The stillage properties depend on the type of feedstock used for ethanol produc­tion as well as the conversion technology employed. The stillage presents high values of biological oxygen demand (BOD5 = 30,000-60,000 mg/L) that is, along with the chemical oxygen demand (COD), a measure of the content of organic matter in an effluent sample. Wilkie et al. (2000) have reviewed the characteris­tics of several types of stillage for various ethanol production processes, includ­ing the stillage yields. The stillage generated in the process employing sugarcane molasses has high polluting loads in terms of both BOD (45,000 mg/L) and COD (113,000 mg/L) as well as in its yield (2.52 liters of stillage per kilogram of molas­ses) and content of potassium, phosphorous, and sulfates. The stillage derived from the process using sugarcane juice exhibits lower levels of BOD, COD, and yield (12,000 mg/L, 25,000 mg/L, and 1.33 L/kg, respectively). The stillage from corn process shows intermediate indicators for BOD and COD (37,000 mg/L and 56,000 mg/L, respectively) but a higher stillage yield (6.20 L/kg). Finally, in the case of lignocellulosic ethanol produced from hardwood, the values of BOD, COD, and yield are 13,200 mg/L, 25,500 mg/L, and 20.4 L/kg, respectively (Wilkie et al., 2000).

If stillage is directly discharged into natural water streams, the microorgan­isms present in them will degrade the organic matter discharged, which uses the oxygen dissolved in the water. This leads to a critical diminution of dissolved oxygen provoking the death of most aquatic organisms. In addition, the utilization of nutrients contained in molasses can lead to the massive propagation of algae on the surface of ponds and lakes causing the blocking of solar light leading to the death of fish and other living organisms. This phenomenon is known as eutrophi­cation. Thus, the stillage discharge can seriously alter the ecological equilibrium of the aquatic ecosystems affected. Residual sugars from fermentation cause still­age COD to reach high values. For every 1% of residual sugar, a stillage COD increment of 16,000 mg/L can be expected (Wilkie et al., 2000). For this reason, full completion of fermentation is desirable in order to reduce stillage volume (greater ethanol concentrations) and minimize the concentration of remaining sugars. For each 1% ethanol left in the stillage (in the case of nonefficient distilla­tion), the stillage COD is incremented by more than 20,000 mg/L.

Due to its elevated organic matter content of stillage, methods for its treatment and economic utilization should be implemented in the industry. Among the most used methods for stillage treatment, irrigation, recycling, evaporation, incinera­tion, and composting should be highlighted. Some of the new treatment methods produce value-added substances. The most important stillage treatment methods are presented in the next sections.

Colombia presents a scenario of a developing country producer and consumer with possibilities of exports. Most of Colombia’s sugarcane is grown in the Cauca Valley, a rich agricultural valley with agroecological characteristics allowing the highest yield of sugarcane in the world (up to 130 tonnes/hectare).

Sugarcane cultivation requires a tropical or subtropical climate, with a mini­mum of 600 mm of annual rainfall. It is one of the most efficient photosynthesiz­ers in the plant kingdom, able to convert up to 2% of incident solar energy into biomass. Sugarcane is cultivated in almost all parts of the world, but only for a few months of the year, a period called safra. The only place in the world where there is no safra and, therefore, sugarcane is cultivated and produced year-round is in Colombia.

Colombian sugar production in 2008 was down 11% to 2,036,134 tons, from total sugar production of 2,277,120 tons in 2007. However, for 2009, an increase of about 15% is expected. Meanwhile, total Colombian sugar exports in 2008 fell 33% from the export volume in 2007 of 720,000 tons (Colombian Ministry of Agriculture and Rural Development, 2008).

TABLE 12.2

Use of Sugarcane Produced in Colombia 2006

Cropped land with cane for sugar-cakes or loaves/ha 340,000

Cropped land with cane for sugar or ethanol/ha

For sugar production/ha

For ethanol production/ha

Sugar production/ton

Sugar consumption/ton

Sugar exports/ton

Fuel ethanol production/L

Source: Colombian Ministry of Agriculture and Rural Development,

2008.

Fuel ethanol production in Colombia in 2008 rose 400 million liters, but the country expects to produce 1 billion liters of ethanol per year by 2010, more than doubling the current output, and the country plans to have enough production by the end of the year for export. It is believed that the U. S. Congress will approve the proposed U. S.-Colombia Free Trade Agreement, which will allow Colombia to permanently ship its ethanol surplus to the United States duty free and not be subjected to any quotas. The Colombian government believes that agricultural exports and food security would not be affected by expanding ethanol output because the cane and other ethanol feedstock for future production will come from new crops and unused land.

