7.1.4.1 Fermentation of Cellulose Hydrolyzates

The classic configuration employed for fermenting biomass hydrolyzates involves a sequential process where the hydrolysis of cellulose and the fermentation are carried out in different units (Sanchez and Cardona, 2008). This configuration is known as separate hydrolysis and fermentation (SHF). When this sequential process is employed, the solid fraction of pretreated lignocellulosic material undergoes hydrolysis (saccharification). This fraction contains the cellulose in a form accessible to acids or enzymes. Once hydrolysis is completed, the resulting cellulose hydrolyzate is fermented and converted into ethanol. S. cerevisiae is the most employed microorganism for fermenting the hydrolyzates of lignocellulosic biomass. This yeast ferments the hexoses contained in the hydrolyzate, but not the pentoses. One of the main features of the SHF process is that each step can be per­formed at its optimal operating conditions (especially temperature and pH). The overall scheme of this process is presented in Figure 7.6. Depending on the type of biomass pretreatment, the lignin can be separated in this step (as in the case of the organosolv process [see Chapter 4]) or remain in the stillage. This scheme involves the fermentation of hemicellulose hydrolyzate by pentose-assimilating yeasts in a way parallel to the fermentation of glucose using S. cerevisiae.

BC International Corporation (Dedham, MA, USA) has operated a pilot plant in Louisiana for producing ethanol from biomass by SHF. This plant has the

).

capacity to process about 500 ton/year of lignocellulosic materials. The University of Florida has granted the license to this company for worldwide commercializa­tion of this technology, including the patent of the microorganism employed in the process, a recombinant strain of Escherichia coli to which Zymomonas mobilis genes, ensuring the biosynthesis of ethanol, has been introduced (Ingram et al., 1991). The company has been seeking investors for a $90 million project for an ethanol production facility using cane bagasse with a capacity of 23.2 mill gal EtOH/year, but the construction has not yet started. In fact, SHF is the technol­ogy with the most possibilities of being implemented at a commercial scale. For instance, the company Abengoa Bioenergia (Spain) is constructing a demonstra­tion facility for ethanol production from lignocellulosic biomass with a capacity of 5 million L/year (Abengoa, 2008). The conversion of wheat and barley straw into ethanol will be done by an SHF scheme, as that shown in Figure 7.7, although this company plans to implement the pentose fermentation step at the midterm.

Some attempts to produce ethanol from municipal solid waste (MSW) have been done considering the cellulosic fraction of this type of materials (Sanchez and Cardona, 2008). Park et al. (2001) have studied the hydrolysis of waste paper contained in MSW obtaining significant sugars yield and evaluating the viscosity as an operating parameter. Bioethanol production from the cellulosic portion of

MSW has been already patented (Titmas, 1999) and some strategies for improv­ing the fermentability of acid hydrolyzates of MSW have been defined. Nguyen et al. (1999) employed a mixed solids waste (construction lumber waste, almond tree prunings, wheat straw, office waste paper, and newsprint) for producing ethanol by SHF using yeasts. In this process, the recycling of enzymes was implemented through microfiltration and ultrafiltration achieving 90% cellulose hydrolysis using a net enzyme loading of 10 filter paper units (FPU)/g cellulose.

S. cerevisiae has demonstrated its elevated resistance to the presence of inhibi­tors in the lignocellulosic hydrolyzate. In the case of the more productive continu­ous regime, one way to enhance this resistance is the increase in the cell retention to prevent washout and maintain high yeast cell density. Brandberg et al. (2005) employed a microfiltration unit to recirculate the cells under microaerobic condi­tions achieving sugar conversion up to 99% for undetoxified dilute-acid pretreated hydrolyzates of softwood (spruce) supplemented with mineral nutrients, although the productivity was low.

This type of reaction-separation integration is carried out by coupling the fer­mentation tank to a vacuum chamber that allows extracting ethanol due to its higher volatility in comparison to the rest of components of the culture broth. It has been reported that a 12-fold increase in ethanol productivity can be reached using vacuum fermentation (Cysewski and Wilke, 1977). However, for reaching this productivity, the addition of oxygen was required that negatively influenced the costs, which were already high enough due to the creation of vacuum con­ditions. Nevertheless, da Silva et al. (1999) point out that vacuum fermentation using a flash chamber coupled to the bioreactor can demonstrate better technical indexes than extractive fermentation or fermentation coupled to pervaporation (see Figure 9.8). Ishida and Shimizu (1996) proposed a novel regime for car­rying out the repeated-batch alcoholic fermentation coupled with batch distilla­tion obtaining ethanol concentrations of 400 g/L. Some examples of this type of reaction-separation integration are shown in Table 9.6.

