Potential bioenergy feedstocks were selected after in-country discussions and government’s indications: sugarcane and cassava for bioethanol and palm oil for biodiesel. The suitability index (resulting from a geographic information system) for the cassava case in the framework of tillage-based production is shown in Figure 12.2. It is seem that very suitable or suitable areas practically do not exist in Tanzania. Potential for cassava growing is in the range of moderately suitable.

To delineate the industrial configurations that would be more adaptable to the Tanzania context, globally available bioethanol technologies were reviewed and analyzed against the technology and human capacity assessment (Sanchez and Cardona, 2008). Three technological configurations were designed as shown in Table 12.5. The configurations are differentiated based on the level of complexity of the technologies involved in each of the main processing steps. Two types of raw materials were considered as shown in Table 12.6.

In the production cost of cassava ethanol, the feedstock was also the high­est factor contributing between 65 to 71% of the total cost. The production cost from fresh cassava feedstock is slightly lower than processed dried chips (see Table 12.7). Taking into consideration the infrastructural limitations in Tanzania

TABLE 12.6

Bioethanol Production Scenarios for Tanzania

Raw Material scenario Parameter Description

Cassava 7 Stand alone Fresh cassava (single plant), increased cassava yield

Cassava 8 Stand alone Dried cassava (single plant), increased cassava yield

and the fact that cassava roots perish quickly after harvesting, this production route may not be the most appropriate. Thus, scenario 8 provides a more viable alternative whereby fresh cassava roots are first dried to extend the shelf life, then are transported to an ethanol processing plant. Scenario 8 facilitates greater opportunity for small farmers in isolated rural areas to participate. The costs pre­sented in Table 12.7 do not include co-generation or use of by-products.

Ruvuma is a representative region in Tanzania located at the southern bound­ary with Mozambique. The welfare effects in this region were analyzed based on a 10% of producer price increase for different crops including cassava. They found positive welfares in specific cases for cassava. For bean and sugarcane, the welfare effects were always negative.

Concluding partial results showed that in order to be able to reap the benefits of bioenergy investments, Tanzania has to consider strengthening and developing local markets, local production capacity, and its infrastructure. Analysis so far does not include all crops and full integration across modules. The results are under discussion in the country.

The preliminary conclusions drawn by the BEFS project indicate that bioen­ergy development, which safeguards food security, is only sustainable in Tanzania when a bioenergy project:

• Does not hinder the natural resource base

• Involves smallholders, increases employment, and takes into account the specific risks for subsistence farmers

• Increases access to markets and infrastructure

• Builds domestic skills and expertise

• Ensures local benefits and sustainability of the industry

• Monitors welfare impacts at the household level

• Respects and protects the livelihoods of women

• Strengthen farmers’ negotiating power

• Further enhances institutional capacity

Although the thermodynamics imposes constraints to different chemical pro­cesses, it also sets itself up as a powerful source of design guidance for generating alternative technological configurations with better performance. For instance, the consideration of the second law of thermodynamics for analyzing the energetic efficiency of technological schemes through the exergy balance allows evaluating and choosing the best alternative configuration for a given process. This balance takes into account that the energy always will be degraded, which implies a lost of work by the mass and energy flows entering or leaving the system, according to the second principle of thermodynamics (Bastianoni and Marchettini, 1997). Sorin et al. (2000) proposed the application of some indexes based on the energy balance, such as the utilizable exergy coefficient and local contribution of exergy for each operation. They used them as criteria for reducing a superstructure of technological schemes in the case of benzene production from cyclohexane. The process synthesis methodology developed allowed obtaining a technological con­figuration with better performance with respect to the feedstock consumption, amount of emissions, and utilizable exergy coefficient compared with process synthesis procedures based on hierarchical decomposition or mathematical pro­gramming for the same benzene production process. In fact, the analysis of the second law of thermodynamics is the base for exploration and generation of tech­nological alternatives in the framework of the algorithmic methods for process design (Seider et al., 1999).

