A Novel Heat Integrated Extractive Dividing Wall Column for Ethanol

Subscriber access provided by UNIV OF LOUISIANA

Process Systems Engineering

A Novel Heat Integrated Extractive Dividing Wall Column for Ethanol Dehydration Md Aurangzeb, and Amiya K. Jana Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.9b00988 • Publication Date (Web): 10 May 2019 Downloaded from http://pubs.acs.org on May 12, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

A Novel Heat Integrated Extractive Dividing Wall Column for Ethanol Dehydration Md Aurangzeb and Amiya K. Jana* Energy and Process Engineering Laboratory, Department of Chemical Engineering, Indian Institute of Technology Kharagpur, West Bengal 721302, India Abstract This work aims at developing a novel extractive dividing wall column (E-DWC) and its heat integrated scheme to produce an anhydrous ethanol. The first proposed E-DWC involves only a bottom reboiler driven by steam, while in the second proposed configuration called heat integrated E-DWC (E-HiDWC), we make thermal coupling between the hot ethylene glycol (a solvent) with the cold overhead ethanol vapor, intermediate water vapor and fresh feed. This strategy leads to reduce high compression ratio, which has a large influence on the capital investment and electricity cost. This results in energy and cost savings. To make the proposed DWC more realistic, the heat transfer through the vertical dividing wall is taken into account, which also leads to cut down the reboiler duty. Further, for a meaningful comparison between the proposed DWC schemes and their conventional column, the input and output specifications of all these configurations are attempted to keep close, if not same, by developing a variable manipulation policy. The optimal process parameters are identified by minimizing the total annual cost for a target purity of more than 99.5 mol% for each component, namely ethanol, water and solvent. The energetic and economic potential are finally evaluated for the optimal E-DWCs with reference to its conventional counterpart. Keywords: Extractive dividing wall column; heat transfer through dividing wall; optimization; energy savings; total annual cost; ethanol dehydration

*

Corresponding author. Tel.: +91-3222-283918; fax: +91-3222-282250. E-mail address: [email protected] (A. K. Jana).

ACS Paragon Plus Environment

1

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

1. Introduction Fossil fuels are the primary and still the most acceptable source of energy worldwide. Their availability in future is uncertain due to their uninterrupted depletion and negative effects on the environment. Therefore, the alternate sources of energy are gaining importance in research to meet the world’s energy demand. In this light, researchers from industries and academics are attempting to produce ethanol, which is eco-friendly as it burns with zero emission of noxious gasses. Additionally, it has high octane number thus improves the combustion efficiency of conventional gasoline when they are blended together. The ethanol is obtained in two sequential steps, namely conversion of biomass to fermentable sugar and trans-esterification reaction of that sugar to bioethanol. Unfortunately, the concentration of ethanol in fermented broth is very low and its tendency of high solubility in water leads to form azeotrope.1 As a consequence, the complete dehydration of ethanol involves an energy intensive process with high capital cost. The common practice to produce bioethanol in large scale is extractive distillation (ED) and azeotropic distillation. These techniques use two conventional columns in a sequence; the first column separates ethanol using a solvent while the second column produces water and pure solvent. However each of these columns consists of an individual reboiler and condenser, involving reasonably high capital and energy cost. The dividing wall column (DWC) has emerged with introducing an innovative idea to replace the two columns by a single one with a vertical dividing wall. As the DWC consists of a single unit of reboiler, condenser and column, the concentration of bioethanol in fermentation broth can be improved to 99−99.8 wt% at lower capital and energy costs in comparison with the conventional method of dehydration.2 In the direction of achieving pure bioethanol in energy efficient DWC, Kiss and coworkers2 have implemented extractive-DWC (E-DWC) and azeotropic-DWC (A-DWC), and compared their performance with extractive conventional distillation sequence (E-CDS). The A-DWC consists of two bottom reboiler while E-DWC has only one reboiler, both of which are steam-driven. They have used sequential quadratic programming (SQP) method in the Aspen flowsheet simulator to obtain optimal design parameters of their E-DWC. Subsequently, Kiss and Ignat3 have developed another E-DWC with two reboilers; one is at the bottom and the other one at the side (side reboiler), both of which are driven by steam. Again, they have configured this process in the Aspen Plus and found the design parameters with minimizing the total heat duty to have almost complete recovery of the bioethanol product and solvent. The efficiency of this extractive dividing wall column3 needs to be


2

Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


improved. In this direction, Loy and co-workers4 have proposed complete vaporization of the solvent in the bottom reboiler and some fraction of that vapor is used to vaporize the water in the side reboiler to reduce steam consumption. Further, Luo and co-workers5 have replaced the steam of side reboiler with the compressed overhead vapor by coupling the column with a compressor and 40% energy reduction is achieved. The controllability of this scheme to handle large change in the disturbances is explored by Luyben6, and Patraşcu and coworkers7. Gordênia and co-workers8 have proposed a strategy for obtaining an optimized DWC configuration based on stage equilibrium for mixtures of ethanol/water and benzene/cyclohexane. They arrived at the conclusion that the reduction of energy costs by DWC depends on the arrangement of conventional distillation sequence (CDS). With this literature survey, we find that the E-DWC for ethanol dehydration is divided either at the bottom, top or intermediate location. In the first case, the DWC is operated with two bottom reboiler, whereas in the last two cases, the DWC has one bottom reboiler and one side reboiler, both of which require external heating source. To make it more energy efficient and cost effective, we propose a modified extractive DWC with dividing wall at the intermediated place with only a bottom reboiler and two reduce-sized condensers. This proposed E-DWC avoids the use of any side reboiler. Compared to the E-DWC with dividing wall either at the top or bottom, this proposed E-DWC with that at the intermediate location leads to a lower temperature difference between the two ends of the column. This, in turn, is beneficial for further intensification through the heat pumping scheme, which is proposed later. With these advantages, the proposed E-DWC is employed to discharge ethanol at the top, water as vapor from the tray just above the dividing wall and ethylene glycol (EG) (a solvent) from the base of the column. Subsequently, we propose further advancement in this E-DWC, in which, the sensible heat of EG is used to heat up ethanol (top) and water (intermediate) vapors. After this, their temperatures are further increased to attain a thermal driving force of 15 K with reference to the bottom reboiler liquid by the use of compressor(s). Additionally, the fresh feed (ethanol/water mixture) is also preheated with the sensible heat of that EG. This scheme is called as extractive heat integrated dividing wall column (E-HiDWC). In both of these E-DWC and E-HiDWC schemes, the heat transfer through the dividing wall is taken into account. Their design parameters are selected by minimizing the total annual cost (TAC) keeping the target purity more than 99.5 mol% for each component. Finally, their performances are compared with the extractive conventional distillation sequence (E-CDS), a reference process for ethanol dehydration, based on two performance indicators, namely energy and total annual cost (TAC) savings. As indicated, for selecting optimal design ACS Paragon Plus Environment

