Levoglucosenone (LGO, 1,6-anhydro-3,4-dideoxy-b-D-pyranosen-2-one, or

6,8-dioxabicyclo[3.2.1]oct-2-en-4-one) is a sugar enone product of cellulose, formed from the combined depolymerization and dehydration reactions, with the possible pyrolytic pathways shown in Fig. 4 [66]. Its structure was firstly confirmed

in 1973 [41], and further confirmed and adopted by other researchers. LGO is an optically active organic compound in which all carbon atoms are different environ­ments and which has easily modifiable functional groups. As a result, LGO can be used in the synthesis of various products (such as tetrodotoxin, thiosugar, ras activa­tion inhibitors). The detailed applications of LGO can be found elsewhere [55].

LGO is formed in very low yield from fast pyrolysis of cellulose or biomass, but can be promoted by the addition of some acid catalysts in the pyrolytic process. A mechanism has been proposed for acid-catalyzed decomposition of LG to form the LGO, as shown in Fig. 5 [41].

Various acid catalysts exhibited the capability to promote the LGO formation, such as the MgCl2 and FeCl3 [45], (NH4)2SO4 and (NH4)2HPO4 [21, 69], CrO3 and CrO3+CuSO4 [34], ZnCl2 [22], M/MCM-41(M=Sn, Zr, Ti, Mg, etc.) [91]. However, almost all of these catalysts did not show high selectivity on the LGO, because they catalyzed the formation of several dehydrated products (LGO, LAC, DGP, FF, etc.), rather than LGO alone.

According to a series of studies performed by Dobele et al. [25-27], fast pyrolysis of cellulose/biomass impregnated with phosphoric acid could produce LGO with very high purity. The highest LGO yield reached 34 wt% from microcrystalline cel­lulose impregnated with 2% phosphoric acid, or 17.5 wt% from birch wood impreg­nated with 2.5% phosphoric acid. Other studies also confirmed the promising catalytic selectivity of the phosphoric acid on the LGO production [35, 64, 82]. Furthermore, Dobele et al. [28] reported that the pretreatment of cellulose/biomass with adsorption of Fe2(SO4)3 provided another way to prepare LGO with high purity, but the selectivity of the Fe2(SO4)3 on the LGO was a little lower than the phosphoric acid.

In a recent study, it was reported that fast pyrolysis of pure cellulose followed with catalytic cracking of the pyrolysis vapors with solid super acids (sulfated metal oxides, SO42-/TiO2, SO42-/ZrO2, SO42-/SnO2 , etc.) allowed the production of LGO with the content reaching 40% (peak area% on the GC/MS ion chromatograms) in the pyrolytic products [49]. In fact, when the solid super acids were mechanically mixed with cellulose, fast pyrolysis of the mixture also produced LGO with high purity. Compared with the impregnation of catalysts (phosphoric acid or Fe2(SO4)3) on the cellulose/biomass, the utilization of solid catalysts avoids the complex pre­treatment process, and will offer a significant advantage on catalyst recycles.

Compared with the LG, the LGO can be easily recovered from pyrolytic liquids by distillation, and a detailed purification method was proposed by Marshall [54].

Biocrude is defined as an aqueous carbohydrate solution (oxygenated hydrocarbon) produced from the liquefaction of biomass. Biocrude derived from the direct lique­faction of biomass can be converted to liquid fuel, hydrogen gas, or chemicals. Aqueous phase reforming processes have been successfully utilized for converting the biomass-derived, water-soluble carbohydrates to liquid alkanes and hydrogen [45, 46, 121]. Preliminary studies on the conversion of various biomass types into liquid fuels have indicated that hydrothermal liquefaction can be more attractive than pyrolysis or gasification. In these studies, typically 25% biomass slurry in water is treated at temperatures of 300-350°C and 12-18 MPa pressures for 5-20 min to yield a mixture of liquid, gas (mainly CO2), and water. The liquid is a mixture with a wide molecular weight distribution and consists of various kinds of molecules. A large proportion of the oxygen is removed as carbon dioxide and the resulting biocrude contains only 10-13% oxygen, as compared to 40% in the dried biomass [35]. Hydrothermal upgrading (HTU) process was first developed by Shell, where biomass was subjected to subcritical water at 330°C to produce biocrude. Biocrude was further upgraded to liquid fuels via hydrodeoxygenation process [35]. In a conceptual process scheme, it was shown that each ton (dry basis) of biomass can produce 300 kg (or 95 gal) of liquid fuel.