For synthesizing sustainable and environmentally friendly processes, the employment of the emergy concept has been proposed. This approach exploits the fact that industrial systems also obey the laws of nature as ecological systems. Since all the materials and services are transformed and stored forms of solar energy, the amount of solar energy directly or indirectly used to make any prod­uct can be employed as a measure of the ecological input or investment in that product or service. In this way, the incorporated solar energy (emergy) is used as a comparison standard for process synthesis (Bakshi, 2000).

Physical-chemical methods of pretreatment are remarkably more effective than physical ones. The most employed physical-chemical methods are presented in Table 4.3. Steam explosion (autohydrolysis) is the most commonly used method along with dilute-acid process for pretreatment of lignocellulosic materials. The use of saturated steam at high pressure causes autohydrolysis reactions in which part of the hemicellulose and lignin are converted into soluble oligomers with the help of some acids released from the biomass itself during the process. The

Physical Methods for Pretreatment of Lignocellulosic Biomass for Ethanol Production

Physical-Chemical Methods for Pretreatment of Lignocellulosic Biomass for Ethanol Production

Physical-Chemical Methods for Pretreatment of Lignocellulosic Biomass for Ethanol Production

factors affecting steam explosion pretreatment are residence time, temperature, chip size, and moisture content (Sanchez and Cardona 2008). In this process, the combined action of temperature and contact time of steam with the biomass is achieved. To quantify this effect, the severity index has been defined (Shahbazi et al., 2005; Soderstrom et al., 2003), and the pretreatment severity is described as a function of time t (in min) and temperature T (in degrees Celsius) related to a reference temperature of 100°C:

(4.1)

If the steam explosion process is carried out under acidic conditions (which increases the efficiency of cellulose hydrolysis), an additional term should be intro­duced in Equation (4.1) to consider the effect of pH on the combined severity (CS):

CS = log R0 — pH

The pH can be calculated from the amount of sulfuric acid added to the material and its water content (Soderstrom et al., 2003).

An important factor to be considered when pretreatment methods are used (including steam explosion) is the particle size of the lignocellulosic materials. The conventional mechanical methods require 70% more energy than steam explosion in order to achieve the same reduction in the particle size. During steam explosion, some inhibitors of the subsequent biological processes (enzymatic hydrolysis, fermentation) are formed. Inhibitors formation during the pretreat­ment requires the washing of pretreated biomass. This decreases the global yield of the saccharification due to the removal of sugars generated during the hemi — cellulose hydrolysis. Typically, 20 to 25% of the initial dry matter is removed by the washing water (Sun and Cheng, 2002). For this reason, the use of very small particles in some cases (e. g., herbaceous wastes) is not desirable if taking into account the economy of the process (Ballesteros et al., 2002).

Steam explosion is recognized as one of the most efficient methods for hard­wood (poplar, oak, birch, maple) and agroindustrial residues, but it is less efficient for softwood (pine, cedar; Sanchez and Cardona, 2008). For instance, an increase of 90% in the efficiency of the subsequent enzymatic hydrolysis of poplar chips pretreated by steam explosion was reported compared to 15% efficiency when pre­treatment of chips is not carried out (Sun and Cheng, 2002). In the case of cane bagasse pretreatment, Kaar et al. (1998) determined the conditions that maximize sugars concentration by varying the temperature within the range of 188 to 243°C and residence time within the range of 0.5 to 44 min. These authors concluded that these conditions strongly depend on the composition of the lignocellulosic mate­rial and it demonstrated the formation of furfural. On the other hand, it has been reported that the susceptibility of the pretreated substrate to the action of cellulases

Process Synthesis for Fuel Ethanol Production

is highly influenced by the steam pressure and vaporization time during the pre­treatment, as was demonstrated for rice straw. In particular, a steam pressure of 3.53 MPa during a short vaporization time (2 min) significantly increases the enzy­matic hydrolysis without any observable inhibitory effect (Moniruzzaman, 1996).