3


parameter, we have used fmincon inbuilt in MATLAB. It is a constrained nonlinear optimization or nonlinear programming approach. This fmincon uses the sequential quadratic programming method or in other word, fmincon is a commercial SQP package inbuilt in MATLAB. This article is organized as follows. Brief literature review and problem statement are reported in Section 1. In Section 2, the production of bioethanol and the process model are discussed in detail. The optimization method is elaborated in Section 3. The results and discussion are illustrated in Section 4. The last Section 5 underlines the outcomes of this study. 2. Production of Bioethanol As stated, ethanol is produced via trans-esterification reaction of fermentable sugar. This step yields dilute bioethanol with ethanol purity in the range of 5-12 wt% that needs to be further concentrated to more than 99.0-99.8 wt% (or above 99.5 mol%).2,9 During the course of increasing ethanol concentration, a binary azeotrope of ethanol/water (95.63 wt% or 89.52 mol% of ethanol) is formed. Figure 1 shows the vapor-liquid equilibrium envelop, generated using the Aspen Plus, which confirms the existence of ethanol/water azeotrope. To obtain the desired alcohol composition at low operational and capital cost, the extractive distillation (ED) is most reliable technology.9 The ED completes the separation in presence of mass separating agent (MSA). The MSA features high boiling point, miscibility and tendency to form no azeotrope with either ethanol or water of the fermented product.2 The common entrainer for ethanol/water system is ethylene glycol,10,11 which is also adopted in this work. 375 372 369

Temperature, K

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 27

366

Liquid composition Vapor composition

363 360 357 354 351 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Liquid/vapor mole fraction, ethanol

Figure 1. Vapor-liquid equilibrium plot of ethanol/water system at 1 atm.


4

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.1. Base case: extractive conventional distillation sequence Figure 2a describes the flowsheet of a typical extractive conventional distillation sequence. It consists of two units of column, reboiler and condenser. The first column (Column 1) is extractive column (EC), receives the fresh feed (ethanol/water) and the mass separating agent (ethylene glycol). With the help of Reboiler 1, the EC discharges vapor from the top that mostly constitutes ethanol and bottom liquid that includes water and EG as major components. This bottom liquid is further treated in a recovery column (i.e., Column 2), for which, water and EG are obtained from the top and bottom, respectively. The pure ethylene glycol is sent to the storage tank for its further use.


5


Page 6 of 27

Figure 2. Process flow diagram of ethanol dehydration: (a) extractive conventional distillation sequence, (b) extractive dividing wall column, and (c) extractive heat integrated dividing wall column. 2.2. Proposed configurations: extractive-DWC and extractive-HiDWC Figure 2b shows the extractive dividing wall column. Unlike E-CDS, it has only one column (with a dividing wall) and a reboiler. This metal wall divides the middle part of the shell into two compartments ( SI and SII ). Compartment SI , like extractive column of E-CDS, receives fresh feed, whereas SII resembles recovery column that discharges pure water. Like E-CDS, the MSA is introduced to a tray at the top of the column. The boilup vapor in the reboiler is distributed to sections SI and SII , while the liquid that flows down the column is completely sent to the trays of SI . The vapors that leave the top trays of the E-DWC and SII contain dehydrated bioethanol and pure water, respectively, and they are sent to their respective condensers (i.e., Condenser 1 and Condenser 2). As ethylene glycol has relatively lowvolatility, it comes out from the base of the column. This extractive dividing wall column is different from the one available in literature2-5 in that it has only a bottom reboiler (instead of two) and no side reboiler. Further, we propose the extractive heat integrated dividing wall column shown in Figure 2c. In contrary to E-DWC, the temperature of the ethanol vapor is increased in two steps. In the first step, the sensible heat of hot EG (solvent) is transferred to the cold ethanol. In the


6

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


next step, by employing Compressor 1, its temperature is again increased by 15 K than that of reboiler liquid. Now, this solvent is further employed as a heating medium in Heater 2 to increase the temperature of the water stream. This water stream is then subjected to compression in Compressor 2, which elevates its temperature to make it 15 K hotter than the reboiler liquid. These compressed vapors are then used to replace the significant (or complete) amount of steam in the reboiler. Once the ethanol and water release their latent heat in the reboiler, their original pressures are recovered via throttling valve (TV 1 and TV 2), and then they are sent to their respective reflux drum. The solvent that leaves Heater 2 is sent to the preheater, where it is coupled with the fresh feed to raise its temperature. Attaining the thermal driving force of 15 K, both the ethanol and water vapors transfer their latent heat to the reboiler liquid of the E-DWC. For this, the compression ratio (CR) needs to be maintained as:

