The nutrient requirement is known to depend on the species but also on the stress that has been induced to stimulate lipid or carbohydrate storage. The nitrogen and phosphorus quota can strongly vary during a starvation period (Geider and La Roche, 2002). The hypotheses on required fertilizers strongly vary according to the species and between the publications for the same species (Lardon et al., 2009; Stephenson et al., 2010; Yang et al., 2011). Needs in ni­trogen vary from 10.9 g kgDM-1 (Lardon et al., 2009) to 20.32 g kgDM-1 (Stephenson et al.,

2010) in limiting conditions and from 9.41 g kgDM-1 (Kadam, 2002) to 77.6 g kgDM-1 (Clarens et al., 2011) without stress. Needs in phosphorus vary from 2.4 g kgDM-1 (Lardon et al., 2009) to 2.58 g kgDM-1 (Khoo et al., 2011) in limiting conditions, and from 0.02 g kgDM-1 (Kadam, 2002) to 71 g kgDM-1 (Yang et al., 2011) without stress. All the au­thors agree on high nutrient consumption for the culture of microalgae, but they differ on the ways to provide them (see Table 13.3). Some authors, such as Sander and Murthy (2010) and Clarens et al. (2010,2011), consider that the needs in nitrogen and phosphorus can be totally or partially covered by the addition of wastewater to the growth medium. But in most of the publications, nutrients are provided by chemical fertilizers. To reduce the nutrient consump­tion, several authors suggest recycling the digestates resulting from the anaerobic digestion of oilcakes (Stephenson et al., 2010; Brentner et al., 2011; Campbell et al., 2011; Clarens et al.,

2011) or of bulk microalgae (Clarens et al., 2011; Collet et al., 2011).

Figure 13.4 shows environmental impacts of various fertilizer sources. As previously dem­onstrated for the energy mix, the source of nutrients can have important consequences on the environmental balance of the energy production from microalgae. Climate-change impact and endpoint impacts on human health and ecosystem can vary by a factor of two, based on the chosen nitrogen fertilizer. For these three impacts, ammonium nitrate is the worst one, and the impacts of ammonium sulphate, calcium nitrate, and urea are quite the same. Concerning resource consumption, urea is the worst, mainly because of the high amount of natural gas used for its production. Clarens et al. (2010, 2011) and Sander and Murthy (2010) suggest using wastewater to grow algae. This assumption allows reducing the con­sumed quantities of freshwater and chemical fertilizers. However, mineral elements’ content in wastewater can strongly vary depending on the place and the period of the year. For these reasons, from our point of view it seems very difficult to rely on such fertilizers.

To enhance algae settling, flat inclined plates are incorporated in a settling tank to promote solids contacting and settling along and down the plates. The slopes of plates are designed for the downgliding of the settled algal particles into a sump from which they are removed by pumping (Mohn, 1980; Shelef et al., 1984). Algae were concentrated to 1.6% solids content, and coagulant dosing was suggested if suspension of tiny algae such as Scenedesmus is fed to the system (Mohn, 1980). Operational reliability of this method was fair, and further thick­ening of algae slurry was required.

A fast pyrolysis system consists basically of a fluidized bed reactor, a cyclone, a condenser, and a combustion chamber, generally constructed as shown in Figure 7.2.

The fluidized bed reactor is where pyrolysis actually occurs. The remaining constituents are responsible for phase separation. The reactor operates at around 450°C. Heating is done by an immersed electrical resistor covered with inert material (silicates, in general). The func­tion of this inert material is to increase the heat transfer between the air and the fluidizing material to be pyrolyzed through abrasive action, increasing the contact surface of the solids (DalmasNeto, 2012).

Once temperature is achieved, air feeding begins. Then heating stops and the material to be pyrolyzed is fed to the reactor. At this point, an initial temperature fall is observed, caused by air and material entrance in much lower temperatures. Reactor temperature can be

immediately reestablished by combustion of the pyrolysis’ incondensable gases or by con­trolled combustion of part of the material fed to the reactor.

The combustion of incondensable gases, such as CO, H2,andCH4 (Cortez et al., 2008), is the best option, generating enough heat for autothermal operation of the reactor, but this entails the acquisition of additional equipment. On the other side, controlled combustion of part of the material fed to the reactor is easier to be handled but means loss of product (about 10% of the material needs to be burned to maintain reactor temperature, according to Mesa-Perez, 2005).