In Colombia today, the food and bioethanol competition based on cane use is not an important issue; however, in the near future, it could be an important topic of discussion. The country uses 340,000 hectares (Table 12.2) with low produc­tivity (30 to 80 ton/ha). The sugar is handcrafted and prepared as brown blocks rather than as a crystalline powder, by pouring sugar and molasses together into molds and allowing the mixture to dry. This results in sugar cakes or loaves called in Colombiapanela (jaggery or gur in India). This product is not important on the international market, but has a very important internal market, as high nutritional raw material for national beverages. The possibility of using part of the existing cane hectares for ethanol instead of panela could be a catastrophic scenario for the food security in Colombia.

Alcohols are an oxygenation alternative for gasoline considering the environmental disadvantages of the ethers mentioned above. The blends of alcohols and gasoline have comparable properties related to the traditional fuels based on oil. Despite their lower combustion heat compared to gasoline, the increase in fuel consumption when alcohols are used as oxygenates is not significant in principle. Moreover, the possi­bility of increasing the conversion of the blend and, therefore, the engine efficiency, represents a great advantage for alcohol-containing gasoline. Furthermore, the emis­sions of hydrocarbons and carbon monoxide are reduced, although a considerable increase in the emission of aldehydes is presented (Rasskazchikova et al., 2004).

In the 1970s, several researches were carried out in countries like Japan, the United States, and Germany aimed at the utilization of methanol (CH3OH) as an additive for enhancing the octane number of gasoline (see Table 1.1). Ethanol was pushed into the background due to its high comparative costs. However, despite its high octane number, methanol usage was limited and even banned in many countries due to its high toxicity, volatility (the highest RVP values of the ana­lyzed oxygenates), and hygroscopicity, which generates a series of technical dif­ficulties for the use of methanol-gasoline blends. Furthermore, the formaldehyde formed during methanol oxidation results in the formation of a substance consid­ered dangerous (Rasskazchikova et al., 2004).

At the beginning of the 1980s, the mixture of methanol and tert-butyl alcohol was commercialized under the trademark Oxynol™. Nevertheless, the high vola­tility of the blends containing methanol, due to the formation of one azeotrope with the gasoline hydrocarbons, caused the market to refuse it (Ancillotti and Fattore, 1998). Currently, methanol is employed as feedstock for the production of MTBE and TAME.

Lignocellulosic biomass is made up of very complex biopolymers that are not used in human food. The main components of lignocellulosic biomass are cellu­lose, hemicellulose, and lignin in addition to a small amount of extractives, acids, and minerals. For its conversion into ethanol, a complex process of pretreatment and hydrolysis is done in order to transform the carbohydrate polymers (cellulose and hemicellulose) into fermentable sugars.

Cellulose is a P-glucan, i. e., a polymer composed of glucose molecules linked by P(1,4) bonds. It can be considered that the cellulose is a linear polymer made up of cellobiose monomers as shown in Figure 3.6. The polymerization degree of cellulose is about 7,000 to 15,000. Due to its linear nature and to the interactions by hydrogen bonds between the OH groups of a same chain or of different chains, cellulose molecules are oriented by length leading to the formation of very stable crystalline structures. These structures allow the bundles of cellulose chains to form rigid, difficult to break microfibers. For this reason, the main function of cellulose in plants is structural, which explains its majority presence in the cell wall. In general, the cellulose composes 40 to 60% of dry matter of lignocellu — losic biomass (Hamelinck et al., 2003).

Hemicellulose composes 20 to 40% of lignocellulosic biomass and consists of short, very branched chains of sugars (200 sugars on average). Among these sugars are, in their order, xylose and arabinose (both 5-carbon sugars or pentoses), and galactose, glucose, and mannose (these latter sugars are hexoses). Other carbo­hydrate-related compounds like glucuronic, methyl glucuronic, and galacturonic acids are also present in hemicellulose structure. Furthermore, hemicellulose con­tains, in a lower proportion, acetyl groups esterified to some OH groups of its dif­ferent sugars. Due to the predominance of xylose, hemicellulose can be considered as a xylan. For lignocellulosic materials derived from hardwood, the xylan back­bone is composed of xylose units linked by P(1,4) bonds that branch through a(1,2) bonds with the methyl glucuronic acid (Figure 3.7a). In the case of xylan from softwood, the acetyl groups are less frequent, but there exist more branches due to the presence of a(1,3) between the xylose backbone and arabinofuranose units (Figure 3.7b). Considering its branched structure, hemicellulose does not form crystalline structures, but amorphous ones. Thus, this biopolymer is more soluble in water and has a higher susceptibility to the hydrolysis (Hamelinck et al., 2003).

