Pretreatment is required to break the crystalline structure of cellulosic biomass to make it more accessible to the enzymes, which can then attach to the cellulose and hydrolyze the carbohydrate polymers into fermentable sugars. The goal of pretreatment is to pre-extract hemicellulose, disrupt the lignin seal and liberate the cellulose from the plant cell wall matrix. Pretreatment is considered to be one of the most expensive processing steps in cellulosic ethanol processes, but it also has great potential to be improved and costs lowered through research and development (Lynd et al., 1996; Lee et al., 1994; Mosier et al., 2005). Many pretreatment techonlogies have been developed and evaluated for various biomass materials. However, each pretreatment method has its own advantages and disadvantages, and one pretreatment approach does not fit all biomass feedstocks. Three widely used pretreatment techologies will be reviewed below.

In low-temperature range (temperature in muffle tm = 250-500оС) from the moment the particle enters the furnace and up to pyrolysis commencement the particles are heated due to muffle irradiation as inert compound (curve a in fig. 1). Further in the case of charcoal the coke-ash residue will ignite (point B), with temperature upsurge (self-heating) exponentially (curve b) and burning in quasi — stationary conditions (curve c). Charcoal has the lowest volatile content and preliminary prepared (mostly) reaction surface (75% porosity and specific surface area of 8.6 m2/ g). In the course of pyrolysis charcoal characteristics change insignificantly (porosity increases up to 77-80%, specific surface up to 29.2 m2/g), self-heating starts at tm и 250оС. Quasi-stationary combustion starts earlier than in the case of wood and seeds. Charcoal self-heating starts about 50 s and 400 s earlier than that of wood and seed respectively (time difference between B points).

In the case of wood having porosity close to coal porosity (65%) and specific surface one order lower (1 m2/ g), heating of the particle after the temperature of intense pyrolysis commencement has been reached (particle temperature tp и 275оС) follows the curve a’ (up to tp и 420оС > tm). The analysis shows that at muffle temperature of 400оС the volatile combustion heat which amounts up to 60% of total heat flux plays a key role. Contribution of bio fuel volatiles combustion to warm-up of their solid residue is of major importance. At muffle temperature of 800оС the main heat flux is muffle irradiation which amounts to approximately 60%.

Intense emission and burning of volatiles hinders the access of oxygen to coke-ash residue, which has been repeatedly described elsewhere, and residue warming is less intense than in the case of charcoal. After the major portion of volatiles has burnt out (point B) the coke-ash residue of wood particle gets heated following the curve b with its intensity close to that of charcoal self-heating but to a higher temperature, and finally it burns out 1.5-2.0 times quicker. It should be noted that porosity and specific surface of wood coke-ash residue exceed that of the coal.

In the case of wood the sources of warming are heat fluxes from muffle furnace, exothermic reactions of pyrolysis and combustion of volatiles. The amount and ratio of these fluxes depend on muffle and particle temperature.

The time of preheating stage and solid residue combustion at 400оС are compatible for wood (~ 1 : 1). With muffle temperature increase this ratio will change towards increase of relative duration of solid residue burn out and at 800оС it will be 1 : 4. Heat emission intensity (W/m2) for volatile combustion and coke oxidation is actually 1 : 1. With temperature increase the intensity of heat emission for volatile combustion will increase significantly compared to coke oxidation, i. e. ~ 4 : 1.

Initial characteristics of seed inner surface are much lower than those of other bio fuels (25% porosity, 0.01 m2/g specific surface). Due to different thermal capacity and under endothermic effects revealed in analysis of the seed, the latter is heated more slowly on segment A’-A" than "model" inert compound. To the right from A" point volatiles release intensively in the form of boiling liquid tar fractions ("cuts") in coke-ash residue pores and on the surface (as is the case with coking coals). During decomposition they partially form soot on porous coke-ash residue surface and partially emit non-ignited as a dense smoke (of fallow color). On segment A"-B within approximately 250-300 sec temperature in seed center actually coincides with muffle temperature. Formally this period is a variety of well — known induction period (Pomerantsev, 1973).

Partial overlap of reaction surface by tar fractions and much less initial porosity (see Table 1) result in notable delay of seed self-warming commencement compared to wood and charcoal self-heating. By moment B which is characterized by disappearance of liquid phase on the surface, the particle becomes accessible to air and ignition and self heating of coke ash residue commence. In this case the speed of temperature increase is almost 10 times lower than that of wood and charcoal which is apparently due to incomplete pyrolysis at previous stage. By the end of self-heating seed coke-ash residue has the greatest porosity and hence the highest overheating temperature and finally the maximum rate of burning.

For seed the duration of thermal pretreatment and solid residue burning at 400 °С is expressed as 3 : 1 (in this case the preparing stage lasts much longer than that of wood). When muffle temperature reaches 800 °С this ratio changes towards increase of relative solid residue burn out period, similarly to the ratio for wood, and becomes 1 : 4. The relation between heat radiation intensity at stages of volatile combustion and seed coke oxidation is 0.3 : 1 (volatile: coke) at 400 °С and 4:1 at 800 °С.

In high temperature range (at muffle temperature above 500 °С) the distinctions between warming and ignition of different bio fuels are smoothed. Seed warming delay relative to charcoal particle decreases actually to zero and overheating by the end of self-heating and burning rate of coke-ash residue in the main segment come closer. Qualitatively varying pyrolysis scenarios for different bio fuels with close quantitative result for burning intensity are of less importance which correlates well with a well-known high-temperature experiment. The processes of wood particle and seed ignition are qualitatively different to a large extent and they both differ from charcoal ignition. Visual examination shows that after wood particle is placed in muffle at 800 °С volatiles release is slow with formation of "faint" burning layer close to surface and slightly fluctuating short flame above it (compatible with particle diameter). Volatile release from seeds has "explosive" intensity with continuous burning substance burst out to the distance of up to 3-4 diameters of seed (Fig. 5)

Fig. 5. Process of inflammation: a) wood chip, b) date seed; 1-4 — stages of inflammation from time of placing particle into the muffle (1) separated by 20 s.

1.2 Kinetics of coke-ash residue conversion in burn out conditions

Visual observations of a single particle show that irrespective of low ash content wood chips burn out inside ash enclosure under natural convention conditions: the size of burning carbon nuclear gradually decreases and remaining ash (soft) enclosure retains the original shape of the particle, actually without changing the size. This effect is reached in intense turbulent fluidizing bed too, but when high-ash flotation tailings with rigid mineral enclosure are burnt (Belyaev A. A., 2009).

