Insight into pressure-swing distillation from azeotropic phenomenon to dynamic control

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Contents lists available at ScienceDirect

Chemical Engineering Research and Design journal homepage: www.elsevier.com/locate/cherd

Review

Insight into pressure-swing distillation from azeotropic phenomenon to dynamic control Shisheng Liang, Yujuan Cao, Xingzhen Liu, Xin Li, Yongteng Zhao, Yongkun Wang, Yinglong Wang ∗ College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao 266042, China

a r t i c l e

i n f o

a b s t r a c t

Article history:

Pressure-swing distillation (PSD) is widely used as an efficient method for separating

Received 16 May 2016

pressure-sensitive azeotropic mixtures in industrial processes. Remarkably, PSD can achieve

Received in revised form 20 October

pure products without introducing a third component compared with extractive distillation

2016

and azeotropic distillation. Heat integration into PSD can save energy and reduce operating

Accepted 21 October 2016

costs, thus relieving the continuous growth of energy consumption in the distillation indus-

Available online 9 November 2016

try. This review paper describes the development of this widely used distillation technique, including all of the main aspects related to thermodynamic analysis, Quantitative structure

Keywords:

property relationship (QSPR), process design, process intensification, and dynamic control.

Pressure-swing distillation

Based on the foundation of research, further development of PSD is proposed for separat-

QSPR

ing multi-component azeotropic mixtures and exploring the process design and dynamic

Process design

control from QSPR, aiming at promoting the industrial application of this environmentally

Process intensification

friendly and well-known separation technique from multi-scale analysis.

Dynamic control

© 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Contents 1. 2. 3. 4.

5.

6.

7.

∗

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Thermodynamic analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 QSPR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Process design of PSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 4.1. Pressure selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 4.2. Determination of the distillation sequence for continuous PSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 4.3. Schemes of PSBD for different azeotropes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 Process intensification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 5.1. Process optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 5.2. Heat integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328 Control scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 6.1. Temperature control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 6.2. Composition-temperature cascade control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 6.3. Pressure-compensated temperature control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332

Corresponding author. E-mail address: [email protected] (Y. Wang). http://dx.doi.org/10.1016/j.cherd.2016.10.040 0263-8762/© 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Nomenclature BIP DCBR DCBS fi oL HIDiC HP HPC Ki LP LPC PID PSD PSBD QSAR QSPR QR /F RR R/F TAC THF VLE VRC ˛ij  iL i v

Binary interaction parameter Double column batch rectifier Double column batch stripper Liquid fugacity in standard state Heat integrated distillation column High pressure High pressure column Phase equilibrium constant Low pressure Low pressure column Proportional-integral-derivative Pressure-swing distillation Pressure-swing batch distillation Quantitative structure activity relationship Quantitative structure property relationship Reboiler heat duty/feed flow rate Reflux ratio Reflux rate/feed flow rate Total annual cost Tetrahydrofuran Vapor liquid equilibrium Vapor recompression column Relative volatility Activity coefficient Vapor fugacity coefficient

319

2015; Horsley, 1947; Knapp, 1991; Wasylkiewicz et al., 2003). Lewis (1928), to our knowledge, is the first to apply this property to distillation of azeotropes. PSD can be divided into three types according to the operating mode: continuous PSD, pressure swing batch distillation (PSBD), and semi-continuous PSD. Many scholars and researchers have focused on the azeotrope separation with PSD in these three operating modes. Table 1 lists the information on PSD separation of azeotropes in published articles. Continuous PSD is widely used to separate binary azeotropes, and the flowsheet includes two columns operating at different pressures or a single shell column that is divided into two sections, the high pressure (HP) section and the low pressure (LP) section (Mulia-Soto and Flores-Tlacuahuac, 2011). High purity products can be obtained at the bottom of distillation columns (minimum boiling azeotropes) or the top of distillation columns (maximum boiling azeotropes). Repke et al. (2007) explored the application of PSD for separating a binary azeotrope experimentally for the first time in a batch rectifier and in a stripper. In PSBD with one column (Klein and Repke, 2009; Repke et al., 2007), the feed is charged into a bottom tank (regular batch) or into a top tank (inverted batch), and the initial feed composition determines whether the first step is LP or HP. High purity products are obtained at the bottom of the PSBD column. The other PSBD mode is the double-column system (Modla, 2010; Modla and Lang, 2008a, 2010; Modla et al., 2010), which is the combination of two rectifying sections or two stripping sections. The double-column system contains a double column batch rectifier (DCBR), which is more suitable for maximum boiling azeotropes from the standpoint of energy consumption with same product quality, and the double column batch stripper (DCBS), which is more economic for minimum boiling azeotropes (Modla and Lang, 2008a). Phimister and Seider (2000) investigated the separation of a minimum boiling azeotrope by semi-continuous PSD. The semicontinuous PSD includes one column, and the distillation column is adjusted from LP to HP in a cyclic campaign.

1.

Introduction

Among all of the separation technologies, distillation is currently the most widely used for separating mixtures in industrial process, and distillation consumes approximately 95% of the total energy used in the chemical separation industry (Mahdi et al., 2015). Despite the high energy consumption, distillation is still a preferred process because of its processing advantages and the abundant theoretical and practical knowledge. Ordinary distillation cannot achieve effective separation when the mixture exhibits the azeotropic phenomenon. If the chemical components are dissimilar and repulsion forces are strong, activity coefficients are greater than unity, thus leading to the formation of minimum boiling azeotropes such as the azeotrope

Another classification method for PSD depends on the number of components in the mixture to be separated. Following this criterion, PSD can be divided into two types, binary mixture PSD (Fulgueras et al., 2016, 2015; Luo et al., 2014) and multi-component PSD (Modla and Lang, 2008b; Modla et al., 2010; Zhu et al., 2016). Some binary close-boiling hydrocarbon mixtures can also be separated by continuous PSD to save energy (Zhang et al., 2014). Many scholars investigated the separation of binary azeotropes with PSD, but the separation of multi-component mixtures (at least one binary azeotrope exists) with PSD is insufficiently studied, especially regarding the continuous PSD. Knapp and Doherty (1992) noted that the continuous PSD will be unlikely to be advantageous compared with extractive distillation for more columns and higher recycle ratios may be needed when more than one distillation boundary is present. Modla (2011b) studied the feasibility and simu-

of ethanol/water, tetrahydrofuran (THF)/water, acetone/methanol. If the chemical components attract each other, activity coefficients are less than unity, and maximum boiling azeotropes,

lation of PSBD separation of ternary homoazeotropic mixtures with

such as water/ethylenediamine, methanol/trimethoxysilane can form

itations and challenges remain for PSD. The limitations of PSD are considered mainly from three aspects. First, PSD is focused mainly on

(Luyben, 2012, 2013a). Azeotropes are also classified in terms of the number of phases (homogeneous and heterogeneous azeotropes) and in terms of the number of components (binary and multi-component azeotrope). Special distillation methods, including azeotropic distillation (Honghai et al., 2015; Li et al., 2015b; Yu et al., 2015), extractive distillation (An et al., 2015; Luyben, 2015; You et al., 2015), and pressure ˜ swing distillation (PSD) (Luyben, 2013a; Munoz et al., 2006; Mulia-Soto and Flores-Tlacuahuac, 2011; Qasim et al., 2015), are common ways for separating azeotropes. PSD has been widely studied and applied in the industry for its superiority qualities of introducing no entrainer, protecting the environment, and saving energy by heat integration. The PSD separation is based on the fact that a mixture of components displays sensitivity to pressure, which means that a simple change in pressure can alter the relative volatilities of the components of the mixture with close boiling points or form an azeotrope. The pressure-sensitivity of azeotropes has been known since the 1860s (Roscoe, 1860, 1862; Roscoe and Dittmar, 1860). Since then, the effect of pressure on the azeotropic mixtures has been explored in many papers from theoretical and experimental studies (Abildskov and O’Connell,

different column configurations in published studies. Despite the above-mentioned advantages to the process, some lim-

pressure-sensitive azeotropes, and it is impossible or energy intensive to separate azeotropes with low sensitivity by the change of pressure (Luyben, 2012). Second, some heat-sensitive azeotropic components may decompose with the increasing pressure, and this heat sensitivity prevents the application of PSD. Third, if a vacuum pressure is used in PSD, some cooling media that are much more expensive than cooling water may be needed, thus causing more operating costs for PSD. In our opinion, the quantitative structure-property relationship (QSPR) resolves the characterization of azeotropic characteristics by using molecular descriptors of azeotropic components well (Katritzky et al., 2011; Solov’ev et al., 2011). However, no papers have focused on the characterization of the vapor liquid equilibrium (VLE), steady state design, and dynamic control of PSD with QSPR. The challenges of PSD are quantitative analysis of the VLE, steady state design, and dynamic control by using QSPR, and these challenges will be described in the following content. Due to the wide use of PSD, there have been several review papers in some journals and chapters in books dealing with the separation

320

Table 1 – Summary of the studies of azeotropes with PSD separation. System

Acetic acid + DMAC Acetone + chloroform Acetone + chloroform + toluene Acetone + methanol

Chloroform + methanol

Cyclohexanone + phenol Diisopropyl ether + isopropyl alcohol Di-n-propyl ether + n-propyl alcohol Ethanol + toluene

Ethanol + water

√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √

Ethyl acetate + ethanol

√ √

Isobutyl alcohol + isobutyl acetate

Isopentane + methanol

Methanol + dimethyl carbonate

√ √ √ √

Model

LP (atm)

HP (atm)

1.0 0.77 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.6

5.0 10.0 10.0 10.0 5.0 10.0 10.0 10.0 10.0 11.1

1.0 1.0 1.0 1.0 1.0 1.0 1.0

3.1 10.0 6.0 3.7 3.5 2.8 2.8

1.0 0.7

10.0 10.9

0.03 1.0 0.3 0.5 0.1 0.1

2.4 10.0 1.0 12.0 1.1 1.1

NRTL Wilson-RK

1.0 1.0

10.0 10.0

Wilson

1.0

10.0

UNIF-HOC UNIQUAC UNIQUAC UNIQUAC UNIQUAC UNIQUAC NRTL Van Laar UNIQUAC UNIQUAC Wilson NRTL NRTL Wilson Wilson Wilson Wilson