Residence time is controlled based on material feeding rate, air flow, and reactor volume. Material characteristics such as density and size are taken into account to avoid dragging out the time. After pyrolysis, the gaseous mixture is sent to a cyclone by pneumatic conveying (by the fluidizing air itself). In the cyclone, gaseous and liquid components are separated by cen­trifugal force. The gaseous products enter the condenser. The condensable fractions are then separated by gravity: In the bottom an output is used for bio-oil gathering, while the acid ex­tract is collected at the middle of the condenser. Gases and very light particles enter a centri­fuge located at the top of the condenser, where some light particles condensate, increasing the yield of the liquid phases.

The condenser effluent gases are formed by four fractions. The first one is composed of inert atmospheric gases that adhered to biomass particles when the reactor was fed; the second one consists of inert gases fed with air in fluidization (nitrogen, CO2). The third fraction involves semioxidized pyrolysis gases such as CO and CH4; the fourth is composed of those gases that are combusted to provide energy to the system. Usually this gas phase is fed back to the system, especially due to the potential of the third fraction to provide energy to the system.

The combustion chamber is responsible for burning all combustible gases generated in the process. It acts as a restorative power cell besides being a security tool (preventing release of flammable gases into the atmosphere).

The following steps and reactions summarize pyrolysis processes (adapted from Gomes et al., 2008):

1. Drying: Humid material! solid material + H2O(g)

2. Pyrolysis: Dry material! coal + volatile products

3. Combustion reactions:

a. C(s) + O2!CO2(g) + energy

b. 2H2(g) + 02(g) ! 2H2O(g) + energy

4. Heat transfer

5. Mass transfer

The smooth operation of a fast pyrolysis system depends very little on the raw material conditions but strongly depends on its composition (organic matter amount). To be pyro — lyzed, the material might be dried and milled into particles smaller than 20 mm (Bridgwater et al., 1999). Low moisture content is desired to avoid wasted energy (or higher energy de­mand) and possible influence on calorific power of the final product. (High-moisture-content materials are frequently pyrolyzed but with the drawback mentioned previously.) Particle size might be big enough to avoid excessive biomass drag by fluidizing air (the flow of which is usually high), causing loss of nonpyrolyzed material, but also small enough to allow easy heat transfer and avoid secondary polymerization and carbonization reactions (this will cause coal yield increase, according to Sanchez, 2003).

Ganesh, 1990 found that both acid and alkaline catalysts tend to increase gas production. The same study noted that desmineralization caused an increase in the superficial area of coal.

Due to the high heating rate to which material is subjected in fast pyrolysis, the residence time might be very short, usually around 1 second (Gomez, 2002). In this condition, advanced stages of undesirable reactions (such as polymerization and/or decomposition) are avoided. Figure 7.3 presents the most probable mechanisms of formation of pyrolysis products.

Osmotic shock or osmotic stress is a sudden change in the solute concentration around a cell, causing a rapid change in the movement of water across its cell membrane (Fajardo et al., 2007). This shock causes a release in the cellular contents of microalgae. The method is more applicable for the strains cultivated in marine environments (eg. Nannochloropsis sp.). Os­motic shock is also induced to release cellular components for biochemical analysis (Mario, 2010). This method is also applied for Halorubrum sp. isolated from saltern ponds. The results showed increased lipid productivities and variations in lipid compositions (Lopalco et al., 2003).

Extraction of lipids is a key aspect involved in biomass-to-biodiesel production, the method directly influences the lipid productivity potential of the process. So far, several methods have been employed for extracting the cellular contents (lipids) of microalgae. Each method has its own advantages and disadvantages for practical applicability. Among the pro­cesses described, solvent extraction is suitable for extracting lipids from mass cultures but requires large volumes of solvent. The Soxhlet extraction method is applicable only when a single solvent is used and is not suitable for binary solvent applications. However, recovery and reusability of the solvent are possible with this method. The ultrasonic extraction method can perform well when coupled with the enzymatic treatment, but both methods lack cost effectiveness and feasibility for large-scale applications. Supercritical carbon dioxide extrac­tion (SC-CO2), pulse electric field procedure, osmotic shock, hydrothermal liquefaction, and wet lipid extraction require more optimization efforts for large-scale applications. A suitable method operatable with both binary and single solvents, applicable at large scales and yield­ing higher lipid productivities, is yet to be optimized for achieving enhanced microalgae lipid yields.

