The accurate determination of lignin Mw is an impor­tant aspect of lignin characterization. The lignin Mw is

typically analyzed by a combination of size-exclusion HPLC and a detection method (refractive index, evapora­tive light scattering, or others) using calibrants which mimic the typical Mws and chemistry of lignins. Howev­er, lignin Mw determination still remains a challenge. For instance, a recent joined study by seven major European lignin research groups reported a reasonably small (<15%) intralab variability in the determination of lignin Mw. However, the interlab differences were very large, up to 49 times(!) difference even under standardized con­ditions (Baumberger et al., 2007).

There are two classes of hydrogenases that commonly present in phototrophic organisms: the [Fe—Fe]-

hydrogenase and the [Ni—Fe]-hydrogenase. The [Fe—Fe]- hydrogenase is found in green algae and some bacteria, and it is the most active H2-forming enzyme. It demon­strates about 100 times higher activity than the [Ni—Fe]- hydrogenase. However, it is irreversibly inactivated when exposed to O2 (see Section Green Algal [Fe—Fe]-Hy — drogenases in this chapter). Cyanobacteria possess two types of [Ni—Fe]-hydrogenases (Houchins, 1984), which are more tolerant to O2 and only temporarily inactivated upon exposure to O2.

Uptake Hydrogenases

In cyanobacteria, the uptake hydrogenases (encoded by hupSL genes) catalyze the consumption of H2 pro­duced by the nitrogenase. Thus, the net H2 evolution by N2-fixing cyanobacteria is barely observed under nat­ural conditions. Uptake hydrogenase has been found in all N2-fixing cyanobacteria studied so far. Nevertheless, a few N2-fixing Synechococcus strains lacking an uptake hydrogenase have been reported (Ludwig et al., 2006; Steunou et al., 2008). It is believed that the uptake hydrogenase transfers electrons from H2 back to the photosynthetic and respiratory electron transport chains, and thus partially regains the energy used for N2 fixation. The cellular/subcellular localization of the uptake hydrogenase is controversial and seems to be species specific. The data obtained from the N2-fixing filamentous nonheterocystous cyanobaterium, Lyngbya majuscule, revealed higher specific labeling associated with the thylakoid membranes, suggesting that the cya — nobacterial uptake hydrogenase is a membrane-bound protein (Seabra et al., 2009). However, it lacks a membrane-spanning region. Therefore, the presence of the third subunit, which would anchor the uptake hy — drogenase to the membrane and link electron transfer from the enzyme to the respiratory or photosynthetic chains, has been suggested (Tamagnini et al., 2007).

In some heterocystous cyanobacteria, such as Anabaena PCC 7120, the uptake hydrogenase enzyme was detected only in heterocysts, while in other cyano­bacteria, such as Nostoc punctiforme, it is localized in both vegetative and heterocyst cells, corresponding most probably to inactive and active pools of the enzyme (Camsund et al., 2011; Seabra et al., 2009).

Since the uptake hydrogenase is an obstacle for H2 production, mutations disrupting the structural hupSL genes have been constructed to improve the H2 produc­tion in N2-fixing cyanobacteria (Happe et al., 2000; Lindberg et al., 2002; Masukawa et al., 2002; Schutz et al., 2004; Yoshino et al., 2007; Khetkorn et al., 2012). These mutants produced about four — to sevenfold more H2 than the control strain. In addition, inactivation of uptake hydrogenase had no major effect on cell growth and heterocyst differentiation. Quantitative shotgun proteomics and physiological approaches on the uptake hydrogenase mutant of N. punctiforme demonstrated that the mutant strain undergoes meta­bolic and structural alterations to compensate for the amount of electrons lost as a release of H2 (Ekman et al., 2011). Construction of mutant strains combining several improvements is likely to be a better approach toward sustainable H2 production. To this end, the single — and the double-mutant strains lacking the homocitrate synthase genes, nifVl and nifV2, were con­structed using the DhupL strain of Anabaena PCC 7120 as the parental strain (Masukawa et al., 2007). The catalytic Mo—Fe center binds homocitrate, which is necessary for N2 fixation, but in the absence of homocitrate gene Mo—Fe center binds citrate: in a Klebsiella mutant this was shown to demonstrate low N2 fixation but high H2 production activity in a N2 atmosphere (Mayer et al., 2002). In line with this result, the DhupLDnifV1 cells also demonstrated high H2 production rate and heterocyst frequency compared to the parental DhupL in N2 atmosphere (Masukawa et al., 2007).

It is nowadays widely recognized that successful implementation of industrial PHA production will only be achieved through satisfaction of sustainability aspects coupled with production of biodegradable poly­mers with desirable properties. Sustainability aspects include cost-competitiveness, environmental benign­ness and production of biodegradable polymers that serve certain market and societal needs. Additional ad­vantages will be provided through the ability to produce PHAs with adjustable properties that could be used in different end-uses by simple modification of fermenta­tion conditions. For instance, the production of different types of PHAs that could be used for both commodity

(e. g. food packaging) and specialty (e. g. scaffolds for tis­sue engineering applications) end-uses by simple modi­fication of fermentation parameters could provide process flexibility.