In low-temperature range the burning of wood and seed coke-ash residues begins after the major portion of volatiles releases at particle temperature tp « 350-400 оС; for charcoal with small volatile content the burning starts at lower temperature tp « 250-300 оС. Coke-ash residue burning at temperature close to above cited values in intra-kinetic mode exhibits significant overheating of particle center relative to environment temperature: AT = 360 оС for seed, 250 оС for wood, 300 оС for charcoal (6,a).

In high temperature range with dominating diffusive resistance the overheating of particle center relative to environment is negligible: at 800 °С it is equal to AT = 145 °С for pellet and seed, 105 °С for wood, 85 °С for charcoal, at 1200 °С it is 75 °С for wood, 40 °С for charcoal. Transfer from kinetic to intra-diffusion conditions is most pronounced for charcoal, as its combustion is not aggravated by volatile release in great amounts and actually represents coke residue burn out. Burning rate curve has a "knee" in the range of environment temperature of 400 °С (tp = 600 °С): steep segment corresponds to kinetic mode and flat segment to diffusion mode (fig. 6,a).

In high temperature range the maximum overheating of particle center at tm = 800 °С is found for seed having maximum porosity after pyrolysis and minimum overheating is exhibited by charcoal which appears to have the least porosity and lowest reactivity by the moment when stationary burning conditions are achieved, compared to any other examined fuel. Overheating of wood particle center is between these two values. Burn out rate ratio for examined fuels are in correlation with the ratio of porosity of examined fuels coke-ash residue porosity relation in point B of thermal curves, similar to overheating relations (table 1, porosity after fast heating).

Fig. (6,b) shows the rate of burning and overheating (fragment) vs blow rate in high temperature range. Air speed variations in the range from 0 to 0.5 m/ s result in approximately two-fold change of burning rate and overheating of examined fuels. Irrespective of extremely low ash content, the wood particle at zero blow speed burns out inside ash envelope: carbon-including portion shrinks and ash enclosure builds up actually retaining the shape and the size of original particle.

At blow speed equal to 0.1 m/ s and more the ash envelope is thrown away by air flow opening the coke-ash residue which is similar to ash enclosure behavior in case of anthracite, charcoal and electrode coal particles combustion. Charcoal burn out rate is lower than that of bio fuels coke-ash residue in the entire range of examined blow speeds.

The mode of fuel thermo chemical conversion which is similar to examined mode without blow is typical of fluidized bed gasifiers and complete combustion furnaces after the blowing has stopped by some reason and low-intense residue burn-out continues for many hours (with fire bed surface intensity qR < 0.3 MW/m2). Examined modes with blow speed about 0.5 m/ s in normal conditions are marginal case of blowing for ordinary FB furnace without bed stability loss and maintain fire bed surface intensity qR « 4 MW/m2. Experimental results were used by the authors for designing the up-to-date gas generator firing the wood fuel. Fig. (6,c) shows the burning rates vs. particle temperature.

Experimental points are located within the "segment" limited by lines of kinetic and diffusion modes. Wood burning rate in high-temperature conditions at tp = 900 оС and w = 0 m/s is less than estimated speed limit in diffusion mode by approximately 5-15%, charcoal rate by 10-30%.

Fig. 7. shows the particle burning rate j, g/(m2s) vs. reverse (1000 / Tp, 1000 / К) and normal (tp, °С) temperatures of particle.

Our data on charcoal at tm = 300-1200 °С are grouped close to calculated curve for diffusion mode, and charcoal points received at tm = 280-335°С (Khitrin, 1955) at lower oxygen content in heating medium with burn out without flame-forming self-heating are closer to kinetic curve. Superimposing of experimental points for coal on one curve show that particle temperature shall be chosen as key value in calculations. Choosing the key temperature is important for processing the experimental data in Arrhenius coordinates, because this determines the slope of curve. Choosing the particle temperature as key temperature is important for the range of medium temperatures from 300 to 700°С where significant overheating occurs. Brown coal particles with diameter dp = 3 mm in FB within temperature range of tm = 400-900 °С burns out in intra diffusion mode. Due to lower reactivity of brown coal coke-ash residue the rate of its burning in intra diffusion mode is almost half of wood coke-ash residue burning rate with dp = 3 mm, and transfer to intra kinetic mode occurs at temperature that is higher by 150-200 °С.

Diffusion behavior of bio fuel combustion at environment temperature above 500 °С is observed in the entire examined range of particle diameters, dp = 3-75 mm. Review of the results obtained in different studies demonstrates that in temperature range from 800 to 950°С fine particles (with sub millimeter diameter) exhibit kinetic mode of burning. Thus, according to fig. 8, the actual overheating of coal dust is below calculated values (line 1). Calculation with actual burning rates (line 2) is close to actual overheating values.

With particle size increase its specific surface decreases (per volume unit) and boundary layer becomes larger. This causes decrease of specific heat emission, overheating and burning rate and the particle approaches to isothermal burning. With diameter changing from 10 to 75 mm at Tm = 800оС the overheating lowers from 120-150оС to 30оС, and burning rate from « 2 to « 0.2 g/ (m2s). In the range of dp = 10-75 mm calculation with accuracy of ±15% is approximated by dependence AT = 1130 • dp0’84, °С and j = 23 • dp1, g/(m2s). Diameter dp is expressed in mm.

The available results for FB were obtained with fluidization rate which is much higher than the rate of natural convection. The structure of flows in FB is quite specific (Sherwood criterion for inert medium at rest tends to 1.0). At Re = 80 and higher the burning of particle in FB and in inert-free medium occurs with close dimensionless rates, at small Reynolds numbers the burning in FB is less intense than in inert-free medium and occurs with overheating approximately half as much. This is why the estimated set of overheating (field 3 in fig. 8) depending on inert substance particle diameter (from 0.4 to 1.5 mm) is in satisfactory agreement with general dependence for individual fuel particle within a wide range of its diameters but is well below than actual data for dust burning by V. I.Babiy (Babiy, 1986). Overheating of pellet in FB (point 5) is greater than that of wood particle of approximately the same size in natural convection which may be associated with higher density of the pellet (Palchonok, 2002).

Fig. 9 demonstrates generalized relations "dimensionless burning rate vs Reynolds number" at medium temperature 400-1500 °С, with particle diameter dp = 0.1-80 mm, blow rate w = 0.01-50 m/ s. Data by V. I. Babiy for dusts can be generalized by his curve if medium temperature is assumed to be key temperature. Calculations based on particle temperature as key temperature (domain 5 in fig. 9) makes his points lie below his own curve. This means that burn out of dusts at these temperatures occurs in kinetic mode. Experimental data by L. I. Khitrin lie close to his curve.

