Process synthesis procedures can be significantly enhanced using process sim­ulation packages as shown in previous case studies. Simulation packages are essential for evaluating the amount of studied alternative process flowsheets. The described approach corresponds to knowledge-based process synthesis (see Chapter 2). In this regard, the task of process synthesis for the biotechnological production of fuel ethanol has mostly been undertaken using the approach based on the present knowledge. The other main approach is the optimization-based process synthesis that relies on the use of optimization for identifying the best configuration. For this, the definition of a superstructure, which considers a sig­nificant amount of variations in the topology of technological configurations of a given process, is required. The evaluation and definition of the best technological flowsheet are carried out through tools such as mixed-integer nonlinear program­ming (MINLP). The advantages and drawbacks of this strategy were disclosed in Chapter 2, Section 2.3.

Several cost-effective flowsheet configurations for the production of fuel etha­nol from renewable resources like biomass have been reported in the literature. Most of the proposed flowsheets have been defined using different heuristic rules and knowledge-based rules, and have involved the use of such tools as commercial process simulators. However, a cost-effective flowsheet for bioethanol production utilizing an optimization-based approach has not been found in the literature. In the following case study, an optimization-based strategy was implemented in order to preliminarily synthesize several technological schemas for ethanol production from lignocellulosic biomass employing a net revenue function as a comparison criterion.

Tetraethyl lead (TEL) is the major oxygenating additive of gasoline that has been used in the world in the recent past (Figure 1.2a). The U. S. scientist Thomas Midgley discovered the excellent properties of TEL in December 1921. One liter of TEL was enough to treat 1,150 liters of gasoline. Thus, the oil companies began the addition of this compound to the gasoline instead of ethanol. In the case of the latter, its use would have reduced the utilization of gasoline by 20 to 30% making the vehicles less oil dependent. Leaded gasoline has an average octane number of 89 (Nadim et al., 2001). However, some concerns were expressed in 1923 about the negative effect of TEL on human health. In particular, it was pointed out that each liter of consumed gasoline would emit 1 g of lead oxide. Later studies demonstrated the negative neurological effects of this compound, especially in children (Thomas and Kwong, 2001). It wasn’t until January 1996 that leaded gas­oline was banned in the United States. In Colombia, leaded gasoline is no longer produced. In fact, its import has been banned since 1994 (Ministerio del Medio Ambiente, 1994). Nevertheless, there exist some African and Mideast countries where leaded gasoline is still being used because of its low cost (Thomas and Kwong, 2001).

One of the ways of avoiding TEL usage is the modernization of refineries in order to elevate the production of aromatic and aliphatic hydrocarbons, but these modifications are costly. In addition, the augmentation in the concentration of aromatic compounds in the gasoline increases the risks related to the benzene exposition. Another option consists in the substitution of TEL with other less toxic compounds. Both options represent additional expenses in the refineries that explain why the phase-out of TEL utilization has been very slow in spite of hav­ing known its toxic properties for a number of years.

As noted above, sugar mills also employ the clarified syrup for ethanol produc­tion. The equivalent of this syrup in the sugar production plants using sugar beet, the dense juice, is also used for bioethanol production. On the other hand, a spe­cial type of syrup called high test molasses is employed for ethanol production especially in the distilleries of the southern United States. These types of molas­ses are usually produced when the sugar prices are very depressed in the world market, thus the sugar mills simply concentrate the cane juice to obtain syrup, to

which acid is added in order to partially hydrolyze the sucrose and avoid its crys­tallization (Murtagh, 1995). In this way, the sugar mills commercialize the high test molasses as a way of exploiting the sugarcane under low sugar price condi­tions. High test molasses contains about 70% sugars and 80° Brix that indicates high sugar content compared to molasses. Therefore, these types of materials, as well as the syrups, require the addition of higher amounts of nitrogen and phos­phates when they are used to prepare the cultivation media for ethanolic fermen­tation using yeasts. Moreover, the pantothenate content of these molasses is low (Murtagh, 1995).

