ABSTRACT Ethanol produced from biomass is one of the most promising renewable fuels.

However, most of the proposed separation methods use volatile organic compounds as solvents. Because such solvents are environmentally hazardous, new, efficient and sustainable ways of alcohol purification process ought to be developed. Ideally, no solvent should be required for the high purity purification of alcohol aqueous solutions. Therefore, in this work the modelling , simulation and control of an internally heat integrated pressure-swing distillation (IHIPSD) process to separate the ethanol/water azeotropic binary system into a high purity ethanol stream are addressed. By means of the proposed separation process, high-purity ethanol, suitable for use in the transportation sector, is obtained. Despite the high interaction between the column sections, the results show that the proposed separation process can be operated smoothly with an array of PI controllers. Moreover, the purity of ethanol is maintained with the control structure proposed in the face of upsets. INTRODUCTION Fossil fuels satisfy most of the primary energy needs in the world, contributing more than 80% in total energy consumption. In fact, in the road transportation sector accounts for over 40% of that energy demand (OPEC, 2009). In spite of the enormous benets that the intense use of fossil fuels has brought to mankind during the last century, the environmental damage has been considerable due to greenhouse gases emission produced during fossil fuel combustion. This environmental situation in addition to the imminent reduction in fossil fuel world production (Escobar et al., 2009) are the main reasons for the development and worldwide implementation of renewable alternative energy sources such as bioethanol and biodiesel. The production of ethanol from renewable feedstocks like sugar cane, corn, or lignocellulosic biomass is a viable option to reduce environmental pollution and dependence on fossil fuels (Eggemand & Atiyeh, 2010; Regalbuto, 2011). In 2006 The United States became the largest bioethanol producer using corn as raw material with around 4.7 billion gallons, Brazil was second place with a slightly smaller production (Demirbas, 2009) using sugar cane as raw material. Although presently most of the ethanol is manufactured from corn and sugar cane there is a growing concern regarding the use of these sources of raw materials because traditionally they have been widely used for feeding human populations. Moreover, large extensions of rainforest have been cleared for sugar cane plantation with a strong negative environmental impact. Presently, there are many serious research efforts for the production of ethanol from cellulosic residues that are world wide available (National Science Foundation, 2008). The success of this research efforts will allow dedicate edible biomass products for human consumption avoiding environmental damage. Therefore, non-edible biomass is the right strategy for the production of sustainably biofuels. Following the biological ethanol production route (Regalbuto, 2011), the transformation of biomass feedstocks into ethanol requires the conditioning or pre-treatment of the feedstocks for fermenting organisms to convert them into ethanol (Cardona & Snchez, 2007). The last step of the process implies ethanol dehydration which is not an easy separation operation owing to the presence of an homogeneous minimum-boiling azeotrope in a width range of pressures. Hence, a conventional
1

distillation scheme is not convenient. One of the aims of the present work is to propose a process conguration to obtain high purity dehydrated ethanol suitable for use in the transportation sector. The separation process should carry out the separation without using entrainers or solvents that do not possess sustainable characteristics. Regulations establish that ethanol used in the transportation sector should feature high purity (99.5 mol fraction). Therefore, because the ethanol/water system forms an azeotropic mixture at around 87.2% ethanol mol fraction and 1 atmosphere (Collura, 1985), the high purity ethanol separation rep- resents a technological challenge. In fact, one of the major hurdles in switching to a biomass-based industry is the production costs (both economic and energetic) of the base alcohols. Purifying an alcohol requires about 6% of the energetic value of the compound itself, with a large portion used in the separation of the alcohol from the fermentation broth, mainly composed of water (Qureshi, Hughes, Maddox, & Cotta, 2005). The average energy requirement per liter of ethanol produced is 2446 kcal (Morris & Ahmed, 1995).

Heterogenous azeotropic distillation (Luyben, 2006), homogeneous azeotropic distillation (Luyben & Abu-Eishah, 1985), membrane processes (Luyben, 2009) and pressure-swing distillation are common methods to separate azeotropic mixtures. Azeotropic distillation processes require addition of a third component as an entrainer to reach complete separation. However, the use of entrainers such as benzene has a negative environmental impact because of toxicity issues. Moreover, they can easily enter into the atmosphere and additional energy is required for recov- ery them. Presently, there is a search for new, sustainable products for the separation of azeotropic mixtures using new kind of sol- vents that display sustainable characteristics (Alvarado-Morales, Terra, Gernaey, Woodley, & Gani, 2009; Chavez-Islas, Vasquez- Medrano, & Flores-Tlacuahuac, in press; Seiler, Jork, Kavarnou, Arlt, & Hirsch, 2004). Distillation/membrane congurations are another way for dehydrating ethanol. Although the relevant results with membranes using simulation tools and pilot plant experiments (Tsuyomoto, Teramoto, & Meares, 1997), the implementation at industrial scale can result difcult because of operating issues.
2