1.3 Kinetics of coke-ash residue conversion in gasification mode

Gasification process was realized with intermittent charging, gas composition changed with bed burn out: first, combustible content raised, but on having reached (at some definite bed depth) maximum value, it began to decrease (fig. 10,d).

Coal bed depth in retort at which H2 and СО contents began to decrease was equal to 6 to 7 initial size (average equivalent diameters) of fuel which corresponds to available reference data. At bed depth equal to 1-3 particle size СО content lowered down to 9-10%, and H2 content approached zero with the accuracy within instrument error range. The depth of 1 to 3 sizes was in compliance with available data on burning zone height.

For two blow types (air and steam-air) charcoal conversion gas composition dependence on air flow rate in maximum concentration point (for bed depth of 6-7 sizes) was studied (fig. 10,a, b). A series of experiments was conducted with sequential blow increase in every next experiment. The greater air flow rate and hence the higher temperature in bed corresponded to higher CO content in gas.

At temperature below 700оС gasification reactions actually did not occur and gas composition corresponded to pyrolysis. СО content tended to estimated limit for pure carbon with air blow (СОс = 34.5%), and blow coefficient defined by effluent gas composition was approaching ac. Maximum concentration limits and minimum blow rate coefficient were reached at blow rate и 4 m3/h (speed of blow in bed calculated for empty cross-section is 1.4 m/ s) and bed temperature of 950оС.

The results explain the phenomenon of the so called rapid burning. It is believed that the higher is the blow rate in combustion process, the lower is solid-carbon based reduction of СО2 to CO beyond combustion zone because the time needed for reduction is insufficient. On reaching a certain rate the reduction process stops and СО and СО2 ratio in gas becomes determined by their ratio that was established in burning zone (primary CO and СО2). Experiments with activated charcoal showed that at blow rate above 0.3 m/ s СО content is 34.5%, and СО2 is 2%. Study results allow to give a simpler explanation of "rapid gasification" phenomenon, i. e. gasification conditions can be achieved at a = ac. The result confirms the possibility of producing gas that does not contain three-atom gases (СО2 and Н2О) and methane, and shows one of the ways of its production in dense bed, by providing optimum depth and temperature not less than 900-950оС.

Fig. 10,c demonstrates the results of experiments on allothermal steam-air gasification in dimensionless form: gas concentration is reduced to that in gas of ideal air-steam conversion of char coal (COid = 43%, H2id = 26%). Steam content (Gs / Gf, kg steam/kg fuel) was increased in every next experiment with constant air supply corresponding to a = 0.7 for dry blow ("semi-gas" mode). In this case gas was enriched with combustible components by reaction C + H2O = CO + H2 . Steam breakthrough was not observed.

At steam supply in the amount of 1-1.2 kg of steam/kg of fuel (moisture in terms of fuel as received W и 60%) СО and Н2 content in the experiment was close to ideal steam-air concentration of 35-42% and 22-25% respectively, with chemical efficiency for steam-air conversion r|chem=1.0.

Fig. 11 demonstrates gas composition change with bed depth for allothermal steam (water) conversion and fig. 12 shows carbon-dioxide conversion of char coal in bed with temperature 950-1000оС and blow at temperature 600-750оС.

Gas composition for steam conversion at temperature Т и 1100оС corresponds by 97.5% to stoichiometric one (СО = 42.6%. H2 = 54.8%), and chemical efficiency of steam conversion qchem =1.26. Ratio of СО content to stoichiometric value for carbon dioxide conversion reached 0.8 and chemical efficiency of carbon dioxide conversion r|chem=1.35.

At bed depth less than 70 sizes a linear variation of CO content with bed height was detected for both carbon dioxide or steam blow. Hydrogen content does not depend on bed height and it was increasing during the entire experiment.

Allothermal technique of the experiment allowed to decompose all water steam fed into the bed due to muffle heat. In the experiments allothermal gasification marginal conditions were achieved for steam-water, steam and carbon dioxide blow. In two latter cases homogeneous gas-water shift reaction CO + H2O = CO2 + H 2 was performed providing for increase of Н2 content above stoichiometric value by heterogeneous reaction and steam conversion of charcoal.

1.4 Using heat recovery (recirculation) for advanced CCGT with thermo chemical conversion of low-grade fuels

Temperature limits and ranges of mass concentration of combustible elements and oxidizing agent were determined, in which the process of low-grade fuel gasification is realized with higher conversion efficiency and which include multi-parametric optimum conditions for hypothetic "ideal" tars free conversion into useful products. Fuels are divided into two groups: "А" group includes fuels that are converted into ideal mixture (СО+Н2) due to their own internal chemical source of heat (in auto-thermal modes) and "B" group that requires heat power source (allo — and auto-thermal mode) for ideal conversion, moreover in case of auto thermal process the gas is not ideal and has complete combustion products (СО2 and Н2О). Wood, peat, and some high-moisture brown coals belong to "B" group. In "A" range the ideal gasification is characterized by constant heating value of gas and constant chemical efficiency of fuel conversion into syngas, moreover, high-metamorphized fuels require considerable amount of water vapor to be added whereas the main source of molecular hydrogen for brown coals is fuel hydrogen.

Allothermal conversion procedure allows to obtain maximum effect of chemical efficiency increase (fig. 13, points 8", 8’" and 8’"’ are results of allothermal steam-air, steam and carbon dioxide gasification). However, conversion efficiency rise due to external sources is rarely used (p. p. 4′, 7′), which is explained by their "high investments" and low production effectiveness.