For the case of softwood that has an increased lignin content and is more difficult to degrade, a two-stage steam pretreatment has been proposed. In the first stage, the operating conditions are defined in such a way that the maximum amount of sugars derived from hemicellulose is obtained. In the second stage, more severe conditions are employed to degrade the solid fraction resulting from the first stage achieving a partial hydrolysis of cellulose. In both stages, the softwood sawdust is impregnated with dilute sulfuric acid. Shahbazi et al. (2005) propose a fractionation procedure for softwood based on steam explo­sion and alkaline delignification in order to produce ethanol and related co­products. An analogous fractionation procedure was utilized by Belkacemi et al. (2002) where the captured hemicellulose-rich liquor was enzymatically treated to produce xylose-rich solutions. Regarding the enzymatic hydrolysis of hemicellulose, Saha (2003) points out that there are no suitable commercial hemicellulase preparations that can efficiently hydrolyze feedstocks like corn fiber to monomeric sugars. This author also briefly reviews the microorganisms and enzymes that could be useful for degrading hemicellulose. Other analogous schemes involve two hydrolysis stages (Nguyen et al., 1999) or an initial treat­ment by steam explosion followed by an acid hydrolysis to completely degrade the xylans with a further acid recovery (Saska and Ozer, 1995).

One of the methods with better indexes is the pretreatment with liquid hot water (LHW) or thermohydrolysis. Laser et al. (2002) mention that under optimal conditions, this method is comparable to the dilute acid pretreatment, but with­out the addition of acids or production of neutralization wastes. In addition, this method presents elevated recovery rates of pentoses and does not generate inhibi­tors (Ogier et al., 1999). Nevertheless, solid load for this method is much less than for the steam explosion method, which is usually greater than 50%.

Another physical-chemical method is the ammonia fiber explosion (AFEX) process whose function is similar to steam explosion. The pretreatment with ammonium does not generate inhibitors for subsequent biological processes, so the washing with water is not necessary. In addition, a small particle size is not required. For this method, Dale et al. (1996) report experimental data correlat­ing conditions of the AFEX process, enzyme doses during cellulose hydrolysis, and the corresponding yields for several agricultural and lignocellulosic residues. Similarly to the AFEX method and steam explosion, CO2 explosion uses the same principle, but the yields are relatively low compared to the other methods (Sanchez and Cardona, 2008; Sun and Cheng, 2002).

One of the proposed technologies for the development of high-performance processes using starchy materials consists of the fermentation of high and very high gravity mashes. During high gravity fermentation, the solids concentration in the medium exceeds 200 g/L, which implies a high substrate load and, con­sequently, high ethanol concentrations at the end of fermentation. Furthermore, a lower amount of process water is required as well as lower energy demands. However, this process implies more prolonged cultivation times and, sometimes, incomplete fermentation due to end-product inhibition, high osmotic pressure, and inadequate nutrition (Barber et al., 2002). To accelerate high gravity fermen­tations, the controlled addition of small amounts of acetaldehyde during the fer­mentation allows the reduction in cultivation time from 790 h to 585 h for initial glucose concentration of 300 g/L without effect on ethanol yield. It is believed that this positive effect may be caused by the ability of acetaldehyde to replenish the intracellular acetaldehyde pool and restore the cellular redox balance (Barber et al., 2002). Continuous operation can improve the performance of high grav­ity fermentations. In particular, continuous fermentation can reach almost the same ethanol production as batch fermentation, although it is likely that there is a threshold concentration of initial glucose above which an ethanol yield no longer increases (Zhao and Lin, 2003).