There exists a significant number of species of yeasts and bacteria having the ability to synthesize ethanol. Main ethanol-producing microorganisms that are currently used in the industry or have potential utilization in the future are shown in Table 6.2. Among the main criteria in choosing a microorganism producing

Ethanol Tolerance Remarks References 150 g/L Lip to 95% of theor. yield Claassen et al. (1999) High osmotic tolerance Bullock (2002); Goyes and Bolanos (2005) 80-90% of theor. yield Kadar et al (2004) 30-45 g/L 94% of theor. yield on xylose; Olsson and Hahn-Hagerdal produces moderate amounts of xylitol (1996); Jeffries and Jin (2000) 35-47 g/L 92% of theor. yield on xylose Olsson and Hahn-Hagerdal (1996); Jeffries and Jin (2000) 37.5-75 g/L 47-55% of theor. yield on Olsson and Hahn-Hagerdal xylose; produces large amounts of xylitol (1996); Jeffries and Jin (2000) 100 g/L Up to 97% of theor. yield Claassen et al. (1999) 10-30 g/L Produces acetic acid Claassen et al. (1999); South et al. (1993) <30 g/L 41 % of theor. yield on xylose Olsson and Hahn-Hagerdal (1996)

ethanol, the ability to assimilate a wide range of substrates should be emphasized. Unfortunately, alcoholic fermentation presents an end-product inhibition, so one of the most desired features of the potential microorganisms is a high ethanol tol­erance. Similarly, the cultivation conditions allow defining some desirable features of the microorganisms to be used. For instance, elevated temperatures allow the reduction of cooling costs as well as the acceleration of the metabolic processes (although this can imply an increase in the inhibitory effect of ethanol). Increased temperatures can be advantageous when the microorganisms are cultivated along with hydrolytic enzymes in simultaneous hydrolysis and fermentation processes of either starch or cellulose since these enzymes have, in general, optimum tem­peratures of enzymatic activity higher than fermentation temperature. Finally, one of the most important selection criteria is the achieved ethanol yield on substrate because the economy of the overall process directly depends on this parameter.

The use of saline extractive agents in extractive distillation processes has been studied in the past few years. Among these agents, potassium acetate should be highlighted (Sanchez and Cardona, 2005). As the salts are nonvolatile compo­nents, the distillate obtained in distillation columns is much easier to separate. Therefore, the energy costs are lower compared to traditional extractive distilla­tion. The salt effect consists of the preferential solvation of the ions formed from the salt dissociation with the less volatile component that leads to the increase of the relative volatility of the most volatile component (ethanol).

Two possible configurations have been simulated for ethanol dehydration using potassium acetate (Ligero and Ravagnani, 2003). In the first case, the starting point is a diluted ethanol solution that is fed to the saline extractive distillation column followed by a multistage evaporation and spray drying for salt recovery. In the column distillate, the anhydrous ethanol is obtained due to the breaking of the azeotrope induced by the salt. The second case has a previous distillation column to concentrate the ethanol solution, which is fed to the saline extractive distillation column with the subsequent salt recovery through spray drying. This second option presents best results from the viewpoint of energy costs. In addi­tion, the salt used is not toxic, which indicates that this process can replace the traditional azeotropic distillation that uses such toxic entrainers as benzene.

Rigorous models for saline extractive distillation for ethanol-water-CaCl2 system have been developed. These models take into account the contribution of the salt to the liquid phase enthalpy (Llano-Restrepo and Aguilar-Arias, 2003). This type of model allows simulating the phase equilibrium and thermody­namic properties of solvent-electrolyte mixed systems having a strong nonlin­ear character that becomes an obstacle for their study. In this way and starting from theoretical considerations, the production of water-free ethanol using CaCl2 as the separation agent was predicted. These rigorous models can make possible the utilization of this type of process again by the industry considering that saline extractive distillation was discarded in the past due to, among other reasons, the difficulty in its design, which was caused by the complexity of the nonideal behavior of the phase equilibrium and the intricate modeling of the thermodynamic properties of these systems. This process was also discarded because of the technical problems derived from the dissolution and subsequent crystallization of the salt, as well as the need to use materials resistant to the corrosion (Pinto et al., 2000).

Besides inorganic salts, the possibility of employing hyperbranched poly­mers like polyesteramides and hyperbranched polyesters during the extractive distillation of ethanol-water mixtures has also been studied. These polymers exhibit a remarkable separation efficiency and selectivity, and their physical — chemical properties can be tailored depending on the required application. The recovery of these polymers can be carried out by washing, evaporation, drying, or crystallization (Sanchez and Cardona, 2005; Seiler et al., 2003).