Karagoez et al. investigated the distribution of products from hydrothermal liq­uefaction (280°C for 15 min) when wood (sawdust), nonwood biomass (rice husk), and model biomass components (e. g. lignin, cellulose) are used as feedstock. The produced bio-oil was characterized for their differences in the hydrocarbon compo­sitions with respect to feedstock. Cellulose showed the highest conversion among the four samples investigated. Sawdust and rice husk had almost similar conver­sions. Liquid products were recovered with various solvents (ether, acetone, and ethyl acetate) and analyzed by GC-MS. The oil (ether extract) from the hydrother­mal treatment of cellulose consisted of furan derivatives whereas lignin-derived oil contained phenolic compounds. The composition of oils (ether extract) from saw­dust and rice husk contained both phenolic compounds and furans; however, pheno­lic compounds were dominant. Rice-husk-derived oil consists of more benzenediols than sawdust-derived oil. The volatility distribution of oxygenated hydrocarbons was carried out by C-NP gram and it showed that the majority of oxygenated hydro­carbons from sawdust, rice husk, and lignin were distributed at n-C11, whereas they were distributed at n-C8 and n-C1 0 in cellulose-derived oil. The gaseous products were carbon dioxide, carbon monoxide, methane in sawdust, rice husk, lignin, and cellulose [53, 54] .

Liquefaction of biomass in subcritical water proceeds through a series of structural and chemical transformations involving [16]

— Solvolysis of biomass resulting in micellar-like structure

— Depolymerization of cellulose, hemicelluloses, and lignin

— Chemical and thermal decomposition of monomers to smaller molecules

Demirbas has also reviewed the possible mechanism of liquefaction. Organic materials are converted to liquefied products through a series of physical and chemical changes such as solvolysis, depolymerization, dehydration, and decarboxylation. Solvolysis is a type of nucleophilic substitution where the nucleophile is a solvent molecule. This reaction results in micellar-like substructures of the feedstock. Depolymerization reactions lead to smaller molecules. Decarboxylation and dehy­dration lead to new molecules and the formation of carbon dioxide through splitting off of carboxyl groups [19].

Switchgrass was effectively liquefied to produce biocrude in subcritical water in a flow-through reactor. Biocrude composed of aqueous phase (water-soluble compound) and solid precipitates. The aqueous phase contained oligomers and monomers of five and six carbon sugars, degradation products (5-HMF and furfural), organic acids (lactic, formic, and acetic acid), 2-furancarboxaldehyde, and other phenolic products containing 5-9 carbon atoms. A small amount of potassium carbonate cata­lyzed the liquefaction and enhanced the decomposition of biomass to water-soluble products. The residual solid contained mainly lignin fractions. Based on the infrared spectroscopy and electron microscopy, it was confirmed that subcritical water treat­ment lead to a breakdown of lignocellulosic structure. Some of the identified compounds by GC-MS analysis (> 85% of confidence level) in biocrude from switchgrass liquefaction are given in Table 3.

Hydrolysis of cellulose in hydrothermal medium has been studied extensively. Several research studies have shown that subcritical and supercritical water can be used under a variety of conditions to rapidly (order of seconds) liquefy cellulose to sugar and its degradation products [18, 76, 102]. Hemicelluloses, an amorphous fraction of lignocellulosic have been successfully extracted up to 95% of its fraction as monomeric sugar and sugar products in hydrothermal medium in the range of 200-230°C in a very short reaction time [9, 80] . Low activation energy of lignin causes substantial degradation of lignin in hydrothermal medium at temperature below 200°C. Reaction proceeds through the cleavage of aryl ether linkages, fragmentation, and dissolution [9] . Lignin depolymerization yields low molecular weight fragments having very reactive functional groups such as syringols, guaiacols, catechols, and phenols [29]. The density of water within the hydrothermal medium has been found to be a key parameter in deciding the product pathways [94].