The lignin comprises from 10 to 25% lignocellulosic biomass. This component is a very complex phenolic polymer composed of phenyl propane units linked by C-C and C-O-C bonds forming a three-dimensional amorphous structure (Lee, 1997). The structural units of lignin are the cinnamyl alcohols, which are differenti­ated by the various substitutions that the aromatic ring presents (Oliva, 2003). Thus, p-hydroxyphenyl units are derived from the p-coumaryl alcohol, the guaiacyl units are derived from the coniferilyc alcohol, and the syringyl units are derived from the sinapyl alcohol (Figure 3.8). The lignin has hydrophobic character and its main func­tion is as incrustive material of a cell wall, i. e., as a sort of cement between the cells.

The interaction and combination between the hemicellulose and lignin provide a covering shell to the cellulose making its degradation more difficult (Figure 3.9). Precisely, the main aim of biomass pretreatment is to break the lignin seal and significantly reduce the proportion of crystalline cellulose in such a way that the enzymes hydrolyzing the cellulose (cellulases) can have greater access to this polysaccharide and convert it into fermentable sugars.

Through recombinant DNA technology, amylolytic yeast strains have been “con­structed.” This allows for the design of ethanol production processes, excluding the liquefaction and saccharification steps using exogenous enzymes, and the uti­lization of only one bioagent during the transformation, the yeast (consolidated bioprocessing, CBP). The savings obtained during the commercial implementa­tion of such a process could offset by far the lower growth rates and the longer fermentation times. In this way, a single microorganism can directly convert the starch into ethanol (Cardona and Sanchez, 2007). Some examples of these efforts are shown in Table 6.3. Lynd et al. (2002) mention, among the saccharolytic genes that have been introduced into microorganisms as S. cerevisiae and Klebsiella oxytoca, those encoding a-amylase, glucoamylase, amylopullulanase, pectate lyase, and polygalacturonase obtained from bacterial and fungal sources.

Many of the investigated recombinant strains have demonstrated the pro­duction of ethanol from starch, but in some cases, the results are not definitive. Surprisingly, reduced starch hydrolysis and fermentation rates have been observed for yeast strains expressing a set of genes previously considered as appropriate, as is the case of the work of Knox et al. (2004). This example shows the difficul­ties that arise during the research using recombinant microorganisms. Although the methods of genetic transformation are relatively developed, the results can be unexpected. This is a key factor when these microorganisms are evaluated from an industrial point of view. For this reason, deep studies on the effects of genetic modifications on engineered strains are required.

The processes with microorganisms modified by genetic engineering involve the optimization not only of microbial physiology parameters, but also of cell culture parameters (retention and stability of plasmids, nutritional factors, cell growth, and protein synthesis). Therefore, the modeling of these processes and the application of the principles of biochemical engineering can be helpful con­sidering the uncertainties and complexities inherent to these biological systems (Cardona and Sanchez, 2007). An example of this type of modeling is the work of Kobayashi and Nakamura (2004), who corroborated experimentally at laboratory scale the higher productivity of the continuous fermentation process from starch using recombinant yeast cells immobilized in calcium alginate beads in compari­son with the free cell system.