As shown in Fig. 2a, conventional pressure-swing distillation is a dual-column process where columns operate at different pres- sures. This process is effective for azeotropic mixtures whose azeotrope composition changes with pressure so that it is possible to recover the feed components. Conventional pressure-swing distillation has been widely studied in the literature (Hamad & Dunn, 2002; Knapp & Doherty, 1992; Luyben, 2008a, 2008b). Besides, there are well-known commercially applications of this process for the separation of tetrahydrofurane/water, acetronitrile/water, methanol/methyl ethyl kenote, and acetone/methanol (Frank, 1997) which are azeotropic mixtures. In the case of ethanol aqueous solutions the ethanol/water azeotrope features low sensibility to pressure changes. However, this slight sensitivity is large enough so as to accomplish separation by pressure-swing, but sepa- rating that mixture using a conventional pressure-swing scheme
3

is an energy demanding process (Black, 1980; Huang, Zhan, Nakaiwa, Nakane, & Takamatsu, 1999). As shown in Fig. 1, the azeotropic composition of the ethanol/water system changes from 87.2 mol% ethanol at 1 atm to 79.4 mol% ethanol at 10 atm (the results shown in this gure were computed using the thermodynamic model described later in the steady-state and simulation section). Even though the shift is not very large, it is sufcient to perform sep- aration using a pressure-swing distillation scheme. Hence, this fact will be used to carry out the high purity ethanol separation without requiring the use of solvents. Of course, high-pressure operation requires the use of compression equipment increasing the processing costs. Even when in the present work we performed process design by a combination of heuristic and evolutive design techniques, the use of formal optimization techniques (Biegler, Grossmann, & Westerberg, 1997) can lead to improved designs featuring better economic prot. In this work an internally heat integrated pressure-swing distil- lation (IHIPSD) process is proposed to separate the ethanol/water system well beyond the azeotropic composition leading to obtain high purity ethanol. The owsheet of the process is shown in Fig. 2a. The operation of the proposed separation scheme is based on pressure-swing distillation and internal heat integration concepts. In comparison with the conventional pressure-swing distillation process, the IHIPSD scheme is composed of a single shell column which is divided in two sections, the high pressure (HP) section and low pressure (LP) section. The former section corresponds to the high pressure column and the latter section does for the low pres- sure column in a conventional pressure-swing distillation scheme. For pressure adjustment, a compressor and a throttling valve are set between the two sections of the IHIPSD process. Owing to pres- sure difference between the sections, a temperature driving force is established between the two sections so that internal heat trans- fer is accomplished. Due to total internal heat integration, a certain amount of heat is transferred from the HP section to the LP section which brings the liquid reux rate for the former and the vapor boilup for the latter. As a consequence, neither the reboiler in the LP section nor the condenser in the HP section are required and hence the energy requirements can be reduced. The internal heat integration phenomenon as occurs in the IHIPSD process is illustrated in Fig. 3. In the IHIPSD process the feed stream enters to the LP section. The bottom ow of this section is one of the products and the distillate, which is compressed, is the feed stream of the HP section. After pressure adjustment, the distillate of the HP section is recycled to the LP section and the corresponding bottom ow is the other product. As mentioned above, during the operation of the IHIPSD process certain amount of energy is transferred from the HP section to the LP section. Internal heat integration in distillation columns was rst pro- posed by Mah, Nicholas, and Wodnik (1977) and since 1985 theoretical studies have been conducted by Nakaiwa et al. (2003), who have conrmed by large-scale experimental evaluation the advantages and the feasibility of that kind of thermal coupling in distillation columns. This way of scheme the feed stream enters to the stripping section. The bottom ow of this section is one of the products and the distillate, which is compressed, is the feed stream of the rectifying section. After pressure adjustment, the bottom ow of the rectifying section is recycled to the stripping section and the distillate is the top product. During the operation of the HIDiC scheme, heat is transferred from the rectifying to the stripping section. It should be stressed that the IHIPSD process operates based on conventional pressure- swing
4

distillation, and the HIDiC operation basis is the conventional distillation process. Therefore, by removing the internal heat integration, the HIDiC and IHIPSD processes turn out to be different separation schemes.