NRTL NRTL UNIQUAC NRTL UNIQUAC NRTL UNIQUAC UNIQUAC

NRTL NRTL UNIQUAC UNIQUAC UNIQUAC UNIFAC UNIFAC Wilson

Pressure selection √

√ √

√

√ √ √ √ √ √

√

√ √ √

0.1 1.0 1.0

5.0 3.5 10.0

0.2 0.2

1.0 1.0

2.0 2.0

10.0 10.0

1.0

12.0

Distillation sequence

Optimization

Dynamic control

Luyben (2012) Luyben (2013a) Modla (2011b) Modla and Lang (2012) Luyben (2012) Luyben (2008a) Fulgueras et al. (2016) Knapp and Doherty (1992) Modla and Lang (2010) Wang et al. (2016a)

Continuous Continuous Continuous Batch Batch Continuous Continuous and batch

Experiment Theory Theory Experiment Theory Experiment Experiment

Repke et al. (2005) Huang et al. (2008) Kim et al. (2013) Repke et al. (2007) Klein and Repke (2009) Repke et al. (2004) Repke and Klein (2005)

Continuous Continuous

Theory Theory

Hosgor et al. (2014) Wang et al. (2016a)

√

Continuous Continuous Continuous Continuous Batch Batch

Theory Theory Theory Theory Theory Theory

Li et al. (2013) Luo et al. (2014) Lladosa et al. (2011) Zhu et al. (2015) Modla and Lang (2008a) Modla and Lang (2007)

√

Continuous Continuous

Theory Theory

Continuous

Theory

Kiran and Jana (2015b) Mulia-Soto and Flores-Tlacuahuac (2011) Knapp and Doherty (1992)

Continuous Continuous Batch

Theory Theory Theory

Fissore et al. (2006) Kiran and Jana (2015a) Modla (2011a)

Continuous Continuous

Theory Theory

˜ Munoz et al. (2006) Luo et al. (2016)

Continuous Continuous

Theory Theory

Luyben (2005) Al-Arfaj and Luyben (2004)

Continuous

Theory

Wei et al. (2013)

√ √

√ √ √

√ √ √ √ √

√

√ √ √

√ √

√

√ √ √ √

√ √

√

√

√

√

√

√

√ √ √ √ √

Reference

Theory Theory Theory Theory Theory Theory Theory Theory Theory Theory

√ √ √ √ √

Theory or experiment

Continuous Continuous Batch Continuous Continuous Continuous Continuous Continuous Batch Continuous

√ √ √

√

Process type

√

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Acetonitrile + water

VLE

– Table 1 (Continued) System

VLE √

Methanol + THF

Methyl acetate + methanol

Methylal + methanol Methanol + benzene + acetonitrile MIBK + butanol N-Pentane + acetone N-Pentane + acetone + cyclohexane N-Heptane + isobutanol THF + ethanol

THF + water

Toluene + 1-butanol Water + ethylenediamine

√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √

Pressure selection

NRTL

LP (atm)

HP (atm)

1.0

10.0

Distillation sequence

Optimization

√

NRTL NRTL

√

1.0 1.0

10.0 10.0

√

√

UNIFAC UNIFAC

√

0.25 0.25

7.0 7.0

√

√

1.0 1.0 1.0 1.0

8.7 3.0 11.0 10.0

1.0 1.0 0.3

11.8 6.0 1

1.0 1.0 1.0

10.0 10.0 10.0

1.0 1.0 1.0

4.0/12.0 10.0 10.0

1.0 1.0 1.1 1.1 1.0 1.0

6.8 6.8 7.9 7.0 7.8 10.0

1.0 0.1 0.1 1.0 0.1 0.1

10.0 2.0 6.6 8.0 8.0 8.0

NRTL UNIQUAC NRTL NRTL NRTL Wilson

√ √ √ √ √ √

UNIQUAC UNIQUAC UNIQUAC NRTL UNIQUAC UNIQUAC Wilson Wilson Wilson Wilson Wilson Wilson

NRTL UNIQUAC NRTL UNIQUAC UNIQUAC UNIQUAC

√ √ √

√ √

√

Dynamic control

√

√

√ √ √

√ √ √

√ √ √

√

√

√

√ √

√ √

√ √

√

√ √

√ √

√ √

√ √ √ √

√

Process type

Theory or experiment

Reference

Continuous

Theory

Wang et al. (2010)

Continuous Continuous

Theory Theory

Wang et al. (2015b) Wang et al. (2014)

Continuous Continuous

Theory Theory

Luyben (2014a) Luyben (2014b)

Continuous Continuous Continuous Continuous

Theory Theory Theory Theory

Bonet et al. (2007) Suo et al. (2016) Zhang et al. (2016) Cao et al. (2016)

Continuous Continuous Continuous

Theory Theory Theory

Yu et al. (2012) Zhu et al. (2016) Li et al. (2016b)

Batch Batch Batch

Theory Theory Theory

Modla (2010) Kopasz et al. (2009) Modla and Lang (2008b)

Continuous Continuous Continuous

Theory Theory Theory

Wang et al. (2016b) Wang et al. (2015a) Wang et al. (2015c)

Continuous Continuous Continuous Continuous Continuous Semicontinuous

Theory Theory Theory Theory Theory Theory

Abu-Eishah and Luyben (1985) Frank (1997) Luyben (2008b) Lee et al. (2011) Hamad and Dunn (2002) Phimister and Seider (2000)

Continuous Continuous Continuous Batch Batch Batch

Theory Theory Theory Theory Theory Theory

Qasim et al. (2015) Li et al. (2015a) Fulgueras et al. (2015) Modla (2010) Modla and Lang (2008a) Modla and Lang (2007)

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Methanol + trimethoxysilane

√

Model

321

322

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Fig. 1 – The classification and study content of PSD. of azeotropic mixtures with PSD that describe the basic theories and applications of PSD. For example, in the published books, Luyben and Chien (2011) concentrated on the design and dynamic control of continuous PSD for separating binary azeotropes with and without heat integration. Lei et al. (2005) addressed some aspects of PSD in the operating principles, the operating mode, and the conceptual design. In his doctoral thesis, Klein (2008) discussed the three operating modes of PSD from model establishment, experimental validation, and process control concepts. Mahdi et al. (2015) reviewed the basic operating principles of three operating modes of continuous PSD, PSBD, and semicontinuous PSD. Although some papers about PSD have been published, a comprehensive description covering the latest technologies of PSD in recent years is not available. The motivation of this article is to provide the present research and future development of PSD from thermodynamic analysis, QSPR, process design, process intensification, and dynamic control. Fig. 1 shows the classification and domain of PSD. Classical PSD can be divided into three types, according to the operating mode, and these types have been described in detail in the above content. The study of PSD is mainly focused on multi-scale analysis, including the exploration of azeotropic characteristics, VLE characterized by a T–x,y or x–y diagram, steady state design, dynamic control, and the analysis of these processes with QSPR. Specifically, the steady state contains mainly pressure selection, distillation sequence, and process intensification. These aspects will be investigated in detail in the following content. The figure is an organizational chart of the paper that gives a general summary of each study module. The connection between classification and study content shown in the figure can yield a better understanding for readers about the main idea of this paper.

2.

Thermodynamic analysis

The azeotropic phenomenon is defined as a state in which mass transfer occurs between phases while the composition of each phase remains constant. Phase equilibrium constant Ki (component i) is a critical property in the design and optimization of the distillation process and it can be defined as Eq. (1),

Ki =

fi0L  Li P˚vi

(1)

where p is the pressure, ˚i v is the vapor fugacity coefficient, fi OL is the liquid fugacity in the standard state, and  i L is the activity coefficient. The activity coefficient  i L is a characterization of the degree of nonideality. When the pressure is low, ˚i v equals 1, and fi oL is equal to the vapor pressure at the system temperature. At the azeotropic point, Ki and relative volatility ˛ij (the ratio of the phase equilibrium constant K for component i and j) show a value of 1, and the system cannot be separated by ordinary distillation (Gmehling et al., 2012). Azeotropes can be classified by their deviations from Raoult’s law (Walas, 1985).

Most azeotropes discovered are minimum boiling azeotrope, for which the deviations from Raoult’s law are positive ( i L > 1), suggesting a lowest point of the T–x curve for which the vapor composition equals the liquid composition. For maximum boiling azeotropes, the deviations from Raoult’s law are negative ( i L < 1), suggesting a peak of the T–x curve, for which the vapor composition equals the liquid composition. If the deviations from Raoult’s law are large enough ( i L  1), a heterogeneous azeotrope occurs with phase separation, for which the vapor phase remains in equilibrium with two liquid phases. VLE data are the foundation of the T–x,y diagrams and the x–y diagrams, which are critical for the analysis of the pressure sensitivity of the azeotrope. Azeotropic data at different pressures can be obtained from experimental measurements (Lladosa et al., 2006; Martínez et al., 2008, 2009), azeotropic data books (Gmehling, 1994), and thermodynamic model prediction (Gmehling et al., 1982; Weidlich and Gmehling, 1987). For the process simulation and the modeling study of PSD with software of Aspen Plus, PRO/II, etc., the first step is to obtain accurate VLE data. However, the VLE plot drawn using thermodynamic models such as UNIQUAC (Abrams and Prausnitz, 1975), Wilson (1964), and NRTL (Renon and Prausnitz, 1968) with the built-in binary interaction parameters (BIPs) in simulation software may show some deviations from the experimental data. Thus, thermodynamic data regression should be performed to obtain the optimal BIPs and minimize the deviations between experimental data and calculated data. Many papers have explored the VLE of binary azeotropes with PSD separation, and the related literature including thermodynamic models is listed in Table 1. According to the types of T–x,y diagrams, binary azeotropes are divided into six types (Frank, 1997; Gmehling et al., 1995), which are shown in Fig. 2. Examples of the different binary azeotropic systems with PSD separation in published studies are listed in Table 2. Azeotropic characteristics of the azeotropes often depend on temperature (pressure). Therefore, understanding the nature and temperature (pressure) dependence of an azeotrope can promote the development of the distillation process. As shown in Fig. 2 and Table 2, homogeneous minimum boiling azeotropes as depicted in Fig. 2(a), and homogeneous maximum boiling azeotropes as shown in Fig. 2(c) have been widely studied with PSD separation because of their common appearance in industry. Large deviations from Raoult’s law often lead to the emergence of heterogeneous azeotropes as shown in Fig. 2(f). However, in some cases, both homogeneous azeotropes and miscibility gaps can be observed, as shown in Fig. 2(b) and (d). Two azeotropic points may occur in a binary system at the given pressure with strong