Another approach employed for modeling this process is the so-called flux balance analysis. Qakir et al. (2004) have employed and experimentally validated this methodology in the case of yeasts. They have determined that if the split ratio in the branch point of the glucose-6-phosphate corresponding to the glucolytic

Some Examples of Recombinant Microorganisms with Potential Use for Fuel Ethanol Production

Glucose and xylose containing medium; acid pretreated starch industry effluents	Saccharomyces sp. strain is the product of fusion between S. diastaticus and S. uvarum and is capable of fermenting xylose to ethanol and utilizing it for aerobic growth; batch cultures; 48 h (sugars medium), 8 d (starch effluents) cultivation; EtOH cone. 60 g/L; yield 84% Introduced genes codify xylose assimilation and metabolism; batch culture; 48 h cultivation; EtOH cone. 62 g/L Hexose and pentose — assimilating E. coli acquires the ability for producing ethanol by introducing ethanol pathway genes; productivity 1.4 g/(L. h) from glucose. 0.64 g/(L. h) from xylose; EtOH cone. 41 g/L from xylose, 57 g/L from glucose; patented strain	Ho et al. (1998); Zaldivar et al. (2005)
Glucose and xylose containing medium
Leksawasdi et al. (2001)
Glucose or xylose containing medium
Ingram et al. (1991)
Continued

Some Examples of Recombinant Microorganisms with Potential Use for Fuel Ethanol Production

S. cerevisiae L26126GC Endo/exoglucanase p-glucosidase Bacillus sp. strain DO4 B. circulans Crystalline cellulose Batch SSF; 12 h cultivation; saccharolytic yeast capable of enzyme production; EtOH conc. 20.4 g/L; a considerable amount of commercial enzymes was reduced; potential strain for DMC of cellulose Cho and Yoo (1999) Clostridium cellulolyticum Pyruvate decarboxylase Alcohol dehydrogenase Z. mobilis Cellulose Native cellulolytic strategy Guedon et al. (2002) S. cerevisiae Cellobiohydrolase Thermoascus aurantiacus Crystalline cellulose Recombinant cellulytic strategy Hong et al. (2003)

metabolic pathway is changed by genetic manipulations, ethanol yield from starch can be considerably affected. This shows that the improvement in ethanol produc­tion goes with a rational design of metabolic pathways. In the future, this infor­mation could be crucial when different bioprocesses are designed, although it is necessary to analyze the costs and the complexity for acquiring this information in comparison with other “more traditional” procedures of design and control.

Similarly, important parameters, such as the stability of the plasmids used for introducing the desired traits to yeast cells, depend on the definition of the best environmental conditions during the cultivation of recombinant microorganisms. For example, Mete Altinta§ et al. (2002) showed that culture media containing specific salts and yeast extract drastically enhance plasmid stability during fed — batch cultures of yeasts using starch as feedstock.

Process integration is gaining more and more interest due to the advantages related to its application in the case of bioethanol production: reduction of energy costs, decrease in the size and number of process units, and intensification of the biological and downstream processes, among others. For instance, the combina­tion in a same unit of the enzymatic hydrolysis and the microbial transformation leads to the reduction of the negative effect due to the inhibition of the enzymes by the product of the reaction catalyzed by them. This corresponds to an integra­tion of the reaction-reaction type.

In general, reaction-reaction integration has been proposed for the integra­tion of different biological transformations taking place during ethanol produc­tion (Cardona and Sanchez, 2007). This type of integration mainly includes the combination of the enzymatic reactions for hydrolysis of starch or cellulose with the microbial conversion of formed sugars into ethyl alcohol. There exist different possibilities for reaction-reaction integration during production of ethanol from starch (Figure 9.1) and lignocellulosic biomass (Figure 9.2).

Considering as the starting point the nonintegrated separate hydrolysis and fer­mentation (SHF) process, several cases of reaction-reaction integration may be analyzed for ethanol production from both starchy and lignocellulosic materials.

The methodology of life cycle assessment (LCA) is a systematic analysis tool considering the environmental impacts of products, processes, or services, and provides a reference structure for the development and application of screen­ing indices and environmental performance indicators, especially during the extension of the system boundaries to the other steps of a life cycle of a product. Therefore, LCA is a technique for assessing the environmental performance of a product, process, or activity from “cradle to grave,” i. e., from extraction of raw materials to final disposal (Azapagic, 1999).

Life cycle assessment is defined as a process to evaluate the environmental burdens associated with a product, process, or activity by identifying and quanti­fying energy and materials used and wastes released to the environment; to assess the impact of those energy and material uses and releases to the environment;

and to identify and evaluate opportunities to effect environmental improvements (Azapagic, 1999). A fundamental feature of LCA compared to other methodolo­gies for environmental evaluation is the analysis of a system (for instance, a pro­cess for production of a particular chemical) during its whole life cycle from the extraction and processing of the feedstocks to the disposal of the target product, co-products, and by-products considering their effect over the whole environment (over the global warming, ozone layer depletion, etc.) and including manufactur­ing, transport, use, reuse, maintenance, and recycling of the different materials involved. In contrast, most methods for environmental assessment are focused only on the immediate effects of the system on the surroundings, such as the effects of the emissions and burdens from the processing plant. In this sense, there exists the possibility that certain measures adopted to reduce these emis­sions and burdens in a given process lead to the increase of other emissions or burdens in other steps of the life cycle of this process, e. g., during the feedstock extraction. Despite these conceptual advantages, the application of LCA during the conceptual design step is quite difficult due to the limitation in the available information as mentioned above.