For industrial ethanol production, fermentation of wheat mashes of very high gravity (VHG) has been proposed. These mashes consist of wheat starch hydro — lyzates containing 300 g or more of dissolved solids per liter of mash. VHG fer­mentation technology enables high ethanol concentrations to be obtained from very concentrated sugar solutions. To this aim, very low levels of dissolved oxygen are required as well as nitrogen sources that do not limit the cell growth, such as urea or ammonium salts (Jones and Ingledew, 1994a). In this way, 21.1% (by volume) ethanol concentrations are obtained in only four days of fermentation from VHG wheat mash. One of the strategies employed is the addition of com­mercial proteases during the VHG fermentation in order to release amino acids from soluble proteins contained in the wheat mash compensating, in this way, the addition of exogenous nitrogen sources (Jones and Ingledew, 1994b). Thomas et al. (1996) emphasize that considerable amounts of water can be saved by applying this technology to fuel ethanol production. Moreover, the implementation of VHG fermentation increases the throughput rate of an ethanol plant without the need of increasing the plant capacity. These authors provide a theoretical method for pre­dicting the maximum concentration of ethanol in fermented mash that takes into account changes in the weight and volume of mash during fermentation. Bayrock and Ingledew (2001) designed and tested a system that combines the multistage continuous culture fermentation and the VHG cultivation for feed containing 150 to 320 g/L of glucose using S. cerevisiae. The maximum ethanol concentration obtained in the process was 132.1 g/L indicating the feasibility of implementing this technology in the industry, particularly in the continuous production of etha­nol from wheat starch (Sanchez and Cardona, 2008).

VHG technology has been tested with successful results for oats, barley, rye, and triticale, as cited by Wang et al. (1999). The pretreatment of feedstock can play an important role when process integration is analyzed during this type of fer­mentation processes. Wang et al. (1999) propose the integration of a pretreatment process, pearling by abrasion of cereal grains such as rye or triticale, with VHG fermentation technology. The pearling of cereals removes approximately 12% of grain dry matter, which increases its starch content to 7 to 8%. This increase in starch content combined with the employment of high concentrations of sugars, which is the main feature of VHG fermentation, allows the increase in the final ethanol concentration of 64% in comparison to the use of nonpearled grains in conventional fermentations.

Reaction-reaction integration allows for the increase of process efficiency through the improvement of reaction processes. However, separation is the step where major costs are generated in the process industry. Therefore, reaction-separation integration could have the highest impact on the overall process in comparison with homogeneous integration of processes (reaction-reaction, separation-sep­aration; Cardona and Sanchez, 2007). The reaction-separation integration is a

TABLE 9.5

Integration of Reaction-Reaction Processes by Consolidated Bioprocessing (CBP) for Fuel Ethanol Production from Different feedstocks

particularly attractive alternative for the intensification of alcoholic fermentation processes. When ethanol is removed from the culture broth, its inhibition effect on growth rate is diminished or neutralized leading to a substantial improvement in the performance of ethanol-producing microorganisms. This improved perfor­mance can permit the increase of substrate conversion into ethanol. In particular, higher conversions make possible the utilization of concentrated culture media (with sugar content greater than 150 g/L) resulting in increased process produc­tivities. From an energy viewpoint, this type of integration allows the increase of ethanol concentration in the culture broth. This fact has a direct effect on distil­lation costs since more concentrated streams feeding the columns imply lower steam demands for the reboilers and, therefore, lower energy costs.

For these reasons, most of the proposed configurations using reaction-sepa­ration integration are related to the ethanol removal by different means including the coupling of different unit operations to the fermentation or the accomplish­ment of simultaneous processes for favoring the in situ removal of ethanol from culture broth.

Through process simulation, the yield of an ethanol production facility from cas­sava can be assessed. This acquires greater importance considering the lack of information published about this type of process. In a previous work (Cardona et al., 2005a), the performance of two processes for bioethanol production employing two different starchy feedstocks using a commercial simulation package (Aspen Plus) and especially developed models were compared. For this, a SSF was con­templated in both processes. For corn, dry-milling technology and an organiza­tion of streams, such as that shown in Figure 11.7, was considered. This flowsheet allows the production of a valuable co-product, the DDGS. In the case of cas­sava, the utilization of fresh roots was considered. This implies the generation of a fibrous residue.

