabolic activities (Figure 8.6). The simpler form of photosynthate, such as simpler carbohydrates, serves as sole energy source for carrying out the metabolic activities of the algal cells (Chang et al., 2011). These carbohydrates, under nutrient-limiting and stress conditions, will favor the lipid biosynthesis, which also helps to cope — up with the stress (Gouveia and Oliveira, 2009). Lipid productivity greatly depends on the photosynthetic activity in terms of atmospheric CO2 fix­ation and microalgae species. Large variations in lipid productivity, ranging from 5% to 68%, were reported under varying operating conditions and species diversity (Murata and Siegenthaler, 2004; Ohlroggeav and Browseb, 1995; Chen et al., 2011; Mata et al., 2010). A major advantage of the autotrophic nutritional mode is the algal oil production at the expense of atmospheric CO2. Large scale microalgae cultivation systems (such as open/raceway ponds) are usually operated under photoautotrophic conditions (Mata et al., 2010). Autotro­phic nutritional mode also has fewer contamination problems compared with other

Autotrophic Nutrition

Calvin

Cycle

FIGURE 8.6 Autotrophic mode of nutrition in microalgal cells towards CO2 fixation and lipid biosynthesis

nutritional modes. Under autotrophic nutrition, the photosynthates also get consumed dur­ing respiration associated with the biomass growth, and hence the lipid productivity repre­sents the combined effects of oil content and biomass production (Chiu et al., 2008).

It has been established that electrons are derived from water upon photochemical oxidation by PSII or so-called water plastoquinone oxidoreductase (PQOR) and are trans­ferred to the [Fe]-hydrogenase, leading to the photosynthetic hydrogen production in the direct photolysis process. Apart from the previously described PSII-dependent hydrogen production, catabolism of endogenous substrate and the associated oxidative carbon metab­olism in green algae may generate electrons for the photosynthetic systems (Gfeller and Gibbs, 1984; Melis, 2002).

Electrons generated from such an endogenous substrate catabolism flow into the PQ pool between photosystems PSI and PSII (Stuart and GaDron, 1972; Godde and Trebst, 1980). An NADPH-PQOR that has been ascertained in vascular plant chloroplasts supplies elec­trons to the PQ pool (Shinozaki et al., 1986; Kubicki et al., 1996; Neyland and Urbatsch, 1996; Endo et al., 1998; Field et al., 1998; Sazanov et al., 1998). Daylight assimilation by PSI and the associated electron transfer raise the redox potential of these electrons to the equivalent level of ferredoxin and the [Fe]-hydrogenase. Functioning as the terminal electron acceptor, the hydrogen ions (protons) would lead to the production of molecular hydrogen (Gfeller and Gibbs, 1984; Bennoun, 2001; Gibbs et al., 1986). It has been found that in the presence of the PSII inhibitors 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU), which impede photosynthetic electron flow from PSII to the PQ pool, the process generates molecular hydrogen and carbon dioxide in a stoichiometric ratio of 2 to 1 (Bamberger et al., 1982). Thus, following a dark incubation of the culture under anaerobic conditions and the ensu­ing induction of the [Fe]-hydrogenase, considerable rates of hydrogen generation can be captured upon illumination of the algae in the presence of DCMU (Happe et al., 1994; Florin et al., 2001).

Commercial fertilizers, used for long periods, have adverse effects on soil productivity and environmental quality, so interest in environmentally friendly, sustainable agricultural practices has been on the rise. In developing and implementing sustainable agriculture tech­niques, biofertilization is of great importance to alleviate deterioration of natural ecosystems and to reduce the impact of environmental pollution while integrating nutrient supply into agriculture. Biofertilizers include mainly nitrogen-fixing, phosphate-solubilizing, and plant growth-promoting microorganisms, as in the case of microalgae.

Marine algae and algae-derived products have been widely used as nutrient supplements and as biostimulants or biofertilizers to increase plant growth and yield. The regulatory sub­stances cytokinins, auxins, gibberellins, and betaines in algae can induce plant growth (Valente, Gouveia et al., 2006), but their roles as macro — and micronutrients also make them valuable components of biofertilizers. A few commercial products based on marine algae are ready available for use in agriculture, but ongoing research has featured several alga species in terms of ascertaining their effects on plant growth. For instance, recent work with Laminaria digitata indicated that this marine macroalga (traditionally used as soil amendment in many parts of the world) improves seed germination and rooting in terrestrial plants (Thorsen, Woodward et al., 2010).