An important innovation on future PHA-based pro­cesses will be the creation of cascade processing schemes in order to increase resource efficiency (Anonymous, 2012b). Cascade processing is based on the reutilization of packaging material after its use (also called postcon­sumer plastics) for other commercial purposes. For instance, hydrolysis into monomers could create value — added platform molecules for the chemical industry. In addition, bioplastics could be used as replacements for coal and heating fuel due to their high calorific value (Anonymous, 2012b). Reutilization of PHA-based pack­aging materials is strongly dependent on the develop­ment of suitable recycling technologies.

Despite their significant advantages, industrial pro­duction of PHAs is hindered by high production cost.

Previous attempts to produce PHAs in large scale had to rely on conventional fermentation technologies that cannot compete with low-cost petroleum-derived plas­tics. As mentioned earlier, raw material supply is one of the most important factors that should be optimized in order to reduce processing costs. For this reason, recent research focuses on the utilization of low-cost feedstock for PHA production (e. g. molasses, crude glycerol, whey, animal fats, and waste cooking oils among others) aiming to substitute for conventional and expensive carbon sources. Table 24.2 presents the re­sults regarding PHA production from various waste and by-product streams. However, even if waste or by­product streams are used as fermentation feedstocks, aerobic cultivation for PHA production in industrial scale operations is still an expensive unit operation. For this reason, integration of PHA production into existing industrial plants or the development of new in­dustrial plants for PHA production should be combined

with the production of value-added co-products. This can be achieved through fractionation of agricultural re­sources or by-product and waste streams from existing industrial processes.

PHA production cost increases further due to down­stream separation and purification of PHAs from resid­ual microbial mass. Several methods have been reported for the recovery of PHAs based on the utilization of organic solvents such as acetone, chloroform or dichlo — roethane. However, these methods are unfavorable for large-scale production since solvents increase opera­tional cost and additional equipment for solvent recov­ery is often needed. Alternative extraction methods have been also proposed including enzymatic lysis of re­sidual microbial mass (Kapritchkoff et al., 2006; Verlin — den et al., 2007), supercritical fluid extraction (Hejazi et al., 2003), mechanical disruption of bacterial cells coupled with chemical treatment, autolysis of bacterial cells, and chemical treatment under acidic or alkaline conditions (Yu and Chen, 2006; Verlinden et al., 2007).

Several studies have also focused on the estimation of PHA production costs from different feedstocks (Choi and Lee, 1997; van Wegen et al., 1998; Posada et al.,

2011) . However, there are limited studies on the evalua­tion of integrated biorefineries focusing on the fraction­ation of the initial raw material combining the production of PHAs with the extraction or production of value-added co-products. In addition, future costing studies should also focus on the evaluation of the poten­tial to integrate PHA production in existing industries.

In recent years, several studies focused on the pro­duction of PHAs from low-cost renewable resources (Akaraonye et al., 2010; Koller et al., 2010; Du et al.,

2012) . This study focuses on the presentation of repre­sentative biorefinery concepts targeting the production of PHAs and other value-added products. In particular, PHA production could be combined with biofuel and bioenergy production.

Hydrolysis is the process by which the lignocellulose polymers are reduced (saccharified) to yield fermentable sugars (hexoses and pentoses) (Harris and DeBolt, 2010). There are two methods of hydrolysis used within the bioenergy and biorefining processes, namely, acid hy­drolysis and enzymatic hydrolysis (Potumarthi et al., 2013, 2012; Dashtban et al., 2009; Ong, 2004).

Acid hydrolysis is the older method of the two and has been implemented on an industrial scale since World War I. In this particular process, dilute or concen­trated acid, normally H2SO4 as it is cheapest, is used to hydrolyze the cellulose with the reaction temperatures dependent upon the molarity; dilute acids require higher temperatures (above 200 ° C) while concentrated acids require lower temperatures. The acid hydrolysis approaches are less attractive due to low yields with dilute acid and the recovery and environmental factors involved with use of concentrated acids (Ong, 2004).

In enzymatic hydrolysis, the lignocellulose is broken down into the corresponding monomeric sugars by spe­cific enzymes produced from bacteria or fungi ( Coyne et al., 2013; Gupta et al., 2013; Ong, 2004; Dashtban et al., 2009). This approach is more complex, expensive and time consuming, in comparison to the acid hydroly­sis approach, but has the advantage of little or no by­products to dispose of at the end of the biorefining process (Ong, 2004) and it can be used for more selective fractionation in a biorefinery context (Menon and Rao, 2012). Pretreatment of lignocelluloses with acid or alkali partially removes the lignin and hemicellulose but also substantially disrupts the fibrillar structure of biomass. Therefore, acid or alkali pretreated lignocellulosic biomass can be saccharified enzymatically to produce fermentable sugars. This results in faster hydrolysis rates and higher glucan enzymatic digestibility. A common belief is that lignin removal in particular promotes faster and more efficient enzymatic cellulose hydrolysis (Zhu et al., 2008).

Pretreatment of lignocellulosic biomass with the white-rot fungi increases biodegradability and leads to high-quality ruminant feed. For example, white-rot fungi-treated cedar wood shows significant improve­ment for rumen digestibility (Okano et al., 2005). When high-lignin forages such as grass, oat straw and alfalfa stems were treated with various white-rot fungi, substantial improvements in digestibilities have also been obtained (Akin et al., 1995,1993; Jung et al., 1992).