The results of liquefaction studies on model compounds and the actual biomass in hydrothermal medium provides an opportunity for converting biomass to biocrude and other important chemicals. The hydrothermal medium (subcritical water) in the range of 250-350°C regions provides a favorable condition for conducting ionic reactions. In general, hydrothermal liquefaction conditions range from 250 to 380°C, 7-30 MPa with liquid water present, often in presence of alkaline catalyst [94]. Crystalline cellulose was successfully converted to monomers (mainly glucose) and oligomers by hydrolysis in subcritical water in a continuous flow reactor. More than 90% of the cellulose converts to water-soluble products above 330°C. The study showed that a high yield of hydrolysis products can be achieved at comparatively lower temperature (335°C) in subcritical water. For example, up to 66.8% of crystalline cellulose was converted to hydrolysis products at 335°C and 27.6 MPa in merely

4.7 s reaction time. With increase in the reaction time, the hydrolysis products degraded to glycoaldehyde, fructose, 1,3 dihydroxyacetone, anhydroglucose, 5-HMF, and furfural. Yield of glycoaldehyde, a retro-aldol condensation product of glucose, increased with a decrease in the density of supercritical water, and the yield of degra­dation products, 5-HMF and organic acids, increased with temperature and residence time. In supercritical water conditions, more than 80% of the cellulose converted into the degradation products (oxygenated hydrocarbons) and organic acids [64].

1.1 Monometallic Systems

Metal-based catalysts containing metal-loading of 10-25 wt% in Fe and 10-30 wt% in Co were prepared by wet impregnation of commercial silica support (BASF DU — 11, surface area (SBET) = 136 m2 g-1) with appropriate amounts of aqueous solution

of Fe(NO3)2 (Aldrich, p. a.) and Co(NO3)2 (Aldrich, p. a.), in a rotary evaporator, respectively. After impregnation, the samples were dried at 110°C for 12 h and then calcined at 450°C for 5 h. Prior to characterization and testing, the catalysts were activated in flowing hydrogen at 500°C for 12 h. The catalysts and the nomenclature used are listed in Table 1.

Table 1 compared reactive performance of catalysts Z4K1C4/FS5-II and Z4K2C4/ FS5-II under CO + H2 and CO2 + H2, respectively. In the case of CO hydrogenation, CO conversion and CO2 selectivity are increased, but CH4 selectivity is decreased with more K added into catalyst. These have been observed by other works [8, 34, 42, 60-62]. For CO2 hydrogenation, catalysts Z4K1C4/FS5-II and Z4K2C4/FS5-II have similar CO2 conversion and CO selectivity, but the latter possesses lower CH4 selectivity than the former. Therefore, catalyst Z4K2C4/FS-II has higher selectivity to C2+ hydrocarbons than Z4K1C4/FS-II. The relation between C2+ hydrocarbons and K content is different to the result found in our previous work [42]. Furthermore, the two catalysts containing SiO2 have higher CO selectivity and CH4 selectivity than the catalysts without SiO2. These changes are due to the addition of SiO2 which decreases the amount of effective potassium [58, 63]. Promoter K is beneficial for carbon chain growth [11, 64] i however, this function is weakened by SiOi. It is required to adjust the contents of promoter Zn, K and Cu with the introduction of SiO2 to precipitated Fe catalyst.

C. thermocellum is a thermophilic anaerobe with a naturally powerful cellulosome — based cellulolytic system that was first isolated in pure culture in 1954 [15]. Its prom­inence as a candidate CBP organism stems from its innate ability to deconstruct and ferment cellulose directly to ethanol, through a very efficient cellulose metabolism at very high temperatures. Despite these positive attributes, native isolates can only generate very low levels of ethanol ranging from 0.08 to 0.29 g ethanol per g glucose equivalents [16-19] and cannot tolerate levels of the fermentation product exceed­ing 1.5% [20-23]. The phenotype of increased tolerance to ethanol has been successfully identified in some strains reportedly being able to tolerate ethanol concentrations approaching 7-8% in certain cases [23] . However, this was not in the presence of hydrolysate which has additional potent inhibitors present that can synergistically decrease tolerance to ethanol [24]. An additional detriment of C. thermocellum is that it lacks the native ability to ferment pentose sugars [7] which would leave the hemicellulose component of biomass unconverted in the process.