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

323

Fig. 2 – Txy diagrams of different types of binary azeotropes: (a) homogeneous minimum boiling azeotrope; (b) homogeneous minimum boiling azeotrope in a system with miscibility gap; (c) homogeneous maximum boiling azeotrope; (d) homogeneous maximum boiling azeotrope in a system with miscibility gap; (e) double azeotrope; (f) heterogeneous azeotrope. real behavior in the vapor phase as shown in Fig. 2(e). A binary double azeotrope is an unusual VLE phenomenon where two azeotropes exist at a given temperature or pressure. Double azeotropes have been found for some organic binary mixtures, such as benzene/hexafluorobenzene (Aucejo et al., 1996), diethylamine/methanol (Srivastava and Smith, 1985), acetic acid/isobutyl acetate (Christensen and Olson, 1992), and n-heptane/isobutanol (Wang et al., 2016b). Taking the n-heptane/isobutanol mixture as an example, the azeotrope exhibits the double azeotropic phenomenon at approximately 7 atm. VLE analysis also indicates that a minimum boiling azeotrope is formed below the pressure of 6 atm while a maximum boiling azeotrope is formed above the pressure of 8 atm. Therefore, the choice of separating nheptane/isobutanol with PSD involves two methods. One of the two PSD processes refers to the conventional PSD, in which minimum boiling azeotropes are formed under both low and high operating pressures. The other process is the unusual PSD, in which minimum and maximum boiling azeotropes are formed under low and high pressures, respectively. For the

double azeotrope of benzene/hexafluorobenzene, the compositions of the two azeotropes approach each other with increasing pressure as reported by Aucejo et al. (1996), which indicates that PSD may be possible. However, there is no PSD study about this type of azeotrope. For the components in a binary mixture, large deviations from ideal behavior often lead to partial miscibility, and heterogeneous azeotropes are formed in most cases. PSD can also be used for separating heterogeneous azeotropes that are pressure-sensitive, but few papers have focused on the separation of heterogeneous azeotropes with PSD. Further PSD study can be expanded to the separation of pressure-sensitive azeotropes, including azeotropes with miscibility gaps, double azeotropes, and heterogeneous azeotropes.

3.

QSPR

The sensitivity analysis of an azeotrope with pressure discloses the dependence of its composition and temperature on pressure. In some mixtures such as ethanol/methyl

324

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Table 2 – Examples of the different binary azeotropic systems with PSD separation in published works. Type of azeotrope (a)

System

Separation type

Acetone + methanol

(a) + (a)

Acetonitrile + water

(a) + (a)

Chloroform + methanol Diisopropyl ether + isopropyl alcohol Di-n-propyl ether + n-propyl alcohol Ethanol + toluene

(a) + (a) (a) + (a) (a) + (a) (a) + (a)

Ethanol + water

(a) + (a)

Ethyl acetate + ethanol Isobutyl alcohol + isobutyl acetate Methanol + THF Methanol + dimethyl carbonate Methyl acetate + methanol

(a) + (a) (a) + (a) (a) + (a) (a) + (a) (a) + (a)

Methylal + methanol MIBK + butanol N-Pentane + acetone THF + ethanol THF + water

(a) + (a) (a) + (a) (a) + (a) (a) + (a) (a) + (a)

Toluene + 1-butanol

(a) + (a)

(b)

/

/

(c)

Acetic acid + DMAC Acetone + chloroform Cyclohexanone + phenol Methanol + trimethoxysilane Water + ethylenediamine

(c) + (c) (c) + (c) (c) + (c) (c) + (c) (c) + (c)

(d)

/

/ a

Reference Modla and Lang (2012), Luyben (2012), Luyben (2008a), Wang et al. (2016a), Fulgueras et al. (2016), Knapp and Doherty (1992), Modla and Lang (2010) Repke et al. (2005), Huang et al. (2008), Kim et al. (2013), Repke et al. (2007), Klein and Repke (2009), Repke et al. (2004), Repke and Kelin (2005) Hosgor et al. (2014), Wang et al. (2016a) Luo et al. (2014) Lladosa et al. (2011) Zhu et al. (2015), Modla and Lang (2007), Modla and Lang (2008a) Kiran and Jana (2015b), Mulia-Soto and Flores-Tlacuahuac (2011), Knapp and Doherty (1992) Fissore et al. (2006), Kiran and Jana (2015a), Modla (2011a) ˜ Munoz et al. (2006), Luo et al. (2016) Wang et al. (2015b), Wang et al. (2014) Wei et al. (2013), Wang et al. (2010) Bonet et al. (2007), Suo et al. (2016), Zhang et al. (2016), Cao et al. (2016) Yu et al. (2012) Li et al. (2016b) Modla (2010), Kopasz et al. (2009) Wang et al. (2015a), Wang et al. (2015c) Abu-Eishah and Luyben (1985), Frank (1997), Luyben (2008b), Lee et al. (2011), Hamad and Dunn (2002), Phimister and Seider (2000) Qasim et al. (2015)

Luyben (2012) Luyben (2013a) Li et al. (2013) Luyben (2014a), Luyben (2014b) Li et al. (2015a), Fulgueras et al. (2015), Modla (2010), Modla and Lang (2007), Modla and Lang (2008a)

(e)

N-Heptane + isobutanol

(a) + (a)/(c)

Wang et al. (2016b)

(f)

Isopentane + methanol

(f) + (f)

Luyben (2005), Al-Arfaj and Luyben (2004)

a

At P = 7 atm.

ethyl ketone (Britton et al., 1943) and water/ethylenediamine (Fulgueras et al., 2015), the azeotrope disappears as the pressure changes. The characteristics of azeotropic composition changing (or azeotrope disappearing) with the variation of pressure leads us to explore the inner rule of the PSD process from QSPR. QSPR or quantitative structure-activity relationship (QSAR), which was initially developed to meet the needs of drug activity design, has been combined with many subjects. The foundation of QSPR is that material properties are determined by material structure. The main idea of QSPR is the establishment of a QSPR model using a chemical theoretical calculation method and a variety of mathematical statistical methods. Once a reliable QSPR model is established, this model can be used to predict the properties of new compounds or unknown properties of existing compounds after being tested on the model predictability. The study content of QSPR contains mainly data collection, calculation and selection of molecular descriptors, model establishment, and validation of the model. The critical point of a QSPR study is selecting the most appropriate molecular descriptors to build a robust model. Some optimization algorithms such as the genetic algorithm and the particle swarm algorithm have been used in the choice of molecular descriptors. Modeling methods

have been divided into two parts, a linear modeling method such as multiple linear regression and a nonlinear modeling method such as an artificial neural network and a support vector machine. Internal and external cross validation are widely used in validating the reliability of a QSPR model. Many papers have predicted the properties of individual compounds with a QSPR model (Gharagheizi et al., 2012; Hemmateenejad et al., 2011; Pourbasheer et al., 2015), and some work has also been performed on QSPR models for mixtures (Gebreyohannes et al., 2013; Mokshyna et al., 2016; Rybinska et al., 2016). Recently, QSPR models have been used to predict azeotropic characteristics of azeotropes (Gaudin et al., 2015; Oprisiu et al., 2013; Oprisiu et al., 2012). Table 3 lists the QSPR model prediction conditions of the boiling points and azeotropic composition of azeotropes that can be separated by PSD in some papers (Katritzky et al., 2011; Solov’ev et al., 2011; Zare-Shahabadi et al., 2013). However, there is no paper focused on the relationships between molecular descriptors and VLE, and steady state design of PSD. Fig. 3 shows the current and future study domain of PSD. QSPR has been used in the characterization of azeotropic characteristics of azeotropes that can be obtained by experiment. The azeotropic phenomenon has important influence

325

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Table 3 – QSPR model prediction conditions of the boiling points and azeotropic composition of azeotropes which can be separated by PSD. System

Tb (K) azeotrope

Tb (K) QSPR prediction

X1W (wt%) experiment

X1W QSPR (wt%) prediction

Acetone + chloroform

337.85 338.15 337.85

330.91/327.42 330.80/324.37 339.85/334.31

21

28

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

328.85 329.15 328.85

314.84/311.62 325.70/324.58 320.61/320.83

88

82

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

Acetonitrile + water

349.65 349.65

309.49/312.02 331.74/350.93

Chloroform + methanol

326.65 326.15 326.65

337.09/332.47 329.90/329.57 327.83/334.31

87

89

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

349.85 350.15 349.85

350.54/360.26 345.90/348.18 351.84/343.60

68

56

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

351.35 351.15 351.35

343.53/336.68 339.8/347.36 325.49/337.74

96

69

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

344.95 345.15 344.95

374.71/366.33 345.50/344.59 342.98/348.79

74

82

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

327.15 327.15 327.15

380.33/359.25 329.40/325.31 329.19/341.06

82

89

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

Methylal + methanol

314.95 314.95

331.56/344.04 305.16/321.69

Pentane + acetone

305.15 306.15 305.15

308.03/305.92 310.70/305.73 310.28/300.60

80

73

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

378.75 379.15 378.75

381.81/375.00 374.90/377.94 364.15/375.59

72

83

Katritzky et al. (2011) Solov’ev et al. (2011) Zare-Shahabadi et al. (2013)

392.15 392.15

364.15/348.75 397.65/409.90

Acetone + methanol

Ethanol + toluene

Ethanol + water

Ethyl acetate + ethanol

Methyl acetate + methanol

Toluene + 1-butanol

Water + ethylenediamine

Reference

Katritzky et al. (2011) Zare-Shahabadi et al. (2013)

Katritzky et al. (2011) Zare-Shahabadi et al. (2013)

Fig. 3 – The current and future study domain of PSD.

Katritzky et al. (2011) Zare-Shahabadi et al. (2013)

326

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

on the separation of azeotrope with PSD. PSD process contains steady state design and dynamic control. So far, no researches have focused on the links between molecular descriptors of azeotropes and PSD optimal operational parameters or PSD dynamic operational conditions. This can be challenge in the future study, which will give guidance on the more effective design of the PSD process.

4.

Process design of PSD

4.1.

