Optimization of Energy and Water Consumption in ... - ACS Publications

7972

Ind. Eng. Chem. Res. 2010, 49, 7972–7982

Optimization of Energy and Water Consumption in Corn-Based Ethanol Plants Elvis Ahmetovic´,† Mariano Martıń,‡ and Ignacio E. Grossmann*,‡ Department of Chemical Engineering, Carnegie Mellon UniVersity, Pittsburgh, PennsylVania 15213, and Department of Process Engineering, Faculty of Technology, UniVersity of Tuzla, 75000 Tuzla, Bosnia and HerzegoVina

In this paper we study the simultaneous energy and water consumption in the conceptual design of cornbased ethanol plants. A major goal is to reduce the freshwater consumption and wastewater discharge. We consider the corn-based ethanol plant reported in Karuppiah, et al. AICHE J. 2008, 54, 1499-1525. We review the major alternatives in the optimization of energy consumption and its impact in water consumption. Next, for each of the alternatives we synthesize an integrated process water network. This requires closing the loops for process and cooling water and steam, and implementing the proper treatment for the water streams. We show that minimizing energy consumption leads to process water networks with minimum water consumption. As a result, freshwater use is reduced to 1.54 galwater/galethanol, revealing that it is potentially possible to achieve levels of freshwater consumption that are significantly lower than the ones in current industrial operation and that wastewater discharged can also be reduced. 1. Introduction Water is the most valuable raw material for life on earth, but its wide availability in many regions in the world has made its price to be inexpensive. Despite an average annual cost increase of 6.7% according to the GWI/OECD,1 the price still remains low. As water is essential for economic development and for maintaining healthy ecosystems, and as world population grows and development requires increased consumption of water for the domestic, agricultural, and industrial sectors, the pressure on water resources intensifies. Figure 1 shows the water availability across the world. Based on the assessment on water resources, two-thirds of the world population will face water stress by year 2025.2 It is also estimated that by 2025, industrial water usage (which includes utility cooling and heating, processing, transportation, air conditioning, cleaning, etc.) will climb to 235 km3, accounting for about 11% of the total world water consumption.2,3 Thus, water consumption has become a major concern,4,5 making water resource management an important operational and environmental issue. In particular, the increasing costs of dependable water supplies and wastewater disposal have increased the economic incentive for implementing technologies that are environmentally friendlier, and that can ensure efficient use of water resources, including the treatment and recycling of wastewater6 as shown later in this paper. The task of synthesizing optimal process water networks has been performed using two different approaches: (a) conceptual engineering approach based on the water pinch heuristics and engineering experience7–13 and (b) systematic methods based on mathematical programming.14–20 Synthesis of heat exchanger networks has also been the topic of extensive research for a long time.21,22 However, research concerning simultaneous synthesis of process water and heat exchanger networks is not at the same level of development. The importance of simultaneous minimization of energy and water was first addressed by Savelski and Bagajewicz (1997),23 although this was only in the context of water networks without considering the process. Conceptual techniques24–29 and mathematical approaches * To whom correspondence should be addressed. Tel.: +1-412-2683642. Fax: +1-412-268-7139. E-mail: [email protected]. † University of Tuzla. ‡ Carnegie Mellon University.

have been used since then.30–37 Recently, a conceptual technique combining numerical and graphical tools has been proposed by Manan et al. (2009).38 Their method consists of three steps, namely, setting the minimum water and wastewater targets; design of minimum water utilization network, and finally, heat recovery network design. However, it is restricted to one contaminant which makes its application to be rather limited. 2. Problem Statement We consider in this paper the corn-based ethanol plant reported by Karuppiah et al (2008)40 to develop a conceptual design to optimize the energy and water consumption. We review first the major alternatives in the optimization of energy consumption. Next, for each of the alternatives we synthesize an integrated process water network. This requires closing the loops for process water and steam in order to establish the actual demand of water in all the process units. In this case study we show that minimizing energy consumption leads to process water networks with minimum water consumption. Furthermore, we also show that it is possible to achieve levels of freshwater consumption that are significantly lower than the ones in current operation. The paper is organized as follows. Section 3 provides an overview of the current consumption of energy and water in plants for the production of bioethanol from corn grains to fuel grade ethanol. In section 4, we review the results of the conceptual design by Karuppiah et al (2008)40 to identify the energy consumption and cooling utilities used in the following stages of the optimization: (a) superstructure optimization with no integration, (b) heat integration of the process, (c) replacement of distillation columns by multieffect columns, and (d), alternative design c with the optimization of reflux ratio in the multieffect columns. In section 5, we design the loops for cooling water and steam to establish the actual demand of water for bioethanol production for cases a to d cited above, introducing the modeling of the boiler and the cooling tower. No treatment for the wastewater is considered at this point. We show in detail the application of the closed loops to cases a and d cited before, but we report the water use for the two other cases too. Finally, in section 6 we implement the global optimization approach for the synthesis of process water networks proposed recently by Ahmetovic´ and Grossmann

10.1021/ie1000955  2010 American Chemical Society Published on Web 05/11/2010

Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010

7973

Figure 1. Water availability across the world. Reprinted with permission from ref 39. Copyright 2006 International Water Management Institute.

(2010)41 to the cases a and d from the paper by Karuppiah et al (2008).40 Three contaminants, total suspended solids (TSS), total dissolved solids (TDS), and organics (e.g., ethanol, organic acids, cells) are considered, and different specific treatment units are used. Only two examples are shown in detail in this section, but we report the results for the application of the water networks to all the four cases a-d. It is shown that the energy optimization achieved in the paper of Karuppiah et al40 together with the implementation of the water network, leads to a minimum water usage value of only 1.54 galwater/galethanol. 3. Corn-Based Ethanol Production Process Ethanol production has become one of the most important alternatives for the production of renewable biofuel owing to its compatibility with the supply chain of gasoline and its capability of being readily used in the current design of automobiles.42 Thus, governmental policies have been supporting its production.43 Currently, more than 95% of U.S. bioethanol is produced using corn. In spite of bioethanol’s environmental benefits like lower emissions, the volume of production of corn ethanol to meet the US policies43 has raised questions regarding its technological feasibility as an alternative fuel. A discussion has been presented on the availability of land for corn production,44–48 the energy demand either reporting favorable values49–60 or negative returns,61–78 and the emissions, where there seems to be an agreement in favor of biofuels compared to gasoline. Apart from these concerns, the National Research Council has recently warned that increases in corn ethanol production may significantly impact water consumption, highlighting ethanol’s dependence on water.79,80 In this paper, we focus on the problem of water consumption. To produce bioethanol, water is needed at two stages of the process. In the first one, irrigation is required to grow the corn.