Several pieces of evidence confirmed that microalgae are beneficial in plant cultivation by producing growth-promoting regulators, vitamins, amino acids, polypeptides, and antibacterial and antifungal substances that exert phytopathogen biocontrol as well as poly­mers, especially exopolysaccharides that improve both plant growth and productivity (de Mule, de Caire et al., 1999). Other indirect growth-promotion effects may be claimed, such as enhancing the water-holding capacity of soils or substrates, improving availability of plant nutrients, and producing antifungal and antibacterial compounds (Schwartz and Krienitz,

2005) . In hydroponic cultivation, microalgae present a few extra benefits: The oxygen pro­duced by photosynthesis avoids anaerobiosis in the root system while releasing such growth-hormones as auxins, cytokins, gibberelins, abscisic acid, and ethylene (Schwartz and Krienitz, 2005). Equally important and promising is the high N:P ratio exhibited by microalgae, which is an extra indicator of its potential as fertilizer.

10.5.1 Animal Feed

The moisture content of fresh marine algae is quite high and can account for up to 94% of their biomass. However, marine algae contain such nutritional elements as proteins, lipids, carbohydrates, vitamins, and minerals that are in high demand for animal feed (Zubia et al., 2008). In particular, the ash content is high compared to that of vegetables (Murata and Nakazoe, 2001) and includes both macrominerals and trace elements.

Fish feeding represents over 50% of the whole operating costs in intensive aquaculture, with protein being the most expensive dietary source (Lovell, 2003). Nowadays 24% of the fish harvested by fisheries worldwide is used to produce fish meal and fish oil, thus putting high pressure on fisheries that aquaculture has attempted to alleviate. This demand promoted extensive efforts to find alternative sources of protein sources for aquatic feed; unfortunately, plants are poor protein sources in fish diets owing to their deficiency in certain essential amino acids, their content of antinutritional compounds, and taste problems.

Conversely, microalgae have been traditionally used to enrich zooplankton, which will in turn be used to feed fish and other larvae. In addition to providing proteins contain essential amino acids, they carry such other key nutrients as vitamins, essential PUFAs, pigments, and sterols, which may then be transferred upward through the food chain (Guedes and Malcata, 2012).

On the other hand, contamination by bacteria that attack fish can potentially devastate aquaculture farms. Microalgal fatty acids longer than 10 carbon atoms can induce lysis of bacterial protoplasts; said ability depends on composition, concentration, and degree of unsaturation of free lipids (Guedes et al., 2011b). The contents of carotenoids are important in aquaculture as well. In fact, artificial diets that lack natural pigments preclude such organisms as salmon or trout to acquire their characteristic red color (muscle), which, in nature, is a result of ingesting microalgae containing red pigments; without such a color, a lower market value will result (Guedes and Malcata, 2012).

The use of a heterogeneous catalyst (base or acid) for biodiesel production has been recently and extensively explored. The main advantage of a heterogeneous catalyst over a homogeneous catalyst is that the former can be recycled and regenerated for repeated reac­tion cycles, resulting in minimum catalyst loss and improvement of the economic feasibility of biodiesel production. Furthermore, the catalyst can easily be separated at the end of the reaction using filtration; therefore, product contamination is reduced and the number of water washing cycles (purification) is minimized (Lam and Lee, 2012).

Heterogeneous catalysts can generally be divided into two categories:

1. Catalysts with basic sites, such as CaO, MgO, ZnO, and waste materials impregnated with KOH or NaOH. This type of catalyst can catalyze transesterification under mild reaction conditions with a high yield of biodiesel usually attained. Nevertheless, due to the basic properties of the catalyst, it is highly sensitive to the FFA content in the oil and results in soap formation instead of biodiesel. Leaching of active sites from the catalyst is another limitation that can cause product contamination and catalyst deactivation.

2. Catalysts with acidic sites, such as SO42~/ZrO2, SO42~/TiO2, SO42~/SnO2, zeolites, sulfonic ion-exchange resins, and sulfonated carbon-based catalysts. These catalysts are insensitive to the FFA content in oil and are able to perform esterification and transesterification simultaneously. However, the reaction rate is exceptionally slow; hence, extreme reaction conditions such as high temperature (more than 100 °C) with high alcohol-to-oil molar ratio (more than 12:1) are necessary to accelerate the overall reaction rate.