White-Rot Fungus Pretreatment in Biological Pulping

White-rot fungi have also been used in biological pulp­ing (biopulping) to reduce the utilization of chemicals in the pulping industry and decrease the environmental hazard caused by the traditional pulping process (Singh et al., 2010). Biopulping process removes not only lignin and hemicellulose but also some of the wood extractives. It can also improve paper quality and significantly reduce the electrical energy and cooking time required for pulp­ing wood chips (Ali and Sreekrishnan, 2001; Hunt et al., 2004; Singh et al., 2010). When C. subvermispora was used for biopulping of agricultural residues including rice, wheat and barley straw samples, the tensile strength and burst factor of hand sheets produced from the bio­pulping process improved significantly compared to the chemical process (Yaghoubi et al., 2008). Blanchette et al. (Blanchette et al., 1992) evaluated the potential application in biopulping of 19 strains of P. chrysosporium and 9 strains of C. subvermispora. For the P. chrysosporium isolates, only a few strains preferentially removed large amounts of lignin from wood while the majority of the isolates removed all cell wall components nonselectively. In contrast, all nine isolates of C. subvermispora led to moderate weight losses and preferential degradation of lignin in aspen, birch and loblolly pine wood.

White-Rot Fungus Pretreatment of Biomass for Biofiber

Microbial pretreatment can also improve the feature of the fiber in biomass for biocomposite production. For example, corn stalk pretreated with the white-rot fungus Trametes hirsuta has been used to produce fiberboard by hot pressing without adhesive. The corn stalk-based fiberboard made of the pretreated biomass has an in­crease of 3.40- and 8.87-fold in moduli of rupture and elasticity, respectively, over the fiberboard made from untreated corn stalk. Further analyses showed that the increase in the mechanical properties of the fiberboard resulted from the pretreated biomass possessing more than twice the number of hydroxyl groups, an 18% higher crystallinity, and twice the polysaccharide content of untreated corn stalk (Wu et al., 2011).

Brown-Rot Fungi

Brown-rot fungi are Basidiomycete fungi that, unlike white-rot fungi, selectively modify and then completely hydrolyze lignocellulose polysaccharides, typically without secreting an exoacting glucanase and without removing lignin (Schilling et al., 2009; Tewalt and Schil­ling, 2010). The wood decay resulting from the action of brown-rot fungi leads to an increased volume of pores in the wood cell wall and decreased degree of polymeriza­tion of holocellulose along with a dramatic weight loss (Flournoy et al., 1991). Depolymerization of holocellu — lose occurs rapidly during the early decay process lead­ing to an extensive degradation of holocellulose in wood (Blanchette, 1995; Irbe et al., 2011; Kumar et al., 2009) and as high as 75% wood strength loss even when only 1% weight loss has occurred (Green and Highley, 1997; Richards, 1954; Wilcox, 1978).

The exact mechanism for brown-rot decay is still un­clear. For the selective removal of polysaccharides, a two-step procedure has been proposed: a nonenzymatic radical-based modification of the wood cell wall through small molecules, followed by secretion of enzymes to catalyze the breakdown of polysaccharides into their sugar monomers (Green and Highley, 1997; Tewalt and Schilling, 2010). However, cellulose and hemicellulose removal by brown-rot fungi does not open up cell walls to facilitate enzyme penetration (Flournoy et al., 1991). Primarily because enzymes are too large to penetrate the decayed wood, attack by cellulolytic enzymes may only be limited to a localized, superficial area (Baldrian and Valaskova, 2008; Flournoy et al., 1991). It has been proposed that Fenton’s reagents and not enzymes are responsible for rapid wood decomposition early in brown-rot decay (Green and Highley, 1997; Jensen et al., 2001; Ray et al., 2010). Other study results also support that hydroxyl radicals (HO’) generated through Fenton chemistry (H2O2—Fe(II)) initiate lignocellulose breakdown (Arantes et al., 2012; Contreras et al., 2007; Hammel et al., 2002; Kaneko et al., 2005; Kramer et al., 2004; Suzuki et al., 2006). Consequently, this suggests that reactive oxygen species play an important role in the early stages of wood degradation by brown-rot fungi (Irbe et al., 2011). In brown-rot wood decay, hemi — cellulose is removed considerably faster than cellulose (Curling et al., 2002; Highley, 1987; Monrroy et al., 2011). Consistently, the total secretome hemicellulase expression and activity for brown-rot fungi peak prior to cellulase activity (Lyr, 1960; Martinez et al., 2009).

Hemicellulose is embedded in cellulose microfibrils and its prior removal may facilitate cellulose degrada­tion and removal (Green and Highley, 1997). Continual degradation of holocellulose by brown-rot fungi leads to gradually increased weight loss but the percent crys­tallinity in decayed wood increases apparently at an early stage, peaks between 2 and 4 weeks and then de­creases implying structural changes of cellulose chains during fungal attack (Howell et al., 2009). Towards the end of brown-rot decay, nearly 100% of carbohydrates can be removed; however, most of the lignin remains (Eriksson et al., 1990). Only a small fraction of the lignin is oxidized, demethylated and depolymerized, often leading to lignin-derived volatile components (Ewen et al., 2004; Irbe et al., 2011; Schilling et al., 2012).