Sterols occur naturally in plants and animals; the most familiar type of the latter is cholesterol, which is vital to cellular functioning due to its role in the fluidity of the cell membrane, besides serving as a secondary messenger in developmental signaling. Further­more, cholesterol is a precursor of fat-soluble vitamins and steroid hormones.

The content and type of sterols vary with the alga species: green algae contain 28-isofucocholesterol, cholesterol, 24-methylene-cholesterol, and p-sitosterol, whereas brown algae contain fucosterol, cholesterol and brassicasterol; red algae contain desmosterol, cholesterol, sitosterol, fucosterol, and chalinasterol. The predominant sterol in brown algae, fucosterol, accounts for 83-97% of the total sterol content, whereas desmosterol, in red algae, accounts for 87-93% (Sanchez-Machado, Lopez-Hernandez et al., 2004; Kumar, Ganesan et al., 2008).

The rapid commercial expansion of the algal biofuel industry is an excellent example of sustainable product development with dramatic future potential for contributions to fuel sup­plies, yet many questions regarding algae production remain unanswered. The state of knowledge regarding the potential environmental impact of the production of algae and algae-derived biofuels continues to be incomplete, fragmented, and largely obscured by proprietary concerns. However, this knowledge is changing rapidly, facilitated by research and industry and driven by economics. Commercialization of the production of algae — derived biofuels as part of the overall biofuel industry will have a profound future impact on society. Waste products that are currently discharged into the environment as contami­nants will be utilized to produce much-needed renewable energy sources. Now is the time to initiate the development of an algae industry evaluation methodology that allows for the advancement of knowledge and evaluation tools for authorities to best understand the potential implications (Menetrez, 2012).

The process of generating biofuel from algae involves the growth, concentration, separa­tion, and conversion of microalgae biomass, some of which can be genetically altered. After separating the desired biofuel product or products from the microalgae biomass, a significant portion of byproduct remains. It is important that the remaining byproducts have a useful and safe purpose for the economic feasibility and environmental sustainability of the process. Post-extraction byproducts must be used efficiently and completely. Since no biofuel is car­bon-neutral in the current scenario, significant fossil-fuel input is needed for growing, processing, and extracting the oil, which might offset the positive aspects of the algal biofuel (U. S. DOE, 2006).

Microalgae hold great promise as starting materials for biofuel production, but signifi­cant challenges exist for the developing industry. Present economy-of-scale differences between the algal oil industry and the petrochemical industry are immense and will require significant investment in the form of government-funded incentives for liquid fuels de­rived from microalgae. The present microalgae manufacturing industry is very small at only 5000 tons y-1 (Pulz and Gross, 2004). It is almost completely devoted to synthesizing high-value nutraceutical products and is not extensively engaged in the mass production of high oil-containing microalgae. In addition, microalgae require significantly higher levels of nitrogen than terrestrial plants to achieve effective growth rates, which increases the cost of production (Brezinski, 2004).

Moreover, microalgae do not achieve concentrations at maturity in their natural aqueous growth media >1 wt%. Therefore, to become a significant industrial commodity in terms of cost and scale, growth and harvesting technologies need to be developed that can econom­ically provide higher concentrations required for industrial-scale processing operations. Actual extraction processes for lipids and lipid-derived materials require considerable improvement. Essentially, any process suitable for commercial consideration must not dry algae by evaporating water. The energy input to evaporate water is significant, and the heat energy input required will, with very few exceptions, be greater than any energy output that can be obtained by combustion of the dried material (Heilmann et al., 2011).

Although algal biofuels possess great potential, profitable production is quite challeng­ing. Much of this challenge is rooted in the thermodynamic constraints associated with producing fuels with high energy, low entropy, and high exergy from dispersed materials (Beal et al., 2012).

One of the difficulties with using conventional gasification technologies for converting high-moisture biomass such as algae is the low thermal efficiency that results from the need to vaporize the water in and with the feedstock. Thus, conventional biomass gasification pro­cesses require a dry feedstock. Of course, energy is still required to do this drying prior to gasification, and the energy needed here offsets some of the energy gained by producing the gaseous product. Gasifying wet biomass in supercritical water is a means of circumventing this energy penalty (Guan et al., 2012).