T CR   out  Tin



  1  

(1)

Here, Tout and Tin denote the outlet and inlet temperature, respectively, with respect to the compressor. The polytropic coefficient (  ) of component j ,  j is the ratio of heat capacities at constant pressure and constant volume. It is a function of temperature and vapor phase composition ( y j ). Accordingly,  of the mixture is estimated from:

NC yj 1    1 j 1  j  1

(2)

For the distillation column operating in continuous mode, the CR is set at a fixed value. The total heat content of that compressed vapor ( QCV ) may be more or less than the actual heat ( QR ) required by the E-DWC. Thus, the following variable manipulation strategy needs to be adopted. Variable manipulation strategy Scenario 1: If the available heat (i.e., QCV ) is more than the heat demand (i.e., QR ), one needs to reject that extra heat in the overhead condenser. To meet the heat demand QR exactly, the fraction of overhead vapor fed to the compressor is obtained. ACS Paragon Plus Environment

7


Vn T C 

QR

Page 8 of 27

(3)

T

in

Here,  is the latent heat of the vapor. From eq 3, it can be inferred that the entire top vapor (

Vn T ) is not required to compress and the remaining amount of vapor, Vn Tr (= Vn T - Vn TC ) is sent to the condenser. Scenario 2: In contrary to Scenario 1, here, QR is more than the available QCV . So, it is mandatory to outsource that extra heat (i.e., QR  QCV ) in the form of steam. The amount of steam ( M s ) can be regulated as:

Ms 

QR

S T

S



VnT  T

S T

(4)

in

S

In eq 4, s is the latent heat of steam at saturation temperature ( Ts ). Now, in both the scenarios, as mentioned, the original pressure of compressed vapor after releasing its latent heat is recovered in the throttling valve. 2.3. Process model The dynamic model of the conventional and proposed configurations is constructed for a sample nth tray, shown in Figure 3, based on the conservation principle of mass and energy, and summation of vapor and liquid mole fractions. These are written as follows:

Figure 3. An equilibrium tray of extractive dividing wall column.


8

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Overall material balance dmn  Ln1  Vn1  Ln  Vn dt

(5)

Component material balance d mn xn , j  dt

 Ln1 xn1, j  Vn1 y n1, j  Ln xn , j  Vn y n , j

(6)

Energy balance



d mn H L,n dt

 L

n 1

H L , n 1  Vn 1H V , n 1  Ln H L , n  Vn H V, n  Qn , w

(7)

Summation of vapor and liquid composition NC

x j 1

n, j

NC

y j 1

n, j

1

(8)

1

(9)

Here, m is the liquid holdup, L the liquid and V the vapor flowrate, H L the liquid and H V the vapor enthalpy, and x j the liquid composition. Suffix j denotes the component index, and Qn ,w the heat transfer through the dividing wall. The above equations are developed with the following assumptions: uniform liquid composition on each tray, negligible vapor holdup, liquid and vapor leaving a tray is at equilibrium, fast energy dynamics, and negligible change in liquid holdup of the reboiler and reflux drum. The equilibrium between vapor and liquid is modelled by assuming ideal vapor, whereas nonideal liquid, for which the liquid activity coefficient is determined from the non-random two-liquid (NRTL) equation. The binary interaction parameters for ethanol, water and ethylene glycol system are obtained from the Aspen Plus simulator, and these are reported in Table 1. To model liquid flow from one tray to another one, we have used the linearized Francis weir formula.12 The vapor flow from each tray is obtained by setting the left hand side of eq 7 to zero. The liquid and vapor enthalpies are calculated from empirical relations.13


9


Page 10 of 27

Heat transfer through the dividing wall The heat transfer, Qn ,w from the liquid of trays in SI to that of trays in SII (or whichever is at higher temperature) is only applicable for DWC. This can be estimated as:

Qn ,w  UAT

(10)

where

A  hliq d c

(11)

Here, U , A , T and d c denote the overall heat transfer coefficient, area of dividing wall, temperature difference and diameter of the column, respectively. The height of liquid, hliq that takes part in heat transfer combines the liquid height between two consecutive trays and the weir height ( hweir ) of the upper tray of these coupled trays. This is represented as: 11  hliq   hTS  hweir  22 

(12)

It is obtained by assuming that the downcomer height equivalent to half of the tray spacing (

hTS ) taken part in heat transfer.14 Table 1. The NRTL model and binary parameters for ethanol, water and ethylene glycol system (Aspen Plus)

x  G  x G

 xm mj Gmj  x j Gij  m NRTL Model: ln  i   ij  j  xk Gkj  k xk Gkj k ki k k  b where : Gij  exp( ij ij );  ij  aij  ij  eij ln T  f ijT T  ij  cij  d ij (T  273.15); and T in K j

ji

ji

j

    

Component ( i )

Ethanol

Ethanol

Water

Component ( j )

aij

Water −0.8009

Ethylene glycol 14.8422

Ethylene glycol 0.3479

a ji

3.4578

−0.1115

−0.0567

bij

246.18

−4664.4058

34.8234

b ji

−586.0809

157.5937

−147.1373

cij

0.3

0.47

0.3


10

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


d ij , eij , f ij

0

0

0

Compressor duty The compressor(s) used in the proposed scheme is assumed to be adiabatic and reversible. The compressor and motor efficiency of 0.8 and 0.9, respectively,15 are chosen in this work. The power required to run the compressor (i.e., compressor duty), Qcomp in hp is computed from the following expression15:

Qcomp  3.03  10 5



 1 Vn T C Pin  CR    1  1  

(13)