1 — fuel oil gasification without regeneration in flow-type gasifier (by VNII NP); 1′ — ditto, calculation with regeneration; 2 — first-generation dense-bed gasifier operating on black coal; 2′ — tuyere gas producer (JSC VTI) operating on black coal; 3 — gas producer by MHI without regeneration (estimated); 3′ — gas generator MHI (expected upgrading of recirculation system); 4 — first-generation dense bed gas producer operating on peat; 4′ — allothermal pyrolyzer for peat OIVTAN; 5, 6, 7, 8 — first-generation dense bed gas generator; 5′ — first-generation gas producer with superadiabatic heating by IPKhF RAN (estimated); 6′ — upgraded downdraft producer by UrFU with regeneration; 7′ — plasma gas generator, IEE RAN; 8′ — three-zone downdraft producer by UrFU, 8» — with steam and air blow, 8»’ — with steam blow at 1000оС to the third zone, 8»» — with carbon dioxide blow at 1000оС to the third zone; 9′ — mini — CHP gasifier Viking, Denmark (estimated); 10′ — mini-CHP gasifier, UrFU (estimated); 11 — brown coal gasification (estimation); 11′ — joint gasification of brown coal and natural gas, VNII NP; 12, 12′ — downdraft producer by ISEM with different blow temperatures

There is a substantial potential to rise chemical efficiency by improving heat recirculation within gas generator or "gas generator-gas consumer" system. This becomes evident when analyzing the tendencies of development of both industrial large-capacity coal gasifiers (from Winkler gasifier to HTW plant), and air gasification plants especially those working on low-grade fuels and bio mass. In simple-design (without recirculation) gasifiers operating on biomass with steam and air blow the percent of fuel needed to provide for complete combustion is great (up to 30 %), and temperature is low, therefore gas contains a considerable amount of intermediate gasification products (tars, СО2 and Н2О) and chemical efficiency is not more than 60 %. Such reactors cannot generate gas with close to ideal characteristics even hypothetically. This is why the development of simple-procedure low — efficiency air reactors does not seem to have any perspective.

2. Conclusion

The kinetics of thermo chemical conversion was studied on low-grade fuels in oxidizing pyrolysis, afterburning and gasification modes in temperature ranges typical of moderate combustion conditions in power reactors. The obtained data were used as the basis for upgrading of effective conversion procedure which will allow to develop gasifying plants with high chemical efficiency in the long term.

The procedure for non renewable resources conversion in fuel gas for gas engines (gas turbine engines, internal combustion engines) and fuel cells has been developed. This procedure provides low tar and hydrocarbon content in gas with high chemical efficiency (80% achieved and 90-95% design efficiency) which is novel in Russia and in prospect it will allow to rise the electrical efficiency of small scale CHP up to 23-25% on bio fuel and up to 50% on fuel cells, and the efficiency of large scale brown-coal CCGT and IGCC up to 50-55% and will make them competitive.

The procedure was implemented at multi-zone dense-bed gasifier with capacity up to 200 kW for bio mass and peat fuels and in a group of pyrolysers having capacity of 10 MW each for pulverized carbon-rich black coal with after-burning of conversion products in coal — fired boiler with capacity of 420 tons of steam per hour.

Small-scale gasification studies are aimed at designing of CHP-ICE with electrical output up 500 kW, using the fluidized-bed gasifier. In the course of works the chemical efficiency of 80% was achieved for gasification process. Gasification of wood fuel having 35% moisture content with air blow resulted in steam-air gas product with heating value of 7.4 MJ / m3. Investigation results were used at pilot mini-CHP with 200 kW capacity.

As mentioned before, the biogas market has enormous possibilities to grow. One of the most important sectors that may trigger large growth of PSA development is within small farms. In such cases, the biogas can be employed for heating and to generate electricity, but a portion of the stream (or the exceeding) can be upgraded to fuel. In such applications, besides the specifications of process performance, six characteristics are desired for any upgrading technology:

1. Economic for small streams,

2. Compact,

3. Automated,

4. Minimal attendance (by non-expert person most of the time),

5. Possible to switch on /off quite fast

6. Deliver product specifications even when subjected to strong variations in feed.

The PSA technology can potentially be employed in such applications since it can satisfy most of the criteria established above. As an example it can be mentioned that some plants of the Molecular Gate technology are operated remotely (automated with minimal attendance) transported in trucks (compact) and they are employed for small streams of natural gas (Molecular Gate, 2011). However, the scale of small biogas application is quite small (smaller than 10 m3/hour). Furthermore, fast switch on/ off a PSA unit for several times was not reported in literature and surely require dedicated research as well as PSA design to handle strong variations in feed streams.

The two major areas where research should be conducted to deliver a PSA unit to tackle such applications are: new adsorbents and design engineering.

The industrial production of biodiesel needs to solve several technical problems in order to obtain this kind of biofuel in an efficient and sustainable way. The physical factors to consider can be summarized by pH, temperature, hydric activity, solvents and supports. Depending on the catalyst used to drive the transesterification reaction, some of the cited factors have different impacts on the global efficiency and feasibility of the process. A non­optimal configuration of the system can reduce significantly the biodiesel yield and compromise the viability of the production plant, especially if the upstream by-products, excess catalyst or auxiliary devices for solvent recovery hinder an easy, clean and rapid downstream processing of the biofuel.

It is generally recognized that the higher the hydrogen content of a petroleum product, especially the fuel products, the better the quality. This knowledge has stimulated the use of a hydrogen-adding process in the refinery, which is called hydrogenation. Currently, the most widely used hydrogenation processes for the conversion of petroleum and petroleum products is hydrotreating.

Hydrotreating (HDT) is a nondestructive, or simple hydrogenation process that is used for the purpose of improving product quality without appreciable alteration of the boiling range. It has become the most common process in modern petroleum refineries. Bio-crude may also be processed by a conventional refinery and potentially augmented with petroleum crude. The oxygen in bio-oils can be removed via hydrotreating. The catalysts commonly used for hydrotreating are sulphide CoMo/ Al2O3, NiMo/ Al2O3 systems. (Nava et al., 2009).

Hydrotreating requires mild conditions, while the yield of bio-oil is relatively low. The process also produces a large amount of char, coke, and tar, which will result in catalyst deactivation and reactor clogging.

The first generation of biofuels demonstrated that energy crops are technically feasible, but that no single solution exists to cover every situation (Venturi & Venturi, 2003). In addition, the production of first-generation biofuels is complicated by issues that are contrary to biofuel philosophy, such as the destruction of tropical rainforests (Kleiner, 2008). Tropical rainforests are the most efficient carbon sinks on earth. Therefore, if biofuels contribute to their destruction, this implies that the carbon balance of biofuels is negative. This consideration limits the viability of first-generation biofuels. It also comes with the corollary that raw materials for biofuel production will have to be diversified over the long term. Second-generation bioethanol is precisely an attempt to overcome this challenge. Second-generation bioethanol will be produced from lignocellulosic biomass, which is a more suitable source of renewable energy (Frondel & Peters, 2007; Tan et al., 2008; Tilman et al., 2007). Lignocellulose is obtained from inexpensive cellulosic biomass that is encountered throughout the world. However, the low-cost transformation of lignocellulose into bioethanol is still challenging. Some possible technologies involve genetic modification of plants, which is a source of concern for society. Whatever the future evolution of the technology, the introduction of energy policies is crucial to ensure that biomass ethanol is effectively developed to become a major source of renewable energy (Tan et al., 2008).