Applying the LCA methodology, the environmental benefits of using the excess of cane bagasse, which remains after employing the bagasse as an energy source in sugar mills, as a feedstock for fuel ethanol production instead of burning it in open fields were demonstrated (Kadam, 2002). Thus, some achievements can be attained such as the emissions reduction, less fossil fuels consumption, diminu­tion in the rate of natural resources depletion, reduction of human toxicity, and less contribution to greenhouse gas effect. Hu et al. (2004b) performed LCA for cars fueled by blends using ethanol from cassava obtaining better results than for conventional cars in China. Kim and Dale (2002) proposed an allocation proce­dure based on the system expansion approach for the net energy analysis of corn ethanol. The allocation procedure is a key factor in LCA when the multi-input/ output process is analyzed as in the case of ethanol production employing wet and dry milling. These same authors also determined the nonrenewable energy con­sumption and greenhouse gas emissions for corn ethanol production in selected counties in the United States, showing positive net energy values and the pos­sibility for reducing greenhouse gas emissions (Kim and Dale, 2005a). Bullock (2002) cites the LCA studies based on data from Australian ethanol plants for gasoline blends with 10% ethanol content. These studies indicate that there is no greenhouse abatement when molasses, wheat starch waste, or wheat are used as feedstocks. In contrast, the hypothetical process from wood waste can lead to CO2 abatement. Gasoline blends with 85% ethanol content present significant environ­mental benefits represented mainly by the greenhouse gas abatement.

Several recent studies applying the LCA methodology have been published demonstrating some advantages of fuel ethanol production even in the polemic case of corn ethanol. For example, using ethanol derived from corn dry milled as liquid fuel (E10 fuel) would reduce nonrenewable energy and greenhouse gas emissions, but would increase acidification, eutrophication, and photochemical smog, compared to using gasoline as liquid fuel (Kim and Dale, 2005a). Other studies using the cereal straw or corn stover as feedstocks for biomass ethanol have also indicated the environmental benefits, but show problems related to soil acidification and eutrophication (Gabrielle and Gagnaire, 2008; Kim and Dale, 2005b).

In a similar way, the LCA performed in France by comparing the ethyl tert — butyl ether (ETBE) obtained from beet ethanol (a partial renewable product) to the methyl tert-butyl ether (MTBE) from fossil origin showed that the energy yield of ethanol (1.18) and ETBE (0.93) are higher than the yields of gasoline (0.74 to 0.80) and MTBE (0.73). Moreover, the ETBE has a lower contribution to the greenhouse gas effect due to its renewable character, and its use as a gasoline oxygenate provokes fewer emissions of nonburnt hydrocarbons than in the case of MTBE. Nevertheless, the ethanol cost for ETBE production in France has been higher than the cost of gasoline and methanol (a feedstock for MTBE production; Poitrat, 1999).

Farmers, refiners, and consumers are the same actors in the food and biofuels market. Any discussion regarding the possible contradictions between the use of crops and residues for food and fuels should not be globalized. The negative energy balance in some countries (consumers) is a problematic issue and fuel ethanol for automotive transport, for example, represents the unique stable alter­native. Additionally, environmental improvements are reached when fuel ethanol is blended with gasoline. On the other hand, other countries can find in biofuels like ethanol the beginning of economic development for rural areas. Here critical positions are related to the possibility of using rural areas for food projects instead of fuel ethanol programs.