As with all current candidate CBP organisms, the creation of improved C. ther — mocellum strains requires the ability of the organism to be genetically manipulated, a capability that at one point remained quite elusive. Recently, however, researchers have developed methods to perform a variety of genetic manipulations [25-27], and very recently a group of researchers has developed the ability to perform targeted gene knockouts in C. thermocellum [26]. which will be an essential tool for per­forming targeted metabolic engineering in this organism that was once seen as genetically rigid. While additional techniques will need to be established and fur­ther developed, this breakthrough makes this highly cellulolytic organism acquies­cent towards genetic manipulations and more importantly shows that C. thermocellum will become a strong candidate organism for CBP. However, major hurdles still need to be overcome with the engineering of this organism including the incorpora­tion of a pentose sugar fermentation pathway, and instilling the ability to produce and tolerate higher levels of ethanol while resisting the toxic environment of a cel — lulosic hydrolysate. It also seems quite likely that given this organism’s ability to deconstruct native biomass, a less severe pretreatment configuration might be an option, thus reducing the toxicity of the hydrolysate. Nonetheless, C. thermocellum remains a prime candidate organism in the field of CBP owing predominately to its strong ability to deconstruct biomass.

Before describing classification schemes, it is worth describing the different enzy­matic activities of cellulases, as it is these properties one seeks to clearly define and delineate in a classification scheme. Indeed, one way to classify cellulases, provided function is known, is directly through the EC number, a classification scheme estab­lished and maintained by the International Union for Biochemistry and Molecular Biology (IUBMB; http://www. chem. qmul. ac. uk/iubmb/). EC classifications sepa­rate enzymes important for cellulose degradation into three groups. The two EC groups that hydrolyze the glycosidic bonds of cellulose are EC 3.2.1.4 and EC 3.2.1.91. From left to right, the 3 designates the hydrolase enzymes, the 2 further specifies glycosylase enzymes, and the 1 specifies glycosidase enzymes, that is, hydrolyzing O-glycosyl compounds. The 4 specifies endocellulolytic activity, that is, hydrolysis of internal (1—4)-b-D-glucosidic linkages in cellulose (Fig. 2a; EC

3.2.1.4 also includes hydrolases that act on lichenin and cereal b-D-glucans). The 91 specifies exocellulolytic activity, that is, hydrolysis of glycosidic bonds at the end of the cellulose chain (Fig. 2b), producing short chain cellooligosacharrides such as cellobiose (two glucose units connected by 1—^4-P linkage).

The third group of enzymes important for cellulose hydrolysis are those of the group EC 3.2.1.21. These enzymes, termed b-glucosidases (and sometimes emulsins), hydrolyze the glycosidic bond of cellobiose, releasing b-D-glucose units (Fig. 2c). Thus, these enzymes are necessary to complete the transformation of cel­lulose to glucose. These enzymes also promote cellulose breakdown by increasing the activity of the exocellulases, which often show product inhibition by cellobiose.

Although the EC system provides a key distinction between the activities of cel­lulolytic enzymes, there are a few shortcomings of this classification. First, the boundaries between endocellulolytic (EC 3.2.1.4) and exocellulolytic (EC 3.2.1.91) activity are sometimes blurry, as some enzymes show both types of activity [71]. Second, the EC groupings include protein families with entirely different structures and, presumably, differing evolutionary origins [29] . Third, as is described in the following section, there are additional mechanistic differences among cellulases that are not resolved using just two categories.

-e—7—•^y-o-‘-*~^—» ‘<’~r-‘-^7~-~a. 4 •«—7~-_^7^0’-^r^—° *>~7~~-^7~~»

EC 3.2.1.4 Г Г — Г Г

.Ср^Г-гЬ^ — tp^ ♦ "г^^-Лр^ГсЬ^.