Algae and cyanobacteria are far more efficient than higher plants in capturing solar energy and will surpass first — and second-generation biofuels in terms of energy capture per unit of surface area (Brennan & Owende, 2010). Algae are already used in pilot CO2-sequestration units for emissions cleaning in some conventional power plants running on fossil fuels. This technique is called CO2 filtration. Unfortunately, algae require capital for investing in reactors that can grow them, making CO2 filtration an excellent opportunity for developing this technology. In that sense, algae can be regarded as a third-generation fuel. New methods and technologies for the production of second — (such as synfuels, Baker & Keisler, 2011) and

third-generation biodiesel fuels are being developed and will result in the modification of the definition of biofuels that is generally used in government regulations (Lois, 2007). Finally, one can also envision the exploration of photosynthetic mechanisms for biohydrogen and bioelectricity production. These would constitute fourth-generation biofuels (Gressel, 2008). The development of effective fourth-generation biofuels is not expected before the second half of the 21st century.

Life cycle assessment (LCA) is a methodology widely used to evaluate environmentally all kind of processes and products production (Hsu et al., 2010; Huo et al., 2009; Lardon., 2009; Schmidt, 2010). Economic assessment based on LCA methodology is also being used in literature (Lee et al., 2009; Huppes et al., 2010; Ouyang et al., 2009; Nassen., et al 2008).

3.2 Environmental analysis

As FAO indicates (FAO, 2008), a policy objective by many countries entails mitigating climate change by means of bioenergy promotion. Conversely, life-cycle analyses — which measure emissions all over the bioenergy production chain — points toward a wide divergence in carbon balances according to technologies used, locations and production paths. Thus, more research should be carried out in this field. As FAO suggests, important sources of emissions seem to be land conversion, mechanization and fertilizer use at the feedstock production stage, as well as the use of non-renewable energy in processing and transport.

To evaluate the environmental impact of the model suggested in this work, a general analysis of different topics can be done: energy and water requirements, biodegradability, equivalent CO2 emissions (global warming), tailpipe engine emissions and deforestation. Moreover, LCA methodology (Schmidt, 2010) is used to comparatively evaluate environmental impacts. Regarding to the use of energy, the proposed method nearly eliminates the impacts related to fuel processing and transport, which allows minimizing energy requirements. Fossil fuels, on the contrary, are transported from remote countries as well as raw materials to produce large-scale first-generation biofuels. Furthermore, both fossil fuels and first — generation biofuels need complex processing, which requires significant amounts of energy. Therefore, the proposed model reduces significantly energy consumption. Additionally, rapeseed cultivation helps crop rotation and direct seeding. This is highly recommended as it reduces the steps of land working, thus minimizing power requirements. This results in less use of fuel for each crop, which is a desirable way to reduce emissions.

As for water requirements, as (FAO., 2008) states, many feedstocks are highly water intensive, meaning that their expansion is likely to create even greater competition for this limited resource, depending upon location and production methods. The method proposed here, moves towards a dry land use, as rapeseed is able to grow in the same conditions as replaced cereals would do. On the other hand, water requirements of SVO production are null whereas as stated by (Pate et al., 2007), water requirements of bioethanol with the current technology are about 4 litres of water per litre of bioethanol produced. Consumptive water use in petroleum refining is about 1.5 l/l and biodiesel refining requires about 1 litre of fresh water per litre of biodiesel produced.

Additionally, concerning biodegradability, commonly used SVOs including rapeseed oil are biodegradable and non-toxic, making them useful for transportation applications in highly sensitive environments, such as marine ecosystems and mining enclosures for example (West et al., 2008). This implies less risk when storing the fuel and less impact to biodiversity if accidentally spoiled.

To compare the CO2 emissions from both models, their differences have to be considered. As long as use of machinery, fertilizer and herbicide requirements are similar, the main variation between the two systems is the use of SVO instead of petrodiesel.

Emissions associated to transport, production and combustion of 1 litre of petrodiesel are 3.16 kg CO2/l (Flessa et al., 2002). Approximately a 10% of these emissions result from the extraction, production and transport of the diesel fuel and the remaining 90% are due to its combustion. The fuel consumption for direct seeding and for traditional seeding, according to local farmers, are respectively 70 and 140 l fuel/ha. Thus, the emitted CO2 due to tractor diesel consumption when using traditional seeding doubles the direct seeding method.

On the other hand, the CO2 emitted when burning SVO in a diesel engine was absorbed by the crop during growth (CO2 neutral). Consequently, these emissions are compensated by the photosynthesis absorption. SVO production is very simple and has a low energy requirement, as already seen. Thus, the CO2 associated emissions of this stage are much lower than the ones from petrodiesel.

According to these results, the proposed system avoids the emission of more than 200 kg CO2/ha. In future studies, a life-cycle assessment of this model will be carried out in order to take into account all the emissions in the studied area. Life-cycle analyses would measure the emissions throughout all the bioenergy production chain.

Regarding tailpipe engine emissions, diverging results are found (Krahl et al., 2007; Thuneke & Emberger, 2007). As concerns CO, CO2 or Particulate Matter (PM) emissions, the SVO is clearly better than petrodiesel. Meanwhile, looking at NOx and HC it is not clear if the use of SVO reduces or increases its emissions. Thus, more research is needed to study this field in greater depth.

In relation to deforestation, the high demands of productive soil in large-scale production of biofuels would produce deforestation especially in tropical forests (Russi, 2008). On the other hand, the small scale production plant presented here deals with a small portion of the amount of available land to produce biofuel, thus avoiding the abovementioned impact.

In order to achieve representative results, the general framework for conducting an LCA is followed in this work (ISO 14040, 2006; ISO 14044, 2006). Taking the cropping model presented in section 6, Gabi 4 software (PE International 2010) has been used to carry out the LCA impact assessment.

The use of diesel or straight vegetable oil (SVO) as the tractor fuel is also included to take into account the consumption and the corresponding fuel emissions. Crop types are considered depending on the crop rotation chosen for each scenario. Data on crop works, fertilizing needs and yields were obtained from the Anoia region, a northeastern dry Mediterranean area in Spain.

Different cropping schemes are studied fixing the functional unit in 109kcal of energy produced, because this is the energy obtained from approximately 100ha of land. Direct cropping technique is assumed. An energy functional unit is the most suitable to evaluate a system where different outputs are found, namely barley grain, wheat grain, rapeseed seed, cake and oil.