Here, the inlet pressure ( Pin ) is in lb f /ft 2 , and the vapor flowrate ( Vn TC ) to the compressor is in ft 3 /min . The model developed for E-CDS, E-DWC and E-HiDWC consists of ordinary differential equations and algebraic correlations. These coupled model equations are coded in MATLAB simulation environment based on the simulation algorithm presented in Supporting Information (Section S1). The algebraic equations are simulated to estimate the tray temperature, and flowrates, enthalpy of vapor and liquid streams, among others. Using the Euler approach, the derivatives for composition and liquid holdup are calculated for each tray. 3. Process Optimization For the optimal design of extractive distillation column, one needs to identify the parameters, define the objective function, and do optimization. For the present case, the parameters for the extractive CDS include the heat duty, reflux ratio, total number of trays, and solvent and feed stages in each column of E-CDS; whereas the E-DWC has one additional parameter, namely vapor flowrate to the two sides of the dividing wall. Here, objective function is concerned with minimizing the total annual cost. While formulating this function, the product purities are considered as inequality constraint (operational constraint). The target purity (≥99.5 mol%) is attempted to achieve with minimum energy consumption. The TAC includes the utility (i.e., steam, coolant and electricity) and installed equipment (i.e., column, trays, reboiler, condenser and compressor) costs. With this, the optimization problem is formulated as:


11


Objective fucntion : TAC  z,u,d  subject to operational contraints : g  z , u , d   0

Page 12 of 27

(14)

The state variables ( z ) for all trays ( N T ) are the molar composition ( xn , j ), liquid holdup ( mn ) and temperature ( Tn ), as shown earlier through Eqs. (5-9). There are two types of independent variables, u and d . The first variable includes the parameters, which can affect the optimization, whereas the last one includes the disturbances that cannot be changed. As TAC and products purity have implicit relation with those variables, some penalty may be given to each of them. Then, one can couple the objective function and operational constraints into a single objective. In this study, we perform the static optimization to find the optimal values of u . In other words, the optimal operating point for the extractive- CDS, -DWC and -HiDWC is obtained in absence of disturbances, d (= 0). Note that, while solving the model equations, the integration of differential form of total material and component balances (i.e., eqs. (5) and (6)) is done using the Euler method. The optimization problem is solved using the MATLAB inbuilt routine fmincon. This routine attempts to find the constrained minimum u described in function, FUN, subject to the constraints of adjustable or optimizing parameters (i.e., u ), and over a set of lower and upper bounds ( lb and ub ) on u . The syntax to use fmincon is:

u  fmincon FUN, u 0, A, b, ulb , uub  subject to Au  b (15) It is important to note that the initial guess u 0 and several iterations have significant effect on the solution thrown by this optimization method. To select a suitable initial guess, u 0 , for the extractive- conventional distillation sequence and -dividing wall column, we first perform a sensitivity test (four cycles), where one of the variables is varied keeping the others fixed and then the optimization procedure is executed.


12

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Optimization steps followed in designing bioethanol dehydration processes. Figure 4 shows the steps followed to get the optimal design of the extractive- CDS and DWC. The initial guess obtained from the sensitivity test is fed to the optimizer, which recalls the process simulator with parameters ( u ). Next, the TAC is calculated and returned to the objective function. The optimizer continues with several iterations until the minimized TAC and target purity check are accomplished. 4. Results and Discussion In this section, we evaluate the performances of the proposed extractive- DWC and -HiDWC on the basis of total annual cost and energy savings with reference to the extractive conventional distillation sequence. As stated earlier, the objective function is TAC, which depends on the operating and the equipment (capital investment (CI)) installed costs. It is expressed as:

TAC  Operaing cost 

Capital investment payback time


(16)

13


Page 14 of 27

The operating cost (OC) consists of the costs of steam, cooling water and electricity as $17 per ton, $0.06 per ton and $0.084 per kW.h, respectively.16 Their total cost is estimated on a yearly working time of 8000 hr. The installed costs of several equipment, namely column, trays, reboiler, condenser and compressor, are calculated from the expression given in the Supporting Information (Section S2). Here, we select a payback time of 10 years for each scheme. For simulating the proposed processes, the feed (ethanol/water) and solvent (ethylene glycol) conditions, such as temperature, composition and flowrate are taken from literature.9 They are reported in Table 2. Table 2. Feed and solvent information for the base and proposed cases Flowrate

Composition

Temperature

(kmol/hr)

(mole frac)

(K)

Ethanol/water (feed)

100.0

0.85/0.15

313.15

EG (pure solvent)

88.94

1.0

353.15

4.1. Optimal design: extractive- CDS and -DWC The optimizing variables for the base case (extractive conventional distillation sequence) include the reboiler duty, reflux ratio, total number of tray, and feed tray for each of the columns. Table 3 encloses the lower and upper values of these variables, which are given to the optimizer. Here, the solvent is fed to the top section of the column (Tray 4, counting made from the condenser as the first tray to the reboiler as the last tray). This apart, the operating pressure in each column of E-CDS and E-DWC is 1 atm with 2.25 mmHg (or 0.0029 atm) tray pressure drop. We adopt the weir height of 5 inch (or 0.127 m) and tray spacing of 24 inch (or 0.6 m).


14

Page 15 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 3. Length of optimizing variables for the extractive-CDS and DWC Configuration

E-CDS

E-DWC

ulb

uub

ulb

uub

Reboiler 1

1000

5000

1000

5000

Reboiler 2

200

1000

N.A.

N.A.

Re

0.01

4

0.01

4

Rw

0.01

4

0.01

4

N.A.†

N.A.

0.01

0.88

Column 1

5

30

2/5/5‡

15/35/20‡

Column 2

5

30

N.A.

N.A.

Column 1

3

10

8

30

Column 2

3

10

N.A.

N.A.