In the second stage, water is used as raw material or utility in the production process. Generally speaking, the biomass needed to produce one liter of biofuel (under currently available conversion techniques) evaporates between 250 and 1000 gal of water.81 In particular, corn production on irrigated lands accounts for a major proportion of water use in agriculture and often involves depletion of aquifers. The corn devoted to ethanol production represents 13.3% of the total harvested corn. It is reported that between 263 and 832 gallons of water per gallon of ethanol are necessary to grow corn.4,5,82,83 For sugar cane, values ranging between 927 and 1391 gallons of water per gallon of ethanol have been reported.84 As potable water becomes less available in developed and developing countries, priorities for water use may affect the development of biofuel production.46,79,83,85 However, not all the results are so pessimistic because irrigation depends heavily on the region. That is the case of the production of sugar cane in Brazil whose climate conditions, a 365-day growing season and ample rainfall at the right times, allows sugar cane production at high yields with minimal or no irrigation.86 Moreover, different feedstocks may reduce water needs as is the case of switchgrass, a perennial warm-season grass that is grown for decades on marginal lands that are not well suited for conventional crop production. Different studies have been carried out to analyze the impact of the current policies in favor of the production of biofuels. According to them, the effect of biofuels production policies can be important in certain regions in terms of water consumption, but on the whole, its impact will be less than 5%. In contrast, if the amount to be produced is far larger in order to fully replace the current consumption of gasoline and diesel, the harvesting regions must be carefully selected not to have a big impact on food production and water availability.87

7974


Figure 2. Dry milling ethanol production process. Table 1. Water Usage for Ethanol Productiona fresh water demands cooling tower make up (%) boiler and process make up (%) overall water demand (gal H2O/gal EtOH)

corn ethanol: cellulosic ethanol: cellulosic ethanol: dry grind biochemical thermo-chemical 68

71

71

32

29

29

3-4

6

1.9

a Reprinted with permission from ref 82. Copyright 2007 SAHRA, University of Arizona.

The second stage in water consumption for the production of ethanol is in the production process itself. The ethanol production processes used so far, like wet mill or dry mill, are water intensive. Sometimes the process water required is more than the one that is available locally. A wide range of water consumption for the conversion process is reported, ranging from about 3 gallons of water per gallon of ethanol to as much as 15 gallons of water per gallon of ethanol.85 Therefore, recycle and reuse of process water is important, as well as the ultimate goal of “zero liquid discharge”. Approximately three-quarters of the bioethanol produced in the United States employs the dry milling process88 whose general flowsheet is shown in Figure 2. Corn grain is first pretreated so that the physical and chemical structures are broken down into the fermentable sugars. Milling breaks the shell of the grain so that saccharification and liquefaction of enzymatic processes can have higher efficiency in liberating the sugars. Fermentation takes place under anaerobic conditions using Saccharomyces cereVisiae, where ethanol concentration must be kept low due to its toxicity for the yeast. The main product of the process is ethanol. The liquid phase is separated from the solid phase by means of mechanical separation. The solids are dried to produce distiller’s dry grains with solubles (DDGS), a byproduct that is sold for animal feed. Ethanol is purified to its fuel grade by distillation and further water removal.89 According to the literature90 the lowest possible water consumption for corn dry grind process is 2.85 gal of water per gal of ethanol, or a more realistic 3-4 gal of water per gal of ethanol in the case of considering the water blowdown and evaporated from the cooling tower as shown in Table 1 from Aden (2007).82

grain to break the physical and chemical structure of the corn making the sugars accessible for fermentation. The process units employed are grinding, direct contact with steam, saccharification, and liquefaction. At the end of this sequence of physical and chemical treatments, sugars are liberated from the grain. The second section is the fermentation of the sugars, mainly glucose, into ethanol using a yeast, Saccharomyces cereVisiae. Water and starch are fed to the batch reactor. The amount of water required is such that the concentration of ethanol at the end of the fermentation is below toxic levels for the yeast. CO2 is also generated in the fermentation. After fermentation, two alternatives were proposed for the separation of solids from the slurry exiting the fermentor: mechanical separation (a) before the beer column (BC1) or (b) after the beer column. The third section comprises the technologies used for the purification and dehydration of ethanol to fuel grade. Three different options were considered: (1) a rectification column that can concentrate ethanol to the azeotropic composition, (2) adsorption of water in corn grits, and (3) molecular sieves. The superstructure is optimized in terms of energy consumption and the resulting flowsheet is shown in Figure 3. The separation of the solids takes place before the beer column, while the dehydration stage consists of the rectification column together with adsorption in corn grits with the final stage in the molecular sieves. Further details on the model can be found in Karuppiah et al. (2008).40 Karuppiah et al (2008)40 showed that a large reduction in energy consumption can be achieved, from 79.0 MW (case a in Figure 4) to 35.9 MW (case d in Figure 4) after performing overall heat integration in the plant and including multieffect distillation columns. The reduction in the energy consumption led to an important reduction of the cooling requirements for the plant, from 59.0 to 21.5 MW as shown in Figure 4. Assuming that no loops are used to recycle and reuse the process and cooling water and assuming a ∆T for the cooling water of 8 °C, the water consumption is reduced from 250 gal/ gal of ethanol (case a in Figure 4) down to 100 gal/gal of ethanol (case d in Figure 4) due to energy optimization. Since these assumptions are not realistic, closed loops for water consumption must be applied as discussed in the next section. 5. Closed Loops for Cooling Water and Steam

4. Review of Energy Optimization in Corn-Based Ethanol Plant Given the dry milling ethanol process in Figure 2, Karuppiah et al. (2008)40 optimized a superstructure for a plant producing 60 M gal ethanol/yr. The plant consists of three different sections. The first section involves the pretreatment of the corn

Karuppiah et al. (2008)40 did not consider closed loops for the cooling water and steam used in the process since they simply considered nominal prices for the utilities in their economic evaluation. To reduce the freshwater consumption, closed loops for cooling tower and steam systems are used in industry. We use the mass and energy balance of the bioethanol


7975

Figure 3. Optimal design of the corn-based bioethanol plant.