The system boundary includes an agricultural exploitation where different crop types are considered. The fate of the obtained products is not considered, only the energy that each obtained product represents. The boundaries comprehend (i) materials inputs which take into account fertilizers, herbicides, insecticides, fungicides, diesel fuel and planting seeds, (ii) cropping stages including fertilizing, herbicide, insecticide and fungicide treatments, sowing, harvesting and seed/grain transportation to cooperative installations, and (iii) rapeseed processing stage which includes transportation, pressing, filtering and degumming processes.

Three scenarios are considered for this environmental assessment, based on grouping three crop types, namely barley, wheat and rapeseed. Barley, wheat and rapeseed models consist on the production of the grain and seed. Additionally, rapeseed model incorporates the seed processing, to obtain rapeseed oil that can be used as biofuel (SVO) in the exploitation. The use of SVO as fuel is also considered in one scenario. Thus, the first scenario is the current exploitation method (current scenario). The second incorporates rapeseed into crop rotation but uses only diesel fuel (diesel scenario). The third additionally includes rapeseed processing and SVO fuel use (SVO scenario).

Emissions of the considered model are aggregated into impact categories according to an international accepted method in the impact assessment phase. CML method from the Environmental Sciences Institute of Leiden University is the method chosen in this study, because it is the one which generates more international consensus and avoids subjectivity (Guinee et al., 2001; Alvaro-Fuentes et al., 2009). It is a cause-effect method that limits the uncertainty in groups according to impact categories (Dreyer et al., 2003). It calculates the increase of damage and quantifies its effects (Garrain, 2009).

Fig. 5 shows the environmental impact category results using 6 CML non-toxicological impact categories, energy consumption and land use for each scenario taking the current one as a basis. The introduction of rapeseed in the classical rotation and its use to produce SVO for fuel self-consumption slightly lessens some of the environmental impacts considered. Crop energy ratio indicator shows a preference for SVO fuelled scenarios, being the ratio 21.6% superior for SVO scenario compared to the current and the diesel seed one. Adverse environmental impacts to SVO scenario (ODP and POCP) are just 8.5% and 9.8% worse than reference scenario. A slight land requirement increase in both diesel and SVO scenarios is obtained, but not much representative, being lower that 1.7%. Favourable environmental impacts to SVO scenario (ADP and GWP) are down to 21.2% and 6.2% lower than reference scenario. AP and EP are 2.7% and 1.9% lower than reference scenario, not being much representative. On the other hand, diesel scenario impacts compared to reference scenario varies less than 4%, being practically the same. The higher diesel scenario impact is global warming potential (GWP), which increases 3.7% whereas in SVO scenario lessens 6.2%. Sensitivity analysis carried out show practically no variation in tendency, however, the impact of the electrical energy use in SVO scenario can be reduced if renewable electrical energy is used.

The microdiesel concept initiated by Steinbuchel et al. can be combined with the management and reutilization of waste waters by the application of microbial lipases to transesterificate the lipids present in the dairy industry or urban wastewater sludges. The lipidic fraction of sludges from urban wastewater treatment represents between 17 and 30% of the dry weight. This lipidic fraction is formed by direct absorption of fats present in the water by the sludge particles and by the phospholipids released from the cell membranes of micro-organisms, as well as from metabolites and cell lysis by-products (Boocock et al., 1992; Shen & Zhang, 2003; Jarde et al., 2005).

Lipid-rich wastewaters require pretreatment in order to reduce the amount of lipids and ease the subsequent conventional treatment. The pretreatment is usually based on physical processes, the most common of which are fat traps, tilted plate separators (TPS), and dissolved air flotation (DAF) units. In addition, centrifuges and electroflotation systems are used occasionally (Willey, 2001). Fat traps are rectangular or circular vessels through which the wastewater passes under laminar-flow conditions, at a rate that allows the lipids to rise to the surface near to the outlet end of the trap. The separation principle is based on Stoke’s law, relating rising velocity of a particle to its diameter, so the theoretical separation efficiency is dependent on depth. In practice, fat traps have a depth of 1.5 m, although if the accumulation of a bottom sludge is expected, then an additional 0.5 m would be added to the total liquid depth. Gravity flow is preferred to pumping when feeding the trap, in order to minimize the wastewater emulsification. Fat traps are used in the food industry and in restaurants (Willey, 2001).

Meanwhile, tilted plate separators were developed in the petrochemical industry and are based on the fact that surface area, rather than depth, determines the oil separation. The introduction of tilted plates into a vessel provides many parallel gravity separators with a high surface to volume ratio in a shallow tank. Typically, TPS can occupy less than 10% of the area needed to install a conventional fat trap, although they have some disadvantages. They are susceptible to fouling if solid or semi-solid fat is present in the effluent and a crane is required to remove the plate pack for cleaning. Besides, the pumping systems have to be carefully selected and controlled to avoid surging and liquid depth fluctuations (Zeevalkink & Brunsmann, 1983; Willey, 2001). Finally, dissolved air flotation units are based on the flotation of lipids by means of microbubble clouds (60-70 pm bubble diameter) created by the injection into water of 6 bar pressure air through nozzles. Microbubbles attach to the surface of the fat and oil particles, increasing their rise rate. These systems are used both in the food industry (Willey, 2001) and in the mining wastewater treatment (Tessele, 1998). Once the pre-treatment has finished, wastewater can receive further treatment prior to its disposal or biological treatment. Thus, chemical treatment may be used to reduce the total fatty matter in wastewater. Such treatment uses aluminium sulphate, ferric chloride, or more usually, lime, to break the emulsion and coagulate the fat particles. Subsequently, the fats can be separated by flotation or sedimentation. The rate of sedimentation can be improved by a second-stage flocculation, involving the addition of low levels of polyelectrolyte (0.5-5.0 mg/l) to the wastewater once coagulation has taken place (Willey, 2001).

The use of sludges to produce biofuels is not a new idea itself, but the available literature focuses mainly on the methane production by anaerobic fermentation, currently applied in the majority of waste water treatment plants (WWTP) to provide energy to these installations, or for fermentative biohydrogen production, which is still not industrially available (van Groenestijn et al., 2002; Wang et al., 2003). Several groups have studied the in situ transesterification of WWTP sludges, but have focused on the chemical catalysis of the transesterification with methanol (Haas & Foglia, 2003; Mondala et al., 2009). However this method still presents the same limiting factors that affect the chemical transesterification of edible vegetable oils (Freedman et al., 1984). In spite of their chemical approach, these works provide useful information about several aspects of the biodiesel production process, especially at the first stages of the process. Thus, one common problem of chemical and enzymatic biodiesel production is the need for the pretreatment of the feedstock to make its lipids easily available to the catalyst. In the case of wastewater treatment sludge, this pretreatment step usually implies the use of organic or non-polar solvents to release the lipids from the organic matter (Antczak et al., 2009; Siddiquee & Rohani, 2011).