LITRETURE SURVEY 1. Modeling, simulation and control of an internally heat integrated pressure-swing distillation process for bioethanol separation. (Francisco et al.,2011) i-HIDiC was conceptually introduced and evaluated by Mah and co-workers (Fitzmorris and Mah, 1980; Mah, Nicholas, & Wodnik, 1977) under the name Secondary Reflux and Vapouri- sation (SRV). Seader (1978/1980), and Glenchur and Govind (1987) introduced different column configurations for imple- menting i-HIDiC. A shell and tube-type packed column was then introduced for i-HIDiC by Aso, Matsuo, Noda, Takada, & Kobayashi (1996/1998). Recently, a group of Japanese researchers (Nakaiwa et al., 1997, 2000, 2001; Naito et al., 2000) studied i-HIDiC, concentrating on the theoretical evaluations and pilot plant testing. These results indicated energy saving up to a 60% with respect to conventional columns.

2. A thermo-hydraulic approach to conceptual design of an internally heat-integrated distillation column (i-HIDiC) (M.Gadalla et al., 2006) This paper introduces an effective method for the conceptual design of internally heat-integrated distillation columns (i-HIDiC). In i-HIDiC, the two column sections are operated at different pressures. The column configuration uses internal heat transfer on column trays between the different sections for heat integration, i.e. they are heat-integrated pressure swing distillation. The new method is based on stage temperature profiles and hydraulic calculations. The paper proposes a tool to assess the feasibility of design and operation of columns combining both the thermodynamic and hydraulic (physical space) capacities. An algorithm is suggested by which the stage temperature profiles are obtained and the pinched stages are identified. Thus, various feasible alternatives for i-HIDiC can be achieved. The work also introduces new terms which are useful for i-HIDiC designs and evaluation of column configurations. Finally, a model, based on hydraulic calculations and geometry analysis, is developed in order to quantify the capability of a column design for heat transfer areas. The model equations can be used in conjunction with commercial software, such as ASPEN+, in order to optimise basic designs for i-HIDiC.

3. Adding rectifying/stripping section type heat integration to a pressure-swing distillation (PSD) process (Huang et al., 2006 ) This paper studies the economical effect of considering rectifying/stripping section type heat integration in a pressure-swing distillation (PSD) process separating a binary homogeneous pressuresensitive azeotrope. The schemes for arranging heat integration between the rectifying section and the stripping section of the high- and low-pressure distillation columns, respectively, are derived and an
5

effective procedure is devised for the conceptual process design of the heat-integrated PSD processes. In terms of the separation of a binary azeotropic mixture of acetonitrile and water, intensive comparisons are made between the conventional and heat-integrated PSD processes. It is demonstrated that breaking a pressure-sensitive azeotropic mixture can be made more economical than the current practice with the conventional PSD process. For boosting further the thermodynamic efficiency of a PSD process, it is strongly suggested to consider simultaneously the condenser/reboiler type heat integration with the rectifying/stripping section type heat integration in process synthesis and design.

4. A New Pressure-Swing-Distillation Process for Separating Homogeneous Azeotropic Mixtures (Knapp et al.,1991) Mixtures in which the desired products lie in different distillation regions can be separated by a new pressure-swing process when one end of the distillation boundary is a pressure-sensitive azeotrope. Thus pressure-insensitive binary azeotropes can be separated using novel entrainers that form pressure-sensitive distillation boundaries. Because the columns already operate at different preasures, such sequences are readily thermally integrated. The separations of ethanol from water and acetone from methanol are used to demonstrate the new pressure-swing technique. These examples exhibit some interesting behavior such as (1) a region of multiplicity in the number of trays required to achieve the same separation at fixed reflux ratio, (2) a maximum reflux above which no feasible column exists, (3) a separation where the unexpected component is the distillate due to a reversal of the relative volatility as the pressure changes, and (4) a nonazeotropic separation that becomes easier as the pressure is increased. . 5. Design and Control of a Fully Heat-Integrated Pressure-Swing Azeotropic Distillation System (Luyben et al., 1985) If the composition of a binary homogeneous azeotrope changes significantly with pressure, the chemical components can be efficiently separated by using two distillation columns that operate at different pressures. This process is called pressure-swing azeotropic distillation. This paper updates the work of Abu-Eishah and Luyben (Ind. Eng. Chem. Process Des. DeV. 1985, 24, 132-140). These authors studied a partially heatintegrated two-column system in which an auxiliary reboiler permitted independent adjustment of vapor boilup in both columns. In the present paper, the two columns operate in the neat mode without any auxiliary reboiler. The details of achieving rigorous steadystate and dynamic simulations of the neat system in Aspen Plus and Aspen Dynamics are discussed. The condenser duty in the high-pressure column must equal thereboiler duty in the low-pressure column. The heat-transfer rate depends on the temperatures on the hot and cold sides in the condenser/reboiler, which change with varying process conditions, both in a dynamic mannerand at steady state. The pressure of the high-pressure column floats with changing conditions. The dynamic controllability of the fully heat-integrated system is compared with that of the partially heat-integrated system, and a significant tradeoff between energy consumption and controllability is demonstrated.