To date, the application of heterogeneous catalysts to algal biodiesel conversion is still scarcely reported in literature, primarily because algal lipids are a relatively new feedstock that is not commercially available in the market. In a recent study carried out by Umdu et al. (2009), CaO supported with Al2O3 was used as a heterogeneous base catalyst in transesterification of Nannochloropsis oculata lipids (Umdu et al., 2009). The highest algal bio­diesel yield attained was 97.5% under the following reaction conditions: reaction temperature of 50 °C, methanol-to-lipid molar ratio of 30:1, catalyst loading of 2 wt%, and reaction time of 4 h. The long reaction time was required mainly because of the initial three immiscible phases (lipid-alcohol-catalyst) that increase mass-transfer limitations in the system. When pure CaO was used as catalyst in the transesterification of algal lipids, insignificant biodiesel yield was recorded, but when CaO supported with Al2O3 (ratio 8:1 w/w) was used instead, signifi­cantly better results were achieved due to the increasing basic density and basic strength of the catalyst. Other heterogeneous catalysts such as Mg-Zr and hierarchical zeolites have also been investigated for transesterification of algal lipids; however, unsatisfactory biodiesel yield (less than 30%) was attained (Carrero et al., 2011; Krohn et al., 2011).

The use of a large variety of impact assessment methodologies can be potentially problem­atic if one wants to compare LCA studies. For instance, the comparison of the results of the "eutrophication" impact is not possible between the study of Kadam (2002) and the study of Brentner et al. (2011). In the first case, the methodology used is CML and the eutrophication impact is expressed in phosphate equivalent, whereas in the second publication the method­ology used is TRACI and the impact is expressed in nitrogen equivalent. Moreover, some studies do not precisely state the impact assessment methodology (Clarens et al., 2010, 2011).

A comparison of the LCA results of bioenergy production from microalgae with results for fossil fuels and other biofuels should be included. The strengths and weaknesses of this new kind of bioenergy production compared to fossil fuel or classical bioenergy production from biomass must be identified. Assessed impacts should include climate change and an energy balance, but impacts that have reduced the interest in first-generation biofuel (such as land use change occupation or impacts linked with the nitrogen flows or the use of chemical prod­ucts) and have motivated the abandonment of fossil fuel (ozone layer depletion or abiotic re­source depletion) should also be presented. A focus should also be made on the quantity and the quality of required water, since evaporation or water spray to cool the process could lead to drastic water consumption (Bechet et al., 2010, 2011).

13.5 CONCLUSION

This chapter presents a critical review of 15 publications about LCA and bioenergy produc­tion from microalgae. The review illustrated the variability of assumptions made about technological and environmental performance of the different processes involved in the pro­duction and transformation of algal biomass. The main conclusion of this analysis is that there is real difficulty in comparing the environmental burdens of the proposed setups, and there is now a need for clear guidelines to ensure that each new LCA study will consolidate the cur­rent knowledge. This is of key importance, since the objective of LCA works will more and more often consist of guiding the design of new biofuel production systems and prove that they lead to actual progress in terms of environmental impact. In this spirit, there is a clear gain for the LCA community to accept a set of rules and guidelines to make any new analysis comparable to the existing ones.

As a consequence, we have proposed some guidelines for the LCA to allow a clearer and sounder comparison between processes and to better estimate the potential and challenges of microalgae for biofuel production.

Acknowledgments

This chapter presents research results supported by the ANR-08-BIOE-011 Symbiose project. Authors P. Collet, A. Helias, and L. Lardon are members of the Environmental Life Cycle and Sustainability Assessment research group (ELSA; www. elsa-lca. org). They thank all the other members of ELSA for their advice.