Numerous studies for developing large-scale production of ethanol from lignocel — lulosic biomass have been carried out in the world. One of the advantages of the use of lignocellulosic biomass is that this feedstock is not directly related to food production, which would implement the extra production of bioethanol without the need of employing vast extensions of cultivable land for cane or corn produc­tion. In addition, lignocellulosics is a resource that can be processed in different ways for the production of many other products, such as synthesis gas, methanol, hydrogen, and electricity (Chum and Overend 2001). However, the main limiting factor is the higher degree of complexity inherent to the processing of this feed­stock. This complexity is related to the nature and composition of lignocellulosic biomass (see Chapter 3, Section 3.3.1). Two of the main biomass polymers need to be broken down into fermentable sugars in order to be converted into ethanol or other valuable products. But this degradation process is complicated, energy-con­suming, and not completely developed. Consequently, the involved technologies are more complex leading to higher ethanol production costs compared to cane, beet, or corn. However, the fact that many lignocellulosic materials are by-prod­ucts of agricultural activities, industrial residues, or domestic wastes offers huge possibilities for the production of fuel ethanol at a large scale as well as its global consumption as a renewable fuel. It is thought that lignocellulosic biomass will become the main feedstock for ethanol production in the near future (Cardona and Sanchez, 2007). According to Berg (2001), the output/input ratio of energy for the production of lignocellulosic ethanol reaches a value of 6, indicating a bet­ter energy efficiency than in the case of corn ethanol.

The classic configuration employed for converting lignocellulosic biomass into ethanol involves a sequential process in which the hydrolysis of cellulose and the fermentation are carried out in different units. As mentioned in Chapter 7, Section 7.1.4.1, the main feature of this configuration known as separate hydrolysis and fermentation (SHF) is that optimum conditions of pH and temperature for both processes can be ensured in an independent way. The general flowsheet of this technology is illustrated in Figure 7.6. Depending on the pretreatment method, the lignin can be recovered in this step or remain in the stillage from which it can be burned to generate steam. The solid fraction obtained in the pretreatment reactor is sent to the hydrolysis bioreactor where it comes in contact with microbial cel — lulases. This scheme involves the fermentation of the hemicellulose hydrolyzate contained in the liquid fraction exiting the pretreatment reactor using pentose — assimilating yeasts (like Candida shehatae or Pichia stipitis) in a parallel way to the glucose fermentation carried out with S. cerevisiae. In the alternative variant involving the simultaneous saccharification and fermentation (SSF), the hydrolysis and fermentation are performed in a single unit as discussed in Chapter 9, Section 9.2.2.3. The most employed microorganism for fermenting the hydrolyzates of the lignocellulosic biomass is S. cerevisiae, which ferments the hexoses contained in the hydrolyzate, but not the pentoses. This configuration is depicted in Figure 9.4.

In general, the biomass releases energy through its conversion into simpler com­pounds. This conversion can be carried out by chemical or biological methods. In this way, the biomass can be employed for producing solid, gaseous, or liquid bio­fuels. The biomass itself can be used as a solid biofuel for electricity production. In this case, the biomass undergoes combustion with or without coal as an auxiliary fuel (co-combustion). The biomass can also be used as a feedstock for producing gaseous fuels. For this, the biomass undergoes thermal treatment in the presence of a reduced amount of oxygen (partial oxidation) or by steam. The aim of these kinds of processes (pyrolysis, thermal gasification) is to obtain a gaseous fuel that can be mixed in a better way with the air leading to a cleaner and more complete combustion than is the case of the solid biomass. Moreover, the biomass can be converted into biogas, a mixture of CH4 and CO2, using anaerobic bacteria that assimilate the organic matter contained in the biomass forming more bacterial cells and releasing methane and carbon dioxide as a result of the methanogenic metabolism in absence of oxygen.