However, the real regulator of land use for food or biofuels is the market. In order to avoid nonequilibrium development of the fuel ethanol market based on energy crops, drastic, but fair, regulations from the governments must exist. Every country has to create appropriate rules and laws for developing fuel etha­nol programs based on its specific supply and demand characteristics. One way to diminish possible impacts of fuel ethanol production on food security is the use of compensation strategies. For example, if residual biomass increases value as a raw material for producing biofuels instead of livestock food, then alterna­tive food crops must be allocated together and simultaneously with the biofuel project. In the case of commercial sugar or starchy feedstocks for bioethanol, accurate calculations and predictions of food and energy consumption should be the basis of any discussion. If sugar, for example, is going to be exported but the prices are not stable and profitable, fuel ethanol production is a sustainable alternative not competing with food. Fuel ethanol producers in some countries having enough land for food have demonstrated that internal and export prices of food are not affected. However, countries like the United States alarmed the world when subsidized corn ethanol production affected neighbors’ food security, especially Mexico.

Finally, the authors of this book consider that the main problem we have today in this important discussion is the existence of much speculative information about biofuels and food security. Most of this information is used incorrectly for political and economical purposes. In this context, technological platforms based on scientific and real information are the only way to consider exactly the influ­ence of biofuels like ethanol on the food security.

The analysis of the statics is another of the thermodynamics-based approaches that has allowed the synthesis of integrated processes of the reaction-separation type (Pisarenko et al., 2001b). This type of analysis is based on the principles of the topologic thermodynamics and has been widely used during the design of reactive distillation processes (Pisarenko et al., 1999; Pisarenko et al., 2001a), although it is also applicable to the synthesis of distillation trains. Thus, the analysis of the statics provides the fundamentals and tools needed for the pre­liminary design of distillation, reactive distillation, and more recently, reactive extraction and extractive fermentation processes through the development of short-cut methods based on a graphic representation that allows the visualization of the process trajectory. With the help of these methods, it is possible to specify the operating conditions and regimes corresponding to stable steady states. This information is used later during the rigorous modeling of the processes or for their simulation using commercial packages in the subsequent design steps. For this, the information on static (not varying with time) properties, not only of phase equilibrium (separation) but also of chemical equilibrium (reaction), is required. While the statics of distillation processes under infinite separability regime have been sufficiently represented through the topologic thermodynam­ics (Serafimov et al., 1971, 1973a, 1973b, 1973c), the statics of the chemical transformations are less developed. Without doubt, this approach is very useful for process design when applying the principle of integration, although its appli­cation has been mainly oriented to the basic organic chemical and petro-chem — ical industries (Cardona et al., 2000, 2002). The application of this approach to biological processes has been very limited, though it is difficult to undervalue the potential of integration in the development of innovative biotechnological processes with a high performance.

Besides the above-mentioned approaches, other types of knowledge-based pro­cess synthesis strategies are being developed as case-based reasoning, axiomatic design, and mean-end analysis (Li and Kraslawski, 2004). Case-based reasoning is supported in very specific data of prior situations and reuses previous results and experience to adjust them to the solution of new design problems. Axiomatic design is based on the principle that a good design maintains the independence of the functional requirements. This approach also applies the axiom that the information content of a good design is minimized. Finally, mean-end analysis considers that the purpose of a chemical process consists in the application of several operations in such a sequence that all the differences between the proper­ties of the feedstocks and products are eliminated. The birth of new paradigms is expected for generating designs boosted, for instance, by the development of the artificial intelligence (Barnicki and Siirola, 2004).

Chemical pretreatments utilize different chemical agents such as ozone, acids, alkalis, peroxide, and organic solvents (Table 4.4). The ozonolysis can be used

to degrade lignin and hemicellulose in many lignocellulosic materials. For the case of poplar sawdust pretreated with ozone, the enzymatic hydrolysis yield is increased from 0 to 57% while the lignin content is reduced from 29 to 8%. Despite its advantages, this method requires high amounts of ozone for an effec­tive pretreatment making the process quite expensive (Sun and Cheng, 2002).

Inorganic acids, especially sulfuric and hydrochloric acids, are the most used agents for biomass pretreatment using acid catalysts. These acids are toxic, dan­gerous, and require reactors resistant to corrosion. Moreover, if concentrated acids are employed, their recycling should be implemented by economic con­siderations. The pretreatment using dilute acids, especially sulfuric acid, has been developed in a successful way to process different lignocellulosic materials whereby high reaction rates can be attained and the subsequent cellulose hydroly­sis can be significantly improved. Nevertheless, the dilute-acid pretreatment costs are usually high related to those of steam explosion or AFEX process (Sanchez and Cardona, 2008; Sun and Cheng, 2002) . The pretreatment of corn stover at pilot plant scale has been studied using dilute sulfuric acid (0.5 to 1.4% p/v) in a continuous reactor for processing 1 ton/day of feedstock (Schell et al., 2003). In this case, high solids load (20%) was utilized unlike those reported in the open literature. Xylose yield reached 77% at 190°C. The digestibility of the pretreated material was evaluated by a simultaneous saccharification and fermentation (SSF) process attaining values of 87%.