Figure 4. Reduction of energy consumption and cooling requirements by successive heat integration in the corn-based ethanol plant: (a) superstrucure optimization, (b) superstructure + HEN, (c) superstructure + HEN + multieffect distillation, (d) superstrucure + HEN + multieffect distillation + optimized reflux ratio).

case study developed by Karuppiah et al (2008),40 and introduce cycles of concentration for the cooling tower and the boiler to calculate the net water consumption for the closed loops for cooling water and steam in the cases described above as cases a-d. The cycles of concentration (COC) are defined as the ratio of the concentration of salts or dissolved solids in the circulating water or blowdown to that in the makeup water.91,92 In industrial practice, the cycles of concentration normally range from three to seven, and they are important in the design and operation of cooling towers.93 Figure 5 shows a closed loop of circulating water between the heat exchanger network and the cooling tower.

Figure 5. Closed loop for circulating water in cooling tower system.

From the cooling requirements in the heat exchanger network (heat rejected in cooling tower) QC(kW), the flow rate of circulating water FREC, in the cooling tower and heat exchanger network can be calculated from the equation: QC ) FRECcp,W∆T

(1)

where cp,W ) specific heat capacity of water (kJ/(kg °C)), ∆T ) temperature difference between inlet and outlet water in the cooling tower (°C). To calculate the evaporation loss in the cooling tower, which is the amount of water evaporated in the tower, an empirical correlation that is often used is the one by Perry and Green (1997):91 FE ) 0.00085∆TFREC (m3/h) × 1.8

(2)

7976


Figure 6. A simplified utility system.

The amount of water lost by drift, which is the liquid water in the tower discharge vapors, typically varies between 0.1 and 0.2% of the water supplied to the tower. However, new developments in the drift-eliminator design make it possible to reduce the drift loss below 0.1%.91 The makeup requirements for the cooling tower consist of the evaporation loss, drift loss, and blowdown: FM ) FE + FDW + FB

Figure 7. Cooling tower water balance.

(3)

As mentioned earlier, the cycles of concentration (COC) are the ratio of the concentration of salts or dissolved solids in the circulating water/blowdown cB (ppm) to that in the makeup water cM (ppm). COC )

cB cM

(4)

According to the literature94 the concentration of total suspended solids (TSS) in the outlet stream of the cooling tower is typically 50 ppm while that of the total dissolved solids is 2500 ppm. Figure 6 shows a simplified utility system consisting of a boiler, heat exchanger network (steam-using operations), and deaerator. For simplicity we assume that a single level of steam is generated. The steam generated in the boiler is used to supply heat in the heat exchanger network, while the steam condensate is returned to the boiler. In the case that there is no steam consumption in the process or no steam loss, fSL, in the heat exchanger network, the flow rate of generated steam in the boiler, and returned steam condensate is the same, fS ) fC. In addition, makeup water requirements for the boiler system will be equal to the discharged blowdown. According to this water balance for the boiler system, the mixer, and the heat exchanger network is given by the equations: fM ) fSL + fB

(5)

fFW ) fM + fC

(6)

fS ) fSL + fC

(7)

fB )

fS COC - 1

(8)

The American Boiler Manufacturers Association95 specifies that the concentration of TSS in the blowdown water from boilers is typically 10 ppm while that of TDS is assumed to be 500 ppm. Assuming the simplified utility system in Figure 6, the generated steam in the boiler can be calculated from the heat requirements in the heat exchanger network as given by eq 9: QH ) fS∆Hv

(9)

Figure 8. Boiler and cooling tower makeup streams versus cycles of concentration.

where ∆Hv ) latent heat of steam condensation (kJ/kg) for a given temperature and pressure. To control the buildup of contaminants in the closed boiler system, the blowdown has to be discharged and fresh makeup water supplied to the boiler so that none of the contaminants exceeds its limit. 5.1. Water Consumption for No Heat Integration Case with Water Loop. Considering the case of the optimized superstructure without heat integration (case a), the cooling requirements are 59.0 MW. We assume that the drift losses are 0.2% of the supplied water to the cooling tower, that the temperature difference between the inlet and outlet water in the cooling tower is 8 °C, and five cycles of concentration, COC, (a typical value that will be used along the paper). According to that, Figure 7 shows the results of the water balance of the cooling tower where it can be seen that a large amount of makeup water is lost by evaporation, about 82%, while the water loss by blowdown is about 18%. The amount of makeup water used and discharged blowdown depends on the cycles of concentration selected as shown in Figure 8. For the optimized superstructure without heat integration (case a), the heat requirements for corn-based bioethanol plant are 79.0 MW. Assuming five cycles of concentration (see Figure 8), steam pressure of 30 bar, and temperature of 233.8 °C, Figure 9 shows the results of the boiler water balance. It should be noted that superheated steam is injected into the Jet1 unit (direct heating) in the process (see Figure 3), and therefore there is no condensate returned to the boiler from this unit (it is considered as steam loss). Assuming closed loops for cooling water and steam but no reuse or recycle (for case a, Figure 4), the overall freshwater


7977

Figure 9. Boiler water balance.

consumption is 331.826 t/h, while the wastewater generation is 243.234 t/h. A large amount of water is lost by evaporation in the cooling tower that cannot be reused in water-using operations. However, as it will be shown in the next section, by heat integration in the corn-based bioethanol plant the cooling and heating requirements can be reduced leading to smaller evaporation loss in the cooling tower as well as lower freshwater consumption. 5.2. Water Consumption in Optimized Process for Corn Ethanol Production. In Figure 10 we present the water balance for the case of complete heat integration in the bioethanol plant (case d) using multieffect distillation columns (three columns for the beer column and two columns for the rectification column). The overall freshwater consumption is reduced from 331.829 t/h to 240.393 t/h and the wastewater generation from 243.234 t/h to 209.255 t/h. It is worth pointing out that with complete heat integration the water loss by evaporation in the cooling tower is reduced about 63% compared to that in case a (from 90.417 to 32.963 t/h). In industrial practice water consumption in corn-based ethanol plants is expressed as gallon of water per gallon of bioethanol produced. Figure 11 shows the results of water consumption

Figure 10. The complete water/steam network for case d.