Variable Reboiler duty (kW)

Reflux ratio

Vapor split Total no. of trays

Feed tray

For E-CDS, following the procedure outlined in Figure 4, we get the optimal reboiler duty (1685.6 kW for Column 1 and 650.8 kW for Column 2), molar reflux ratio ( R e = 0.34 and

R w = 2.17), total number of trays (26 for Column 1 and 13 for Column 2 ) and feed tray (18 for Column 1 and 6 for Column 2). These parameters are obtained with minimum TAC of $0.737 million per year with 99.63 mol% purity of ethanol at the top of first column of ECDS, 99.85 mol% of water as distillate in second tower and 99.99 mol% of ethylene glycol as bottom product from the same second column. Figure 2b shows that the extractive dividing wall column has two additional parameters, namely vapor split to the two sides (i.e., SI and SII ) and overall heat transfer coefficient ( U ) of the dividing wall. The range of U (refer eq 10) given to the optimizer is from 228 to 426 W/(m2 K). Unlike E-CDS, the total number of trays for each section (bottom, SI , SII and top sections) of the E-DWC needs to be optimized. The span of these variables, u are listed in Table 3. Hereby, at steady state, we get 99.81 mol% of ethanol and 99.9 mol% water with complete recovery of ethylene glycol. The steady state composition and flowrate of the three † ‡

N.A. denote ‘not applicable’ Each figure correspond to total number of trays in bottom, feed and top sections ACS Paragon Plus Environment

15


Page 16 of 27

outlet streams enriched with ethanol, water and EG are shown in Table 4. The optimizer finally converges the TAC to a minimum value of $0.657 million per year. With respect to this, the steady state optimal parameters for E-DWC are obtained as: reboiler duty of 2018.24 kW; reflux ratios R e and R w as 0.34 and 0.242, respectively; total number of trays in bottom, middle and top sections are 6, 24 and 9, respectively; vapor fraction to SI is 83.54%; feed tray is 23; overall heat transfer coefficient, U is 306.90 W/(m2 K). Table 4. Steady state composition and flowrate of three outlet streams of E-DWC. Stream

Composition

Flowrate

(mol%)

(kmol/hr)

ethanol/water/EG Ethanol stream

99.81/0.16/0.03

85.18

Water stream

0.1/99.9/0.0§

14.90

EG stream

0.0/0.0/100

88.86

4.2. Energetic and economic evaluation 4.2.1. Extractive– CDS and –DWC Table 5 evaluates the energetic potential of all concerned configurations with estimating the utility consumption in the reboiler, condenser and compressor. In E-CDS and E-DWC, reboiler is the only entity where heat is supplied by external steam, and condenser rejects that heat. For extractive conventional distillation sequence, Reboiler 1 and Reboiler 2 need 1685.6 kW and 650.8 kW, respectively; thus, total amount of heat requirement is 2336.4 kW. While for the same bioethanol dehydration, the E-DWC uses 2018.24 kW of reboiler duty with 78.86 kW of heat obtained via heat transfer through the dividing wall. Therefore, the proposed extractive dividing wall column secures a 13.62% energy savings. Before calculating the costs of the individual elements (i.e., reboiler, condenser, column and trays) of the E-CDS and E-DWC, one must have in hand the information of condenser duty, reboiler duty and column diameter. These are reported in Table 5 except the column diameter. This diameter is estimated from17:

§

It is actually 0.01 ppm. ACS Paragon Plus Environment

16

Page 17 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


 

d c  0.15 Vn T

0.5

(17)

In eq 17, column diameter ( d c ) and top vapor flowrate ( Vn T ) must be in m and kmol/hr, respectively. For E-CDS, the respective diameter of Column 1 and Column 2 are 1.60 m and 1.02 m, whereas, that of E-DWC is 1.60 m. It becomes obvious that the E-DWC significantly reduces the column size. Now, following the correlations given in the Supporting Information (Section S2), the installed costs of reboiler, condenser, column and its trays, and the costs of steam and cooling water are calculated in Table 6 for each of the schemes. It is observed that E-DWC reduces the capital investment and operating cost respectively by $0.049 million and $0.075 million per year. Overall, the reduction in total annual cost by the proposed extractive dividing wall column is 10.86% with reference to the extractive conventional distillation sequence (or conventional method of ethanol dehydration). Table 5. Evaluating energetic potential of extractive- CDS, -DWC and -HiDWC E-CDS

E-DWC

E-HiDWC

Condenser duty (kW) Condenser 1

1292.53

1287.15

N.A.

Condenser 2

532.93

211.05

N.A.

Reboiler duty (kW) Reboiler 1

1685.6

2018.24

469.34

Reboiler 2

650.8

N.A.

N.A.

Compressor duty (kW) Compressor 1

N.A.

N.A.

187.78

Compressor 2

N.A.

N.A.

12.33

Total energy consumptions (kW) 2336.4

2018.24

1069.67

Energy savings (%) N.A.

13.62


54.22

17


470

Main column (Bottom-SI-Top) SII section

450

Temperature, K

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 27

430 410 390 370 350 1

6

11

16

21

Tray no.