CHEMCAD DESIGN OF PRESSURE SWING DISTILLATION

Results :

Job Name: Project Chemcad Design `Stream No. Stream Name Temp (C) Pressure (atm) Enthalpy (MJ/h) Vapor mole fraction Total (kmol/h) Total (kg/h) Total std L (m3/h) Total std V (m3/h) Flowrates in (kmol/h) Ethanol Water 1 Feed 35.0000* 1.0000* -5.6E+005 0.00000 2000.0000 39396.4783 40.8172 44827.29 120.0000 1880.0000 2 85.0000 1.0000 -5.6E+005 0.00000 2000.0000 39396.4783 40.8172 44827.29 120.0000 1880.0000 3 150.3148 10.0000 -2.04E+005 0.00000 780.8359 32387.1650 40.0894 17501.38 653.0408 127.7950 4 78.0762 1.0000 -2.4510E+005 0.00000 899.7419 37848.3675 46.9460 20166.50 771.3524 128.3895

Stream No. Stream Name Temp (C) Pres (atm) Enthalpy (MJ/h) Vapor mole fraction Total ( kmol/h) Total (kg/h) Total std L (m3/h) Total std V (m3/h) Flowrates in (kmol/h) Ethanol Water

5 78.5181 11.0000 -2.45E+005 0.00000 899.7423 37848.3710 46.9460 20166.51 771.3524 128.3895

6 90.1917 10.6600 -2.43E+005 0.00000 899.7419 37848.3675 46.9460 20166.50 771.3524 128.3895

7 99.6131 1.0000 -5.26E+005 0.00000 1881.3231 33944.8118 33.9725 42167.31 1.8813 1879.4416

8 Water Product 44.3583 1.0000 -5.3E+005 0.00000 1881.3231 33944.8153 33.9725 42167.31 1.8813 1879.4418

Stream No. Stream Name Temp (C) Pres (atm) Enthalpy (MJ/h) Vapor mole fraction Total (kmol/h) Total (kg/h) Total std L (m3/h) Total std V (m3/h) Flowrates in (kmol/h) Ethanol Water

9 151.3180 10.0000 -30790 0.00000 118.9060 5461.2043 6.8566 2665.12 118.3116 0.5944

10 Ethanol Pdt 80.0000* 10.0000 -32191. 0.00000 118.9372 5462.6359 6.8584 2665.82 118.3426 0.5946

REFERENCE :

1. Jorge Francisco Mulia-Soto, Antonio Flores-Tlacuahuac(2011). Modeling, simulation and control of an internally heat integrated pressure-swing distillation process for bioethanol separation. Computers and Chemical Engineering, 35 ,1532 1546
2. Luyben, W. L. (2008a). Design and control of a fully heat-integrated pressure-swing azeotropic distillation system. Industrial & Engineering Chemistry Research, 47, 26812695. 3. Mah, R. S. H., Nicholas, J. J., & Wodnik, R. B. (1977). Distillation with secondary reflux and vaporization: A comparative evaluation. AIChE J., 23, 651658. 4. Luyben, W. L., & Abu-Eishah. (1985). Design and control of a two-column azeotropic distillation system. aI. Industrial & Engineering Chemistry Research, 24(1), 132140. 5. Knapp, J. P., & Doherty, M. F. (1992). A new pressure-swing distillation process for separating homogeneuos azeotropic mixtures. Industrial & Engineering Chemistr Research, 31, 346357. 6. K. Huang, K. Matsuda, K. Iwakabe, T. Takamatsu, M. Nakaiwa, Interpreting design of an ideal heatintegrated distillation column through exergy analysis, J. Chem. Eng. Jpn. 39 (9) (2006) 963970.

10