Fuel ethanol production is directly linked to the production of sugar. In fact, gen­erally speaking, ethanol can be considered as a co-product in the sugar industry, but it also can be the main end product obtained from sugarcane. In the for­mer case, cane molasses formed during sugar processing is employed for ethanol production. However, the requirements of fuel ethanol in cane sugar-producing countries with gasoline oxygenation programs cannot be covered with the sole molasses. Therefore, autonomous (stand-alone) distilleries not co-located near sugar mills have been put into operation in countries like Brazil in order to meet the required amounts of fuel ethanol. These distilleries use the sugarcane to produce ethanol exclusively. Besides molasses, some sugar mills use the cane juice for producing ethanol in their co-located distilleries. This means a reduced amount of produced sugar compared to when the mill does not produce ethanol. This is the case in Colombia whose sugar production has been decreased due to the implementation of the gasoline oxygenation program using E10 blends. The rise in the domestic price of sugar during 2006 through 2008 has been explained not only by the increase in the international oil price, but also by the onset of this program (Sanchez and Cardona, 2008a).

The technology for fuel ethanol production from sugar-rich materials (mostly sucrose) offers multiple alternatives regarding the use of feedstocks generated within the sugar mills. For instance, and excluding the cane juice and molasses, Colombian distilleries use part of the clarified syrup. In this way, a great flexibil­ity for production of both sugar and ethanol is attained, exploiting the integration of the different sugar-rich streams during sugar processing. For sugar mills, this flexibility allows them to respond in a suitable way to the changes and needs of both ethanol and sugar markets.

In the case of ethanol production facilities using sugar beet, both the diffusion juice and beet molasses can be employed for producing bioethanol. The produc­tion of fuel ethanol directly from the beet juice is a nonviable technological option for most countries producing sugar from sugar beet due to the higher costs of the juice produced and to the need of covering their domestic sugar demands as in the case of European countries and North America. For these countries, ethanol is mostly produced from corn, wheat, and other grains. In fact and according to Decloux et al. (2002), there exist no stand-alone distilleries producing ethanol from beet juice in France, the first ethanol producer from sugar beet. Instead of that, the distilleries are co-located next to sugar mills. Some important issues concerning the main sucrose-containing feedstocks for ethanol production are highlighted below.

TABLE 3.5

Composition of the Raw Sugarcane Juice

5.2.3.1 Efficiency of Cellulases

Cellulase utilization plays a crucial role considering the global costs of the bio — mass-to-ethanol process. The cellulases available for ethanol industry account for 36 to 45% of the costs of bioethanol produced from lignocellulosic materi­als. According to different evaluations (Cardona and Sanchez, 2007; Reith et al., 2002), a 30% reduction in capital costs and 10-fold decrease in the cost of current cellulases are required for this process to be competitive in relation to ethanol produced from starchy materials. These analyses evidence the need of improving the cellulase performance in the following aspects:

• Increase of thermal stability

• Improvement of the binding to cellulose

• Increase of specific activity

• Reduction of the nonspecific binding to lignin

The increase of the thermal stability of cellulases is important since the tem­perature increase implies an increase of the cellulose hydrolysis rate (Mielenz, 2001). The augment of specific activity, in turn, can lead to significantly reduced costs. It is estimated that a 10-fold increase in specific activity could lead to nearly 16 cents (U. S.) savings per liter of ethanol produced. Among the strategies to attain the increase of specific activity are the increase in the efficiency of active sites through protein engineering or random mutagenesis, augment of thermal tolerance, improvement in the degradation of the crystalline structure of cellu­lose, enhancement of the synergism among the cellulases from different sources, and reduction of nonspecific bindings (Cardona and Sanchez, 2007; Sheehan and Himmel, 1999).

In general, the costs of cellulases are considered high. According to prelimi­nary evaluations of the National Renewable Energy Laboratory (NREL) cited by Tengerdy and Szakacs (2003), the cost of cellulase production in situ by sub­merged culture is U. S.$0.38/100,000 FPU. Hence, cellulase costs make up 20% of ethanol production costs, assuming them at U. S.$1.5/gallon. On the other hand, commercial cellulase cost (U. S.$16/100,000 FPU) is prohibitive for this process. In contrast, these authors indicate that the cost for producing cellulases by solid — state fermentation of corn stover would be U. S.$0.15/100,000 FPU that would correspond to U. S.$0.118/gal EtOH, i. e., nearly 8% of total costs. The analysis of process integration via simulation of not only the technological scheme, but also the costs structure, can provide key elements allowing a deep evaluation of all of these alternatives in order to choose the more convenient option for in situ cellu — lase production. On the other hand, the mathematical description of cellulase pro­duction can allow the definition of useful relationships to assess the performance and quality of the production of such enzymes using different process analysis approaches. These approaches undoubtedly will contribute to the definition of strategies to lower the costs of cellulases. This is the case of cellulase production by Trichoderma reesei using kinetic and neural networks approaches (Tholudur et al., 1999), which allowed the optimization of operating conditions considering two performance indexes based on the estimated protein value and volumetric productivity.