26

31

36

Figure 5. Tray temperature profile of extractive dividing wall column. 4.2.2. Extractive HiDWC Figure 5 shows the temperature profile of the extractive dividing wall column. It is observed that the temperature of the bottom reboiler liquid, overhead ethanol and intermediate water (top tray of section SII ) vapor are 473.99 K, 351.59 K and 373.88 K, respectively. With this, when the vapor of ethanol and water is only compressed in Compressor 1 and Compressor 2, respectively, the compression ratio obtained is around 18 and 3. Although the CR for water is reasonable but that for ethanol is insignificantly large. At this situation, at first we increase their temperature by the employment of Heater 1 and 2 (Figure 2c), respectively, with using hot EG, which is produced at the time of separation. It leads to decrease both the CRs. It is important to observe that the outlet temperature of the ethanol and water vapor must be 15 K more than 473 K (i.e., the reboiler liquid). Thus, the outlet temperature at the compressor must be 488 K. As reported18, the maximum allowable discharge temperature for a single stage (i.e., single impeller mounted on rotor) reciprocating compressor is 300 0F (or 422 K). On the other hand, a centrifugal compressor with special design (e.g., placing multiple impeller in series on the rotor, center supported diaphragms and high temperature sealants) deals with higher outlet vapor temperature, typically in the range of 400 to 450 0F (477 K to 505.37 K).19


18

Page 19 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 6. Cost of elements associated to extractive- CDS, -DWC and -HiDWC Items

E-CDS

E-DWC

E-HiDWC

Column 1

0.480

0.755

0.777

Column 2 Column 1

0.140

N.A.

N.A.

0.045

0.070

0.073

Column 2

0.009

N.A.

N.A.

Condenser 1

0.301

0.300

N.A.

Condenser 2

0.169

0.093

N.A.

Reboiler 1

0.296

0.333

0.375

Reboiler 2

0.160

N.A.

N.A.

Compressor 1

N.A.

N.A.

1.171

Compressor 2

N.A.

N.A.

0.126

Heater 1

N.A.

N.A.

0.084

Heater 2

N.A.

N.A.

0.025

Preheater Total CI (in million of USD ($))

N.A.

N.A.

0.088

1.6

1.551

2.719

Reboiler 1

0.378

0.461

0.107

Reboiler 2

0.148

N.A.

N.A.

Condenser 1

0.036

0.035

N.A.

Condenser 2

0.015

0.006

N.A.

Compressor 1

N.A.

N.A.

0.126

Compressor 2

N.A.

N.A.

0.008

Total OC (in million of USD ($)/year)

0.577

0.502

0.241

TAC (in million of USD ($)/year)

0.737

0.657

0.513

OC savings (%)

N.A.

13.00

58.23

TAC savings (%)

N.A.

10.86

30.39

Payback time (year)

N.A.

N.A.

3.33

Column (in million of USD ($)) Tray (in million of USD ($)) Condenser (in million of USD ($)) Reboiler (in million of USD ($)) Compressor cost (in million of USD ($)) Heater cost (in million of USD ($))

Steam (in million of USD ($)/year) Coolant (in million of USD ($)/year) Electricity (in million of USD ($)/year)

The respective heat capacities of hot fluid (EG) (at 473.99 K) and cold fluid (ethanol) (at 351.59 K) are 138.11 kJ/(kmol.K) and 73.96 kJ/(kmol.K). Using these information along with the molar flowrates (hot EG: 88.86 kmol/hr; cold ethanol vapor: 120.27 kmol/hr) in calorimetric equation, the thermal equilibrium temperature obtained in Heater 1 is 422.56 K.


19


Page 20 of 27

Thus the temperature of hot ethylene glycol drops from 473.99 K to 422.56 K while that of cold ethanol vapor rises from 351.59 K to 422.56 K. Now, with this elevated temperature ethanol is compressed in Compressor 1 to a temperature of 488.99 K to ensure a thermal driving force of 15 K with reference to the reboiler liquid. This is achieved at CR equals 4.42 which is reasonably low as compared with the direct compression (with CR = 18) from 351.59 K to 488.99 K. The total heat received from this ethanol vapor is 1262.7 kW. During the process of heating, the vapor shifts from saturated to superheated state, and after compression it is coupled with the reboiler. Accordingly, the vapor first loses the sensible heat followed by latent heat. Here, we neglect this sensible heat because of its negligible amount compared to latent heat. As shown in Figure 2c, the hot ethylene glycol releases the sensible heat in Heater 1 and thereafter it is further coupled with the cold water vapor in Heater 2. This time the heat capacity of EG at 422.56 K is 135.66 kJ/(kmol.K) and that of water vapor at 373.88 K is 34.27 kJ/(kmol.K). Again feeding these information with molar flowrates (hot EG: 88.86 kmol/hr; cold water vapor: 18.58 kmol/hr) in equation of calorimetry, the respective temperature of water is increased and that of EG is decreased to 420.12 K. Next, this water vapor temperature is raised in Compressor 2 with a small CR of 1.88 (earlier it was 3). The water vapor reduces the external supply of heat (i.e. steam) amounting 199.69 kW. The total amount of heat generated in extractive HiDWC from internal sources is 1462.39 kW (1262.7 kW from ethanol vapor and 199.69 kW from water vapor). Hence, the total amount of heat, 2018.24 kW, which is supplied from external source in E-DWC, is now reduced to 555.85 kW in E-HiDWC. To further reduce this external amount of heat, we propose to preheat the fresh feed by coupling it with the EG, coming out of Heater 2 (see Figure 2c). In doing so, the feed temperature is enhanced to 370.7 K from 353.15 K. With this, now the external energy requirement is further reduced to 469.34 kW from 555.85 kW. In comparison to E-DWC, the parameters of E-HiDWC are slightly changed because the dynamics of the column is changed when feed is introduced at higher temperature. The reflux ratio, R e is changed to 0.41 (for E-HiDWC) from 0.34 (for E-DWC), vapor split to SI is decreased to 83.38% (for E-HiDWC) from 83.54% (for E-DWC). Also, the diameter of the E-HiDWC is slightly increased to 1.64 m from 1.60 m in E-DWC. Comparatively, the purity of ethanol is improved from 99.81 mol% to 99.95 mol%, and that of water from 99.9 mol% to 99.96 mol%. The E-HiDWC requires 469.34 kW of heat from steam and 200.11 kW to run the compressor (Compressor 1: 187.78 kW; Compressor 2: 12.33 kW). A factor of 3 is multiplied