The characteristics of microalgae and the state in which they thrive can greatly affect the choice of algae-harvesting technology and its performance (Cooney et al., 2009). Separation of tiny and loosely suspended algal particles from the broth can be cumbersome because algal

cells normally carry negative charge and excess extracellular polymeric substances (EPS) to maintain algal stability in a dispersed state (Gudin and Therpenier, 1986). The stability of microalgae in the growth medium is mainly associated with algal surface charge, size, and density of the algal cells, which in turn influence their separability from aqueous suspensions. Both the electric interactions between algal cells and cell interactions with the surrounding culture broth contribute to the stability of the algal suspension (Tenney et al., 1969), whereas size and density of algal cells dictate their settling rate, which is an important consideration for sedimentation process design. Harvesting cost can be high since the mass fractions in culture broth are generally low.

Studies of the effect on surface charge of particles by various treatment methods have been extensively documented. Farvardin and Collins (1989) noted that pre-ozonation increases surface charge of humic substances. In another study, Chheda et al. (1992) noted an increase in stability of suspension of Na-montmorillonite particles at increased ozone dose, attributed to the increase in surface charge as a result of disruption of metal-oxygen bonds in crystal lattice. These studies, however, reported the existence of an optimal ozone dose whereby the coagulation of particles can be improved.

Conversely, Chheda and Grasso (1994) revealed that ozonation reduced stability of the Na-montmorillonite particles coated with natural organic matters (NOM) in river waters. It was postulated that the adsorption of NOM on Na-montmorillonite particles per se would render the particles more hydrophilic. Subsequent ozonation, however, turns the particle surface less negatively charged. This, in turn, resulted in partial dealuminization of Na-montmorillonite and transformation of coated NOM to increase the hydrophobicity of the particle surface, hence destabilizing the particles. It can be deduced from these studies that appropriate treatment of ozonation would help destabilize particles, leading to improved separation from the medium. Possible mechanisms for enhanced coagulation of suspended particles caused by ozonation were proposed (Reckhow et al., 1986; Plummer and Edzwald, 2002). These mechanisms include:

1. Increase in carboxylic content to enhance adsorption to alum floc and calcium and

magnesium precipitates

2. Reduction in molecular weight of adsorbed organics to reduce steric hindrance of

particles

3. Breakdown of organometallic bonds to release ions such as Fe3+ for organics precipitation

4. Rupture and lyses of algal cells to release biopolymers for coagulation

5. Polymeriazation to large particles for sedimentation

Henderson et al. (2009) noted that bubbles with surfaces modified using chemicals of both a hydrophobic long tail and a hydrophilic high charge head can yield sufficient algal removal without upstream coagulation and flocculation. In an earlier study by Henderson et al. (2008), it was reported that the algogenic organic matters (AOM) extracted from four algal species (Chlorella vulgaris, Microcystis aeruginosa, Asterionella formosa, and Melosira sp.) were domi­nated by hydrophilic polysaccharides and hydrophobic proteins of low specific UV absor­bance and negative zeta potentials. The hydrophobicity of AOM was attributable to the hydrophobic proteins of molecular mass greater than 500 kDa. Additionally, the charge density for the AOM, being attributable to hydrophilic and acidic carbohydrates and not hydrophobic humic acids, decreases inversely with hydrophobicity. On the other hand, inhibition on ferric chloride coagulation of algae by isolated AOM secreted by cyanobacgterium Aphanothece halophytica was reported (Chen et al., 2009). It was hypothe­sized that the AOM can form complex compounds with ferrum thereby inhibits the coagu­lation. As discussed earlier, ozonation is able to increase hydrophobicity of NOM, thereby enhancing its coagulation. The impact of ozonation on AOM, however, is unclear and yet to be investigated.

Just like planktonic cells, algal cells normally carry negative surface charge. Whereas algal surface charges are derived from ionization of ionogenic functional groups at the algal cell wall (Golueke and Oswald 1970) and selective adsorption of ions from the culture medium, the intensity of the charge is influenced by algal species, ionic concentration of medium, pH, and other environmental conditions.

Based on the principles of the Deyaguin-Landau-Verwey-Overbeek (DLVO) theory of col­loid stability, the interactions between colloidal particles are influenced by various interacting forces such as electrostatic double-layer repulsion, van der Waals attraction, and steric inter­action. There is a potential energy barrier to be overcome if coagulation of the minute charge particles is to be attained. It can be exceeded by the kinetic energy of the particles or, alter­natively, by the reduction of the energetic barrier. This is done by compressing the double layer through either by increasing the counter-ion concentration or by using counter-ions of higher valency (Ives, 1959). Although the double-layer theory is of great theoretical impor­tance, its use is restricted to cases in which specific chemical interactions do not play a role in colloid stability.