The comparison results are summarized in Table 11.10. Presented data show that the higher yields correspond to the corn due to its higher starch content mea­sured in wet basis. These data are close to ethanol yields from corn and cassava reported by the FAO (for instance, these data were published by Observatorio Agrocadenas Colombia, 2006). The production and protein content of the material to be used as a protein supplement in animal feed in the corn case (DDGS) are higher than in the case of cassava (fibrous residue) due to its low protein content. On the other hand, the high moisture content of the cassava implies that the con­densates of the stillage evaporation step cannot be recycled to the cooking step. In fact, this explains the need for employing a higher amount of feedstock to achieve the same starting starch mass as in the case of corn. However, for some countries, cassava represents a better option considering the high agronomic yields com­pared to corn. For instance, the cassava yield in Colombia reaches 20 ton/ha while average corn yield only reaches 6 ton/ha (Observatorio Agrocadenas Colombia, 2006). This situation would represent ethanol yields per hectare of 3,336 L EtOH/ ha cassava and 2,105 L EtOH/ha corn. These estimations clearly favor the utiliza­tion of the tuber especially if the agroecological conditions of this country are taken into account.

Note: The mass flowrate of starch in each process is 30,675 kg/h. a Solids equivalent to DDGS.

Considering the very probable depletion of liquid fossil fuels toward 2090 and the start of declination of oil production in 2020 through 2030, humankind will face the huge challenge of maintaining its economic growth and stable technological development without compromising the welfare of the future generations (sus­tainable development). In addition, the quality of life for people from developing and underdeveloped countries should be improved without compromising the life level of those people in developed nations. Kosaric and Velikonja (1995) point out that the solution to this problematic situation depends on how mankind develops and implements viable technologies for the industry, transport sector, and heat­ing based on alternative (renewable) fuels and feedstocks as well as ensuring the availability of sufficient amounts of renewable resources (energy and raw materi­als). Furthermore, man should develop and implement technologies for reducing the environmental pollution and CO2 emissions. For these reasons, the renewable energies may partially or totally replace the fossil fuels, especially if humankind does not choose the dangerous pathway toward the global development of nuclear energy as a primary source of energy.

The renewable energy sources correspond to those kinds of energy that are obtained from natural sources, which are practically inexhaustible due to the huge amount of energy that they contain and to their ability to regenerate themselves by natural means. Among the energy sources of one type, solar, wind, hydraulic, geothermal, and tide energy should be highlighted. The energy sources of a sec­ond type correspond to the bioenergy or energy from biomass. One of the main features of the renewable energies is that their utilization does not imply the net generation of polluting emissions, which contribute to the greenhouse gas effect or the destruction of the ozone layer. The renewable energy sources represent 7.68% of the total energy consumed in the world, with the biomass being the most exploited resource (Energy Information Administration, 2008). Two thirds of the biomass is used for food cooking and heating in developing countries (tra­ditional use of the biomass), e. g., through the use of firewood. The remaining third corresponds to the commercial use of the biomass for the industry (e. g., cane bagasse as energy source in sugar mills), generation of electricity (e. g., wood chips feeding small thermal plants), and transport sector (liquid biofuels produc­tion). The hydroelectric energy is the second most important renewable resource, whereas the contribution to the global energy consumption from such sources as the sun, winds, tides, and geothermal energy is marginal. A 59.2% increase in the consumption of renewable energy (corresponding to 828 million tons of oil equivalent) is expected in the period 2002 to 2030 (IEA, 2004).

One renewable solution in the search for alternative sources of energy for the world populace is the use of solar energy in the form of biomass (bioenergy). The global potential of bioenergy is represented by the energy-rich crops (mostly rep­resented by agroenergy) and lignocellulosic biomass (including the dendroenergy from forest activities). The conversion of these feedstocks into biofuels, either for electricity generation or for their use in vehicles, is an important option for exploi­tation of alternative energy sources and reduction of polluting gases (Sanchez and Cardona, 2008b), mainly CO2. The emissions generated by the combustion of biofuels are offset by the CO2 absorption during the growth of plants and other plant materials from which these biofuels are produced. In this way, the biomass utilization releases the carbon dioxide that was fixed during its growth, compen­sating the emissions generated in the current scale of time. In contrast, fossil fuel usage releases into the atmosphere the carbon dioxide that was fixed by the plants million of years ago, which implies a net increase in the amount of atmospheric CO2, provoking global warming.