Recently, the potential application of brown-rot fungi for the pretreatment of biomass to increase downstream enzymatic hydrolysis has been explored. When spruce and pine woods were treated with one of two brown — rot fungi, Gloeophyllum trabeum or Fomitopsis pinicola, saccharification efficiency was increased significantly even though total sugar yield was low, probably due to low enzyme loading (Schilling et al., 2009). In another effort, G. trabeum-treated pine wood block only led to a maximum 22% glucose release even though 60 FPU Cel — luclast was loaded, suggesting brown-rot fungus G. tra — beum modification of pine wood may not be sufficient to increase cellulose accessibility (Tewalt and Schilling,

2010) . Similarly, when the brown-rot fungi G. trabeum and Laetoporeus sulphureus were used for the pretreat­ment of the wood Pinus radiate and Eucalyptus globules, the highest glucose yield was 14% after 8 weeks of biodegradation (Monrroy et al., 2011). On the other hand, when G. trabeum was used to pretreat different biomass including aspen, spruce, or corn stover, sugar yield was significantly increased up to threefold. In the best case, a 2-week pretreatment of aspen by G. trabeum led to a 72% cellulose-to-glucose yield corresponding to 51% yield relative to original glucan. For corn stover, a weak colonization with minor degradation by another tested brown-rot fungus, Postia placenta, resulted in more than a twofold increase in sugar yield (Schilling et al., 2012). Similar to wood biomass, when corn stover is pretreated with the brown-rot fungus Fomitopsis sp. IMER2, the amorphous regions of the cellulose are pref­erentially degraded in contrast to the significant lignin degradation by the white-rot fungus I. lacteus CD2 (Zeng et al., 2011). In another successful case, simple pre­treatment of Scots pine (Pinus sylvestris) with the brown rot fungus Coniophora puteana for 15 days permitted re­covery of greater than 70% of the glucose present in the biomass, with a total wood mass loss of 9%, suggest­ing great potential for use of this specific group of fungi in lignocellulosic biomass pretreatment (Ray et al., 2010). Brown-rot fungi therefore hold significant potential for practical application in biological pretreatment.

Electrode Reactions in MFC

A typical MFC reactor contains an anodic chamber, a cathodic chamber and a PEM partitioning the two cham­bers. Figure 9.1 shows a dual-chamber MFC.

Anode Reaction

The microbes in the anodic chamber oxidize sub­strates such as glucose, acetate and some refractory organics. For example, glucose is oxidized as follows to generate electrons, protons and carbon dioxide (Pham et al., 2006):

C6H12O6 + 6H2O/6CO2 + 24H+ + 24e~ (9.1)

Because electrons cannot "swim" in an aqueous solu­tion, the oxidation reaction must occur in a biofilm that is capable of transferring electrons to the anode. In the absence of a suitable oxidant in the anodic chamber to absorb the electrons, electrons will be transferred to the anode by the biofilm (Zhao et al., 2009). The electrons reach the cathode via an external circuit linking the anode and the cathode, where they are used to reduce an oxidant such as oxygen (Figure 9.1). A load is placed on the external circuit to harvest the electricity. To main­tain electroneutrality, protons must carry an equal amount of positive charges from the anodic chamber to the cathodic chamber usually through a PEM. Inefficient proton migration will result in accumulation of protons that causes acidity in the anodic chamber (Xu et al., 2012).

In the anodic chamber, anaerobic conditions are very important to guarantee the substrate oxidation by the microbes through anaerobic respiration (Liu et al., 2005b; Logan et al., 2006). Oxygen leaked into the anodic chamber from outside air or through diffusion from the cathodic chamber (Figure 9.1) would reduce Coulombic efficiency of the MFC by directly oxidizing the organic matter in the anodic chamber. In this case, energy will be released as low-grade heat instead of electricity. A PEM plays an important role of preventing oxygen diffusion from the cathodic chamber to the anodic cham­ber (Li et al., 2011), while allowing positive charges to go through it via a proton exchange process. If nonoxygen oxidants such as sulfate and nitrate are present in suffi­cient quantities in the anodic chamber feed stream, the biofilm on the anode must not be able to catalyze their reduction because it would divert the electrons released from oxidation of organic matters for the local reduction of sulfate or nitrate. A buffer solution that usually con­tains NH4Q, NaH2PO4, Na2HPO4, KCl, and so on is often used to enhance the proton transfer in laboratory MFC investigations (Liu et al., 2011). The presence of a buffer solution increases the conductivity, thus reducing internal resistance of the MFC (Liu et al., 2005a).

Cathode Reaction

The cathode reaction has a major impact on MFC performance. The electrons coming from the anode via the external circuit, the protons coming from the anodic chamber via the PEM and the electron acceptors (e. g. O2) will react with the help of catalysts on the cathode (Pham et al., 2006):

24H+ + 24e~ + 6O2 / 12H2O (9.2)

Reactions (9.1) and (9.2) form a thermodynamically favorable redox reaction, that is, the aerobic oxidation of glucose. However, a thermodynamically favorable reaction may not proceed at an appreciable rate if the kinetics is too slow. In an MFC, anode and cathode reac­tions almost always require catalysis. For the anode reaction, a biofilm is required to catalyze organic carbon oxidation and electron transfer. For the cathode, oxygen reduction rate is very slow without catalysis. The cathodic reaction efficiency depends on the concentra­tion and type of electron acceptors, proton concentra­tion, electrode structure and its catalytic ability (Zhou et al., 2012).