The distillate product from the rectification column represents the stream enter­ing the ethanol dehydration scheme, where the fuel ethanol with ethanol content greater than 99.5% should be produced. To achieve this high purity from streams containing 90 to 92% ethanol, it is necessary to employ nonconventional separation

operations like pressure-swing distillation, azeotropic distillation, extractive distil­lation, adsorption, and pervaporation (Sanchez and Cardona, 2005). All of these operations have found industrial application in the fuel ethanol industry.

Usually, in ethanol producing plants, the stillage obtained from the distillation step (whole stillage) undergoes centrifugation in order to recover organic solids, especially yeast debris. The liquid fraction is known as thin stillage. In a typi­cal distillery, more than 20 liters of thin stillage can be generated per each liter of ethanol produced. To minimize the effluent treatment costs, a portion of thin stillage (from 25 to 75%) is recycled in different process steps as fermentation or saccharification (in the case of conversion of starchy materials). In a simi­lar way, the whole stillage can be recycled back to the fermentation step when cane molasses is used. Although this procedure decreases the volume of fresh water employed, it decreases, in turn, the volume of stillage generated; the total amount of organic matter in the stillage (measured as COD) does not change because its concentration increases with the amount of recycled stillage (Wilkie et al., 2000).

The North American ethanol industry that uses corn as feedstock makes use of a fraction of thin stillage to replace a percentage of water required during the mashing process (saccharification step). The thin stillage stream recycle is called backset. The drawback to this practice is that there is an accumulation of undesirable substances, such as lactic acid, produced by contaminant bacteria as well as minerals and nonutilized substrates. In this way, the continuous backset usage results in the buildup of compounds inhibiting the yeast growth and etha­nol production. On a laboratory scale, it has been demonstrated that the use of a backset in recycling 50% of thin stillage during the wheat mashing for five suc­cessive fermentation batches will produce a 60% loss of yeast viability, though the fermenting ability of these microorganisms and the glucoamylase activity is not affected (Chin and Ingledew, 1994). The backset utilization has been evalu­ated in the case of very high gravity (VHG) fermentations of wheat starch. As noted in Chapter 7, Section 7.1.3.2, VHG fermentation requires an available nitro­gen source, but the recycled thin stillage contains no significant amounts of free amino nitrogen. In this sense, the use of fresh yeast autolyzates are obtained by centrifugation of the fermented wort before its distillation. The lysis of yeasts is induced through their resuspension in water and by increasing the temperature up to 48°C for 48 h, followed by pasteurization at 75°C. In this way, the costs associated with the treatment of cell biomass can be reduced in ethanol-producing plants as well (Jones and Ingledew, 1994).

The possibility of implementing a system with zero-discharge of stillage for the process employing starch or starch residues from wet-milling process has been proposed. The stillage is decanted and the resulting thin stillage undergoes ultrafiltration. The solids produced are recycled back to the decanter while the permeate is recirculated to the cooking step of starch. After eight-fold recycling of the permeate, the yield is similar to that of the conventional process, though this system increases the fermentation time from 60 to 90 h (Nguyen, 2003).

For the process based on sugarcane molasses, a bioconcentration system of vinasses has been proposed. This system consists of the recycling of 60% stillage to the fermentation step by replacing part of the water used during the preparation of the culture medium (Navarro et al., 2000). This stillage recycling percentage can increase the ethanol production without provoking inhibitory effects result­ing from the accumulation of by-products released by the yeasts. In the process, 46.2% reduction of fresh water, 66% decrease of nutrients, and 50% reduction of sulfuric acid are attained. The energy balance of the process can be improved if subsequent stillage incineration is implemented since the liquid effluent contains about 24% solids. Moreover, the obtained stillage contains significant amounts of glycerol. In this way, it is possible to generate only five liters of stillage per liter of ethanol produced. These systems require the use of osmotolerant yeasts, such as Schizosaccharomyces pombe, able to grow under high solids concentration in culture medium (Goyes and Bolanos, 2005). The distilleries of CSR (Australia) utilize the biostill process (see Chapter 7, Section 7.1.2.3) implementing such types of recycling and using S. pombe. With this technology, one can obtain still­age with 28 to 30% solids content making it suitable for its direct utilization as a fertilizer (Bullock, 2002). In the distilleries co-located in cane sugar mills, the vinasses can be alternatively used as part of the water for washing the sugarcane or recycled to the molasses dilution step (Sheehan and Greenfield, 1980).