Figure 11. Freshwater consumption with closed loops for cooling water and steam but without water networks for cases given in Figure 4.

for the four cases given in Figure 4. These values are in the range of the earlier values presented on water consumption in ethanol plants.82 To further reduce the water usage given in Figure 11, integrated process water networks are synthesized in the following section. 6. Water Consumption Optimization by Implementation of Water Networks 6.1. Water Network Superstructure and Model. To synthesize water networks in the corn-based bioethanol plant together with the cooling water and stream loops, we use the global optimization approach of the superstructure of integrated process water networks shown in Figure 12, which has been recently proposed by Ahmetovic´ and Grossmann (2010).41 The superstructure consists of one or multiple sources of fresh water of different quality, water-using processes, and wastewater treatment operations. The unique feature is that

7978


Figure 12. Generalized superstructure for the design of integrated water networks: PU, process unit; DU, demand unit; SU, source unit; TU, treatment unit. Reproduced with permission from ref 41. Copyright 2010 AIChE.

all feasible connections are considered between them, including water reuse, water regeneration, recycling, local recycling around process, and treatment units and pretreatment of feedwater streams. Multiple sources of freshwater include water of different quality that can be used in the various operations, and which may be sent first for pretreatment. The superstructure incorporates both the mass transfer and nonmass transfer operations. The mathematical model of the generalized superstructure consists of mass balance equations for water and the contaminants for every unit in Figure 12. The model is formulated as a nonconvex nonlinear programming (NLP) that is solved to global optimality. The objective function is to minimize the total network cost consisting of the cost of freshwater, the investment cost of treatment units, and the operating cost for the treatment units as given by the NLP model: min Z ) H

∑ FW

s

s∈SW

· CFWs + AR

∑ IC

t

R · (FTUout t ) +

t∈TU

∑

1 R IC · (FTUout t ) (10) 3 t∈TU t subject to splitter mass balances, mixer mass balances (bilinear), process units mass balances, treatment units mass balances. We assume the operating cost of the treatment units to be one-third of the investment cost.96 The details of the mathematical model can be found in Ahmetovic´ and Grossmann (2010).41 6.2. Design of Water Network. To synthesize the water network for the bioethanol plant using the superstructure optimization approach described above, the process and treatment units must be identified. First, the units that can be considered as water-using operations are the washing unit, the fermentor, the boiler, and the cooling tower. In the washing unit there is direct contact between the process stream (corn kernels) and the freshwater. The fermentor requires water, so

this unit is considered as a water demand unit. The beer and rectification columns recover water by separation and, thus, they are water source units. In addition to this, the flow rate of the boiler and cooling tower makeup water and blowdown is different because of the loss of steam in Jet 1 unit, and water by evaporation in the cooling tower. The specification of the treatment units must be done in accordance with the contaminants in the corn-ethanol process. Wastewater streams are generated from the boiler, cooling tower, and beer and rectification columns. Three main contaminants are considered: total suspended solids (TSS), total dissolved solids (TDS), and organics. Suspended solids are present in the water that is used for washing the corn, the organics are the main contaminants in the streams coming out of the distillation columns, while the dissolved solids include the concentration of salts as a result of the evaporation processes in the boiler and cooling tower. Furthermore, the water fed to the fermentor must have no ethanol, which is toxic for the yeast. We assume that there are three different wastewater treatment units, each one for removing one of the contaminants. To remove the solids, screens are widely used. Relatively large solids, like straw (0.7 mm or larger), can be removed in a primary screening facility. The simplest configuration is that of flow-through static screens, which have openings of about 1 mm. The removal rates vary depending on the size of the solids.97 We assume 99.9% removal for suspended solids. To purify the water from the distillation columns, a system of anaerobic and aerobic treatment units is required. The anaerobic stage will remove 90% of the organics generating biogas rich in methane that can be reused to obtain energy. Later, the water is treated in an aerated lagoon to obtain relatively clean water (100% removal) that can be recycled to the process according to the results presented by Zhang et al (2009).98 For this study, both treatments are integrated and modeled as a single treatment unit.


7979

Figure 13. Optimal water network for the optimized process production of corn-based bioethanol. Case study d: TU1, solids removal: TU2, organic removal.

Finally, to partially remove the total dissolved solids and to avoid any build up of salts in the system due to the operation of the boiler and the cooling tower, a reverse osmosis system is included. The literature reports removal up to 90%, which is better than ion exchange or nanofiltration.99 We will assume that the fresh water has no contaminants. The regulations require that the concentration of TDS in the effluent be at most 500 ppm.100 The cost correlations for the equipment involved in the network, screens,101 aerobic and anaerobic treatment units,102 boiler,103 cooling tower,104 and the reverse osmosis (RO)105 are as follows: Ccoolingtower ) 3229(Qcooling (kW))0.59

(11)

Cfurnace ) 2328.3(Qboiler (kW))0.7

(12)

Cscreens ) 10085A (m2)0.43

(13)

3

Q (m /s) , V (m/s) ) 1.6 V (m/s) (depends on sedimentation velocity)

(14)

Cscreens ) 4750(m (ton/h))

(15)

A)

0.43

Cbiological treatment ) Caeration tank + Canaerobic treatment = 1500(m (ton/h))1.13

(16)

CRO ) 3024m (ton/h)

(17)

The annualized factor (AR) for investment on the treatment units is taken to be 0.1, and the total time for the network plant operation in a year is assumed to be 8640 h. To compare the

results with the ones published by Karuppiah et al (2008),40 we used the same freshwater cost ($8.71 × 10-3/ton) as given in their paper. The relative optimality tolerance was set to zero, and we used the general purpose optimization software BARON106 to solve the global optimization of the nonconvex NLP problem. 6.2.1. Water Network for the Optimized Superstructure with No Heat Integration. For case a the freshwater consumption is reduced from 331.826 to 108.562 t/h when the optimal water network is design. This value represents a reduction of 73% compared to the same case without application of water reuse, regeneration, and recycling. The loss of water is due to evaporation in the cooling tower (79.613 t/h) and the discharge is 24.488 t/h. The water use for this case is 3.60 gallons of water per gallon of ethanol, which is a value in the range of the ones reported in the literature for new bioethanol plants.107 The total water network cost is $206,263/year. The presence of salts does not allow zero discharge since it expensive to treat TDS and the percent removal using reverse osmosis is not that high. 6.2.2. Water Network for the Optimized Superstructure with Heat Integration and Multieffect Columns. Figure 13 shows the optimal design of water network for case d, the optimized heat integrated corn-based ethanol plant. The freshwater consumption is reduced in this case from 240.393 to 40.822 t/h. This represents an 87% reduction compared to the same case without application of water networks. The evaporation loss in the cooling tower is 31.311 t/h and the wastewater discharged is reduced to 4.088 t/h as a result of the lower flow rates treated in the cooling tower and in the boiler. The total network cost ($201,548/year) is similar to the previous case