Like steam explosion, dilute-acid pretreatment can be combined with other pretreatment methods to carry out a two-stage process. In particular, lignin removal can be greater if the biomass is treated with a dilute acid in the first stage followed by the addition of a concentrated acid plus ethanol in the second stage. In this way, the biomass fractionation can be accomplished, i. e., the separation of its three main components:

1. Sugars generated from hemicellulose hydrolysis that remain in the liquid fraction after the first stage,

2. Cellulose with a higher susceptibility to the enzymatic attack that remains in the solid fraction, and

3. Oligomers of lignin that result from the combined action of the con­centrated acid and ethanol and that are solubilized in the second stage (delignification) releasing the cellulosic fiber.

These oligomers can be precipitated after biomass fractionation (Papatheofanous et al., 1995). Another variant of the two-stage dilute-acid pretreatment consists of conducting the hemicellulose hydrolysis at 140°C during 15 min in a first stage to reduce the formation of furans and carboxylic acids, and then increasing the tem­perature to 190°C for 10 min to make the cellulose more accessible to the cellulase attack (Saha et al., 2005a, 2005b; Sanchez and Cardona, 2008). If dilute-acid pre­treatment is performed at a lower temperature (121°C), the degradation of sugars into furans (furfural and hydroxymethylfurfural), which may have an inhibitory effect on the fermentation, can be prevented, but sugar yields are reduced.

TABLE 4.4 Chemical Methods for Pretreatment of Lignocellulosic Biomass for Ethanol Production
examples of Pretreated Materials Poplar sawdust Pine Bagasse, wheat straw, cotton straw, green hay, peanut Poplar wood Bagasse, corn stover, wheat straw, rye straw, rice hulls Switchgrass, Bermuda grass
Methods ProcedureMgents Ozonolysis Ozone, room temperature, and pressure	Remarks No inhibitors formation Further cellulose conversion can be >57% Lignin degradation pH neutralization is required that generates gypsum as a residue 80-100% hemicellulose hydrolysis, 75-90% xylose recovery Cellulose depolymerization occurs at certain degree High temperature favors further cellulose hydrolysis Lignin is not solubilized, but it is redistributed Acid recovery is required Residence time greater compared to dilute-acid hydrolysis Peracetic acid provokes lignin oxidation	References Sun and Cheng (2002)
Dilute-acid 0.75-5% H2SO4, HCl, or HNO3, hydrolysis p~1 MPa; continuous process for low solids loads (5-10 wt.% dry substrate/mixture): T = 160-200°C; batch process for high solids loads (10^0 wt.% dry substrate/mixture): T = 120-160°C	Esteghlalian et al. (1997); Hamelinck et al. (2005); Lynd et al. (2002); Martinez et al. (2000); Rodriguez-Chong et al. (2004); Saha et al. (2005a, 2005b); Schell et al. (2003); Sun and Cheng (2002); Sun and Cheng (2005); Wooley et al. (1999)
Concentrated-acid 10-30% H2SO4, 170-190°C, hydrolysis 1:1.6 solid-liquid ratio 21-60% peracetic acid, silo-type system	Poplar sawdust Bagasse	Cuzens and Miller (1997); Teixeira et al. (1999a, 1999b)