The stillage that cannot be used as animal feed and that has high BOD and COD values can be treated by anaerobic digestion. During this process, the transforma­tion of the organic matter by a mixed bacterial culture is carried out in such a way that the BOD is reduced and the resulting sludge can be more easily disposed of, i. e., it is stabilized. Wilkie et al. (2000) have reviewed the conditions for accom­plishing the anaerobic digestion of different stillage types under both mesophilic and thermophilic conditions and have reported the main configurations of the equipment required by this process. COD reductions of 97% for corn stillage in an up-flow anaerobic sludge blanket (UASB) reactor, as well as 94% reduction of the soluble COD of sugarcane stillage, have been achieved. For stillage derived from wheat starch, some studies at pilot plant scale have shown 89 to 92% reduc­tions of COD (Nguyen, 2003).

The production of sludge by anaerobic digestion is about 10% lower than the sludge produced by aerobic digestion. This anaerobic sludge can be used as a feed­stock for producing components for balanced animal feed (Wilkie et al., 2000). The effluents from anaerobic digestion contain plant macronutrients (nitrogen, phosphorous, and potassium), as well as micronutrients (iron, zinc, manganese, copper, and magnesium), so they could be applied to plantations ensuring an appropriate dosage. These nutrients should be removed when the liquid efflu­ents are to be directly discharged into the natural water streams. If they are not removed, eutrophication problems can arise. These nutrients can be removed through the cultivation of algae in special ponds containing these effluents. In general, the stillage treated by anaerobic digestion should undergo color removal through different methods, including flocculation and coagulation, photovoltaic removal, or microbial removal (Wilkie et al., 2000). One alternative for degrada­tion of the organic matter contained in the stillage is the use of aerobic processes like high-rate oxidation ponds for those cases when low cost lands are available (Olguin et al., 1995).

The productivity and other characteristics of biofuel crops are factors to be con­sidered as a potential response to food security concerns. Plants have not been domesticated for high-scale biofuel production, although the biotechnology advance is the quickest and most efficient way to rationally transform plants to biofuel feedstock. Here the differences between highly developed and developing countries are so big that influence of biofuels on food security is not a priority. For example, the United States has up to 60 million hectares of transgenic (genetic modified) crops mainly for livestock and biofuels; at the same time, Colombia uses 50,000 hectares for textile and other uses. So, for the transgenic soy and corn, where productivity per hectare could be four times more than in the case of the same native crops, one hectare in the United States replaces four hectares in Colombia. The challenges or opportunities in biotechnology for improving biofuel crops have been studied in detail by Gressel (2008). The author of this work, as many other scientists in the world, considers that transgenic crops are imperative for the production of biofuel. Based on this work and other reviews, the expected contribution from biotechnology to bioethanol production, from the point of view of food security, is presented in Table 12.4.

Global population may possibly exceed 6 billion by 2050. Approximately 90% of the global population will reside in Asia, Africa, and Latin America. Today, the population of these countries suffers from malnutrition problems and energy insecurities. Transgenic crops represent promising technologies that can make a vital contribution to global food and biofuels security. However, harmony and high level research (to avoid the possible dangerous aspects of using transgenics directly for human food) are the key words to reach these purposes.

In this chapter, the importance of conceptual process design during the develop­

ment of innovative process configurations is analyzed. The utilization of some useful tools for process design is highlighted, such as mathematical modeling and simulation. The key role played by process synthesis methodologies is empha­sized. The two major strategies for accomplishing process synthesis are discussed. Main trends in knowledge-based process synthesis are briefly presented as well as the main approach to carrying out optimization-based process synthesis.

2.1 CONCEPTUAL PROCESS DESIGN

Process systems engineering deals with the development of procedures, techniques, and tools to undertake the generic problems of design, operation, and control of productive processes related to the different sector of process and chemical indus­try (Perkins, 2002). As part of process systems engineering, process design plays a fundamental role during the development of efficient technologies, especially in technoeconomic and environmental terms, in order to produce a wide range of industrial products. In this way, its main objective consists of the selection and def­inition of a process configuration that makes possible the conversion of feedstocks into the end product. This should be done in such a way that the products meet given specifications and that the configuration performance is superior to existing ones, or nonexisting ones in the case of a new product introduced into the market. In general, process design can be accomplished from the perspective of sequential engineering, concurrent engineering, reverse engineering, or reengineering.