In addition to these holistic challenges, some of the other challenges with respect to the reactors and catalyst are as follows: Due to the high pressures needed for processing, special reactor and separator designs are required. The process has to be designed in a way that can handle solids loading in excess of 15-20 wt% and handle feedstock with impurities. Proper heat-recovery systems have to be designed because the reactors operate at very high temper­atures in pressurized conditions. Feeding at high pressures into the reactor is always a chal­lenge and is a major problem in small-scale plants. In the case of heterogeneous catalysts, the catalyst has to be robust and should not deactivate easily due to the formation of coke. If ho­mogeneous catalysts are used, they must be recovered at the end of the process and reused again. Another most important phenomenon that occurs is the wall effect, which can cause serious problems after scaling up if not understood at lab-scale level (Peterson et al., 2008).

With respect to the conversion methods, effective heat and mass transfer is required for the proper conversion of feedstock into the desired products. This requires advanced design of reactors used for conversion and preparation of hybrid catalysts.

Acknowledgments

The authors thank the Director, Indian Institute of Petroleum, Dehradun, for his constant encouragement and sup­port. RS thanks the Council of Scientific and Industrial Research (CSIR), New Delhi, India, for providing a senior research fellowship (SRF). The authors also thank the CSIR for financial support.

One of the statements of the LCA methodology is to link every economic and environmen­tal flow to the reference flow of the FU. However, several processes implied in the production of the FU can lead to the production of several products. Two approaches are possible to han­dle the multifunctionality of the system: allocation or substitution. The allocation approach consists of distributing the environmental burden of the upstream between all the coproducts of the multi-output process. This distribution should be based on the most sensitive criterion, e. g., mass, economic value, or energy content of the products. The perimeter expansion (or sub­stitution) option consists of adding the coproduct to the FU. The ISO norm for LCA stipulates that perimeter expansion should be preferred when possible. When substitution is not pos­sible, energy allocation should be preferred for processes leading to the production of energy.

Among the 15 publications, 3 publications (Kadam, 2002; Clarens et al., 2010; Jorquera et al., 2010) analyze systems without coproducts. Table 13.8 presents the various coproducts and the choices between allocation and substitution. Several processes can lead to coproducts:

• The oil extraction process leads to the production of an extraction residue (oilcake); only Lardon et al. (2009) chose to use an energy-based allocation at this level. Other authors (Stephenson et al., 2010; Brentner et al., 2011; Campbell et al., 2011; Clarens et al., 2011) chose to directly treat the oilcake by anaerobic digestion. Oilcakes can also replace other products: aquaculture or livestock food, carbohydrates’ source for bioethanol production.

• Oil esterification produces methylester and glycerol; here economic and energy allocation are often used. In case of substitution, glycerol is mainly used as a source of heat.

• Anaerobic digestion produces biogas and solid and liquid digestates; these digestates can be considered waste (and hence cannot support a part of the environmental burden of the process), fertilizer, or soil conditioner. The liquid digestate can be recirculated to the culturing device and hence substituted to a fraction of the mineral fertilizer required for the algae. The produced biogas is transformed into heat used on site to heat the digesters and/ or converted into electricity. Electricity is also consumed on site, and the surplus is injected into the network (Stephenson et al., 2010; Clarens et al., 2011).

A solid-ejecting disc centrifuge provides intermittent solids ejection by regulating its valve — controlled peripheral ports using a timer or an automatic triggering device. The advantage of this centrifuge for algae harvesting is its ability to produce algal cake in a single step without chemical dosing (Mohn and Soeder, 1978; Mohn, 1980; Shelef et al., 1984). This centrifuge con­centrated various types of microalgae effectively, achieving algal cake of 12-25% solids (Mohn, 1980; Moraine et al., 1980). The extent of the algae suspension separation increases with increas­ing residence time (decreasing feed rate), and the ejected cake concentration is affected by the intervals between successive desludging (Shelef et al., 1984). A solid-ejecting disc centrifuge was found very reliable. The only reported setback was that solids finer than algae may be retained in the overflow, which reduces the separation efficiency (Moraine et al., 1980). High capital and energy costs render this separation method unappealing.