Destabilization of colloidal suspension such as that in algal culture as a result of specific chemical interactions is attainable by the presence of organic polymers (Shelef et al., 1984). Commercial polymers, usually those of high molecular weights such as polyelectrolytes or polyhydroxyl complexes, are considered superior coagulants or flocculants. The polymeric coagulation-flocculation is explained by the bridging model, postulating that a polymer can attach itself to the surface of an algal particle by several segments with remainder segments extended into solutions. These segments are then able to attach to vacant sites of other algal particles, forming a three-dimensional floc network (Gregory, 1977).

A planktonic algal cell can be considered a very minute spherical object that falls in a continuous viscous fluid medium at velocities governed by gravity’s downward force and the upward drag (or frictional) and buoyancy forces. If the algal particle is falling in the vis­cous fluid by its own weight due to gravity, then a terminal velocity, also known as the settling velocity, is reached when this frictional force, combined with the buoyant force, exactly bal­ances the gravitational force, as described by Stokes’ law. In actual fact, the settling velocity of planktonic algae in natural habitat is dictated by a variety of complex factors, which include cell mobility, water turbulence and flow, and upwelling caused by winds and temperature stratification (Hutchinson, 1967). The settling velocity of planktonic algae can be reduced in an ecosystem by the following:

1. Motility

2. Reducing cell dimensions

3. Increment of the drag forces as in the Scenedesmus species, which contain seta (Conway

and Trainer, 1972)

4. Reducing cell density, as in many blue-green algae, which contain gas vacuoles (Fogg, 1975;

Paerl and Ustach, 1982)

Hence, the settling velocity of an algal cell can be increased by increasing cell dimensions,

i. e., by cell aggregation into a larger body. This principle is applied in algal separation pro­cesses where chemical coagulants are added to form large algal flocs which settle rapidly to the reactor or tank bottom. Conversely, air bubbles, which may attach to the already formed algal flocs, will reduce drastically the floc density, causing the floc to float atop the vessel. Increasing the gravity force will increase the settling velocity of algal cells, which is attainable by applying centrifugal forces on algal suspensions.

In summary, destabilization and flocculation of algal suspension are important consider­ations in most of the various algal separation and harvesting processes, which are described separately in the following section.

In the heterotrophic batch cultures, high initial concentration of substrates, e. g., sugars, is usually used to provide sufficient carbons for obtaining high cell density. Accompanying the high substrate concentration, however, is the occurrence of possible growth inhibition. For instance, the optimal sugar concentration for growing C. zofingiensis was reported below 20 g L-1, above which the inhibition of algal growth was observed (Ip and Chen, 2005; Liu et al., 2012a). The substrate-based inhibition caused not only the decreased specific growth rate but also the lowered biomass yield coefficient based on sugars (Sun et al., 2008; Liu et al., 2012a), contributing accordingly to the increased cost input. To overcome the inhibition issue associated with batch cultures, fed-batch cultivation is a commonly used strategy in which the substrate is fed into the algal cultures step by step to maintain it sufficiently for cell growth but below the level of inhibition threshold. There have been a number of reports employing fed-batch strategy to grow algae heterotrophically with the aim of avoiding the possible inhibition caused by the initial high substrate and improving the production poten­tial of biomass as well as of oils (Xu et al., 2006; Li et al., 2007a; Sun et al., 2008; Xiong et al., 2008; Liu et al., 2010; 2012a; De la Hoz Siegler et al., 2011; Yan et al., 2011; Chen and Walker,

2012) . Liu et al. (2010) investigated the heterotrophic oil production by C. zofingiensis using fed-batch cultures in a 3.7-L bioreactor. A two-stage feeding was adopted: three times of feed­ing with glucose-containing nutrients (to maintain linear growth) followed by four times of glucose feeding alone (to further increase biomass and induce oil accumulation; Figure 6.4). Glucose concentration of the cultures was maintained between 5 and 20 g L-1. The maximum lipid yield and lipid productivity achieved in the fed-batch cultures were 20.7 g L-1 and 1.38 g L-1 day-1, respectively, representing around a 2.9-fold increase of the those obtained in batch cultures.