Energy-rich crops comprise those crops that could be exclusively addressed to the energy production either as solid fuels for electricity generation or as liquid biofuels that can substitute for the fossil fuels (bioethanol, biodiesel). It is esti­mated that one hectare of energy-rich crops employed for liquid biofuel production can avoid the emissions of 0.2 to 2.0 tons of carbon into the atmosphere compared to the use of fossil fuels (Cannell, 2003). For the case of ethanol obtained from sugarcane in Brazil, its use may offset carbon dioxide emissions at a rate of 2 t C/ (Ha/year) related to the oil. This substitution is more appreciable in tropical coun­tries, whereas the offset is more effective for European countries if the electricity is produced from biomass. Kheshgi and Prince (2005) indicate that if the CO2 released during the alcoholic fermentation is captured and injected into the sub­soil or deeply in the ocean (where it is dissolved), ethanol production may lead to a net carbon dioxide removal from the atmosphere (CO2 sequestration) avoiding the emissions generated by the gasoline usage. Eventually, the environmental benefits can be enhanced if the feedstock employed is made up of residues or wastes.

The so-called lignocellulosic biomass includes agricultural, forestry, and municipal solid residues as well as different residues from agro-industry, the food industry, and other industries. The lignocellulosic biomass is made up of complex biopolymers that are not used for food purposes. The main polymeric components of biomass are cellulose, hemicelluloses, and lignin. For their conversion into a liquid biofuel such as ethanol, a complex pretreatment process is required in order to transform the carbohydrate polymers (cellulose and hemicellulose) into fermentable sugars. In contrast, for electricity generation, only the combustion of the biomass is needed.

It is evident that lignocellulosic biomass as feedstock for energy production is important. The lignocellulosic complex is the most abundant biopolymer on Earth and is present in such profuse materials as wood, sawdust, paper residues, straw, and grasses (Sanchez and Cardona, 2008b). It is estimated that the lignocel­lulosic biomass makes up about 50% of world biomass and its annual production has been estimated at 10 to 50 billion tons (Claassen et al., 1999). For instance, 35% of the material collected during the total wheat harvest in Europe corre­sponds to the straw, whereas 45% corresponds to the grain. In addition to the energy generation, the biomass utilization allows the economic exploitation of a wide range of residues from domestic, agricultural, and industrial activities. One of the main advantages of using lignocellulosic biomass is that this feedstock is not related to food production, which would permit the energy production without the utilization of a great number of hectares of land for cane, corn, or cassava pro­duction. Furthermore, the biomass is a resource that can be processed in different ways to produce a significant variety of products such as ethanol, synthesis gas, methanol, hydrogen, and electricity (Chum and Overend, 2001). However, some authors, such as Berndes et al. (2001), consider that the great scale implementa­tion of biomass energy would create serious social, economic, and environmental consequences, especially if dedicated energy crops are employed. For example, these authors estimate that the labor requirements for bioenergy production at a great scale in any country should not exceed 1% of the total labor force in order to make its production feasible. Grassi (1999) points out that the development of bioenergy production technologies in the European Union (EU) can represent the creation of 200,000 direct and indirect jobs as well as the reduction of 255 million annual tons of CO2 by 2010. Thornley (2006) summarizes the main environmen­tal, social, and economic advantages of employing biomass as an energy source. Some of these advantages are applicable to both developed and developing coun­tries. Among them, the reduction of greenhouse gas emissions, reduced use of agro-chemicals, diversification of rural economies, the potential for low-cost heat supply, and the potential income streams for farmers should be highlighted. On the other hand, this author presents some of the main consequences of bioen­ergy development: impacts on particular native species, the visual impact of crop growth and conversion plants, environmental emissions associated with thermal conversion plants, the uptake of significant amounts of water from below ground, and the requirement for policy support, among others. The need for policies that stimulate the development of technologies based on biomass through tax exemp­tions or subsidies should be emphasized considering the lower production and transport costs of fossil fuels.