The most extended protocols rely on chloroform:methanol mixtures, as used in the Folch’s method (Folch, 1957), in which a 2:1 chloroform:methanol reactant is mixed with the sample, where water acts as ternary component to form an emulsion. After equilibration with a fourth volume in saline solution, the emulsion separates in two phases: the lower one containing chloroform:methanol:water in the proportions 86:14:1 alongside the lipids; and the upper one containing the same solvents in proportions 3:48:47 and carrying the non — lipidic components of the sample. Bligh and Dyer’s method is a simplified variant of the former, but requires the re-extraction of the sample residue with chloroform (Bligh & Dyer, 1959). Nevertheless, there are some methods with near to Folch’s reagent yielding which use less toxic reagents, such as pure hexane or different combinations of hexane and other solvents, such as the hexane-isopropanol (3:2) blend proposed by Hara and Radin (Hara & Radin, 1978), or the ethyl acetate-ethanol (2:1) mixture used by Lin et al. (Lin et al., 2004). For a detailed revision of the solvents based extraction protocols, see Kuksis, 1994; Murphy, 1994; Kates, 1996.

In spite of being slightly less toxic than chloroform, the cited solvents are hazardous and present enough management risks to consider other extraction strategies. Several authors propose solvent-free methods based on ionic liquids (Ha et al., 2007), boiling the sludge or subjecting it to supercritical gases, mainly f-butanol (Wang et al., 2006; Royon et al., 2007), propane (Rosa et al., 2008), syngas (Tirado-Acevedo et al., 2010) and CO2 (Helwani et al., 2009), or even to extreme pressures and temperatures (cracking) (Saka & Kusdiana, 2001). All of them are costly and not feasible with the current technology (Siddiquee et al., 2011). A more realistic and ready-to-use option is extraction using hot ethanol, which can be used to perform the lipids’ extraction without using coadyuvant solvents. This approach to extraction can be illustrated with the works developed by Holser and Akin (2008) or Nielsen and Shukla (2004), among others. Although these authors have focused on the ethanol — based extraction of high value lipids from flax processing wastewater and egg yolk powder, respectively, their findings could be scaled and applied to biodiesel production from wastewater sludges. Nielsen and Shukla found that the use of ethanol at room temperature led to the extraction of nearly all the phospholipids, together with cholesterol and a minor part of the triacylglycerols, without special extraction and filtration devices. On the other hand, Holser and Akin performed a serial extraction of the lipids present in flax wastewater in three steps, under different temperature values (50, 80, 90 and 100°C). They found that the most efficient extraction was achieved when the sample-ethanol mixture was heated to 90°C and the reaction time was 15 minutes (Holser & Akin, 2008).

Considering the above findings, and taking into account the fact that the enzymatic production of biodiesel generally requires high alcohol to oil ratios, to improve the solubilization of the lipids and the formation of water-oil, enzyme-oil and enzyme-alcohol — oil interfaces, we propose that a suitable scheme for the production of biodiesel from WWTP sludge could be as simple as using a pressurized pretreatment tank, where the sludge is soaked in ethanol, kept at 90°C under stirring and refluxed to subject the mixture to three extraction cycles. This is followed by incubation in a reaction tank where the extracted lipids, alongside with part of the ethanol used in the previous step, are added to a reaction mix containing the enzyme (free, immobilized or whole cell cafalysf) and kept at the optimal temperature and pH conditions, to ensure both enzyme stability and an acceptable microdiesel production rate. Heat exchangers between the two tanks could serve to save energy, using the heat released before entering the second tank to preheat the sludge before entering the first one.

The system could even be autonomous in terms of ethanol requirements if the engineered microorganism used to produce the lipase was able to produce ethanol simultaneously, or if the cited tanks were coupled with a third reactor where bioethanol was produced from sugars present in the non-lipidic products obtained in the pretreatment tank by means of ethanol-producing yeast or bacteria strains. In the case of economic restrictions, some short­term cost reduction could be achieved by replacing the pretreatment pressurized tank with a non-pressurized unit, and keeping the temperature of the extraction mix below 79°C, although it would imply medium-term economic losses because the lower extraction efficiency must be compensated by performing more extraction cycles at a higher reflux rate and a greater ethanol volume in the pretreatment tank, or even by the use of at least two serial pretreatment tanks.

3. Conclusion

As a short-term response to the consequences of greenhouse gas emissions and the unsustainability of the fossil fuel-based energy model, the industry has developed ready-to — use substitutes for traditional fossil fuels, delivered generally and ambiguously under the commercial ‘bio’ denomination. However, the first — and second-generation of so-called biofuels are neither of completely biological origin nor based on renewable and environmentally friendly feedstocks. In addition, the production techniques rely on high energy inputs, both in feedstock production (as is the case for rapeseed, soybean or palm oil) and in the biofuel synthesis (acid catalyzed biodiesel or corn bioethanol perfectly illustrate the neat energy gain problems). Alongside these problems, new and complex problems have emerged. Firstly, the increase in the prices of grain and vegetable oils used both to produce biofuels and for human nourishment and livestock feeding; and secondly, the expansion of agricultural land to increase production of sugar cane or vegetable oils to satisfy the huge demand for these sugar and lipid sources, generated by the abrupt increase in biofuels production. Thus, the development of cleaner and more sustainable biofuels is required to achieve the challenge to totally replace traditional fossil fuels by third-generation biofuels, independent of non-renewable precursors or inefficient industrial processes, that damage the environment directly and indirectly and threaten biodiversity and food security (UNCTAD, 2010).

A great variety of domestic, agricultural and industrial residues, from lignocellulosic forestry and agriculture waste to fatty acid rich waste waters, generated by the dairy, poultry or vegetable oil refinery industries, as well as the sludges from urban waste waters, can be used as precursors of biofuels. The treatment of these residues could be combined with the production of third-generation biofuels by enzymatic catalysis because the high cost of enzymes could be compensated by the low cost of the residues (or even the presence of incentives for residue reduction and management). But the massive application of these concepts requires a series of technical and biotechnological improvements, such as the optimization of lipids and sugars extraction, feedstock pretreatment processes, biofuels production plant design, heterogeneous catalysts and enzyme immobilization techniques, protein engineering of lipases, alcohol dehydrogenases or hydrolases to increase their activity and reusability, genetic engineering of microbes to facilitate both the pretreatment of precursors, and the synthesis and purification of the biofuels.