Although the employment of fed-batch culture strategy proves able to eliminate the sub­strate inhibition, it cannot overcome the inhibition caused by the toxic metabolites that would be produced by the algal cultures and accumulate as the cells build up, preventing further enhancement of cell density.

FIGURE 6.4 (A) Growth and glucose consump­

tion and (B) lipid production in a two-stage fed-batch A» fermentation of C. zofingiensis in a 3.7-L fermentor. (O) cell biomass; (□) glucose concentration; (column) lipid content; (△) lipid yield; (#) glucose-containing medium feeding; (##) glucose feeding alone. Adapted E from Liu et al. (2010) and the permission for reprint requested.

6.5.1 Continuous Cultivation

The term continuous cultivation refers to the fresh medium being continuously added to a well-mixed culture while cells or products are simultaneously removed to keep the culture volume constant. It allows the steady state of kinetic parameters such as specific growth rate, cell density, and productivity and is thus considered an important system for studying the basic physiological behavior of heterotrophic algal cells. Figure 6.5a shows the schematic di­agram of the continuous cultivation system. This system is capable of effectively eliminating the metabolite-driven inhibition. There are several reports of continuous cultivation of algae in both photoautotrophic (Molina Grima et al., 1994; Otero et al., 1997) and heterotrophic (Wen and Chen, 2002b; Ethier et al., 2011) growth modes. Ethier et al (2011) investigated the continuous production of oils by the microalga Schizochytrium limacinum with various di­lution rates (D) and feed glycerol concentrations (S0). The yields and productivities of bio­mass, total fatty acids (TFA), and docosahexaenoic acid (DHA), shown in Figure 6.6, were over the range of D from 0.2 to 0.6 day-1 (S0 fixed at 90 g L-1) and the range of S0 from 15 to 120 g L-1 (D fixed at 0.3 day-1). The highest biomass productivity is 3.9 gL-1 day-1, obtained with the 0.3 day-1 of D and 60 g L-1 of S0 (Figure 6.6b). The maximum productivities of both TFA and DHA were also achieved at the same D but with a higher S0 of 90 g L-1 (Figures 6.6d and 6.6f). Liu et al (2012a) surveyed the feasibility of using a semicontinuous C. zofingiensis culture fed with waste molasses for oil production. The waste molasses contains relatively high levels of metal ions and salt that are inhibitory to algal growth, causing the

FIGURE 6.5 Schematic diagram of (A) continuous, (B) perfusion, and (C) perfusion­bleeding culture systems. X, cell concentration; V, culture volume; S, carbon concentration in medium; F, flow rate of feed; F1, flow rate of per­fusion; F2, flow rate of bleeding; S0, carbon con­centration in feed. The flow rates are controlled to keep the culture volume constant.

failure of molasses-based fed-batch cultivation when molasses was not pretreated; in con­trast, C. zofingiensis in the semicontinuous culture fed with diluted raw molasses showed comparable growth rate and sugar utilization to that with pretreated molasses (Liu et al., 2012a). Although continuous cultivation can promote the productivity, it is worth to mention that accompanying the increase of dilution rate is the drop of cell density as well as of sub­strate utilization efficiency (Wen and Chen, 2002b). From a cost-effectiveness point of view, this is undesirable in that the residual substrate is wasted with the effluent and more energy input is required to harvest the diluted cells.

FIGURE 6.6 Algal growth, TFA and DHA production of continuous Schizochytrium limacinum in a 7.5-L fermentor with various dilution rates (D) (A, C, E; S0 = 90 gL-1) and feed glycerol concentrations (S0) (B, D, F; D = 0.3 day-1). Adapted from Ethier et al. (2011) and the permission for reprint requested.

It has been reported that with flat panel/plate photobioreactors, high photosynthetic effi­ciencies can be achieved (Hu et al., 1996; Richmond, 2000). Accumulation of dissolved oxygen concentrations in flat plate photobioreactors is relatively low compared to horizontal tubular photobioreactors. Milner’s (1953) work paved the way to the use of flat culture vessels for cultivation of algae. Flat panel photo bioreactors were used extensively for mass cultivation of different algae (Tredici and Materassi, 1992; Hu et al., 1996; Zhang et al., 2002; Hoekema.,

2002) . Lack of temperature control and gas engagement zones are some of the inherent dis­advantages observed with this type of photobioreactor.