Among the world’s foremost sugar producers (presented in Table 3.4), the United States, France, Germany, and Russia produce their sugar mostly from sugar beet. In the case of the United States, 53% of its sugar production comes from sugar beet and the balance comes from sugarcane. China produces 7% of its sugar from sugar beet (700,000 ton on average). The remaining percentage corresponds to the cane mostly cropped in southeastern provinces of the country (10,000,000 ton on average; FAS, 2005).

Sugar beet contains from 12 to 15% sucrose. The beets arrive at the produc­tion plant without crown and are loaded in silos by mechanical means through

channels with circulating water. The beets are washed and passed through sys­tems retaining diverse solid materials, such as stones, leaves, and small roots. Once washed, the beets are transported to the choppers where they are shredded into very thin slices (3 mm), called cossettes, and passed to a machine called a diffuser to extract the sugar content into a water solution. The diffusers are large rotary drums where the cossettes are put in contact with a hot water stream flow­ing in the opposite direction. The sucrose is extracted from the vacuoles of the beet cells into the flowing water generating the raw juice or diffusion juice. The spent cossettes are called pulp and leave the diffuser with about a 95% moisture content, but with low sucrose content. To recover part of the sucrose contained in the pulp, it is pressed in screw presses reducing the moisture to 75%. After dry­ing in rotary drums, the pressed pulp is sold as animal feed due to its high fiber content, while the resulting liquid is added to the diffusers. The diffusion juice is

an impure sucrose solution also containing gums, pectin, amino acids, mineral salts, nitrogenous compounds, etc. This juice has about 16° Brix and 85% purity. It can be clarified by a process very similar to that of cane sugar. Thus, a clear juice with 15° Brix is obtained. From this point, the process is practically the same as for cane sugar.

From the viewpoint of the conversion process, the most important factors to be taken into account for hydrolysis of cellulose contained in lignocellulosic materi­als are the reaction time, temperature, pH, enzyme dosage, and substrate load (Sanchez and Cardona, 2008). By testing lignocellulosic material from sugarcane leaves, Hari Krishna et al. (1998) have found the best values of all these param­eters varying in each experimental series the value of one of the factors while fix­ing the other ones. Sixty-five to seventy percent cellulose conversion was achieved at 50°C and pH of 4.5. Although enzyme doses of 100 FPU[1]/g cellulose caused almost a 100% hydrolysis, this amount is not economically justifiable. Hence, a 40 FPU/g cellulose dosage was proposed for which a 13% reduction in conversion was observed. Regarding the substrate concentration, solids loads of 10% were defined as the most adequate considering rising mixing difficulties and accumu­lation of inhibitors in the reaction medium. Though these conditions were deter­mined for a specific material pretreated by overliming and extrapolations to other lignocellulosic feedstocks are risky, found values are within the reported ones in the literature for a wide range of materials. Hydrolysis tests for steam-pretreated spruce also indicate the need of high enzyme loadings of both cellulases and P-glucosidase in order to achieve cellulose conversions greater than 70% due to the lower degradability of the softwood (Tengborg et al., 2001). Similar studies were carried out for saccharification of dilute-acid pretreated Douglas fir showing that enzyme dosage has a significant effect on glucose yield (Schell et al., 1999). Saha and Cotta (2006) obtained 96.7% yield of monomeric sugars using an enzy­matic cocktail of cellulase, P-glucosidase, and xylanase for saccharification of wheat straw pretreated by alkaline peroxide method. An ethanol concentration of 18.9 g/L and a yield of 0.46 g/g available sugars were achieved in the subsequent fermentation using a recombinant Escherichia coli strain capable of assimilat­ing both hexoses and pentoses. Jeffries and Schartman (1999) propose the use of sequential enzyme addition in a countercurrent mode for the saccharification of fiber fines from a paper recycling plant. Sequential addition of enzyme to the pulp in small aliquots produced a higher overall sugar yield per activity unit of enzyme than the addition of the same total amount of enzyme in a single dose. During the succeeding fermentation employing different yeasts, 78% ethanol yield of the theoretical maximum was obtained.