Pressure selection

To use PSD, the azeotropic composition must vary by at least 5% (preferably over 10%) when the pressure changes no more than 10 atmospheres (Seader et al., 1997). Because the two columns or two sections in PSD operate at different pressures, it is critical to set the appropriate pressure for PSD. Many published studies have described the pressure selection in PSD (Abu-Eishah and Luyben, 1985; Lladosa et al., 2011; Luo et al., 2014; Luyben, 2013a, 2014b; Modla and Lang, 2010; Yu et al., 2012; Zhu et al., 2015). Overall material balances have been conducted (Abu-Eishah and Luyben, 1985) in continuous PSD for separating binary minimum boiling azeotropes, and an analogous determination can be obtained for PSBD. Luyben (2012) noted that recycling liquid bottoms was essentially the same in energy investment as recycling distillate streams that must be boiled up in the column. The results of material balance equations of PSD for separating binary minimum and maximum boiling azeotropes indicate that when XD1 (distillate composition of the LP column for separating minimum-boiling azeotrope) approaches XD2 (distillate composition of the HP column for separating minimum-boiling azeotrope) or XB1 (the bottom composition of the LP column for separating maximum-boiling azeotropes) approaches XB2 (the bottom composition of the HP column for separating maximum-boiling azeotropes), the recycle flow rates between the two columns become very large, which will result in high energy consumption for PSD. Because XD1 and XD2 (XB1 and XB2 ) are very close to the azeotropic compositions at LP and HP, the pressure differences between LP and HP should be maximized to maximize the difference between XD1 and XD2 (XB1 and XB2 ), which can save energy. In addition to the maximum pressure difference, the choice of pressure of LP is from the perspective of using cooling water in its condenser while the pressure of HP is selected from the consideration that steam heat can be used in the reboiler. For PSBD, Repke et al. (2007) noted that when the pressure difference between the LP step and the HP step increases, the amount of bottom product rises. Modla and Lang (2008a, 2010) reported that the azeotropes can be separated when the feed composition is located in a common section of the feasibility intervals of the two different components. Hence, when the feed composition is specified, the pressure should be selected to provide the common section.

4.2. Determination of the distillation sequence for continuous PSD The distillation sequence affects the separation efficiency and total annual cost (TAC) of PSD. Azeotropic characteristic and feed composition play an essential role in deciding the distillation sequence. Fig. 4 shows the T–x,y diagram of a homogeneous minimum boiling azeotrope. The system is a

special case. Component A is a light component, which has a lower boiling point at LP; component B is a heavy component with a higher boiling point. At HP, the mole fraction of A decreases in the azeotrope, and its boiling point becomes higher than component B compared with the LP condition. This performance is attributed to the nature of the Bancroft point in this system. The Bancroft point is the temperature and pressure at which the saturation vapor pressure of the components is the same, and the existence of the Bancroft point can indicate the composition dependence of an azeotrope (Elliott and Rainwater, 2000). Three different feed compositions and the corresponding distillation sequences are explored. In the case of Fig. 4(a), the feed stream F1, which is a combination of the fresh feed F and the recycle stream D2, has a mole fraction of component A that is much lower than the azeotropic composition of D2, so the stream F1 is fed into LP first. In the bottom stream B1 of LP, pure B can be achieved, and in the bottom stream B2 of HP, pure A can be obtained. In the case of Fig. 4(b) and (c), the mole fraction of component A in feed stream F1 is between the azeotropic composition of D1 and D2, and the distillation sequence can be LP + HP and HP + LP. In both distillation sequences, pure A and pure B are achieved at the bottom of HP and LP, respectively. Several papers have explored the selection of the LP + HP and HP + LP distillation sequence for separating binary azeotropes (Fulgueras et al., 2016; Fulgueras et al., 2015; Kim et al., 2013; Lladosa et al., 2011; Wang et al., 2015a, 2014; Zhu et al., 2015), and the optimal choice is based on the minimal TAC. For the case of Fig. 4(d), the component A mole fraction in stream F1 is much larger than the azeotropic composition of D2, so the feed stream should be fed into HP first. As shown in Fig. 4(d), the bottom product of HP is pure A while the bottom product of LP is pure B.

4.3.

Schemes of PSBD for different azeotropes

PSBD can be used to separate different azeotropes that exist in the pharmaceutical and some other industries. Different schemes of PSBD such as single column or double column configurations have been designed to suit different conditions (Modla and Lang, 2007). Repke et al. (2007) studied the performance of the inverted and regular PSBD to separate a minimum azeotrope. There are two steps, where one step is the LP step, and the other step is the HP step in the process modes. The main difference between the modes is feed location. In the inverted PSBD, the feed of the two steps is in the top tank of the column so that pure products can be obtained at the bottom while the azeotropes at different pressures are distilled back to the tank. In the regular PSBD, the feed of the steps is charged into the bottom tank, and the azeotropes are distilled at the top. At the end of each step, pure products are collected in the bottom tank, and the azeotropic distillate is assembled for the next step. The single column configuration costs less for the equipment, but pure products cannot be obtained simultaneously. Double column configuration, where one column is operated at low pressure and the other column is operated at high pressure, can produce two pure components at the same time. Two modes of the double column configuration are used to separate azeotropes (Modla and Lang, 2008a). DCBR has a feed vessel at the bottom of the two columns, and it is used to separate maximum azeotropes. The mixture is boiled up by the reboilers, and the vapor enters into the two columns. Then, pure components are distilled at the top,

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

327

Fig. 4 – Distillation sequence of different feed conditions. and the azeotropes return to the feed vessel. The operation of DCBS is reversed compared with DCBR, but DCBS is used to separate minimum azeotropes. The mixture is fed at the top of the two columns, and pure components are collected at the bottom while azeotropes at different columns are fed back to the feed vessel. The double column configuration can produce high purity products simultaneously, and the operation is simplified.

5.

Process intensification

Ramshaw (1995) first defined process intensification as a strategy for making dramatic reductions in the size of a chemical plant to reach a given production objective. With the development of the chemical industry, process intensification mainly emphasizes the sustainable development issues in process industry and presents the core element of Green Chemical Engineering. The main idea of process intensification is to reduce energy, material, and equipment costs, increase process flexibility and process safety, improve production quality, and obtain better environmental performance (Reay et al., 2013). Capital investment reduction and energy use reduction are two key themes of process intensification. For PSD, the flowsheet needs to be optimized, and heat integration can be applied to reduce energy consumption and capital costs. The minimum duration of the startup for mass- and heatintegrated PSD, which can save resources and reduce waste and costs, was also investigated in published paper (Varbanov et al., 2008). Fig. 5 shows a schematic diagram of optimization and heat integration of PSD, which involves optimization of heat integration. A typical and commonly used optimization method is the sequential iterative method, and some more advanced algorithms such as the simulated annealing optimization algorithm can be used for the optimization process.

A detailed introduction about the optimization process and heat integration is shown in the following content.

5.1.

Process optimization

Optimization is used to find the values of decision variables, which minimizes or maximizes the value of a given objective function and also satisfies the specified constraints. The steady state flow sheet of PSD can be simulated with some software from Aspen Plus, PRO/II, ChemCAD, etc. From the standpoint of economy, the flowsheet variables such as theoretical trays, feed locations, and reflux ratios (RRs) should be optimized to obtain a minimal TAC. The parameters and formulations about the capital and energy costs in TAC can refer to the Douglas book (Douglas, 1988) and the Turton book (Turton et al., 2008). Several optimization methods are used in the published literature to find the optimal design variables with the minimum energy consumption and equipment costs (Fulgueras et al., 2016; Hamad and Dunn, 2002; Lee et al., 2011). The sequential iteration optimization method is the most widely used in continuous PSD to obtain the optimal operation parameters, and the detailed optimization steps have been described in many papers (Li et al., 2015a; Wang et al., 2014, 2016b; Zhu et al., 2015). Variables such as theoretical trays can be selected as outer iterative loop variables while reflux ratios and feed locations can be selected as inner iterative loop variables. After a series of iterative calculation, the minimal TAC can be obtained. A genetic algorithm was also used to determine the optimal size and operation parameters of continuous PSD for separating acetone and methanol (Modla and Lang, 2012) and optimize the semi-continuous PSD process for separating THF and water (Phimister and Seider, 2000). The optimization process of PSBD is focused mainly on the double column system. The influence of some parameters such as feed location, RR, division ratio, and column profiles on the energy consumption and capital costs in the double

328

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Fig. 5 – Optimization and heat integration of PSD. column system have been explored in several papers (Modla, 2010, 2011a,b; Modla and Lang, 2010). Different from continuous PSD, the optimization process for PSBD is conducted in a dynamic state with software such as ChemCAD. Some optimization methods such as the simulated annealing algorithm and the colony optimization method can be used to optimize the continuous PSD process (Wang et al., 2016a) and make up the shortcomings of massive calculation and easily being a local optimum in the sequential iteration optimization method. A computer program using the optimization algorithm as shown in Fig. 6 can improve the optimization precision of continuous PSD and promote the development of a continuous PSD optimization process. The optimization procedure consists of the following steps. First, some initial values and algorithm parameters are given to the computer program, and then they are passed to simulation software as design variables for the process simulation. At the completion of the simulation, the results are transferred back to the computer program which calculates the TAC, and the design variables are updated according to the optimization algorithm. The steps are repeated until minimal TAC is obtained, and the optimal results can be found in the Microsoft Excel file that contains the optimal results. Currently, PSD optimization is focused mainly on single objective optimization, and TAC is often the sole optimization goal. However, PSD optimization may have more than one objective related to performance, cost, energy, safety, controllability, etc. Sometimes, a trade-off between cost, safety, and controllability needs to be considered. In such situations, it is better to perform multi-objective optimization. Unlike the single optimal solution from a single objective optimization,

multi-objective optimization gives many optimal solutions if the objectives are conflicting.

5.2.