Alkaline Dilute NaOH, 24 h, 60°C; Reactor costs are lower compared Hardwood Hamelinck et al. (2005); Hari hydrolysis Ca(OH)2, 4 h, 120°C; it can be complemented by adding H2O2 (0.5-2.15 vol.%) at lower temperature (35°C) to acid pretreatment >50% hemicellulose hydrolysis, 60-75% xylose recovery Low inhibitors formation Cellulose swelling Further cellulose conversion can be >65% 24-55% lignin removal for hardwood, lower for softwood Bagasse, corn stover, straws with low lignin content (10-18%), cane leaves Krishna et al. (1998); Kaar and Holtzapple (2000); Lynd et al. (2002); Rivers and Emert (1988); Saha and Cotta (2006); Sun and Cheng (2002); Teixeira et al. (1999b) Oxidative delignification Peroxidase and 2% H2O2, 20°C, 8 h Almost total solubilization of hemicellulose Further cellulose conversion can be 95% 50% lignin solubilization Bagasse Sun and Cheng (2002) Wet oxidation 1.2 MPa oxygen pressure, 195°C, 15 min; addition of water and small amounts of Na2CO3 or H2SO4 Solubilization of major part of hemicellulose Inhibitors formation Lignin degradation Corn stover, wheat straw Bjerre eta al. (1996); Varga et al. (2004) Organosolv Organic solvents (methanol, Solvent recovery required Poplar wood Lynd et al. (2002); Pan et al. process ethanol, acetone, ethylene glycol, triethylene glycol) or their mixture with 1% of H2SO4 or HCl; 185-198°C, 30-60 min, pH = 2.0-3.4 Almost total hydrolysis of hemicellulose, high yield of xylose Almost total lignin solubilization and breakdown of internal lignin and hemicellulose bonds Mixed softwood (spruce, pine, Douglas fir) (2005); Rezzoug and Capart (1996); Sun and Cheng (2002) Source: Adapted from Sanchez, O. J., and C. A. Cardona. 2008. Bioresource Technology 99:5270-5295. Elsevier Ltd.

Process Synthesis for Fuel Ethanol Production

Pretreatment, using concentrated acids, for fuel ethanol production also has been proposed. Arkenol, Inc. (USA) has reported the development of a fuel ethanol production process from cane bagasse through pretreatment with concentrated sulfuric acid, which has been patented (Farone and Cuzens, 1996). This technology requires the retrofitting of sugar mills in order to pro­duce ethanol and improve energetic indexes of this kind of processes (Cuzens and Miller, 1997; Sanchez and Cardona, 2008). This company is evaluating Zymomonas bacteria for use in its concentrated-acid process (Mielenz, 2001). An alternative approach has been tested by Teixeira et al. (1999a, 1999b), which employed a silo-type system that introduced the feedstock (bagasse or hybrid poplar) into plastic bags to which a peracetic acid solution was added with a concentration range from 0 to 60% (by weight). To enhance the process effi­ciency, sodium or ammonium hydroxide was added before the acid treatment, which allowed the use of lower amounts of peracetic acid. Cellulose conversion of pretreated material reached 93.1% in 120 h using 21% acid concentration or in 24 h using 60% acid concentration. This system requires low energy because the process is carried out at room temperature. Other methods involve the conversion of both cellulose and hemicellulose into fermentable sugars, which eliminates the necessity of adding cellulases, but the operation condi­tions are far from economically viable (Iranmahboob et al., 2002). In addition, there exists an additional problem related to the oxidation of glucose, which is obtained because of the high acid concentration and relatively prolonged times for biomass heating.

Alkaline pretreatment is based on the effects of the addition of dilute bases on the biomass: increase of internal surface by swelling, decrease of polymeriza­tion degree and crystallinity, destruction of links between lignin and other poly­mers, and breakdown of lignin. The effectiveness of this method depends on the lignin content of the biomass (Sun and Cheng, 2002). This type of pretreatment has been applied to corn stover obtaining 60 to 80% delignification efficiency employing 2.5 to 20% ammonium at 170°C for 1 h (Sun and Cheng, 2002), as well as to sugarcane bagasse and rice straw (Rivers and Emert, 1988). The addi­tion of an alkali can be combined with the addition of hydrogen peroxide as reported by Hari Krishna et al. (1998). Lignin degradation can also be carried out using the peroxidase enzyme in the presence of hydrogen peroxide through a process called oxidative delignification (see Table 4.4). Another pretreatment method involving lignin degradation is wet oxidation that is based on the addi­tion of oxygen and water at high temperatures and pressure leading to the open­ing of crystalline cellulose and the breakdown of lignin into simpler compounds, such as CO2, water, and carboxylic acids (Bjerre et al., 1996). Lignin oxidation can also be carried out with KMnO4, although with low cellulose conversions (below 50% for rice straw and cane bagasse; Rivers and Emert, 1988). In gen­eral, the utilization of bases like sodium hydroxide or solvents like ethanol or methanol (organosolv process) allows the dissolution of lignin, but their costs are so high that these methods are not competitive for large-scale plants (Lynd et al., 1999).