From the viewpoint of the life cycle of an industrial process, the sequential engineering in its classical version involves all the steps presented in Table 2.1 in a sequential manner. Reverse engineering is based on obtaining technical infor­mation of a product in order to determine what it is made of, what makes it work, and how it was constructed. This approach is particularly useful in the case of pharmaceuticals production, especially during the production of generic drugs. For ethanol, this approach is not applicable because its chemical structure and ways of production are relatively well known. Reengineering looks for the radi­cal change of process design through its fundamental revision to achieve decisive improvements in terms of quality, costs, celerity, flexibility, customer satisfac­tion, etc., all of them simultaneously. Reengineering does not look for incremental improvements, but drastic changes that allow reaching the predefined targets. On the other hand, concurrent engineering proposes the creation of a design environ­ment where all the actors involved in the development of a product participate,

TABLE 2.1

Steps Involved in the Development of an Industrial Process during its life Cycle

not only the designers. In this way, processing features and market demands will be taken into account during early stages of design, when the changes are easier and less expensive to implement. Therefore, the problems, conflicts, and change needs can be detected in time to carry out the necessary modifications with sub­stantially less effort than by means of sequential engineering. In this framework, concurrent engineering applied to design implies the integration of all life cycle steps of a process in the early stages, attaining the achievements of several goals. Thus, the research and development activities and conceptual design have to con­sider not only the building of a plant, its operation, control, and maintenance, but also the achievement of technoeconomic, market, environmental, and even social objectives. Practically, and for the case of commodities, the three first stages of the process life cycle are accomplished in a concurrent (simultaneous) way. If the synthesis pathways for a given product are already known, as in the case of fuel
ethanol, process design procedures are focused on the second and third stages in Table 2.1.

In this book, these two steps are analyzed for fuel ethanol production empha­sizing the related integrated processes. In fact, concurrent engineering elements are considered when taking into account as evaluation criteria not only technical indexes (yield, productivity, energy consumption), but also financial and envi­ronmental indicators in the framework of process intensification. Financial and environmental criteria correspond to the macro and mega scale levels of analysis (plant and unit integration and interaction between market conditions and envi­ronmental impact, respectively), as reported by Li and Kraslawski (2004). Just these two levels of analysis have been developed with more intensity in the past 15 years as a result of the globalization of the economy and worsening environment. On the other hand, process and apparatus integration corresponds to the micro scale level of analysis. At this level, process intensification through integrated and hybrid processes with higher efficiency and less size has become the most important development trend. This forced the change of the old paradigm of a chemical process made up of a series of unit operations where the processes and apparatus are coupled (meso scale). Finally, the nano scale (molecular design and new materials) has become crucial for designing processes to obtain very high value-added products.

The task of defining an appropriate process configuration requires the genera­tion and evaluation of many technological schemes (process flowsheets) in order to find those exhibiting better performance indicators. This task is called process synthesis. In a process synthesis problem, system inputs (type, composition, con­ditions, and flowrates of raw materials) and outputs (product flowrate and speci­fications, effluent streams constraints) are given and the task consists of defining the configuration of the process flowsheet or, in other words, the topology of the technological scheme, which allows the synthesis of the product from the feed­stocks entering the process. For this, at least one comparison criterion should be established with the aim of evaluating different alternative process flowsheets proposed in order to choose that with the better performance.

The configuration comprises the type and amount of unit processes and opera­tions required by the overall process as well as their interconnection (intermedi­ate, recycle, and purge streams) and the parameters of that configuration (mostly those ones related to operating conditions: flowrates, temperatures, pressure, compositions). Process synthesis procedures can be applied not only to the con­ceptual design of new processes, but also to the retrofitting of existing ones. Some approaches for process synthesis involve and apply fundamental concepts of ther­modynamics as the starting point for generating new alternative process configu­rations. Thus, energy consumption (calculated by enthalpy balances) of different flowsheets can be helpful for selection of the best alternatives. In a similar way, the concept of useful energy or exergy (widely employed in mechanical engineer­ing) can also be employed as a criterion for selection of alternatives. Recently, more global concepts from the ecology field, such as emergy, have been used for choosing the best configuration of a process (see Section 2.2.5).