20

Page 21 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with that compressor duty to produce an equivalent amount of thermal energy. This factor is determined with respect to Japan,20 which takes into account the cost associated with fuel, electricity and production technology. Table 5 shows that total energy consumption by the extractive heat integrated dividing wall column is reduced to 1069.67 kW. Thus, in comparison to E-CDS, the amount of energy reduced by this E-HiDWC is 54.22%. On the other hand, if we make comparison between E-DWC (energy consumed is 2018.24 kW) and its hybrid scheme (energy consumed is 1069.67 kW), it is observed that the E-HiDWC reduces the steam consumption by 47.0%. Thus the extractive-HiDWC is efficient in terms of energy savings. It is important to note here that the total heat generated from the entire ethanol and water vapor is less than the actual heat demand. Thus, the E-HiDWC scheme comes under the Scenario 2. Table 6 shows the costs of different elements fused with the extractive heat integrated dividing wall column. The installed costs of tower and trays are slightly away from the costs obtained for E-DWC because of slight increase in the column diameter due to change in the feed temperature. Figure 2c shows that E-HiDWC must have larger reboiler (of area 85.19 m2) than that of E-DWC (of area 68.02 m2) to provide space for ethanol and water vapors. Due to this reason, the reboiler cost for the E-HiDWC case is increased by 12.61%. Because of involvement of compressors, the total capital investment is significantly higher compared to the extractive- CDS and -DWC. To evaluate the capital cost of Heaters, one requires information about the overall heat transfer coefficient. For Heater 1, where the heat exchange occurs between ethylene glycol (a heavy organic because its viscosity is more than 1.0 centipoise21) and ethanol (a light organic as viscosity is less than 0.5 centipoise), the heat transfer coefficient, U is adopted as 60 Btu/(hr.ft2.oF) (or 340.70 W/(m2.K))22. The same value is taken for feed preheater, as the ethanol content is much higher than the water. On the other hand, for Heater 2, U between ethylene glycol and water vapor is 75 Btu/(hr.ft2.oF) (or 425.87 W/(m2.K))22. Table 6 shows that the cost of these heaters is much less than to the costs of the compressors and reboiler. The extractive heat integrated dividing wall column is impressive enough both in terms of energy (= 54.22%) and total annual cost (30.39%) savings with respect to extractive conventional distillation sequence. In comparison to extractive dividing wall column, the EHiDWC improves the TAC savings by three folds (from 10.86% for E-DWC to 30.39% for E-HiDWC) with a payback time of 3.33 years.


21


Page 22 of 27

5. Conclusions In this work, we propose dividing wall column based extractive distillation configurations for bioethanol dehydration. At first, we develop an E-DWC that involves a single reboiler compared to the two reboilers used in the available DWC in literature. Optimizing the process parameters of this E-DWC, its performance is quantified in terms of energy savings and TAC. It is observed that this E-DWC secures a 13.62% and 10.86% savings in energy consumption and TAC with respect to is conventional analogous. To further improve its potential, vapor recompression in introduced in the E-DWC between overhead vapor and reboiler content as well as intermediate water vapor and same reboiler content. It is investigated that the compression ratio in the former case is 18 (CR1), whereas it is 3 (CR2) in the later case. Thus, we subsequently propose the thermal coupling between ethylene glycol with the overhead vapor, intermediate water vapor and fresh feed. This leads to reduce the CR1 from 18 to 4.42 and the CR2 from 3 to 1.88. To make a fair comparison between the proposed E-DWCs with reference to its conventional counterpart, their input and output specifications are attempted to keep close, if not same, by developing a variable manipulation strategy. With this, it is investigated that the proposed extractive- HiDWC provides a 54.22% energy and 30.39% TAC savings with reference to the conventional extractive distillation, and 47% energy and 21.92% TAC savings compared to the E-DWC. Acknowledgment The authors are thankful to Niraj Thakre and Sidharth Sankar Parhi. Supporting Information Available: It is divided into two sections. In the first Section S1, the extractive dividing wall column is presented with its simulation algorithm, start-up composition profile and controllability. Section S2 presents the performance indices. This material is available free of charge via the Internet at http://pubs.acs.org. Notation Abbreviations A-DWC

Azeotropic dividing wall column

CDS

Conventional distillation sequence

CI

Capital investment

CR

Compression ratio

DWC

Dividing wall column


22

Page 23 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EC

Extractive column

E-CDS

Extractive conventional distillation sequence

ED

Extractive distillation

E-DWC

Extractive dividing wall column

EG

Ethylene glycol

E-HiDWC

Extractive heat integrated dividing wall column

M&S

Marshal and Swift inflation index

MSA

Mass separating agent

OC

Operating cost

SQP

Sequential quadratic programming

TAC

Total annual cost

TV

Throttle valve

Symbol A

Area of heat transfer, m 2

d

Disturbance variable

dc

Column diameter, m

H

Enthalpy, kJ/kmol

hliq

Liquid height, m

hTS

Height of tray spacing, m

hweir

Weir height, m

j

Component index

L

Liquid flowrate, kmol/hr

MS

Molar flowrate of steam, kmol/hr

mn

Tray liquid holdup, kmol

NT

Total number of trays

nC

Number of component

P

Pressure, atm

Qcomp

Compressor duty, hp

QCV

Heat content of compressed vapor, kW

Qn,w

Heat transferred through dividing wall, kW

QR

Reboiler duty, kW


23


Re

Molar reflux ratio of ethanol

Rw

Molar reflux ratio of water

T

Temperature, K

U

Overall heat transfer coefficient, W/ m 2 .K

u

Design variable index

V

Vapor flowrate, kmol/hr

VnT

Flowrate of vapor at the top, kmol/hr

VnTC

Inlet vapor flowrate to the compressor, kmol/hr

VnT r

Vapor flowrate to the condenser, kmol/hr

x

Liquid composition, mole frac

y

Vapor composition, mole frac

z

State variable



Cost factor for dividing wall



Liquid activity coefficient

 

Latent heat, kJ/kmol

T

Temperature difference, K



Page 24 of 27



Polytropic coefficient

Subscript L

Liquid phase

n

Tray index

S

Steam

V

Vapor phase

Author Information Corresponding author *E-mail

address: [email protected]

ORCID Md. Aurangzeb: 0000-0002-9163-9234 A. K. Jana: 0000-0003-1367-5480 Notes The authors declare no competing financial interest.