In order to improve electricity generation, a good cat­alytic cathode is crucial since the catalysts can reduce the activation energy and thus greatly increase the reaction rate. Currently, for oxygen reduction, platinum (Pt) appears to be most effective. However, it is extremely expensive and, thus, unrealistic for most practical appli­cations even when only Pt coating is used. Some alterna­tive catalysts have been explored such as MnOx, CoTMPP, PbO2, iron(II) phthalocyanine (FePc) and recently the biocathode (Roche and Scott, 2008; Zhou et al., 2011).

Oxygen is the most popular acceptor because of its high standard potential (0.818 mV), low cost and envi­ronmental "friendliness". However, the rate of oxygen reduction is very low on the cathode surface, resulting in a high overpotential, which is one of the most impor­tant limiting factors in MFCs (Gil et al., 2003). Potassium ferricyanide (K3[Fe(CN)6]) can overcome this handicap (Logan et al., 2006; Nevin et al., 2008; Park and Zeikus,

2003) . However, the regeneration of K3[Fe(CN)6]is a problem because it usually is not sufficiently oxidized by oxygen. It needs to be replenished periodically (Franks and Nevin, 2010). In addition, K3[Fe(CN)g] can diffuse into the anodic chamber through the PEM, thereby influencing the desired anaerobic conditions of anodic chamber (Logan et al., 2006). Potassium perman­ganate is also used as an acceptor, and the power density was reported to be higher than that with K3[Fe(CN)g] and oxygen (You et al., 2006). In practice, wastewater streams are low-grade energy sources that are pale in comparison to pure fuels such as hydrogen or ethanol as a fuel. This inherently means that a large volume of water must be treated to harvest a sufficient amount of electricity. This makes all externally added soluble catalysts impractical, limiting them to academic investigations.

To overcome the requirement for catalysis by oxygen oxidation on the cathode, biocathodes have been explored (Biocathodes Section). Various biofilms have been tested on cathodes to biocatalyze oxygen or a nonoxygen oxidant such as nitrate and perchlorate (Shea et al., 2008; Srikanth et al., 2012; Zhang and Angelidaki, 2012).

Biolipids can be derived from plant, animal, oleagi­nous microorganisms and algal sources. The composi­tion of biolipids derived from each of these sources differs greatly and has varying degrees of suitability to the biofuel production industry. The major lipids produced from each of these sources are listed below and the degree of suitability to the production of biofuel production is discussed.

Plant-Derived Biolipids

In 2007, 95% of world biodiesel was produced via edible plant oils, which were supplied by the agricul­tural industry, with the vast majority supplied by rapeseed oil, 84% (Food and Agriculture Organization,

2008) . Overall, plant lipids are divided into three major categories: edible, nonedible and waste vegetable oils described below.

Edible Lipids

The main edible oils used for biofuel production are rapeseed, palm and soy bean oils. Edible oils have the disadvantage of competing directly with food produc­tion. The use of edible oils for the production of biodiesel competes directly with the use of land for the produc­tion of food and without proper planning results in reduced food production (Gui et al., 2008). However, the productivity from edible oils is high in terms of oil yield and the quality of the resulting biofuel. The oil yield from palm is the highest of the commonly grown edible oil crops at 5 tons per hectare while rapeseed produces 1 ton per hectare and soy bean 0.52 tons per hectare. A high lipid yield is vital for the economical production of biofuel from these plants. Although the productivity from palm oil is particularly high its use as a biofuel is limited as it is the world’s most commonly used edible lipid and thus competition for the oil between the food and biofuel industry would result in an increase in the price of this oil (Lam et al., 2009). In terms of the suitability for biofuel, palm oil has a high degree of saturation and thus is not the most suitable for biofuel production with the resulting fuel having poor cold flow properties. However, the cold flow prop­erties of a lipid can be altered by the use of cold filtration (Kerschbaum et al., 2008) or alternatively the use of alco­hols such as ethanol, isopropanol or isobutanol, which results in the production of fatty alkyl esters with lower freezing points and therefore improved cold flow prop­erties (Dunn, 2009). There are also some environmental and ecological concerns surrounding palm oil produc­tion, with the clearing of rain forests to make way for palm plantations. The plantation costs of edible oil crops are relatively low with the exception of palm oil, which has a higher cost; however, this is offset by the high oil yield from the crop. The overall estimated energy balance of rapeseed and soybean is similar at 3.7 and 3.4, respectively, while palm oil is significantly higher at 9.6 due to the high yields (Food and Agriculture Organization, 2008). Currently rapeseed oil is the most commonly used plant oil used in biodiesel production because it makes an excellent biofuel with excellent cold flow properties. The main disadvantage of using rapeseed oil is the growth of rapes is difficult and unsus­tainable as it must be part of a one in five rotation due to the large quantity of nutrients required for the growth of the organism and the buildup of pathogens and disease in the environment targeting rapeseed if grown annually.