Heat integration

In the PSD process, the reboilers of both columns need to be heated by steam, and the top vapors of both columns need to be cooled with cooling media Additionally, owing to pressure differences between the two columns (sections), a temperature driving force is established between the two columns (sections) so that heat transfer can be accomplished. To reduce heating and cooling loads of thermal utilities and improve heat recovery capacity of the system, some methods such as heat integration that features heat exchange between the hot stream and the cold stream are proposed. Heat integration is a systematic methodology that can achieve partial or global energy utilization of the process to minimize energy costs through the design of the heat exchange networks. Heat integration to reduce the energy consumption in PSD has been explored in many papers. Table 4 lists some published papers about the heat integration of PSD. In the continuous PSD process with two columns, the temperature difference between the condenser of the high pressure column (HPC) and the reboiler of the low pressure column (LPC) makes the condenser/reboiler type of heat integration possible. The overhead stream of HPC can be used to supply energy for the reboiler of LPC on the basis of no heat integration. For the full heat integration, the condenser duty of HPC can be equal to the heat duty of the LPC reboiler through adjusting the RRs of two columns and can supply the heat demand of the LPC reboiler correctly, which means that the LPC reboiler is not needed at

329

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Fig. 6 – Optimization algorithm interfaces with PSD simulation software. Table 4 – Heat integration of PSD for separating azeotropes in published works. Heat integration type Condenser/reboiler

System

Process type

Reference

Acetone + chloroform Acetone + chloroform + toluene Acetone + methanol

Continuous Batch Continuous

Acetone + methanol Acetonitrile + water Chloroform + methanol Cyclohexanone + phenol Diisopropyl ether + isopropyl alcohol Ethanol + toluene Ethyl acetate + ethanol Isobutyl alcohol + isobutyl acetate Methyl acetate + methanol Methanol + THF Methanol + trimethoxysilane Methylal + methanol MIBK + butanol N-Heptane + isobutanol THF + ethanol THF + water

Batch Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous

Toluene + 1-butanol Water + ethylenediamine

Continuous Continuous

Luyben (2013a) Modla (2011b) Luyben (2012), Luyben (2008a), Fulgueras et al. (2016), Wang et al. (2016a) Modla and Lang (2010) Repke et al. (2005), Repke et al. (2004) Hosgor et al. (2014), Wang et al. (2016a) Li et al. (2013) Luo et al. (2014) Zhu et al. (2015) Fissore et al. (2006), Kiran and Jana (2015a) Luo et al. (2016) Suo et al. (2016), Zhang et al. (2016) Wang et al. (2015b),Wang et al. (2014) Luyben (2014a), Luyben (2014b) Yu et al. (2012) Li et al. (2016b) Wang et al. (2016b) Wang et al. (2015a), Wang et al. (2015c) Abu-Eishah and Luyben (1985), Luyben (2008b), Hamad and Dunn (2002) Qasim et al. (2015) Li et al. (2015a), Fulgueras et al. (2015)

Rectifying/stripping

Acetonitrile + water Ethanol + water

Continuous Continuous

Huang et al. (2008) Mulia-Soto and Flores-Tlacuahuac (2011)

Hybrid mode

Ethanol + water Ethyl acetate + ethanol

Continuous Continuous

Kiran and Jana (2015b) Kiran and Jana (2015a)

all, and the cost of the heat exchanger can be reduced. For the partial heat integration, the heat duty of the LPC reboiler can be reduced for partial heat provided by the top steam of the HPC, thus causing the decrease of the heat exchange area, which means the reduction of heat exchanger capital costs. The rectifying/stripping type of heat integration in continuous PSD, which means arranging heat integration between the rectifying section and the stripping section of the HPC and

the LPC, respectively, has also been investigated (Huang et al., 2008). A hybrid thermal integration scheme for PSD by combining an internal heat integrated distillation column (HIDiC) with a vapor recompression column (VRC) process is explored in the published literature (Kiran and Jana, 2015b). In continuous PSD with two sections, several papers have studied the heat integration between the LP and the HP section. An internally heat-integrated PSD scheme for separating

330

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

ethanol and water, which is composed of a single shell column divided into HP section and LP section, can reduce the energy requirements due to the total internal heat integration between the two sections (Mulia-Soto and Flores-Tlacuahuac, 2011). Some close-boiling mixtures can be separated with pressure swing thermally coupled distillation, which means the reboiler at the bottom of the stripping section can be heated by the vapor from top of the rectifying section. Recently, the combination of reactive process and continuous PSD, including the heat integration, has aroused the attention of many scholars (Bonet et al., 2007; Kiran and Jana, 2015a; Suo et al., 2016; Wang et al., 2010), and the reactive PSD process can be further intensified in future studies. Currently, the study of heat integration in PSBD is focused mainly on the condenser/reboiler type (Modla, 2011b; Modla and Lang, 2010). By thermal coupling of the condenser of the high pressure column with the reboiler of the low pressure column, energy can also be saved for PSBD.

6.

Control scheme

Dynamic control is an essential factor that should be taken into consideration for the industrial application of PSD, and many papers have explored the dynamic control of continuous PSD and PSBD for separating azeotropes (Al-Arfaj and Luyben, 2004; Kopasz et al., 2009; Luo et al., 2014; Luyben, 2014a; Repke and Klein, 2005; Repke et al., 2004, 2005; Wang et al., 2015b,c; Wei et al., 2013). Process dynamics conducted in dynamic simulation software can give guidance for optimal control structure selection (Cao et al., 2016). In continuous PSD, two columns operate at different pressures at different sides of the azeotrope. To ensure the transition between the two pressure stages, it is necessary to set control structures to keep the distillate compositions at their set points (Klein, 2008). The first step is the choice of control variables and manipulated variables. Generally, control variables contain mainly bottom compositions or distillate compositions, bottom levels and reflux drum levels, feed flow rates and sensitive stage temperatures. Manipulated variables are usually bottom and distillate flow rates and heat duties of reboilers. The choice of optimal pairing should be considered from two aspects. First, the influence of manipulated variables on the control variables should be large enough for little meddling, and second, manipulated variables and control variables should be better in the same column to reduce the time constants of the controllers. For PSBD, the process operates without reaching a steady state, so there are some differences with continuous process. The distillation composition of the regular process and the bottom compositions of the inverted process are controlled by adjusting reflux and reboiler heat duty, respectively. The typical control system for continuous PSD including controller tuning conditions and a disturbance response curve, which play important roles in estimating the control effect, is also shown in Fig. 7. The proportional-integralderivative (PID) control method is widely used in the PSD control. Fig. 7 shows the controllers and control schemes available for dynamic control of PSD. For some PSD processes, the basic control which contains flow rate controllers, pressure controllers, liquid level controllers, and temperature controllers can achieve robust control and this condition is desirable for industrial production. However, some other PSD processes need more complex controllers or control schemes such as composition-temperature cascade control,

pressure-compensated temperature control, and fixed RR scheme to obtain stable controllability. In the development of the dynamic control of PSD, some ratio control structures such as reflux rate/feed flow rate (R/F) (Luyben, 2002), reboiler heat duty/feed flow rate (QR /F) (Luyben, 1992), and fixed RR make contributions to the improvement of the dynamic control of PSD. The tuning method for controllers has an important influence on the effective control structure of PSD. The calculation results of conventional Ziegler–Nichols (Ziegler and Nichols, 1942) and Tyreus–Luyben (Luyben, 1996) tuning rules that are used in the tuning of the temperature and composition controllers are conservative, and there is a deviation from practical application. Therefore, tuning methods of the PID controller arouse significant attention from scholars and researchers (Bouallègue et al., 2012; Jeng et al., 2014; Pavkovic´ et al., 2014; Zheng et al., 2014). The tuning rules of the PID controllers based on the n-th order lag process model (the so-called PTn model) (Pavkovic´ et al., 2014) can be used in tuning the composition and temperature controllers to obtain an optimal control structure. A control structure with good performance should make the process operate robustly in a very large feed concentration range. Commonly used feed disturbances are feed flow rate disturbances and feed composition disturbances. Evaluation indices of controllability in PSD contain mainly the product purities, the sensitive stage temperature, the bottom flow rates, and the distillate flow rates. When feed disturbances are introduced, the response conditions of these evaluation indices can be plotted as curves as depicted in Fig. 7 to show the change process clearly. For a robust control structure, the dynamic response curve of these evaluation indices should have small transient deviations, little time reaching the new steady state, and small final deviations. The challenges in PSD dynamic control are focused mainly on multi-component azeotropic PSD separation, including the heat integration process and the quantitative analysis of dynamic control with QSPR. The increased number of columns used for the PSD process brings more manipulated variables and control variables, which makes the dynamic control more complex compared with binary azeotropes. For the heat integrated PSD process of the multi-component azeotrope, the existence of more than one heat integration method makes the dynamic control challenging. In our opinion, quantitative analysis of the PSD dynamic control can give guidance on the more effective design of the PSD process. In addition to the type of PID control, predictive control (He et al., 2015), which is more suitable for the actual requirements of the industrial process, is also used in continuous PSD for recovering ethyl acetate (Fissore et al., 2006). Some other control methods such as fuzzy control, optimal control, and neural network control can be used in controlling the PSD process. Fuzzy control has the advantages of good real-time performance and strong anti-interference ability and has been widely studied (Castillo et al., 2015; Liu and Tong, 2015; Xu et al., 2015) and can be extended to the PSD control. Optimal control, which is based on optimization techniques (Kim et al., 2016), can also be used in the control scheme of PSD. Neural network control with strong adaptability and learning ability, has been used in a membrane distillation unit (Porrazzo et al., 2013), a distillation column (Rani et al., 2013), and a chemical reactive process (Li, 2014; Li et al., 2016a). Further study of neural network control can be extended to the PSD process.

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

331

Fig. 7 – Flowsheet of the typical control system for continuous PSD.

6.1.

Temperature control

Temperature control is widely used for its advantages of high reliability, little dynamic lag and downtime, and low investment. Sensitivity, correlation with products, and dynamic responses are three major factors used for determining the location of a temperature control point. For the selection of temperature-sensitive stages in PSD, Luyben (2006, 2013b) described in detail the methods of slope criterion, sensitivity criterion, singular value decomposition criterion, constant temperature criterion, and criterion for minimum variation of product purity. For some azeotropic systems where the temperature profiles have more than one large temperature change from stage to stage, different PSD heat integration schemes need different temperature-sensitive stages to obtain a robust control structure (Wang et al., 2015b). Therefore, the method for selecting temperature-sensitive stages with more than one large temperature change from stage to stage needs a systematic study on the temperature distribution of stages and the influence of feeding, heat duty, and RR on the dynamic responses of stage temperature. Temperature selection has important significance for understanding the dynamic characteristics of the related system and realizing the effective dynamic control of PSD. Temperature controllers are important parts for establishing stable control structures for separating azeotropes in PSD. The PSD separation of C5s and methanol was controlled by the temperature control structure (Luyben, 2005). An internally heat integrated PSD

process for bioethanol separation achieved robust control with a temperature controller (Mulia-Soto and Flores-Tlacuahuac, 2011). Temperature controllers can also be combined with some ratio control structures in control systems. The QR /F ratio control scheme, which shows some superiority in reducing transient deviation of product purity, is combined with a temperature controller in controlling the azeotropic systems of THF and ethanol (Wang et al., 2015c), methanol and THF (Wang et al., 2015b), methanol and dimethyl carbonate (Wei et al., 2013), THF and water (Luyben, 2008b), etc. In the design of a PSD control scheme for separating chloroform and methanol (Hosgor et al., 2014) and cyclohexanone and phenol (Li et al., 2013), a temperature controller performs well with the fixed RR.