24

Page 25 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Literature Cited (1) Haider, J.; Qyyum, M. A.; Hussain, A.; Yasin, M.; Lee, M. Techno-economic analysis of various process schemes for the production of fuel grade 2, 3-butanediol from fermentation broth. Biochem. Eng. J. 2018, 140, 93-107. (2) Kiss, A. A.; David, J.; Suszwalak, P. C. Enhanced bioethanol dehydration by extractive and azeotropic distillation in dividing-wall columns. Sep. Purif. Technol. 2012, 86, 7078. (3) Kiss, A. A.; Ignat, R. M. Innovative single step bioethanol dehydration in an extractive dividing-wall column. Sep. Purif. Technol. 2012, 98, 290-297. (4) Loy, Y. Y.; Lee, X. L.; Rangaiah, G. P. Bioethanol recovery and purification using extractive dividing-wall column and pressure swing adsorption: An economic comparison after heat integration and optimization. Sep. Purif. Technol. 2015, 149, 413427. (5) Luo, H.; Bildea, C. S.; Kiss, A. A. Novel heat-pump-assisted extractive distillation for bioethanol purification. Ind. Eng. Chem. Res. 2015, 54 (7), 2208-2213. (6) Luyben, W. L. Improved plantwide control structure for extractive divided-wall columns with vapor recompression. Chem. Eng. Res. Des. 2017, 123, 152-164. (7) Patraşcu, I.; Bildea, C. S.; Kiss, A. A. Dynamics and control of a heat pump assisted extractive dividing-wall column for bioethanol dehydration. Chem. Eng. Res. Des. 2017, 119, 66-74. (8) Cordeiro, G. M.; de Figueirêdo, M. F.; Ramos, W. B.; Sales, F. A.; Brito, K. D.; Brito, R. P. Systematic strategy for obtaining a dividing-wall column applied to an extractive distillation process. Ind. Eng. Chem. Res. 2017, 56 (14), 4083-4094. (9) Ramos, W. B.; Figueirêdo, M. F.; Brito, K. D.; Ciannella, S.; Vasconcelos, L. G.; Brito, R. P. Effect of solvent content and heat integration on the controllability of extractive distillation process for anhydrous ethanol production. Ind. Eng. Chem. Res. 2016, 55 (43), 11315-11328. (10) Kotai, B.; Lang, P.; Modla, G. Batch extractive distillation as a hybrid process: separation of minimum boiling azeotropes. Chem. Eng. Sci. 2007, 62 (23), 6816-6826. (11) Ravagnani, M. A. S. S.; Reis, M. H. M.; Maciel Filho, R.; Wolf-Maciel, M. R. Anhydrous ethanol production by extractive distillation: A solvent case study. Process Saf. Environ. Prot. 2010, 88 (1), 67-73. (12) Luyben, W. L. Process Modeling, Simulation and Control for Chemical Engineers, 2nd ed.; McGraw-Hill: Singapore, 1990.


25


Page 26 of 27

(13) Reid, R. C.; Prausnitz, J. M; Sherwood, T. K. The Properties of Gases and Liquids, 3rd ed.; McGraw-Hill: New York, 1977. (14) Ludwig, E. E. Applied Process Design for Chemical and Petrochemical Plants, 3rd ed. Vol. 2; Gulf Publishing Company: Houston (TX), 2002. (15) Douglas, J. M. Conceptual Design of Chemical Processes, 1st ed.; McGraw-Hill: New York, 1988. (16) Huang, K.; Shan, L.; Zhu, Q.; Qian, J. Adding rectifying/stripping section type heat integration to a pressure-swing distillation (PSD) process. Appl. Therm. Eng. 2008, 28, 923–932. (17) Turton, R.; Bailie, R. C.; Whiting, W. B.; Shaeiwitz, J. A. Analysis, synthesis and design of chemical processes. Prentice Hall PTR: New Jersey. 1998. (18) Gallick, P.; Phillippi, G.; Williams, B. F. What's Correct For My Application-A Centrifugal Or Reciprocating Compressor?. In Proc. 35th Turbomachinery Symposium, September 2006, Texas A&M University, Houston, TX, 113-122. (19) API Standard 618. Reciprocating compressors for Petroleum, chemical, and gas industry services, 4th ed.; American petroleum institute: Washington, D.C, 1995. (20) Iwakabe, K.; Nakaiwa, M.; Huang, K.; Nakanishi, T.; Røsjorde, A.; Ohmori, T.; et al. Energy saving in multicomponent separation using an internally heat-integrated distillation column (HIDiC). Appl. Therm. Eng. 2006, 26 (13), 1362-1368. (21) Jerome, F. S.; Tseng, J. T.; Fan, L. T. Viscosities of aqueous glycol solutions. J. Chem. Eng. Data 1968, 13 (4), 496-496. (22) Kern, D. Q. Process Heat Transfer, 1st ed.; McGraw-Hill: New York, 2001. Table of Contents (TOC)


26

Page 27 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


…..


27

A Novel Heat Integrated Extractive Dividing Wall Column for Ethanol

Recommend Documents