Bioresource use in the forms of new and waste biomass is a great opportunity and a challenge for the future since it offers the chance of replacing fossil fuels for the produc­tion of energy carriers, materials and specialty chemicals and diminishing the market pressure in an almost carbon-neutral way. Industrial biorefineries are seen as one of the most promising directions toward a sustainable bio-based economy. Fully developed biorefineries combine biological and physicochemical processes.

A weakness of biorefineries as an alternative to con­ventional oil refineries consists in the fact that the former is based on biofeedstock, which can require an intensive cultivation and land use.

Moreover, biorefineries could compete with food re­quirements and needs, which would limit the land allo­cated to biomass for biorefineries. As a result, the future of biorefineries should consider the use of nonedible biomass and the advanced processing of biomass waste, as well as land which could not normally be used for agriculture. This type of land could be used for microal­gae cultures or renewable plants. Other sources of raw material for biorefineries could be found on waste from the food industry and urban organic waste. The processing of this raw matter can be successfully and eco-efficiently carried out through the development of enzymatic systems and engineered microorganisms capable of separating useful compounds from waste.

The development of these technologies should also consider the important issue of costs, since, currently, oil-based refineries offer more cost-effective solutions at the expense of environmental degradation and pollution.

Acknowledgments

This work was partially supported by the grant of the Romanian National Authority for Scientific Research, CNCS—UEFISCDI, project number PN-n-ro-PCE-2011-3-0559, Contract 265/2011.

Supercritical fluids (SCFs; conditions where the sol­vent is both above the critical temperature and critical pressure of the chemical) show unique properties that are different from those of either gases or liquids under standard conditions. SCFs have liquidlike densities and gaslike transport properties of diffusivity and viscosity. So, SCFs have the ability to penetrate the crystalline structure of lignocellulosic biomass overcoming the mass transfer limitations encountered with other pre­treatments. Another important advantage is the fact that SCFs have tunable properties such as partition coef­ficients and solubility. Small changes in temperature or pressure close to critical point can result in up to 100-fold changes in solubility, which can simplify sepa­ration. Supercritical carbon dioxide (CO2) with a critical temperature (Tc) of 31 °C and a critical pressure (Pc) of

7.4 MPa, as well as supercritical water has been used for biomass pretreatment. REAC fuels and Renmatix are examples of companies employing this kind of tech­nology (Table 17.6).

Other technologies such as gamma rays, ozonolysis, biological pretreatment (mainly with fungi) are still in an earlier phase and currently face challenges in scaling up and commercialization (Agbor et al., 2011; Alvira et al., 2010).

Summary of Lignocellulosic Biomass Pretreatments

Recently technoeconomic comparisons of some of the different pretreatment technologies have been done using identical feedstocks, and analytical methods to generate comparable data (Wyman et al., 2005, 2011; Eggeman and Elander, 2005). The results indicated that no clear winning pretreatment technology could be identified and that further optimization potential is available in the pretreatment methods. It is also clear that the optimal pretreatment technology is very much substrate dependent further hampering the surfacing of a predominant technology (Table 17.7). The effect of pH on solubilization of the different lignocellulosic

components was nicely illustrated by Garlock et al.

(2011) as depicted in Figure 17.1. Table 17.6 summarizes the effect of various pretreatment methods on the chemi­cal composition and chemical/physical structure of lignocellulosic biomass. It can be concluded that at the moment there is no clearly winning technology also because each subsequent conversion process (e. g. fermentative, chemocatalytic) has its own set of require­ments. Therefore, a wide range of technologies are currently in the progress of being scaled-up. In Table 17.7 an overview of currently worldwide devel­oped demonstration and pilot plant facilities is pre­sented for production of bioethanol and other chemicals.

LIGNOCELLULOSIC
BIOREFINERIES—CLASSIFICATION

Biorefineries can be classified on the basis of a num­ber of their key characteristics. Major feedstocks include perennial grasses, starch crops (e. g. wheat and maize), sugar crops (e. g. beet and cane), lignocel — lulosic crops (e. g. managed forest, short rotation coppice, and switchgrass), lignocellulosic residues (e. g. stover and straw), oil crops (e. g. palm and oilseed rape), aquatic biomass (e. g. algae and seaweeds), and organic residues (e. g. industrial, commercial and post­consumer waste).

These feedstocks can be processed to a range of bio­refinery streams termed platforms. These platforms include single carbon molecules such as biogas and syngas, five — and six-carbon carbohydrates from starch, sucrose or cellulose; a mixed five — and six-carbon carbo-hydrates stream derived from hemicelluloses, lignin, oils (plant-based or algal); organic solutions from grasses; and pyrolytic liquids. These primary plat­forms can be converted to a wide range of marketable products using combinations of thermal, biological and chemical processes (Table 17.8).

Knowledge of a biorefinery’s feedstock, platform and product allows it to be classified in a systematic manner (Cherubini et al., 2009). The classification of biorefineries enables the comparisons of biorefinery systems, im­proves the understanding of global biorefinery develop­ment and allows the identification of technology gaps.