——— — "-Їй-. •♦-•^/ ••

Fig. 2 Enzymatic breakdown of cellulose, (a) Reactions catalyzed by endocellulases (EC 3.2.1.4). both with retention (upper) and inversion (lower) at the anomeric carbon. The endocellulase reaction results in the formation of a new reducing and a new nonreducing end. (b) Reactions catalyzed by exocellulases (EC 3.2.1.91) on reducing (top) and nonreducing (bottom) ends. Both exocellulase reactions produce a new reducing and a new nonreducing end. and a cel — lobiose molecule. In nonreducing end reactivity, the new reducing end is on the soluble cellobiose product, whereas for reducing-end reactivity, the new reduc­ing end is on the cellulose chain and likely remains part of the insoluble cellulose fiber, (c) Reactions catalyzed by |3-glucosidases (EC 3.2.1.21). These enzymes (also called emulsins) indirectly accelerate exocellulase activity (b) by relieving product inhibition

The present invention [2] is directed to a revolutionary photosynthetic ethanol — production technology based on designer transgenic plants (e. g., algae) or plant cells. The designer plants and plant cells are created using genetic engineering tech­niques such that the endogenous photosynthesis regulation mechanism is tamed, and the reducing power (NADPH) and energy (ATP) acquired from the photosyn­thetic water splitting and proton gradient-coupled electron transport process can be used for immediate synthesis of ethanol (CH3CH2OH) directly from carbon dioxide (CO2) and water (H2O) according to the following process reaction:

2CO2 + 3H2O ^ CH3CH2OH + 3O2 (2)

The ethanol-production methods of the present invention completely eliminate the problem of recalcitrant lignocellulosics by bypassing the bottleneck problem of the biomass technology. As shown in Fig. 2, the photosynthetic process in a designer organism effectively uses the reducing power (NADPH) and energy (ATP) from the photosynthetic water splitting and proton gradient-coupled electron transport process for immediate synthesis of ethanol (CH3 CH2OH) directly from carbon dioxide (CO2) and water (H2O) without being drained into the other pathways for synthesis of the undesirable lignocellulosic materials that are very hard and often inefficient for the biorefinery industry to use. This approach is also different from the existing “cornstarch ethanol-production” process. In accordance with this invention, ethanol will be produced directly from carbon dioxide (CO2) and water (H2O) without having to go through many of the energy-consuming steps that the cornstarch ethanol — production process has to go through, including corn-crop cultivation, corn-grain harvesting, corn-grain cornstarch processing, and starch-to-sugar-to-ethanol fermentation. As a result, the photosynthetic ethanol-production technology of the present invention is expected to have a much (more than ten times) higher solar — to-ethanol energy-conversion efficiency than the current technology. Assuming a 10% solar energy-conversion efficiency for the proposed photosynthetic ethanol — production process, the maximal theoretical productivity (yield) could be about

88,700 kg of ethanol per acre per year, which could support about 70 cars (per year per acre). Therefore, this invention has the potential to bring a significant capability to the society in helping to ensure energy security. The present invention could also help protect the Earth’s environment from the dangerous accumulation of CO2 in the atmosphere, because the present methods convert CO2 directly into clean ethanol energy.

A fundamental feature of the present methodology is utilizing a plant (e. g., an alga) or plant cells, introducing into the plant or plant cells nucleic acid molecules coding for a set of enzymes that can act on an intermediate product of the Calvin cycle and convert the intermediate product into ethanol as illustrated in Fig. 2, instead of making starch and other complicated cellular (biomass) materials as the end products by the wild-type photosynthetic pathway (Fig. 1). Accordingly, the

STROMA

Fig. 2 A designer organism such as a designer alga becomes a “green machine” for production of ethanol directly from CO2 and H2O when the designer photosynthetic ethanol-producing pathway(s) is turned on

present invention provides, inter alia, methods for producing ethanol based on a designer plant (such as a designer alga), designer plant tissue, or designer plant cells, DNA constructs encoding genes of a designer ethanol-production pathway, as well as the designer algae, designer plants, designer plant tissues, and designer plant cells created. The various aspects of the present invention are described in further detail herein below.