10.1.2.1 Stillage Evaporation

The evaporation represents an intermediate step during the stillage treatment. The evaporated stillage is the starting point for solids recovery, fertilization, and incineration of stillage. Thus, during ethanol production from corn by the dry-milling technology, the thin stillage obtained can be evaporated in order to produce syrup (also called distiller’s solubles). This syrup is combined with the solids from corn to produce a co-product used for animal feed (DDGS). The con­densed water generated in the multiple-effect evaporators contains small amounts of volatile organic compounds and can be employed during cooking and liquefac­tion steps of the corn suspension, but the accumulation of inhibitors impedes the 100% recycling of these condensates. The condensates from evaporators can also undergo aerobic or anaerobic treatment, thus providing all the nutrients required (Wilkie et al., 2000). Palmqvist et al. (1996) proposed, from bench-scale data, to fraction the stillage by evaporation and recirculate only those fractions that have demonstrated having no inhibitory effects on fermentation using Saccharomyces cerevisiae in the case of willow wood hydrolyzates pretreated by steam explo­sion. The nonvolatile fraction has high inhibitory effects presumably caused by the presence of lignin degradation products. Other fractions have low levels of BOD and COD so they can be directly discharged without any treatment. Similar studies were carried out for pine and spruce (softwood) performing the simula­tion of evaporation step using a six-effect evaporation train to optimize the energy consumption (Larsson et al., 1997).

Brazil is a developing country that is a high producer and high consumer of bioethanol, and one that today exports biofuel to more than 10 countries in the world. Additionally, although more than 90% of all electrical energy in Brazil comes from hydroelectric sources (Colombia), sugar mills and distilleries sell the excess electrical energy they produce back to the grid. This electricity comes

TABLE 12.3

Use of Sugar Cane Produced in Brazil 2007

Cropped land with cane for sugar and other products/ha

Cropped land with cane for ethanol/ha

Sugar production/ton

Sugar consumption/ton

Sugar exports/ton

Fuel ethanol production/L

Fuel ethanol export/L

Source: Based on Foreign Agricultural Service. 2008. GAIN report BR8013. U. S. Dept. of Agriculture, Washington, D. C.; Food Outlook. 2007. Global market analysis 2007. FAO.

from burning the sugarcane trash and bagasse. Today, in Latin America, Brazil is a major producer (Table 12.3) extending its expertise to other countries, includ­ing technology and investments. Sugarcane in Brazil has a life cycle of 12 to 18 months and yields a range of 50 to 130 ton/ha.

Brazil is also the world’s largest consumer of sugar, with per capita consump­tion around 55 kg/year, just beating out Mexico and the United States. But, the reality today in Brazil is that sugarcane bioethanol production, as in other coun­tries with available land, has little effect on food production. This is explained by the fact that there is enough capacity for supporting new requirements in tool or expanded agricultural activities during the coming years. Another important fact to be considered is that today Brazil is the second largest producer of fuel ethanol in the world and simultaneously one of the largest food suppliers in the international market.

Ethanol (C2H5OH), also known as ethyl alcohol, fuel ethanol (when it is used as a fuel or gasoline oxygenate), or bioethanol (when it is obtained from biomass [energy-rich crops or lignocellulosic materials]), is the most widespread alcohol employed in the transport sector. Ethanol usage has many advantages and some disadvantages. An objective proof on advantages of this biofuel against its dis­advantages is the rising number of countries that have chosen it as a gasoline oxygenate or even as a direct fuel. Sometimes, the disadvantages of using fuel ethanol are emphasized with arguments that have been the subject of controver­sial debates between supporters and detractors of biofuels. This polemic is even more exacerbated when considering the temporal character of the governments, the main promoters of gasoline oxygenation programs using fuel ethanol. This discussion goes beyond the academic field (which is desirable) and falls into the political debate with the corresponding manipulation doses that it implies.