6.2.

Composition-temperature cascade control

The purpose of composition control is to satisfy the constraints defined by product quality specifications. The constraints must be satisfied at all times, particularly in the face of disturbances. Composition control has the advantage of measuring the product purity directly although the limitation of a long dynamic lag and a high maintenance cost. A composition controller is often used as a cascade structure with a temperature controller. A composition-temperature cascade control structure can combine the advantages of temperature control and composition control. The methanol and THF azeotrope with partial heat integration PSD separation (Wang

332

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

et al., 2015b) and the cyclohexanone/phenol azeotrope with new PSD separation via introducing a heavy entrainer (Li et al., 2013) achieved robust control with a composition-temperature cascade control structure.

Financial supports from the National Natural Science Foundation of China (Project 21306093) are gratefully acknowledged.

6.3.

References

Pressure-compensated temperature control

Pressure-compensated temperature has been described briefly by Buckley et al. (1985). The technology is to measure pressure deviations from the flowsheet pressure, calculate temperature changes at a constant composition due to the pressure deviations, and subtract the temperature changes from the actual temperature. Luyben and Chien (2011) illustrated the implementation of pressure-compensated temperature control for the partial heat integrated PSD process in detail. Pressure-compensated temperature is widely used in controlling the heat-integrated PSD, which becomes more difficult to control compared with no heat integrated process due to the decrease of degrees of freedom. Several azeotropic systems separated by heat-integrated PSD use the pressure-compensated temperature control including ethanol and toluene (Zhu et al., 2015), THF and ethanol (Wang et al., 2015c), methanol and THF (Wang et al., 2015b), acetone and methanol (Luyben, 2008a), chloroform and methanol (Hosgor et al., 2014), etc. Several papers have explored the composition control of PSBD, focusing mainly on the double column systems. For double-column reactive PSBD, the RR and reboil ratio of the two columns can be adjusted by manipulating the distillate and bottom flow rates, respectively, thus maintaining the purity requirement (Modla, 2011a). For DCBS and DCBR in the open mode, the product purities are maintained by manipulating the corresponding product flow rates.

7.

Conclusions

A comprehensive description of the current research developments for PSD from thermodynamic analysis, QSPR, process design, process intensification, and dynamic control is reviewed in this paper. PSD has been a well-known technology used for separating pressure-sensitive azeotropic mixtures, with opportunities for separating pressure-insensitive azeotropes via introducing an entrainer. Currently, PSD studies are focused mainly on binary azeotropic systems. Multi-component azeotropes are also common phenomena in industry. The development of PSD for separating multicomponent azeotropes should be a focus, especially involving the heat integration and other technologies such as a reactive process and a batch process. Because of the high energy consumption of the distillation technology, process intensification of PSD should be further explored to improve energy utilization efficiency in PSD. In our opinion, the key issue for PSD is the exploration of the nature of PSD, which means that the VLE, steady state design, and dynamic control of PSD can be quantitatively analyzed by QSPR, and this analysis can give guidance on the more effective design of the PSD process. In the face of the requirement for sustainable development and environmental protection, research in PSD should be given sufficient attention by researchers and scholars.

Acknowledgments Comments and suggestions from anonymous reviewers are acknowledged.

Abildskov, J., O’Connell, J.P., 2015. Responses of azeotropes and relative volatilities to pressure variations. Chem. Eng. Res. Des. 99, 97–110. Abrams, D.S., Prausnitz, J.M., 1975. Statistical thermodynamics of liquid mixtures: a new expression for the excess Gibbs energy of partly or completely miscible systems. AIChE J. 21, 116–128. Abu-Eishah, S.I., Luyben, W.L., 1985. Design and control of a two-column azeotropic distillation system. Ind. Eng. Chem. Process Des. Dev. 24, 132–140. Al-Arfaj, M.A., Luyben, W.L., 2004. Plantwide control for TAME production using reactive distillation. AIChE J. 50, 1462–1473. An, Y., Li, W., Li, Y., Huang, S., Ma, J., Shen, C., Xu, C., 2015. Design/optimization of energy-saving extractive distillation process by combining preconcentration column and extractive distillation column. Chem. Eng. Sci. 135, 166–178. ˜ Aucejo, A., Montón, J.B., Munoz, R., Wisniak, J., 1996. Double azeotropy in the benzene + hexafluorobenzene system. J. Chem. Eng. Data 41, 21–24. Bonet, J., Thery, R., Meyer, X.-M., Meyer, M., Reneaume, J.-M., Galan, M.-I., Costa, J., 2007. Infinite/infinite analysis as a tool for an early oriented synthesis of a reactive pressure swing distillation. Comput. Chem. Eng. 31, 487–495. Bouallègue, S., Haggège, J., Ayadi, M., Benrejeb, M., 2012. PID-type fuzzy logic controller tuning based on particle swarm optimization. Eng. Appl. Artif. Intell. 25, 484–493. Britton, E.C., Horsley, L.H., Nutting, H.S., 1943. Separation of ketones and monohydric alcohols from mixtures thereof. U.S. Patent, 2,324,255. Buckley, P., Shunta, J., Luyben, W., 1985. Design of Distillation Column Control Systems. Instrument Society of America, Research Triangle Park, USA. Castillo, O., Lizárraga, E., Soria, J., Melin, P., Valdez, F., 2015. New approach using ant colony optimization with ant set partition for fuzzy control design applied to the ball and beam system. Inf. Sci. 294, 203–215. Cao, Y., Li, M., Wang, Y., Zhao, T., Li, X., Zhu, Z., Wang, Y., 2016. Effect of feed temperature on economics and controllability of pressure-swing distillation for separating binary azeotrope. Chem. Eng. Process., http://dx.doi.org/10.1016/j.cep.2016.10.011. Christensen, S.P., Olson, J.D., 1992. Phase equilibria and multiple azeotropy of the acetic acid-isobutyl acetate system. Fluid Phase Equilib. 79, 187–199. Douglas, J.M., 1988. Conceptual Design of Chemical Processes. McGraw-Hill, New York. Elliott, J.R., Rainwater, J.C., 2000. The Bancroft point and vapor–liquid equilibria in the system benzene + isopropanol. Fluid Phase Equilib. 175, 229–236. Fissore, D., Pin, M., Barresi, A.A., 2006. On the use of detailed models in the MPC algorithm: the pressure-swing distillation case. AIChE J. 52, 3491–3500. Frank, T.C., 1997. Break azeotropes with pressure-sensitive distillation. Chem. Eng. Prog. 93, 52–63. Fulgueras, A.M., Kim, D.S., Cho, J., 2016. Modeling and optimization study of pressure-swing distillation for the separation process of acetone–methanol mixture with vapor–liquid equilibrium analysis. J. Chem. Eng. Jpn. 49, 84–96. Fulgueras, A.M., Poudel, J., Kim, D.S., Cho, J., 2015. Optimization study of pressure-swing distillation for the separation process of a maximum-boiling azeotropic system of water-ethylenediamine. Korean J. Chem. Eng. 33, 46–56. Gaudin, T., Rotureau, P., Fayet, G., 2015. Mixture descriptors toward the development of quantitative structure–property relationship models for the flash points of organic mixtures. Ind. Eng. Chem. Res. 54, 6596–6604.

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Gebreyohannes, S., Yerramsetty, K., Neely, B.J., Gasem, K.A.M., 2013. Improved QSPR generalized interaction parameters for the nonrandom two-liquid activity coefficient model. Fluid Phase Equilib. 339, 20–30. Gharagheizi, F., Eslamimanesh, A., Ilani-Kashkouli, P., Mohammadi, A.H., Richon, D., 2012. QSPR molecular approach for representation/prediction of very large vapor pressure dataset. Chem. Eng. Sci. 76, 99–107. Gmehling, J., 1994. Azeotropic Data. Wiley-VCH, Weinheim. Gmehling, J., Kolbe, B., Kleiber, M., Rarey, J., 2012. Chemical Thermodynamics for Process Simulation. Wiley-VCH, Weinheim. Gmehling, J., Menke, J., Krafczyk, J., Fischer, K., 1995. A data bank for azeotropic data—status and applications. Fluid Phase Equilib. 103, 51–76. Gmehling, J., Rasmussen, P., Fredenslund, A., 1982. Vapor–liquid equilibriums by UNIFAC group contribution. Revision and extension. 2. Ind. Eng. Chem. Process Des. Dev. 21, 118–127. Hamad, A., Dunn, R.F., 2002. Energy optimization of pressure-swing azeotropic distillation systems. Ind. Eng. Chem. Res. 41, 6082–6093. He, D., Yu, S., Yu, L., 2015. Multi-objective nonlinear model predictive control through switching cost functions and its applications to chemical processes. Chin. J. Chem. Eng. 23, 1662–1669. Hemmateenejad, B., Shamsipur, M., Zare-Shahabadi, V., Akhond, M., 2011. Building optimal regression tree by ant colony system-genetic algorithm: application to modeling of melting points. Anal. Chim. Acta 704, 57–62. Honghai, L., Yi, J., Shaohui, T., 2015. The application of MPSO algorithm to the optimal design of azeotropic distillation column. Chem. Ind. Eng. 5, 017. Horsley, L., 1947. Graphical method for predicting azeotropism and effect of pressure on azeotropic constants. Anal. Chem. 19, 603–609. Hosgor, E., Kucuk, T., Oksal, I.N., Kaymak, D.B., 2014. Design and control of distillation processes for methanol–chloroform separation. Comput. Chem. Eng. 67, 166–177. Huang, K., Shan, L., Zhu, Q., Qian, J., 2008. Adding rectifying/stripping section type heat integration to a pressure-swing distillation (PSD) process. Appl. Therm. Eng. 28, 923–932. Jeng, J.-C., Tseng, W.-L., Chiu, M.-S., 2014. A one-step tuning method for PID controllers with robustness specification using plant step-response data. Chem. Eng. Res. Des. 92, 545–558. ¨ Katritzky, A.R., Stoyanova-Slavova, I.B., Tamm, K., Tamm, T., Karelson, M., 2011. Application of the QSPR approach to the boiling points of azeotropes. J. Phys. Chem. A 115, 3475–3479. Kim, K.W., Shin, J.S., Kim, S.H., Hong, S.K., Cho, J.H., Park, S.J., 2013. A Computational study on the separation of acetonitrile and water azeotropic mixture using pressure swing distillation. J. Chem. Eng. Jpn. 46, 347–352. Kim, W., Jeon, S.W., Kim, Y., 2016. Model-based multi-objective optimal control of a VRF (variable refrigerant flow) combined system with DOAS (dedicated outdoor air system) using genetic algorithm under heating conditions. Energy 107, 196–204. Kiran, B., Jana, A.K., 2015a. Assessing the performance improvement of an intensified heat integration scheme: reactive pressure-swing distillation. Appl. Therm. Eng. 76, 509–520. Kiran, B., Jana, A.K., 2015b. A hybrid heat integration scheme for bioethanol separation through pressure-swing distillation route. Sep. Purif. Technol. 142, 307–315. Klein, A., 2008. Azeotropic Pressure Swing Distillation (Dissertation). TU, Berlin. Klein, A., Repke, J.-U., 2009. Regular and inverted batch process structures for pressure swing distillation: a case study. Asia Pac. J. Chem. Eng. 4, 893–904. Knapp, J.P., 1991. Exploiting Pressure Effects in the Distillation of Homogeneous Azeotropic Mixtures (Dissertation). University of Massachusetts.