FIGURE 17.1 Cell wall model showing the general effect of pH on solu­bilization of hemicellulose and lignin. (A) Untreated cell wall and (B) cell wall during pretreatment. Cellulose can also be degraded under extremely acidic condi­tions; however, that is not portrayed in this diagram. Source: Designed by Garlock et al., 2011 based on figures from Mosier et al., 2005 and Pedersen and Meyer, 2010. (For color version of this figure, the reader is referred to the online version of this book.)

TABLE 17.8 Biomass-Derived Chemical Building Blocks

Derived Chemical Building cont’d

An overview of current feedstocks, platforms and prod­ucts is given in Figure 17.2.

C6 AND C6/C5 SUGAR PLATFORM

Six-carbon sugar platforms can be accessed from su­crose or through the hydrolysis of starch or cellulose to give glucose. Glucose serves as feedstock for (biological) fermentation processes providing access to a variety of important chemical building blocks. Glucose can also be converted by chemical processing to useful chemical building blocks.

Mixed six — and five-carbon platforms are produced from the hydrolysis of hemicelluloses. The fermentation of these carbohydrate streams can in theory produce the same products as six-carbon sugar streams; however, technical, biological and economic barriers need to be overcome before these opportunities can be exploited. Chemical manipulation of these streams can provide a range of useful molecules.

Fermentation Products

The number of chemical building blocks accessible through fermentation is considerable. Fermentation has been used extensively by the chemical industry to produce a number of products with chemical produc­tion through fermentation starting around the turn of the twentieth century. Around 8 million tons of fermen­tation products are currently produced annually (Bakker et al., 2010).

• Fermentation-derived fine chemicals are largely manufactured from starch and sugar (wheat, corn, sugarcane, etc.)

• The global market for fermentation-derived fine chemicals in 2009 was $16 billion and is forecast to increase to $22 billion by 2013 (Frost and Sullivan,

2011) .

• The market is broken down as follows:

Modern biotechnology is allowing industry to target new and previously abandoned fermentation products and improve the economics of products with commer­cial potential. Coupled with increasing fossil feedstock costs, cost reductions in the production of traditional fermentation products such as ethanol and lactic acid will allow derivative products to capture new or

increased market shares. Improving cost structures will also allow previously abandoned products such as butanol to reenter the market. Many see the future abundant availability of carbohydrates derived from lignocellulosic biomass as the main driver. However, carbohydrate costs are increasing strongly in recent years and its use for nonfood products is under pres­sure even in China. Fermentation also gives the indus­try access to new chemical building blocks previously inaccessible due to cost constraints. The development of cost-effective fermentation processes to succinic, ita — conic and glutamic acids promises the potential for novel chemical development.

Chemical Transformation Products

Six — and five-carbon carbohydrates can undergo selec­tive dehydration, hydrogenation and oxidation reactions to give useful products, such as sorbitol, furfural,
glucaric acid, HMF and levulinic acid. Over 1 million tons of sorbitol are produced per year as a food ingre­dient, personal care ingredient (e. g. toothpaste) and for industrial use (ERRMA, 2011; Vlachos et al., 2010).

In a microbial (or higher) system, proteins are synthe­sized with the help of an information template in the form of an mRNA molecule and the translatory machin­ery of ribosomes and amino acylated tRNA; however, poly(amino acid)s are synthesized enzymatically without the requirement of an information template.

While proteins owe their functionality to the tertiary structure that they attain, the poly(amino acid)s owe it normally to their physical properties, usually the num­ber of repeating units more dominant in a pool of poly (amino acid)s. Another striking difference between pro­teins and poly(amino acid)s is the fact that one type of a protein will contain an exact number of amino acids, while poly(amino acid)s display wide polydispersity, i. e. in a single organism, the size of the same poly (amino acid) will vary. The fact that these polymers are biocompatible with human physiology and for applica­tions other than pharmacology, the polymer is biode­gradable, makes it an attractive alternative to the widely used polymers obtained through the petrochem­ical route, which are often nonbiodegradable, and end up accumulating in the environment. These poly(amino acid)s have been shown to be useful in multitude of applications, like controlled drug release, preparation of bioplastics, use as antimicrobial additive in food, as well as superabsorbers (replacement of polyacrylate gels) (Obst and Steinbuchel, 2004). They may also be used as a viable source of dipeptide neutraceuticals (Sallam and Steinbuchel, 2010).

There are three identified poly(amino acid)s so far that have been reported to be produced from microbial source; they are cyanophycin, poly-g-glutamate (PGA)
and e-poly lysine (Figure 19.3). In poly(glutamic acid) the amide linkages are formed on the a-amino group to then g-carboxyl group in the polymer backbone, whereas in poly(lysine), the a-carboxyl group is linked to the є-amino group of lysine. In the case of cyanophy­cin, almost equivocal amounts of arginine and aspartic acid are arranged as a polyaspartate backbone, with arginine moieties linked to the b-carboxyl group of almost every aspartic acid residue.