During the next step of the life cycle of an industrial process, process analy­sis, the structure of the selected technological scheme is established in order to improve the global process through its more comprehensive insight. The type of problems undertaken by process analysis is summarized as follows: given the process inputs and once determined the technological configuration of the pro­cess that includes each one of the unit operations and processes involved, as well as their parameters, find the system outputs. The analysis is aimed at predicting how the given process behaves. It involves the process decomposition into its con­stituent elements for the individual study of each unit performance. The detailed features of the process (flowrates, pressures, temperatures, compositions) are pre­dicted by using mathematical models, empiric correlations, and computer-aided process simulation. Moreover, experimental methods are employed to study the system behavior as well as validate the theoretical approach used for its descrip­tion. It could be considered that the conceptual design stage corresponds to pro­cess synthesis activities and the development stage to process analysis activities, although there is no clear boundary between these two activities of process sys­tem engineering.

Though process synthesis is carried out prior to process analysis, both tasks interact with each other in order to achieve their goals. Thus, for evaluating the performance of an alternative technological configuration during process synthe­sis, mathematical models are required in order to predict the behavior of process units. This involves a distinctive task of process analysis. Usually, aggregated models are employed in process synthesis. Such models simplify the synthe­sis problem in a considerable way through the representation of one aspect or objective that tends to dominate the problem. Furthermore, short-cut models are utilized as well. In this kind of model, the description of the units involved in each proposed design is done through relatively simple nonlinear models with the aim of reducing the computational costs or exploiting the algebraic structure of the equations. During process analysis, more rigorous and complex models are involved to predict the performance of the different units, which make up part of a technological scheme (Grossmann et al., 2000). On the other hand, process opti­mization plays a very important role during process design. Once the structure with the best performance (this can, in turn, imply the employment of optimiza­tion tools) is defined, and knowing the structure of the system components that allow the prediction of its behavior, the optimization of the technological scheme can be accomplished in order to find the optimal operating parameters making possible the maximization or minimization of an objective function that evaluates the performance of the overall system based on one or more criteria (technical, economic, environmental).

Approximately 1.6 million ton per year of municipal solid waste (MSW) is pro­duced in the world. This waste incurs serious problems due to the progressive deterioration of the environment and also to the costs associated with its recollec­tion, transport, and disposal. It is estimated that the services for disposal, treat­ment, and economic utilization of MSW accounts for an annual market of $100 million worldwide, from which $42 million corresponds to North America, $42 million to the European Union, and only $6 million to South America, having the corresponding MSW volumes of 250, 200, and 150 million ton per year, respec­tively (Skinner, 2000). The organic fraction of MSW (that can reach more than 65%) is an extraordinarily heterogeneous mixture of materials in which materials with a lignocellulosic nature predominate: paper, cardboard, fruit and vegetable peels, garden residues, wood items, etc. The production of ethanol from MSW has already been patented (Titmas, 1999). MSW presents high amounts of inhibi­tors of its conversion process, which not only originate during the processing, but also come from the wastes themselves. Despite this, the possibility of producing ethyl alcohol from this waste has been demonstrated (Green et al. 1988; Green and Shelef, 1989). Many of these residues have significant starch contents especially in the case of wastes from marketplaces. In this case, the presence of discarded fruits with different maturity degrees provides the starch, which can be hydrolyzed using enzymes that degrade this polysaccharide and the cellulose coming from the lignocellulosic biomass present in these residues as well (Cardona et al., 2004).

The technology for fermentation of sucrose-based media, mostly sugarcane juice, cane molasses, or beet molasses, can be considered as a mature technology espe­cially if the process is accomplished in batch regime. However, many research efforts are being made worldwide to improve the efficiency of this process, par­ticularly to increase the conversion of these feedstocks into ethanol as well as to gauge its productivity. Ethanolic fermentation can be carried out by discon­tinuous, semicontinuous, and continuous processes. In general, temperature and medium pH are quite similar in the different types of ethanolic fermentation: about 30°C and pH in the range of 4.0 to 4.5. Nevertheless, each one of these cul­tivation regimes presents very different performance indicators, especially con­sidering the ethanol volumetric productivity in terms of g EtOH/(L x h). Another performance indicator is ethanol yield defined as the grams of ethanol produced from one gram of substrate consumed (usually the carbon source, i. e., the sugars). Finally, substrate conversion is defined as the ratio between the amount of sub­strate consumed and the initial amount of substrate either loaded at the start of
fermentation for batch regime or contained in the feed stream for continuous regime. This indicator can be expressed in percentage. It is worth emphasizing that substrate conversion implies that the microorganisms consume the substrate not only for ethanol biosynthesis, but also for formation of new cellular biomass and other substances generated as fermentation by-products. S. cerevisiae is the most employed microorganism for industrial ethanol production from sucrose — based media.