Microalgae can also function in mixotrophic nutrition mode by combining both the auto­trophic and the heterotrophic mechanisms. It facilitates fixing atmospheric CO2 as well as con­suming the organic molecules and micronutrients from the growing environment (Figure 8.8). Microalgae can assimilate available organic compounds as well as atmospheric CO2 as a car­bon source in mixotrophic mode. The CO2 released by microalgae via respiration will again be trapped and reused in mixotrophic nutritional mode. It differs from photoheterotrophic nutrition mode in terms of CO2 utilization. The mixotrophs have the ability to utilize organic carbon; therefore, light energy is not a limiting factor for biomass growth (Chang et al., 2011). The acetyl-CoA pool will be maintained from both carbon sources—that is, by the CO2 fixation (Calvin cycle) and intake from outside the cell, which can further make malonyl-CoA. The photosynthetic metabolism utilizes light and CO2 for growth and organic photosynthate pro­duction, whereas respiration uses the organic photosynthates produced during photosynthe­sis. If an external carbon source is available in the system, there is a less loss of photosynthate during respiration, and the algae utilize the available excess photosynthates for biomass development. Mixotrophic cultures show reduced photoinhibition and improved growth rates over autotrophic and heterotrophic cultures (Chojnacka and Noworyta, 2004).

8.3 NUTRITIONAL MODE OF MICROALGAE

Mixotrophic Nutrition

Lipid

Glucose

FIGURE 8.8 Mixotrophic mode of nutrition in algal cells towards CO2 fixation and glucose assimilation for lipid biosynthesis

Algae have the flexibility to switch their nutritional mode based on substrate availability and light condition. If simpler carbohydrates are present in the system, algae shift towards heterotrophic nutrition from autotrophic mode to save energy. Scenedesmus obliquus readily adapted to heterotrophic growth in dark conditions utilizing glucose (Abeliovich and Weisman, 1978). Heterotrophic cells differed significantly from photoautotrophic cells with respect to several physiological properties such as the rate of photoassimilation of CO2 and the rate of incorporation of carbon and chlorophyll a concentration. Algal cells in an oxidation pond shared features common to both photoautotrophic and heterotrophic cells (Abeliovich and Weisman, 1978), associating with the mixotrophic mode of operation. Bacteria seem to play a minor role in biological oxygen demand reduction in high-rate oxidation ponds, and their role is probably confined to degradation of biopolymers, thus producing substrates for algal consumption.

The advantages of mixotrophic nutrition are its independence in terms of both photosyn­thesis and growth substrates (Kong et al., 2012). The mixotrophic growth regime is a variant of the heterotrophic growth regime, where CO2 and organic carbon are simultaneously assim­ilated and both respiratory and photosynthetic metabolism operates concurrently (Kaplan et al., 1986; Lee, 2004; Perez-Garcia et al., 2011). Mixotrophism is often observed in ecological water bodies, where the homeostatic structure and function of living systems are supported by chemical, physical, and organic activity in biota that balance the ecological status. Water ecosystems generally consist of nutrients and organic carbon as integral parts (Venkata Mohan et al., 2009), where microalgae, along with other living components, function together symbiotically. Some microalgal species are not truly mixotrophs but have the ability to switch between phototrophic and heterotrophic metabolisms, depending on environmental condi­tions (Kaplan et al., 1986). Microalgae-accumulating lipids are generally grown in natural

water bodies; therefore, ecological water bodies embedded with diverse microalgae species can be considered as potential reservoirs for harnessing biodiesel. In this regard, an attempt was made to explore the ability and potential of mixed microalgae cultures derived from dif­ferent water bodies in extracting lipids, which can be further transesterified to biodiesel. The study also focused on the economic mode of lipid production from the treatment of domestic sewage. The growth of algae was shown to be highest under mixotrophic conditions, with higher biomass productivity under photoautotrophic conditions (Bhatnagar et al., 2010; 2011). Mixotrophic cultivation was shown to be a good strategy to obtain a large biomass and high growth rates (Ogawa and Aiba, 1981; Lee and Lee, 2002), with the additional benefit of producing photosynthetic metabolites (Chen, 1996; Perez-Garcia et al., 2011). Solazyme, a renewable oil company in the United States, has developed an integrated algal cultivation process by dark heterotrophic mechanisms, giving carbon sources externally. The company is using various forms of waste material as feedstock for the cultivation of algae in fermenters and harnessing as much as 75% of oil on the basis of dry cell weight. The company is antic­ipating in selling algal oil to commercial refineries by the end of 2013.