Cyanophycin/Cyanophycin Granule Polypeptide

Cyanophycin is the ideal nitrogen storing molecule, because every repeating unit has about five atoms of ni­trogen, and it is insoluble at physiological conditions in­side the cell protoplasm. Due to its insolubility, it does not cause detrimental shifts in the osmolarity of the cell, hence helps in cell survival (Oppermann-Sanio and SteinbUchel, 2002). Cyanobacteria normally pro­duce cyanophycin when the organism senses a decrease in sulfur, phosphate and significantly by the decrease in nitrogen concentration in the surrounding milieu (Lin et al., 2012). Apart from cyanobacteria, cyanophycin granule polypeptide (CGP) has been found in some strains of Synechococcus sp. (Hai et al., 1999).

FIGURE 19.3 Amino compounds from microbial sources. (For color version of this figure, the reader is referred to the online version of this book.)

Production of Cyanophycin

Cyanophycin is enzymatically produced by the action of cyanophicin synthases on small primers of cyanophy­cin. Because of the slow growth of cyanobacteria in pho­tobioreactors, large-scale production of CGP is hindered by the lower cell densities and thus used to get only low yield of CGP with respect to the cell dry mass (CDM). To solve this problem cyanophycin synthase gene or CphA gene has been identified and cloned into a wide range of organisms, from E. coli to eukaryotic microbes like S. cer — evisiae (Steinle et al., 2008), commercially important strains like Ralstonia eutropha, C. glutamicum, Pseudo­monas spp., etc. (Aboulmagd et al., 2001) and even higher plants like potato and tobacco (Neumann et al., 2005). Recombinant CGP produced were shorter in length (21—35 repeating units), along with a small per­centage of arginine replaced by incorporated lysine. Till some time, the highest CGP production was attained by using Acinetobacter calcoaceticus, with a value of 48% CDM, with the addition of exogenous arginine, along with the addition of other carbon and nitrogen sources (Elbahoul et al., 2005). The economic viability of recom­binant strains is always a problem for large—scale usage, due to the sheer amount of antibiotics that have to be added to the medium. However, the hunt for a commer­cially compatible strain for economical production of CGP led to a method for obtaining high cell densities and high yield of CGP with Ralstonia eutropha. The strain is poly hydroxy butyrate (PHB) negative (the wild type produces PHB) and was devoid of the 2-keto-3-deoxy — phosphogluconate aldolase (eda) gene (Lin et al.,

2012) . The plasmid carrying the cphA gene was con­structed along with the eda gene; for the microbe to sur­vive, the plasmid had to be retained, and in other words, a selective pressure for plasmid retention was achieved, without the use of antibiotic resistance genes, thus making the process economically viable.

In general, the production was optimized with a basic mineral medium, Mineral Salts Medium (MSM), with sufficient supplements of fructose, NH3, K2SO4, MgSO4.7H2O, Fe(III) NH4-citrate, CaCl2.2H2O, and trace elements. A 30 l pilot study gave promising yields of water-insoluble CGP and water-soluble CGP, contrib­uting to 47.5% and 5.8% (w/w) of CDM, and a cell den­sity as high as 57 g/l CDM was obtained (Elbahloul et al., 2005). CGP is normally purified by acid extraction, which involves the solubilization of CGP in acidic solu­tions of pH 1, followed by washes with distilled water, which renders it insoluble again.

Alternative production strategies include the use of molecular farming approach, and expression of the cphA gene in certain specific tissues of selected plants. Potato and tobacco have been successfully transformed with the cphA gene; however, production of CGP within the plant tissues would lead to slow growth and fleshy leaves. Use of the stable cyanophycin synthase for direct synthesis of cyanophycin has been suggested by certain authors as an alternative to the intricately controlled fermentative production of cyanophycin (Hai et al., 2002).

Biodegradability of Cyanophycin

Biodegradation of cyanophycin is observed in all the organisms that naturally produce the polymer, as it serves as a reserve carbon and nitrogen pool. Cyanophy — cin can be depolymerized by intracellular cyanophyci — nase, which also has been isolated from Synechocystis sp. strain PCC6308 (Hai et al., 2002). The cyanophyci — nase gene or the cphE gene has been found to be located downstream of the cphA gene. The cyanophycinase enzyme does not cleave the polymer into arginine and aspartate. Recent studies have shown that cyanophycin is degraded by most of the gut bacteria through the anaerobic route within a time period ranging from 1 day to 7 days (Sallam and Steinbuchel, 2009). The above studies have opened the doors to the use of cyano — phycin directly as nutrient supplement.

Applications for Cyanophycin

Cyanophycin can be hydrolyzed to its constituent amino acids, aspartic acid, and arginine. These amino acids may be utilized directly in food and pharmaceu­tical applications. Cyanophycin can be stripped off of arginine through chemical modifications, so as to pro­duce polyaspartate. Polyaspartate is a polyanionic poly­mer that can be utilized for production of biodegradable surfactants, and can be utilized for applications pertain­ing to polyacrylate (Schwamborn M, 1998). It has been discovered recently that cyanophycin can be degraded by the gut bacteria obtained from a diverse group of or­ganisms ranging from mammals, birds, and fishes. This opens routes for the use of cyanophycin directly as a nutritional substrate, instead of constituent dipeptides or amino acids, which would require additional invest­ments of time and money (Sallam A and Steinbuchel A

2009) . Even though a considerable amount of research has been carried out for the viable production of cyano — phycin, the complete potential for the various bulk chemicals that may be obtained from cyanophycin is not attained yet.