4. Acknowledgement

We thank the Junta de Andalucia (Spain) for funding this study through project reference P08-RNM-04180 and the Spanish Ministry of Science and Technology for funding through project reference CTM2009-09270. M. Manzanera received grants from the Programa Ramon y Cajal, (Ministerio de Educacion y Ciencia MEC, Spain, and ERDF, European Union).

It would be ideal to realize biogas production from the liquid phase only — it would be possible to introduce high performance UASB (Upflow Anaerobic Sludge Blanket) digesters and to achieve the large saving of technological volumes but the concentration of substances in the liquid phase should have to be increased. The solid phase of substrates, which cannot be applied as an organic fertilizer after the fermentation process, would be used as biomass for the production of solid biofuels in the form of pellets or briquettes. But it would be necessary to reduce its chlorine content to avoid the generation of noxious dioxins and dibenzofurans during the burning of biofuel pellets or briquettes at low burning temperatures of household boilers and other low-capacity heating units. Wachendorf et al. (2007, 2009) were interested in this idea and tried to solve this problem in a complex way by the hot-water extraction of the raw material (at temperatures of 5°C, 60°C and 80°C) followed by the separation of the solid and liquid phase by means of mechanical dehydration when a screw press was used. This procedure is designated by the abbreviation IFBB (Integrated Generation of Solid Fuel and Biogas from Biomass). These researchers successfully reached the transfer ratio of crude fibre from original material (grass silage) to liquid phase only 0.18, which is desirable for biogas production, but for more easily available organic substances influencing biogas production, e. g. nitrogen-free extract, the ratio is 0.31. The transfer of potassium, magnesium and phosphorus to the liquid phase ranged from 0.52 to 0.85 of the amount in fresh matter, calcium transformation was lower, at the transfer ratio 0.44 — 0.48 (Wachendorf et al. 2009). Transformation to the liquid phase was highest in chlorine, 0.86 of the amount in original fresh matter, already at a low temperature (5°C). The transfer of mineral nitrogen to the liquid phase before the process of anaerobic digestion is very low because there is a minute amount of mineral N in plant biomass and the major part of organic matter nitrogen is bound to low-soluble proteins of the cell walls. Nitrogen from these structures toughened up by lignin and polysaccharides is released just in the process of anaerobic digestion. Because in the IFBB process also organic nitrogen compounds (crude protein — nitrogen of acid detergent fibre ADF) are transferred to the liquid phase approximately at a ratio 0.40, the liquid phase, subjected to anaerobic digestion, is enriched with mineral nitrogen.

Like Wachendorf et al. (2009), we proceeded in the same way applying the IFBB system for the parallel production of biogas and solid biofuels from crops grown on arable land. The IFBB technological procedure is based on a high degree of cell wall maceration as a result of the axial pressure and abrasion induced with a screw press.

Carlos A. Grande

SINTEF Materials and Chemistry, Oslo

Norway

1. Introduction

Biogas is a raw gaseous stream produced by anaerobic decomposition of organic matter. The main component of biogas is methane, reason why this stream is considered to be a renewable source of energy and fuel. The most positive aspects of biogas rely on its worldwide decentralized production and on the environmental benefits of avoiding methane emissions to atmosphere while using bio-methane it to replace fossil fuels.

In order to use the energy obtained in biogas, its production should be controlled. The production of biogas from organic matter is a complex process involving many different bacterial groups. In a simple way, the entire biogas conversion from organic matter can be divided into four steps (Gavala et al., 2003; Demirbas et al., 2011):

1. Hydrolisis: complex organic molecules are hydrolyzed into smaller units (sugars, amino-acids, alcohols, fatty acids, etc.

2. Acidogenese: acidogenic bacteria further break down the molecules into volatile fatty acids, NH3, H2S and H2.

3. Acetanogese: the acetanogens transform the molecules into CO2, H2 and mainly acetic acid.

4. Metanogese: at the end of the process, the methanogenic archaea transform the H2 and acetic acid molecules into a mixture of CO2, CH4 and water.

These production steps can be controlled in reactors (digesters) or are naturally occurring in landfills that can be optimized for collection of biogas (see www. epa. gov/lmop). The digesters can operate in mesophilic and thermophilic modes, which means that the biogas is generated at 293-313 K and 323-333 K, respectively (Gavala et al., 2003). Biogas generation in landfills mainly operates in psychrophilic conditions (285-290 K) (Monteiro et al., 2011). Biogas main constituents are methane, carbon dioxide, sulphur compounds (H2S, siloxanes), water and minor contaminants (O2, N2, ammonia, chlorine, fluorines, etc) (Wellinger, 2009; Pettersson and Wellinger, 2009, www. epa. gov/lmop). The final composition of biogas is variable and strongly depends on the source of organic matter (Pettersson and Wellinger, 2009). Major sources of biogas production are landfills, waste-water treatment plants, manure fermentation and fermentation of energy crops. The composition of the biogas obtained from these sources is given in Table 1. For comparison, the composition and properties of natural gas are included in Table 1. The methane content vary strongly due to the different kind of molecules processed: i. e, fat has a much higher bio-methane yield than carbohydrates. The biogas yield of cereal residues is also high (around 200 m3 CHi/ton)

(Pettersson and Wellinger, 2009), representing an interesting opportunity for fermentation in farms around the world.

Since the content of methane in biogas streams is higher than 50%, its emissions to atmosphere result in a waste of an efficient hydrocarbon which also has a greenhouse warming potential 23 times higher than CO2. For this reason, adequate collection and utilization of the bio-methane contained in biogas avoid important emissions of methane to atmosphere and also transforms this stream in a valuable source of renewable energy. The bio-methane can be directly burned and converted to electric power. In fact, biogas utilization for production of electric energy is increasing worldwide.

An alternative application of biogas is its upgrading to obtain purified bio-methane that can be either injected in the natural gas grid or be directly used as vehicle fuel. The purification of biogas involves several steps: removal of impurities (sulphur compounds mainly), water removal and the upgrading process which consist in bulk removal of CO2. Normally the removal of sulphur compounds is the first step in biogas cleaning. The order of separating H2O and CO2 depends on the specific technology employed for upgrading. The bio-methane obtained as product should hold certain purity and the maximum quantity of CO2 allowed is between 2-4% depending on the legislation of each country. From all the purification steps, the bulk removal of CO2 is the most expensive one and is the one that will be considered in this Chapter.