7980


Figure 14. Water usage after the application of water network to cases a-d.

operating costs. Two main utilities are minimized, steam, and freshwater. However, the cost for the equipment as well as the natural gas to feed the boiler will have to be added. At this point it is important to see whether including the new equipment has a large impact on the operating costs of ethanol. To be consistent with the results presented in Karuppiah et al. (2008),40 the cost for natural gas is calculated as 0.167$/kg (the price at the time that paper was prepared). If we take the production cost of corn ethanol reported in Karuppiah et al. (2008)40 for case a, it is 1.34 $/gallon, for case d, it is 1.24 $/gallon. If we add the cost of the water network (screens, biological treatment, furnace with 80% efficiency,110 cooling tower) and the utilities required (natural gas to feed the furnance) to these values, the operating costs for both cases increase only by a modest amount to 1.37 $/gallon and 1.30 $/gallon, respectively. The higher the cost of water, the more significant the application of water networks will be in terms of process economics since the decrease of water consumption is quite large as shown in this paper along the text and in Figures 11 and 14. 7. Conclusions

Figure 15. Reported data of water usage. Figure adapted from ref 107. Copyright 2008 Minnesota Technical Assistance Program, University of Minnesota.

($206,263/year). The reason is that the contribution of the cost of the freshwater to the total cost is rather small. The low cost of fresh water also results from avoiding the use of reverse osmosis for treating TDS. It is worth pointing out that the water consumption for this case is only 1.54 gallon of water per gallon of ethanol, which is much less compared to the data published in the literature and matches the industry goal of 1.5 gal/gal.108 Thus, this result is of great practical significance. Figure 14 shows water use and the wastewater discharged for cases a-d with the integrated water networks. Thus, we clearly show that minimizing energy consumption leads to process water networks with minimum water consumption and reduced wastewater discharge. The results presented in Figure 14 are very promising and allow an explanation of the current water consumptions. In industrial practice, the water usage in ethanol plants has improved in the past decade as shown in Figure 15107 where it can be seen that the newest plants show values in the range of the ones calculated in section 6.2.1. In the case a neither heat integration nor multieffect columns are used to minimize energy consumption, although water recycle and reuse is in operation. This value, however, can be reduced further to meet the claims of Delta T Company that states that values of 1.5 gal/gal are possible.108 There is discussion whether 1.5 galwater/galethanol is achievable but there is no demonstration yet of this in an operating plant. On the other hand, values of 2 galwater/galethanol have already been demonstrated for the production of ethanol from lignocellulosic materials.109 Our results suggest that it should be possible to achieve a value of 1.54 gal/gal for case d. 6.3. Costs and Benefits. The implementation of the integrated water networks results in a cost that has to be added to the

We have studied the energy and water consumption of cornbased bioethanol plants. As has been shown in this paper, water consumption can be reduced by energy optimization together with the recycle and reuse of process and cooling water and steam. Mathematical programming techniques have been used to optimize energy consumption and to synthesize an optimal process water network for corn-based bioethanol plants. The optimization of water consumption has been presented in three steps to show the contribution of energy optimization, water reuse, and recycle. The optimized water network yields the very promising value of water consumption of 1.54 galwater/ galethanol, which is the lowest value that has been reported to the knowledge of the authors, matches the industry goal of 1.5 galwater/galethanol . Thus, it has been shown that the process for corn-based ethanol can be designed so that it is not as water intensive as it is was in the past. Wastewater discharge is also reduced. However, zero discharge is only possible by improving the water purification technologies to reduce the operating costs and increase their efficiencies. Further decrease in water consumption can also be achieved by improving the performance of the cooling towers. Acknowledgment The authors gratefully acknowledge the Center for Advanced Process Decision-making at Carnegie Mellon University. Dr. Mariano Martıń acknowledges the Ministry of Education and Science of Spain and Fulbright commission for providing a MICINN-Fulbright Posdoctoral fellowship. Dr. Elvis Ahmetovic´ would like to express his gratitude to the Fulbright Program for Fulbright Scholar Grant and support throughout this work. Literature Cited (1) World Water Prices Rise by 6.7%. Global Water Intell. [Online], 2008, 9; http://www.globalwaterintel.com/archive/9/9/analysis/world-waterprices-rise-by-67.html accessed November 20, 2009. (2) Rosegrant, M. W.; Cai, X.; Cline, S. A. Food Policy Report; Global Water Outlook to 2025 (AVerting an Impending Crisis); International Food Policy Research Institute: Washington, DC, 2002. (3) Global Water and Food Outlook; SPARKS Companies, Inc.: Alexandria, VA, 2003. (4) Elcock, D. Baseline and Projected Water Demand Data for Energy and Competing Water Use Sectors, ANL/EVS/TM/08-8; Environmental Science Division, Argonne National Laboratory: Argonne, IL, 2008.

Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010 (5) Chiu, Yi-Wen; Walseth, B.; Suh, S. Water embodied in bioethanol in the United States. EnViron. Sci. Technol. 2009, 43, 2688–2692. (6) Petrakis, S. Reduce cooling water consumption. New closed loop cooling method improves process cooling tower operations. Hydrocarbon Process. 2008, 95–98. (7) Wang, Y. P.; Smith, R. Wastewater minimisation. Chem. Eng. Sci. 1994, 49, 981–1006. (8) Wang, Y. P.; Smith, R. Design of distributed effluent treatment systems. Chem. Eng. Sci. 1994, 49, 3127–3145. (9) Wang, Y. P.; Smith, R. Wastewater minimization with flowrate constraints. Chem. Eng. Res. Des. 1995, 73, 889. (10) Smith, R. Wastewater minimization. Waste Minimization through Process Design; Rossiter, A., Ed.; McGraw-Hill: New York, 1995; pp 93108. (11) Kuo, W. J.; Smith, R. Designing for the interactions between wateruse and effluent treatment. Trans. IChemE 1998, 76A, 287–301. (12) Forstmeier, M.; Goers, B.; Wozny, G. Water network optimization in the process industry case study of a liquid detergent plant. J. Clean. Prod. 2005, 13, 495–498. (13) Foo, D. C. Y. State-of-the-art review of pinch analysis techniques for water network synthesis. Ind. Eng. Chem. Res. 2009, 489, 5125–5159. (14) Takama, N.; Kuriyama, T.; Shiroko, K.; Umeda, T. Optimal water allocation in a petroleum refinery. Comput. Chem. Eng. 1980, 4, 251–258. (15) Doyle, S. J.; Smith, R. Targeting water reuse with multiple constraints. Trans. IChemE 1997, 75, 181–189. (16) Alva-Argaez, A. Integrated design of water systems. Ph.D. Thesis. UMIST, Manchester, U.K., 1999. (17) Bagajewicz, M. A review of recent design procedures for water networks in refineries and process plants. Comput. Chem. Eng. 2000, 24, 2093–2113. (18) Bagajewicz, M., Rodera, H., Savelski, M. A robust method to obtain optimal and sub optimal design and retrofit solutions of water utilization systems with multiple contaminants in process plants. Proceeding of the 7th International Symposium on Process System Engineering (PSE), 2000. (19) Saeedi, M.; Hosseinzadeh, M. Optimization of water consumption in industrial systems using linear and nonlinear programming. J Appl. Sci. 2006, 6, 2386–2393. (20) Karuppiah, R.; Grossmann, I. E. Global optimization for the synthesis of integrated water systems in chemical processes. Comput. Chem. Eng. 2006, 3, 650–673. (21) Linnhoff, B.; Townsend, D. W.; Boland, D. A User Guide on Process Integration for the Efficient Use of Energy; Institute of Chemical Engineers: U.K., 1982. (22) Papoulias, S. A.; Grossmann, I. E. A structural optimization approach in process synthesis. II. Heat recovery networks. Comput. Chem. Eng. 1983, 7, 707–721. (23) Savelski, M. J. Bagejewicz, M. J. Design and retrofit of water utilization systems in refineries and process plants. AIChE Annual Meeting, Los Angeles, 1997, Paper No. 188g. (24) Savulescu, L. E. Smith, R. Simultaneous energy and water minimisation. AIChE Annual Meeting, November 15-20, 1998, Miami Beach, FL. (25) Savulescu, L.; Sorin, M.; Smith, R. Direct and indirect heat transfer in water network systems. Appl. Therm. Eng. 2002, 22, 981–988. (26) Savulescu, L.; Kim, J. K.; Smith, R. Studies on simultaneous energy and water minimisation. Part 1. Systems with no water re-use. Chem. Eng. Sci. 2005, 60, 3279–3290. (27) Savulescu, L.; Kim, J. K.; Smith, R. Studies on simultaneous energy and water minimisation. Part 2. Systems with maximum re-use of water. Chem. Eng. Sci. 2005, 60, 3291–3308. (28) Zheng, X. S.; Feng, X.; Cao, D. L. Design water allocation network with minimum freshwater and energy consumption. Comput. Aided Chem. Eng. 2003, 15, 388–393. (29) Feng, X.; Li, Y.; Yu, X. Improving energy performance of water allocation networks through appropriate stream merging. Chin. J. Chem. Eng. 2008, 16, 480–484. (30) Bagajewicz, M. J.; Pham, R.; Manousiouthakis, V. On the state space approach to mass/heat exchanger network design. Chem. Eng. Sci. 1998, 53, 2595–2621. (31) Du, J.; Meng, X.; Du, H.; Yu, H.; Fan, X.; Yao, P. Optimal design of water network with energy integration in process industries. Chin. J. Chem. Eng. 2004, 12, 247–255. (32) Liao, Z.; Wu, J.; Jiang, B.; Wang, J.; Yang, Y. Design energy efficient utilization systems allowing operation split. Chin. J. Chem. Eng. 2008, 1, 16–20. (33) Dong, H. G.; Lin, C. Y.; Chang, C. T. Simultaneous optimization approach for integrated water allocation and heat-exchange networks. Chem. Eng. Sci. 2008, 63, 3664–3678.