Open-pond culture systems and enclosed bioreactor facilities have been used commer­cially in the recently evolved alga biomass biotechnology, but bioreactor design in algal hy­drogen production is still in the research and development stage. Because biohydrogen metabolism is primarily the domain of microalgae, diatoms, or cyanobacteria, the design of a photobioreactor depends on microbiological processes associated with bacteria and microalgae (Show et al., 2008; 2011; 2012). Although these photoheterotrophic bacteria differ in photochemical efficiency, absorption coefficient, and size, the light regime, including light and dark cycles, is assumed to be much more determining than biological factors (Akkerman et al., 2003). Hence the productivity of a photobioreactor is light-dependent, and a large surface-to-volume ratio is a prerequisite for a productive photobioreactor for optimal light exposure of the algae. Provisions for thermal control and monitoring of factors, including flow rates, pH, and dissolved oxygen, sulfur, and hydrogen are essential. Technical develop­ment is now moving toward devising gas-tight systems, engineered microalgae culturing, and computer-controlled systems for monitoring and automatic nutrient delivery and culture dilution.

Photobioreactors have been designed to achieve an economical, rapid multiplication and high algal biomass density (>1012 cells per cubic meter of culture) (Evens et al., 2000). Various photobioreactor designs, including flat plate, tubular, pond, and pool type, have been inves­tigated (Akkerman et al., 2003). The photobioreactor process, whether of batch or continuous flow, should be designed for optimal light exposure to the algae. Sufficient light supply is vital for adequate biomass growth in achieving high-density culture and for photosynthetic generation of hydrogen. Light conversion efficiencies are low (limited to 10% theoretically) and tend to decrease at higher light intensities because of the light saturation effect (Akkerman et al., 2003). The reason for this inefficiency is that, at high solar intensities, the rate of photon absorption by the chlorophyll antenna of the upper layers of algal biomass far exceeds the rate at which photosynthesis can consume. This phenomenon is attributable to the fact that algae have an intrinsic tendency to accumulate a large assembly of photon­absorbing chlorophyll antenna molecules as a survival strategy. The overabsorption of light by the chlorophyll antenna results in loss of excess photon as heat or other rays. Moreover, cells at the upper layers of the algal mass are subject to severe photoinhibition of photosyn­thesis due to the high rate of photon absorption (Baroli and Melis, 1996; Melis, 1999).

Research is underway in improving further algal photosynthetic capacity using an molec­ular engineering approach, whereas algal strains have been manipulated to increase hydro­gen production (Hankamer et al., 2007; Beer et al., 2009). It has been reported that a truncated chlorophyll antenna size of the photosystems in the chloroplast of the microalgae could alle­viate the optical shortcomings associated with a fully pigmented chlorophyll antenna (Melis et al., 1999; Neidhardt et al., 1998). The work on the truncated chlorophyll antenna size in maximizing solar conversion efficiencies is delineated in Section 9.5. Mutant algae with less chlorophyll were cultured and are able to distribute more sunlight to deeper layers in the algal biomass for large-scale applications (Hankamer et al., 2007). In this manner, sunlight is made available for more algal cells to generate hydrogen. Hence, for efficient photoproduction of hydrogen, it is critical to dilute the light and distribute it over the entire reactor volume and to mix the culture at high rates so that cells are exposed to the light for only a short time.

Algal photobioreactors can be designed to regulate light inputs to the algal culture to im­prove its photon conversion efficiency. A substantial increase in light utilization efficiency of up to 15% has been reported (Tetali et al., 2007; Laurinavichene et al., 2008). Conversion ef­ficiency between 10% and 13% is feasible using engineered microbial culture to better utilize the solar energy (Turner et al., 2008). However, improvements must be made to the solar con­version efficiency of the algae for commercial purposes. Critical issues such as the optical shortcomings associated with the chlorophyll antenna size and the light saturation of photo­synthesis must be addressed under mass culture conditions (Melis et al., 1999). Technological advancement addressing these issues is discussed in Section 9.5.