Even though the Jatropha seeds are rich in oil and crude protein, they are highly toxic and unsuitable for human or animal consumption. The toxic nature of oil and Jatropha seedcake has been demonstrated in several studies [1,2, 5, 6, 46]. The toxic or irritant compounds found in Jatropha seedcake are phorbol ester (2.43 mg/g ker­nel in toxic varieties and 0.11 mg/g kernel of nontoxic varieties), lectin (102 mg/g kernel in toxic varieties and 51 mg/g kernel on nontoxic varieties), trypsin inhibitor activity (21.2 mg inhibitory/g meal in toxic varieties and inhibition of 26.5 mg/g meal in nontoxic varieties), phytate (9.7% in the Jatropha seedcake of toxic varieties and 8.9% in nontoxic varieties), and Saponin (equivalent to 2.3% diosgenin in Jatropha seedcake of toxic varieties and 3.4% in nontoxic varieties).

Curcin, a toxic protein isolated from the seeds, was found to inhibit protein syn­thesis in in vitro studies. The high concentration of phorbol esters in Jatropha seed has been identified as the main toxic agent of Jatropha which is responsible for the toxicity ([52,4]. Several cases of poisoning J. curcas in humans after consumption of seeds by chance have been reported with symptoms of dizziness, vomiting, and diarrhea and in extreme conditions has been noted even cause death [13].

Lectin is also predicted to cause toxicity in J. curcas [16]. However, Aderibigbe et al. [3] and Aregheore et al. [9] show that the lectin is not the main toxic com­pounds in Jatropha seedcakes. Successful utilization of Jatropha seedcakes cannot be achieved without the removal of all of toxic and antinutritional compounds. Toxic-removal processes for seedcakes of low FFA Jatropha can be done directly by in situ transesterification [61] and through detoxification process using heat and chemical treatments for seedcakes of high FFA [9, 68].

The 2009 Copenhagen Climate Conference highlighted the growing importance of developing economic CO2 neutral energy systems, and also recognised that the global temperature rise should be limited to less than 2°C to avoid severe climate change [105]. The global annual primary energy consumption in 2009 was estimated at 11,164 million tonnes of oil equivalent (mtoe), a reduction of 1.1% on the 2008 figures, which was attributed to the global economic recession [22]. Fossil fuels accounted for 88% of the primary energy consumption,[9] comprising of oil (35%), coal (29%) and natural gas (24%) as the major fuels, while nuclear energy and hydroelectricity account for 5 and 7% of total primary energy consumption, respec­tively [22]. It has been predicted that the global primary energy demand will increase by 40% for the period 2007-2030 [95, 103] , mainly due to increases in demand from developing countries. Over the same period, it is also predicted that CO2 emis­sion reduction of 25-40% range will be required to maintain a 2°C limit in global temperature rise [105]. High dependence on fossil fuels is the largest contributor of greenhouse gas (GHG) emissions to the biosphere, with estimated global CO2 emissions of 29 Gtonnes in 2006 [65]. It is therefore imperative to identify compat­ible mitigation strategies to minimise the excess CO2 emissions as the associated climate change projections could have major consequences for nature as well as human systems 2 17, 962 • Consequently, renewable and carbon neutral fuels are essential to both environmental and economic sustainability; hence, biofuels will be key energy sources [152].

Biofuels refer to liquid, gas and solid fuels derived from biomass, the collective term for plant-derived matter (terrestrial and aquatic) other than that which has been fossilised. Biomass includes dedicated energy crops (viz. sugarcane, rapeseed, short-rotation woody crops, etc.), agricultural and forestry residues and agri-food waste. Classification includes first-, second — and third-generation biofuels. First — generation biofuels refer to those derived from sources like starch, sugar, animal fats and vegetable oil, usually extracted by conventional techniques of production (e. g. biodiesel, vegetable oil, biogas, bio-alcohols and syngas). Second-generation biofuels are derived from lignocellulosic biomass by thermochemical conversion processes such as gasification and Fischer-Tropsch process. The second-generation biofuels offer key advantages over the first-generation, including: possible use of a much wider range of raw material, especially waste, which may lower the cost of the feedstock significantly; the resulting fuels are of high-quality and clean-burning, with potentially a much lower well-to-wheels CO2 profile; cultivation process (if any) could be less environmentally intensive than for ordinary agricultural crops (e. g. reduced or zero cultivation result in even lower GHG emissions from cultivation), and can be co-produced with electricity. Third-generation biofuels in this context refer to fuel extracts from micro — and macroalgae feedstock and from biomass crops that have been genetically modified in such a way that their structure or properties conform to the requirements of a particular bioconversion process. Technology frontiers for plant-based third-generation biofuels include the development of tree crops with weakened lignin content to ease conversion to fuels [101], and modified corn varieties with enzymes necessary to break down cellulose and hemi-cellulose into simple sugars in the leaves, allowing for more cost-effective, efficient production of ethanol from corn residue [200] .