7981

(34) Bogataj, M.; Bagajewicz, M. J. Synthesis of non-isothermal heat integrated water networks in chemical Processes. Comput. Chem. Eng. 2008, 32, 3130–3142. (35) Leewongtanawit, B.; Kim, J. K. Synthesis and optimization of heatintegrated multiple-contaminant water systems. Chem. Eng. Process. 2008, 27, 670–694. (36) Kim, J. Y.; Kim, J. K.; Kim, J. H.; Yoo, C. K.; Moon, I. A simultaneous optimization approach for the design of wastewater and heat exchange networks based on cost estimation. J. Cleaner. Prod. 2009, 17, 162–171. (37) Feng, X.; Li, Y.; Shen, R. A new approach to design energy efficient water allocation networks. Appl. Therm. Eng. 2009, 29, 2302–2307. (38) Manan, Z. A., Tea, S. Y., Alwi, S. R. W. A new technique for simultaneous water and energy minimisation in process plant, Chem. Eng. Res. Des., published online 2009, http://dx.doi.org/10.1016/j.cherd.2009.04.013. (39) Annual Report 2006/07; International Water Management Institute: Colombo, Sri Lanka, 2006; http://www.iwmi.cgiar.org/About_IWMI/ Strategic_Documents/Annual_Reports/2006_2007/theme1.html, accessed November 15, 2009. (40) Karuppiah, R.; Peschel, A.; Grossmann, I. E.; Martıń, M.; Martinson, W.; Zullo, L. Energy optimization for the design of corn-based ethanol plants. AIChE J. 2008, 54, 1499–1525. (41) Ahmetovic´, E.; Grossmann, I. E. Global superstructure optimization for the design of integrated process water networks. AIChE J. 2010, in press. (42) Cole, D. E. Issues facing the auto industry: Alternative fuels, technologies, and policies. ACP Meeting, Eagle Crest Conference Center, June 20, 2007. (43) Bush, G. W. Presidential State of the Union Address. Washington DC, 2007. (44) Hart, D.; Bauen, A.; Chase, A.; Howes J. Liquid biofuels and hydrogen from renewable resources in the U.K. to 2050: A technical analysis. E4tech (UK) Ltd., Study carried out for the UK Department for Transport, 2003, http://www.senternovem.nl/mmfiles/TechanalBiofuelsandH22050_tcm24187065.pdf (accessed December 2, 2009). (45) Hammond, G. P.; Kallu, S.; McManus, M. C. Development of biofuels for the UK automotive market. Appl. Energy. 2008, 85, 506–515. (46) Keeney, D. Ethanol USA. EnViron. Sci. Technol. 2009, 43, 8–11. (47) EU 2007 Energy Policy for Europe. http://eur-lex.europa.eu/ LexUriServ/site/en/com/2007/com2007_0001en01.pdf, accessed October 12, 2009. (48) USDA Agricultural Projections to 2018. www.ntis.gov, accessed November 10, 2009. (49) Shapouri H. Duffield, J. A.; Wang, M. Estimating the net energy balance of corn ethanol. U.S. Department of Agriculture (USDA), Economic Research Service, Agricultural Economic Report No. 721. 1995. (50) Shapouri, H.; Duffield, J. A.; Wang, M.; The energy balance of corn ethanol: an update: USDA, Office of Energy Policy and New Uses, Agricultural Economics. Rept. No. 813. 2002, 14 p. (51) Shapouri, H.; Duffield, J.; McAloon, A.; Wang, M. The 2001 Net Energy Balance of Corn-Ethanol (Preliminary); US Dept. Agriculture: Washington, DC, 2004. (52) McCain, J. Statement of Senator McCain on the Energy Bill: Press Release. Wednesday, November, 2003. (53) Marland G. & Turhollow A. F. CO2 emissions from the production and combustion of fuel ethanol from corn. Oak Ridge National Laboratory. Report No. 3301. 1990. (54) Moore, R., Producing Greener Ethanol, Ethanol Producer Magazine, September 2007. (55) Morris D.; Ahmed I. How Much Energy Does It Take to Make a Gallon of Ethanol; Institute for Self Reliance: Minneapolis, MN, 1992; http:// www.p2pays.org/ref/24/23824.pdf, accessed May 15, 2009. (56) Assessment of Net Emissions of Greenhouse Gases from EthanolGasoline Blends in Southern Ontario; Levelton Engineering Ltd. and (S&T)2 Consulting, Inc.: Ottawa, Canada, 2000. (57) Lorenz, D.; Morris, D. How Much Energy Does It Take to Make a Gallon of Ethanol? (Updated and revised); Institute for Local Self-Reliance, Minneapolis, MN, 1995; http://www.ethanol.org/pdf/contentmgmt/ILSR_ energy_balance.pdf, accessed May 15, 2009. (58) Graboski, M. Fossil Energy Use in the Manufacture of Corn Ethanol; Colorado School of Mines: Golden, CO, 2002. (59) Andress, D. Ethanol Energy Balances; David Andress & Associates, Inc.: Kensington, MD, 2002. (60) Wang, M.; Saricks, C.; Santini, D. Effects of Fuel Ethanol Use on Fuel-Cycle Energy and Greenhouse Gas Emissions, ANL/ESD-38; USDOE Argonne National Laboratory, Center for Transportation Research: Argonne, IL, 1999; http://www.transportation.anl.gov/pdfs/TA/58.pdf, accessed January 12, 2010.

7982


(61) Ho, S. P. Global warming impact of ethanol versus gasoline. Presented at 1989 National Conference for Clean Air Issues and American Motor Fuel Business, Washington, DC, October, 1989. (62) Keeney, D. R.; DeLuca, T. H. Biomass as an energy source for the mid-western US. Am. J. AlternatiVe Agric. 1992, 7, 137–143. (63) Citizens for Tax Justice. More corporate giveaways high on congressional agenda, July 22, 1997, http://www.ctj.org/html/cgive97.htm (accessed September 17, 2002). (64) Giampietro, M.; Ulgiati, S.; Pimentel, D. Feasibility of large-scale biofuel production. BioScience 1997, 47, 587–600. (65) Youngquist, W. GeoDestinies: The IneVitable Control of Earth Resources oVer Nations and IndiViduals; National Book Company: Portland, OR, 1997. (66) Pimentel, D. Ethanol fuels: energy security, economics, and the environment. J. Agric. EnViron. Ethics 1991, 4, 1–13. (67) Pimentel, D. Pimentel, M., Food, Energy and Society; Colorado Univ. Press: Boulder, CO, 1996; p 363. (68) Pimentel, D. Energy and Dollar Costs of Ethanol Production with Corn; Hubbert Center Newsletter No. 98/2; M. King Hubbert Center for Petroleum Supply Studies: Colorado School of Mines: Golden CO, 1998; p 7. (69) Pimentel, D. The limitations of biomass energy. Encyclopedia of Physical Science and Technology, 3rd ed.; Meyers, R., Ed.; Academic: San Diego, CA, 2001; Vol. 2, pp 159-171. (70) Pimentel, D. Ethanol fuels: energy balance, economics, and environmental impacts are negative. Nat. Resour. Res. 2003, 12, 127–134. (71) Pimentel, D.; Patzek, T. W. Ethanol Production Using Corn, Switchgrass, and Wood; Biodiesel Production Using Soybean and Sunflower. Nat. Resour. Res., 2005, 14, 65–76, 1. (72) CalGasoline. Ethanol is not a suitable replacement for MTBE. www.calgasoline.com/factetha.htm (accessed October 20, 2006). (73) Croysdale, D. Belatedly, DNR concedes our air is clean. The Daily Reporter, November 6, 2001; http://www.dailyreporter.com/editorials/eds/ nov07/asa11-7.shtml (accessed October 20, 2006). (74) NPRA Opposes Ethanol Mandate; Asks Congress Not To Hinder Efforts To Maintain Supply; National Petrochemical and Refiners Association: Washington, DC, 2002; http://www.npradc.org/news/releases/ detail.cfm?docid)164&archive)1 (accessed October 20, 2006). (75) Lieberman, B. The Ethanol Mistake: One Bad Mandate Replaced by Another. Competitive Enterprise Ins. 2002 http://www.nationalreview. com/comment/comment-lieberman 031202.shtml (accessed October 20, 2006). (76) Ferguson, A. R. B. Implications of the USDA 2002 Update on Ethanol from Corn; The Optimum Population Trust: Manchester, U.K., 2003; p 11-15. (77) Ferguson, A. R. B. Further implications concerning ethanol from corn: Draft manuscript for the Optimum Population Trust, 2004. (78) Patzek, P. Thermodynamics of the corn-ethanol biofuel cycle. Crit. ReV. Plant Sci. 2004, 23, 519–567. (79) Water Implications of Biofuels Production in the United States; National Research Council, National Academies Press: Washington, DC, 2008; pp 19-25. (80) http://www.greatlakesbioenergy.org/wp-content/uploads/2009/07/ fuel-standards-image.jpg. (81) CA 2007. http://www.iwmi.cgiar.org/WhatsNew/Biofuels/index.htm. (82) Aden, A. Water usage for current and future ethanol production. Southwest Hydrol. 2007, 22–23. (83) Jacobson, M. Z. Review of solutions to global warming, air pollution, and energy security. Energy EnViron. Sci. 2009, 2, 148–173. (84) SIU Sugarcane Industry Union; UNICA: Sao Paulo, Brazil, 2007. (85) Keeney, D. R.; Muller, M. Water Use by Ethanol Plants: Potential Challenges. www.agobservatory.org/library.cfm?refid)89449, 2006; accessed May 20, 2009. (86) de Cerqueira Leite, R. C.; Leal, M. R. L. V.; Cortez, L. A. B.; Griffin, W. M.; Scandiffio, M. I. G. Can Brazil replace 5% of the 2025 gasoline world demand with ethanol? Energy 2009, 34, 655–661.