Appropriate configuration of the bioreactor needs to be established for the most effective use of light and surface area. Biomass mixing is hence significant to ensure uniform disper­sion of nutrients and light illumination in the culture as well as to prevent agglomeration and sedimentation of algal biomass (Melis, 2002). Modular design of experimental systems should be allowed for possible scale-up. Such commercial scale should achieve sustainable gas out­put and high hydrogen yields with compact configuration. Trapping and withdrawal of hy­drogen gas in the system are also important design considerations for photobioreactors. Given the current advancement in photobiohydrogen production, technical and economic strategies for cycling the microalgae between sulfur deprivation and supply must be devel­oped (Laurinavichene et al., 2008).

Various types of photobioreactors had been investigated in a study by Janssen (2002). Small-scale flat panel reactors consisting of a rectangular transparent box were mixed with gas introduced via a perforated tube at the bottom of the reactor. To create a high degree of turbulence, 3 to 4 liters of air per liter of reactor volume per minute must be provided. The panels were illuminated from one side by direct sunlight, and the panels are placed vertically or inclined toward the sun. Light/dark cycles were short in flat panel reactors, and this is probably the key factor leading to high photochemical efficiency. A disadvantage of flat panel reactors systems is that the power consumption of aeration (or mixing with another gas) is high, although mixing is always necessary in any reactor. The large-scale flat plate reactor consists of a rectangular airlift photobioreactor with a large number of light — redistributing plates fixed a few centimeters from each other. Mixing was provided by air injected between adjacent plates, and the culture liquid rises in between.

Tubular photobioreactors consist of long transparent tubes with diameters ranging from 3 to 6 cm and lengths ranging from 10 to 100 m (Janssen, 2002). The culture liquid is pumped through these tubes by means of mechanical or airlift pumps. The tubes can be positioned in many different ways: in a horizontal plane as straight tubes with a small or large number of U-bends; vertical, coiled as a cylinder or a cone; in a vertical plane, positioned in a fence-like structure using U-bends or connected by manifolds; or horizontal or inclined, parallel tubes connected by manifolds. In addition, horizontal tubes can be placed on different reflective surfaces with a certain distance between the tubes. Although tubular reactor design is rather diverse, the predominant effect of the specific designs on the light regime is a difference in the photon flux density incident on the reactor surface. The shape of the light gradient in the tubes is similar in most designs. Also with respect to liquid mixing, the circumstances in most de­signs are similar. The length of the tubes is limited because of accumulation of gas, though this might not be so important for nitrogenase-based processes, since they may be less inhibited by hydrogen. The way to scale up is to connect a number of tubes via manifolds. Flat panel reactors normally show a high photochemical efficiency or biomass yield on light energy, while biomass density is also high. Tubular bioreactors in theory should show better efficiencies because of the shorter average light/dark cycles.

Although much of the research has been focused on single-stage photobioreactor systems, multistage bioreactors entailing three or even four bioreactors in biohydrogen production have also been examined (see Figure 9.2) (U. S. DOE, 2007; Wang et al., 2011; Show et al., 2011). Sunlight is first filtered through first-stage direct photolysis, in which visible light is utilized by blue-green algae, and the unfiltered infrared ray is used by photosynthetic mi­crobes in the second-stage photofermentative reactor. The effluent from the second-stage photofermentation, together with the biomass feedstock, is fed into a third-stage dark fermen­tation reactor, where the microorganisms convert the substrate into hydrogen and organic acids. As the effluent is enriched with organic acids, a supply of external organic acids for the photofermentative process can be eliminated. The fourth stage involves the use of a microbial electrolysis cell to convert the organic acids generated from the dark fermentation into hydrogen in a light-independent process. This stage thus can be operated during the night or in low-light conditions.

The increasing attention on hythane has led to research interest in hydrogen production by dark fermentation of biomass in hybrid or multistage bioreactors. Hythane, a mixture of hydrogen and methane, is a highly efficient and ultraclean-burning alternative fuel that is probably the most promising biogas for industrial applications (Cavinato et al., 2009). How­ever, there are issues to be addressed before multistage bioreactors can be put to practical applications. Integration of multiple biochemical conversion processes poses significant chal­lenges for multistage reactor engineering, system design, process control, operation, and maintenance. Major challenges with the simultaneous production of hydrogen and oxygen from photolytic hydrogen production include respiration-to-photosynthetic-capacity ratio, co-culture balance, and concentration and processing of cell biomass (Holladay et al., 2009).