Through specific proven biochemical or thermochemical conversion processes (viz., esterification, anaerobic digestion, pyrolysis and thermochemical liquefaction), biomass can be converted to a range of biofuels such as ethanol, methane, hydrogen, biodiesel, and Fischer-Tropsch diesel [55]. Sustainable deployment of biofuels has therefore become a crucial strategy for limiting or reducing GHG emissions, enhancement of regional security of energy supply (including employment creation), and controlling price volatility in fossil fuel markets [23].

Supercritical fluid extraction (SFE) is an emerging green technology that is a poten­tial substitute to traditional organic solvent extraction.

5.2.1 Basic Principles

When the temperature and pressure of a fluid is raised over its critical values (Tc and Pc), the fluid is transformed to a supercritical state and exhibits properties similar to both a liquid and a gas (Fig. 9) [50, 59]. The intermediate properties of this state (Table 3) are found to significantly enhance the propensity of the fluid to extract lipids from living tissues. In comparison to organic solvent extraction, SFE is known to have the following advantages [40-42, 59]:

• Tunable selectivity

The solvating power of a supercritical fluid is a function of its density which can be continuously adjusted by changing pressures and temperatures. Therefore, its selectivity can be tuned such that it interacts only with neutral lipids.

• More favourable mass transfer

High diffusion coefficient (similar to a liquid) and low viscosity (similar to a gas) enable a supercritical fluid to penetrate through cellular matrices much more rapidly than an organic solvent, leading to a higher extraction rate and a shorter extraction time.

• Production of solvent-free lipids

Since the lipids obtained at the end of SFE are free from any solvent, no energy

needs to be expended for solvent removal.

CO2 is used as the primary solvent in most SFE applications for numerous reasons. Its moderate critical pressure (72 bar) enables a modest compression cost, while its low critical temperature (32°C) avoids possible degradation of thermally sensitive lipids. Its low toxicity, low flammability and lack of reactivity also facilitate for a safer SFE operation [37, 59]. In cases where microalgal cells are cultivated in synergy to a coal power station, the CO2 for extraction can be conveniently supplied from the scrubbed flue gas of the station.

Figure 10 shows the schematics of a laboratory-scale supercritical carbon dioxide (SCCO2) extraction system [3] . A mixture of microalgal cells (either as a wet paste or dried biomass) and packing materials (normally diatomaceous earth) in a specific ratio is packed into the extraction vessel equipped with a heating element. A feed pump delivers the CO2 from its reservoir to the extraction vessel at a pressure greater than Pc. Once the vessel is heated (T> T), the compressed CO. is trans­formed to its supercritical state and performs lipid extraction on the microalgal cells. Lipid molecules are dissociated from the cellular structure and transferred into the eluting SCCO. in a diffusion-driven mechanism. The lipid-saturated SCCO. then leaves the extraction vessel and enters the collection vessel. The micrometering valve at the entry to the vessel is opened to rapidly depressurize the incoming SCCO2. Once completely decompressed, the SCCO2 returns to its gaseous state and evaporates to the ambience, thereby forcing the extracted lipids to precipitate in the collection vessel. Even though SCCO2 extraction can be operated as either a batch (static) or a continuous (dynamic) process, the dynamic operation is preferred as it generally leads to an improved yield [59]. Even though SCCO2 process has been used to extract lipids from microalgal cells in a laboratory scale [1, 2, 7, 10, 28, 40-42, 53] its energy efficiency still needs to be assessed and is a subject of future research endeavour.