(87) de Fraiture, Charlotte; Giordano, M.; Yongsong, L. Biofuels and implications for agricultural water use: Blue impacts of green energy. Water Policy 2008, 10, 67–81. (88) Risk assessment for bioenergy manufacturing facilities in Alberta. GenSolutions Final Report, 2007; Project No. E06-019, p 51. (89) Jacques, K.; Lyons, T. P.; Kelsall, D. R. The Alcohol Textbook, 3rd ed.; Nottingham University Press: Nottingham, UK, 1999. (90) Pfromm, P. H. The minimum water consumption of ethanol production via biomass fermentation. Chem. Eng. J. 2008, 2, 1–5. (91) Perry, R. H.; Green, D. W. Perry’s Chemical Engineer’s Handbook; McGraw-Hill: New York, 1997. (92) Mann, J. G.; Liu, Y. A. Industrial Water Reuse and Wastewater Minimization; McGraw-Hill: New York, 1999. (93) Fatigati, M. Conserving Water in Ethanol Plants. Wester Region BBI Biofuels Workshop, San Diego CA, 2006, http://www.bbibiofuels.com/ biofuelsworkshop/2006/docs/speakerpapers/west/bww/BWW06-14-Fatigati. pdf, accessed October 17, 2009. (94) American Public Health Association Standard Methods for the Examination of Water and Wastewater, 17th ed.; American Public Health Association, Washington, DC, 1989. (95) American Boiler Manufacturers Association. http://www.abma. com/. (96) Cargill. Personal communication, 2006. (97) Wang, L. K., Hung, Y.-T., Howard, H., Lo, H. H., Yapijakis, C., Li, K. H., Eds. Handbook of Industrial and Hazardous Wastes Treatment, 2nd ed.; Marcel Dekker, Inc.: New York, 2004. (98) Zhang, S.; Marechal, F.,; Gassner, M.; Perin-Levasseur, Z.,; Qi, W.; Ren, Z.; Yan, Y.; Favrat, D. Process modeling and integration of fuel ethanol production from lignocellulosic biomass based on double acid hydrolysis. Energy Fuels 2009, 23, 1759–1765. (99) Water and Wastewater. http://www.waterandwastewater.com/cgibin/yabb/YaBB.pl?board)general;action)display;num)1190280253, accessed March 10, 2010. (100) Patoczka, J. Impact of chemicals addition in water/wastewater treatment on TDS concentration and sludge generation. http://www. patoczka.net/Jurek%20Pages/Papers/Florida%2007%20Paper.pdf, 2007; accessed March 10, 2010. (101) Matches. http://www.matche.com/EquipCost/Screen.htm, accessed October 18, 2009. (102) Construction costs for municipal wastewater treatment plants: 1973-1977. Technical Report MCD-37; U.S. E.P.A.: Washington, DC, 1978 (adapted in http://www.fao.org/DOCREP/003/V9922E/V9922E09.htm), accessed October 18, 2009. (103) Matches. http://www.matche.com/EquipCost/Boiler.htm, accessed October 18, 2009. (104) Matches. http://www.matche.com/EquipCost/Cooling.htm, accessed October 18, 2009. (105) Matches. http://www.oas.org/usde/publications/unit/oea59e/ch20.htm. (106) Sahinidis, N. V. BARON: A general purpose global optimization software package. J. Global Optim. 1996, 8, 201–205. (107) Ethanol benchmarking and best practices. The production process and potential for improVement; Minnesota Technical Assistance Program, University of Minnesota: Minneapolis, MN, 2008, www.mntap.umn.edu/ MnTAP%20Ethanol%20Report.pdf, accessed August 15, 2009. (108) ThomasNet News. http://news.thomasnet.com/companystory/ 525435, accessed August 15, 2009. (109) Tao, L.; Aden, A.; Humbird, D.; Dutta, A.; Kinchin, C.; Inman, D.; Hsu, D. D.; Dugeon, D.; Lukas, J.; Olthof, B.; Schoen, P. Worley, M. Lignocellulosic biomass to ethanol. Process design and economics for corn stover. AICHE Annual Meeting, Nashville, Tennesse, 2009, Paper 239c. (110) Wallas, S. M. Chemical Process Equipment: Selection and Design, 3rd ed.; Butterworth-Heinemann: Oxford, UK, 1990.

ReceiVed for reView January 14, 2010 ReVised manuscript receiVed March 31, 2010 Accepted April 5, 2010 IE1000955

Optimization of Energy and Water Consumption in ... - ACS Publications

Recommend Documents