The culture broth can contain or not cells of yeast or other ethanol-producing microorganisms. This stream is called fermented wort (especially when sugarcane is the feedstock employed), fermented mash (when cereal grains are used), or sim­ply wine or beer. Besides ethanol and depending on the raw material employed, other substances can be found in a culture broth, such as:

• Nonfermented sugars

• Oligosaccharides resulting from the incomplete saccharification of starch or cellulose

• Ground and spent cereal grains

• Lignin (in dependence on the biomass pretreatment method)

• Other fermentation products like glycerol

• Lactic acid produced from contaminant bacteria

• Small amounts of acetic acid released during hemicellulose hydrolysis

• Dissolved carbon dioxide, salts, and excretion products from microbial cells metabolism

• Other compounds and materials

However, the two main components are water (80 to 90%) and ethyl alcohol. Making use of the higher volatility and lower boiling point of ethanol, the unit operation used as a rule for its separation is the conventional distillation at a pres­sure equal or higher than atmospheric pressure.

Usually, two distillation columns are utilized to elevate the ethanol concentra­tion up to 90 to 92%. Concentrations higher than 95.6% (or 89.4 molar %) are impossible to obtain by conventional distillation due to the similar composition of both the saturated vapor and saturated liquid achieved in the top of the distillation column. This composition is named azeotropic and is one of the main thermo­dynamic limits imposed on ethanol purification process. To produce anhydrous ethanol (99.5% or more), nonconventional separation technologies are required (see Section 8.2).

The technological scheme of the concentration and rectification steps of the ethanol contained in the culture broth is shown in Figure 8.1. During fermentation, carbon dioxide is generated as a result of the microbial metabolism. In the gas outlet stream and, along with CO2, small amounts of volatilized ethanol, as well as even much smaller amounts of water and other volatile substances, can be found. To avoid the ethanol loss, this gaseous stream is fed to a scrubber where a counter-current stream of water absorbs more than 98% ethanol. The scrubber is filled with a plastic-packed bed favoring the contact between the rising gaseous stream and the downward liquid stream. The gaseous ethanol-stripped stream is released into the atmosphere while the liquid stream containing about 2.5% etha­nol is unified with the culture broth coming from fermenter in order to be fed to the distillation column (Wooley et al., 1999).

The first distillation column is called the concentration or beer column. This column has a determined number of plates that can be of various types. It has been suggested that the fixed-valve Nutter-type plates are the most suitable to handle streams containing solids exhibiting a relatively good efficiency (about 48%). The concentrated ethanol stream (35 to 50%) is removed from the column by a side stream. The overhead vapors contain mostly CO2 (approximately 84%), a significant amount of ethanol (12%), and a small amount of water.

The bottoms of this column, called stillage or vinasses, concentrate the non­volatile substances and suspended solids that enter along with the culture broth.

Stillage composition depends on the feedstock employed for fuel ethanol produc­tion. The heat of this bottom’s stream is utilized to preheat the stream feeding the same distillation column. As an example of the location along this column of these streams, the design information for a concentration column corresponding to the biomass-to-ethanol process is provided (Wooley et al., 1999). This column has 32 plates, the feed stream is supplied to the fourth plate from the top, the side ethanol stream is removed from the eighth plate, and the reflux ratio required is 6.1.

The second column or rectification column is fed with the side ethanol stream from the concentration column. The operation of this column allows for a distil­late with 90 to 92% ethanol concentration, which is sent to the dehydration step. The bottoms of this column have a very low ethanol content (less than 1%), being mostly water. For illustration purposes, the design data of the rectification col­umn for the above-mentioned lignocellulosic ethanol process are provided as well (Wooley et al., 1999). The column has 69 plates and an additional feed stream corresponding to the recycle stream from the dehydration step, which is fed into the nineteenth plate from the top. The feed stream from the concentration column is supplied in the forty-fourth plate; fixed-valve Nutter-type plates are employed and the reflux ratio required is 3.2.