333

Knapp, J.P., Doherty, M.F., 1992. A new pressure-swing-distillation process for separating homogeneous azeotropic mixtures. Ind. Eng. Chem. Res. 31, 346–357. Kopasz, A., Modla, G., Lang, P., 2009. Operation and control of a new pressure swing batch distillation system. Chem. Eng. Trans. 18, 707–712. Lee, J., Cho, J., Kim, D.M., Park, S., 2011. Separation of tetrahydrofuran and water using pressure swing distillation: modeling and optimization. Korean J. Chem. Eng. 28, 591–596. Lei, Z., Chen, B., Ding, Z., 2005. Special Distillation Processes. Elsevier Science Publishers, Amsterdam. Lewis, W., 1928. Dehydrating alcohol and the like, U.S. Patent, 1,676,700. Li, D.-J., 2014. Neural network control for a class of continuous stirred tank reactor process with dead-zone input. Neurocomputing 131, 453–459. Li, R., Ye, Q., Suo, X., Dai, X., Yu, H., 2015a. Heat-integrated pressure-swing distillation process for separation of a maximum-boiling azeotrope ethylenediamine/water. Chem. Eng. Res. Des. 105, 1–15. Li, S., Gong, M., Liu, Y., 2016a. Neural network-based adaptive control for a class of chemical reactor systems with non-symmetric dead-zone. Neurocomputing 174, 597–604. Li, C., Song, Y., Fang, J., Liu, Y., Su, W., Hu, Y., 2016b. Separation process of butanol-butyl acetate-methyl isobutyl ketone system by the analysis to residual curve and the double effect pressure-swing distillation. Chin. J. Chem. Eng., http://dx.doi.org/10.1016/j.cjche.2016.08.011. Li, W., Shi, L., Yu, B., Xia, M., Luo, J., Shi, H., Xu, C., 2013. New pressure-swing distillation for separating pressure-insensitive maximum boiling azeotrope via introducing a heavy entrainer: design and control. Ind. Eng. Chem. Res. 52, 7836–7853. Li, W., Zhong, L., He, Y., Meng, J., Yao, F., Guo, Y., Xu, C., 2015b. Multiple steady-states analysis and unstable operating point stabilization in homogeneous azeotropic distillation with intermediate entrainer. Ind. Eng. Chem. Res. 54, 7668–7686. Liu, Y.-J., Tong, S., 2015. Adaptive fuzzy control for a class of unknown nonlinear dynamical systems. Fuzzy Set. Syst. 263, 49–70. Lladosa, E., Montón, J.B., Burguet, M., 2011. Separation of di-n-propyl ether and n-propyl alcohol by extractive distillation and pressure-swing distillation: computer simulation and economic optimization. Chem. Eng. Process. 50, 1266–1274. ˜ Lladosa, E., Montón, J.B., Burguet, M.C., Munoz, R., 2006. Isobaric vapor–liquid equilibria for the binary systems 1-propyl alcohol + dipropyl ether and 1-butyl alcohol + dibutyl ether at 20 and 101.3 kPa. Fluid Phase Equilib. 247, 47–53. Luo, B., Feng, H., Sun, D., Zhong, X., 2016. Control of fully heat-integrated pressure swing distillation for separating isobutyl alcohol and isobutyl acetate. Chem. Eng. Process. 110, 9–20. Luo, H., Liang, K., Li, W., Li, Y., Xia, M., Xu, C., 2014. Comparison of pressure-swing distillation and extractive distillation methods for isopropyl alcohol/diisopropyl ether separation. Ind. Eng. Chem. Res. 53, 15167–15182. Luyben, W., 2002. Plantwide Dynamic Simulators in Chemical Processing and Control. CRC Press, New York. Luyben, W.L., 1996. Tuning proportional-integral-derivative controllers for integrator/deadtime processes. Ind. Eng. Chem. Res. 35, 3480–3483. Luyben, W.L., 2005. Comparison of pressure-swing and extractive-distillation methods for methanol-recovery systems in the tame reactive-distillation process. Ind. Eng. Chem. Res. 44, 5715–5725. Luyben, W.L., 2006. Evaluation of criteria for selecting temperature control trays in distillation columns. J. Process Control 16, 115–134. Luyben, W.L., 2008a. Comparison of extractive distillation and pressure-swing distillation for acetone-methanol separation. Ind. Eng. Chem. Res. 47, 2696–2707.

334

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

Luyben, W.L., 2008b. Design and control of a fully heat-integrated pressure-swing azeotropic distillation system. Ind. Eng. Chem. Res. 47, 2681–2695. Luyben, W.L., 1992. Practical Distillation Control. Van Nostrand Reinhold, New York. Luyben, W.L., 2012. Pressure-swing distillation for minimum- and maximum-boiling homogeneous azeotropes. Ind. Eng. Chem. Res. 51, 10881–10886. Luyben, W.L., 2013a. Comparison of extractive distillation and pressure-swing distillation for acetone/chloroform separation. Comput. Chem. Eng. 50, 1–7. Luyben, W.L., 2013b. Distillation Design and Control Using Aspen Simulation. John Wiley & Sons, New York. Luyben, W.L., 2014a. Control of a heat-integrated pressure-swing distillation process for the separation of a maximum-boiling azeotrope. Ind. Eng. Chem. Res. 53, 18042–18053. Luyben, W.L., 2014b. Methanol/trimethoxysilane azeotrope separation using pressure-swing distillation. Ind. Eng. Chem. Res. 53, 5590–5597. Luyben, W.L., 2015. Improved design of an extractive distillation system with an intermediate-boiling solvent. Sep. Purif. Technol. 156, 336–347. Luyben, W.L., Chien, I.-L., 2011. Design and Control of Distillation Systems for Separating Azeotropes. John Wiley & Sons, New York. Mahdi, T., Ahmad, A., Nasef, M.M., Ripin, A., 2015. State-of-the-art technologies for separation of azeotropic mixtures. Sep. Purif. Rev. 44, 308–330. Martínez, N.F., Lladosa, E., Burguet, M., Montón, J.B., 2008. Isobaric vapour–liquid equilibria for binary systems of 2-butanone with ethanol, 1-propanol, and 2-propanol at 20 and 101.3 kPa. Fluid Phase Equilib. 270, 62–68. Martínez, N.F., Lladosa, E., Burguet, M.C., Montón, J.B., Yazimon, M., 2009. Isobaric vapour–liquid equilibria for the binary systems 4-methyl-2-pentanone + 1-butanol and + 2-butanol at 20 and 101.3 kPa. Fluid Phase Equilib. 277, 49–54. Modla, G., 2010. Pressure swing batch distillation by double column systems in closed mode. Comput. Chem. Eng. 34, 1640–1654. Modla, G., 2011a. Reactive pressure swing batch distillation by a new double column system. Comput. Chem. Eng. 35, 2401–2410. Modla, G., 2011b. Separation of a chloroform–acetone–toluene mixture by pressure-swing batch distillation in different column configurations. Ind. Eng. Chem. Res. 50, 8204–8215. Modla, G., Lang, P., 2007. New column configurations for pressure swing batch distillation II. Rigorous simulation calculations. IFAC Proceedings 40, 361–366. Modla, G., Lang, P., 2008a. Feasibility of new pressure swing batch distillation methods. Chem. Eng. Sci. 63, 2856–2874. Modla, G., Lang, P., 2008b. Separation of a ternary homoazeotropic mixture by pressure swing batch distillation. Hung. J. Ind. Chem. 36, 1–2. Modla, G., Lang, P., 2010. Separation of an acetone- methanol mixture by pressure-swing batch distillation in a double-column system with and without thermal integration. Ind. Eng. Chem. Res. 49, 3785–3793. Modla, G., Lang, P., 2012. Removal and recovery of organic solvents from aqueous waste mixtures by extractive and pressure swing distillation. Ind. Eng. Chem. Res. 51, 11473–11481. Modla, G., Lang, P., Denes, F., 2010. Feasibility of separation of ternary mixtures by pressure swing batch distillation. Chem. Eng. Sci. 65, 870–881. Mokshyna, E., Polishchuk, P., Nedostup, V., Kuz’min, V., 2016. QSPR-modeling for the second virial cross-coefficients of binary organic mixtures. IJQSPR 1, 72–84. ˜ Munoz, R., Montón, J.B., Burguet, M.C., de la Torre, J., 2006. Separation of isobutyl alcohol and isobutyl acetate by extractive distillation and pressure-swing distillation: simulation and optimization. Sep. Purif. Technol. 50, 175–183. Mulia-Soto, J.F., Flores-Tlacuahuac, A., 2011. Modeling, simulation and control of an internally heat integrated pressure-swing

distillation process for bioethanol separation. Comput. Chem. Eng. 35, 1532–1546. Oprisiu, I., Novotarskyi, S., Tetko, I.V., 2013. Modeling of non-additive mixture properties using the online CHEmical database and modeling environment (OCHEM). J. Cheminf. 5, 4. Oprisiu, I., Varlamova, E., Muratov, E., Artemenko, A., Marcou, G., Polishchuk, P., Kuz’min, V., Varnek, A., 2012. QSPR approach to predict nonadditive properties of mixtures. application to bubble point temperatures of binary mixtures of liquids. Mol. Inf. 31, 491–502. ´ D., Polak, S., Zorc, D., 2014. PID controller auto-tuning Pavkovic, based on process step response and damping optimum criterion. ISA Trans. 53, 85–96. Phimister, J.R., Seider, W.D., 2000. Semicontinuous, pressure-swing distillation. Ind. Eng. Chem. Res. 39, 122–130. Porrazzo, R., Cipollina, A., Galluzzo, M., Micale, G., 2013. A neural network-based optimizing control system for a seawater-desalination solar-powered membrane distillation unit. Comput. Chem. Eng. 54, 79–96. Pourbasheer, E., Beheshti, A., Vahdani, S., Nekoei, M., Danandeh, M., Abbasghorbani, M., Ganjali, M.R., 2015. Simple QSPR modeling for prediction of the GC retention indices of essential oil compounds. J. Essent. Oil Bear. Plants 18, 1298–1309. Qasim, F., Shin, J.S., Cho, S.J., Park, S.J., 2015. Optimizations and heat integrations on the separation of toluene and 1-butanol azeotropic mixture by pressure swing distillation. Separ. Sci. Technol. 51, 316–326. Ramshaw, C., 1995. The incentive for process intensification. Proceedings, 1st International Conference on Process Intensification in the Chemical Industry 18, 1–4. Rani, A., Singh, V., Gupta, J., 2013. Development of soft sensor for neural network based control of distillation column. ISA Trans. 52, 438–449. Reay, D., Ramshaw, C., Harvey, A., 2013. Process Intensification: Engineering for Efficiency, Sustainability and Flexibility, 2nd ed. Butterworth-Heinemann, Oxford. Renon, H., Prausnitz, J.M., 1968. Local compositions in thermodynamic excess functions for liquid mixtures. AIChE J. 14, 135–144. Repke, J.-U., Klein, A., 2005. Homogeneous azeotropic pressure swing distillation: continuous and batch process. ESCAPE 15, 721–726. Repke, J.-U., Klein, A., Forner, F., Wozny, G., 2004. Pressure swing distillation for separation of homogeneous azeotropic mixtures in a mass-and heat-integrated column system: operation performance. Intelligent control and automation. WCICA 2004 Fifth World Congress on. IEEE, 3543–3546. Repke, J.U., Forner, F., Klein, A., 2005. Separation of homogeneous azeotropic mixtures by pressure swing distillation–analysis of the operation performance. Chem. Eng. Technol. 28, 1151–1157. Repke, J.U., Klein, A., Bogle, D., Wozny, G., 2007. Pressure swing batch distillation for homogeneous azeotropic separation. Chem. Eng. Res. Des. 85, 492–501. Roscoe, H.E., 1860. On the composition of the aqueous acids of constant boiling point. J. Chem. Soc. 13, 146. Roscoe, H.E., 1862. XXXVI:—on the composition of the aqueous acids of constant boiling point—second communication. J. Chem. Soc. 15, 270–276. Roscoe, H.E., Dittmar, W., 1860. XV:—on the absorption of hydrochloric acid and ammonia in water. J. Chem. Soc. 12, 128–151. Rybinska, A., Sosnowska, A., Grzonkowska, M., Barycki, M., Puzyn, T., 2016. Filling environmental data gaps with QSPR for ionic liquids: modeling n-octanol/water coefficient. J. Hazard. Mater. 303, 137–144. Seader, J., Siirola, J.J., Barnicki, S.D., 1997. Perry’s Chemical Engineer’s Handbook, 7th ed. McGraw-Hill, New York. Solov’ev, V.P., Oprisiu, I., Marcou, G., Varnek, A., 2011. Quantitative structure–property relationship (QSPR) modeling

chemical engineering research and design 1 1 7 ( 2 0 1 7 ) 318–335

of normal boiling point temperature and composition of binary azeotropes. Ind. Eng. Chem. Res. 50, 14162–14167. Srivastava, R., Smith, B.D., 1985. Total-pressure vapor-liquid equilibrium data for binary systems of diethylamine with acetone, acetonitrile, and methanol. J. Chem. Eng. Data 30, 308–313. Suo, X., Ye, Q., Li, R., Dai, X., Yu, H., 2016. The partial heat-integrated pressure-swing reactive distillation process for transesterification of methyl acetate with isopropanol. Chem. Eng. Process. 107, 42–57. Turton, R., Bailie, R.C., Whiting, W.B., Shaeiwitz, J.A., 2008. Analysis, Synthesis and Design of Chemical Processes, 3rd ed. Prentice Hall PTR, New Jersey. Varbanov, P., Klein, A., Repke, J.U., Wozny, G., 2008. Minimising the startup duration for mass- and heat-integrated two-column distillation systems: a conceptual approach. Chem. Eng. Process. 47, 1456–1469. Walas, S.M., 1985. Phase Equilibria in Chemical Engineering. Butterworth-Heinemann, Boston. Wang, S.-J., Yu, C.-C., Huang, H.-P., 2010. Plant-wide design and control of DMC synthesis process via reactive distillation and thermally coupled extractive distillation. Comput. Chem. Eng. 34, 361–373. Wang, Y., Bu, G., Wang, Y., Zhao, T., Zhang, Z., Zhu, Z., 2016a. Application of a simulated annealing algorithm to design and optimize a pressure-swing distillation process. Comput. Chem. Eng. 95, 97–107. Wang, Y., Cui, P., Ma, Y., Zhang, Z., 2015a. Extractive distillation and pressure-swing distillation for THF/ethanol separation. J. Chem. Technol. Biotechnol. 90, 1463–1472. Wang, Y., Cui, P., Zhang, Z., 2014. Heat-integrated pressure-swing-distillation process for separation of tetrahydrofuran/methanol with different feed compositions. Ind. Eng. Chem. Res. 53, 7186–7194. Wang, Y., Zhang, Z., Xu, D., Liu, W., Zhu, Z., 2016b. Design and control of pressure-swing distillation for azeotropes with different types of boiling behavior at different pressures. J. Process Control 42, 59–76. Wang, Y., Zhang, Z., Zhang, H., Zhang, Q., 2015b. Control of heat integrated pressure-swing-distillation process for separating azeotropic mixture of tetrahydrofuran and methanol. Ind. Eng. Chem. Res. 54, 1646–1655. Wang, Y., Zhang, Z., Zhao, Y., Liang, S., Bu, G., 2015c. Control of extractive distillation and partially heat-integrated pressure-swing distillation for separating azeotropic mixture of ethanol and tetrahydrofuran. Ind. Eng. Chem. Res. 54, 8533–8545. Wasylkiewicz, S.K., Kobylka, L.C., Castillo, F.J., 2003. Pressure sensitivity analysis of azeotropes. Ind. Eng. Chem. Res. 42, 207–213.

335

Wei, H.-M., Wang, F., Zhang, J.-L., Liao, B., Zhao, N., Xiao, F.-k., Wei, W., Sun, Y.-H., 2013. Design and control of dimethyl carbonate–methanol separation via pressure-swing distillation. Ind. Eng. Chem. Res. 52, 11463–11478. Weidlich, U., Gmehling, J., 1987. A modified UNIFAC model. 1. Prediction of VLE, hE, and.gamma.infin. Ind. Eng. Chem. Res. 26, 1372–1381. Wilson, G.M., 1964. Vapor–liquid equilibrium. XI. A new expression for the excess free energy of mixing. J. Am. Chem. Soc. 86, 127–130. Xu, B., Shi, Z., Yang, C., 2015. Composite fuzzy control of a class of uncertain nonlinear systems with disturbance observer. Nonlinear Dyn. 80, 341–351. You, X., Rodriguez-Donis, I., Gerbaud, V., 2015. Improved design and efficiency of the extractive distillation process for acetone–methanol with water. Ind. Eng. Chem. Res. 54, 491–501. Yu, B., Wang, Q., Xu, C., 2012. Design and control of distillation system for methylal/methanol separation. Part 2: pressure swing distillation with full heat integration. Ind. Eng. Chem. Res. 51, 1293–1310. Yu, H., Ye, Q., Xu, H., Zhang, H., Dai, X., 2015. Design and control of dividing-wall column for tert-butanol dehydration system via heterogeneous azeotropic distillation. Ind. Eng. Chem. Res. 54, 3384–3397. Zare-Shahabadi, V., Lotfizadeh, M., Gandomani, A.R.A., Papari, M.M., 2013. Determination of boiling points of azeotropic mixtures using quantitative structure–property relationship (QSPR) strategy. J. Mol. Liq. 188, 222–229. Zhang, L., Liu, J., Li, X., Li, H., Jiang, B., Xiao, X., 2014. Separations of different binary hydrocarbon mixtures using pressure swing thermally coupled distillation process. Sep. Sci. Technol. 50, 148–157. Zhang, Z., Zhang, Q., Li, G., Liu, M., Gao, J., 2016. Design and control of methyl acetate-methanol separation via heat-integrated pressure-swing distillation. Chin. J. Chem. Eng., http://dx.doi.org/10.1016/j.cjche.2016.06.013. Zheng, S., Tang, X., Song, B., 2014. A graphical tuning method of fractional order proportional integral derivative controllers for interval fractional order plant. J. Process Control 24, 1691–1709. Zhu, Z., Wang, L., Ma, Y., Wang, W., Wang, Y., 2015. Separating an azeotropic mixture of toluene and ethanol via heat integration pressure swing distillation. Comput. Chem. Eng. 76, 137–149. Zhu, Z., Xu, D., Liu, X., Zhang, Z., Wang, Y., 2016. Separation of acetonitrile/methanol/benzene ternary azeotrope via triple column pressure-swing distillation. Sep. Purif. Technol. 169, 66–77. Ziegler, J.G., Nichols, N.B., 1942. Optimum settings for automatic controllers. Trans. ASME, 64.