Towards energy efficient styrene distillation scheme: From grassroots design to retrofit

Accepted Manuscript Towards better styrene distillation scheme: From grassroots design to retrofit

Chengtian Cui, Xingang Li, Dongrong Guo, Jinsheng Sun PII:

S0360-5442(17)31025-3

DOI:

10.1016/j.energy.2017.06.031

Reference:

EGY 11039

To appear in:

Energy

Received Date:

20 May 2016

Revised Date:

14 April 2017

Accepted Date:

06 June 2017

Please cite this article as: Chengtian Cui, Xingang Li, Dongrong Guo, Jinsheng Sun, Towards better styrene distillation scheme: From grassroots design to retrofit, Energy (2017), doi: 10.1016/j. energy.2017.06.031

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Towards better styrene distillation scheme: From grassroots design to retrofit Chengtian Cuia, Xingang Lia, b, c, Dongrong Guod, Jinsheng Suna, * aSchool

of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, P. R. China

bNational

Engineering Research Center of Distillation Technology, Tianjin, 300072, P. R. China

cCollaborative

Innovation Center of Chemical Science and Engineering, Tianjin, 300072, P. R. China

dShandong

Qilu Petrochemical Engineering Co. Ltd, Zibo, Shandong, 255400, P. R. China

Abstract The current state-of-the-art commercial styrene distillation schemes, featured by conventional distillation columns to purify styrene, are introduced with energy, exergy and economic analyses. Amongst all the procedures the separation of ethylbenzene/styrene, the critical close-boiling system, accounts for ~65% of the total energy requirement. To improve the energetic efficiency, double-effect distillation (DED) and heat pump distillation (HPD) are suggested as competitive improvements on conventional distillation schemes (CDSs), which give birth to advanced distillation schemes (ADSs). In addition, sensitivity analysis is carried out to determine the optimal operational parameters of columns in styrene distillation process. Taking the CDSs as benchmark processes, the ADSs with DED and HPD can lower operating costs by up to 30% and 40%, respectively. The synergistic effect makes retrofit proposals’ payback period very attractive, through considerably energy costs reduction and uttermost equipment reuse. In the view of total annualized cost (TAC), the ADSs can cut a corner of ~35-40% from the CDSs. Specifically, the ADS using HPD slightly outperforms its DED counterpart in TAC comparison. Despite energetic or monetary advantage, the ADSs also show their environmental drawback of higher exergy losses than the CDSs. Keywords: Styrene distillation scheme; Conventional distillation sequence; Grassroots process design; Energy-saving distillation technology; Advanced distillation scheme; Retrofit proposals 1. Introduction Styrene monomer (SM) is the simplest and by far the most important member of a series of aromatic monomers that is used in the plastics industry [1]. It is widely used in synthesizing various polymers and copolymers, such as polystyrene (PS), acrylonitrile-butadiene-styrene copolymers (ABS), styrene-acrylonitrile copolymers (SAN), styrene-butadiene rubber (SBR), styrenic block copolymers (SBC), styrene-butadiene latex (SBL) and adhesives [2]. The capacity of producing SM has increased constantly with the commercial demand. It is estimated over 25 million tons of SM is produced worldwide annually [1-3]. Currently, catalytic dehydrogenation of ethylbenzene (EB), shown in Fig. 1, overwhelms in styrene industry [4]. When styrene production happens in dehydrogenation reactor, steam is added to lower partial pressure of EB, boosting styrene conversion and suppressing undesirable byproducts, such as lights (hydrogen, ethylene, carbon monoxide, carbon dioxide etc.), benzene, toluene and other heavy components known as styrene tar [5]. After removal of lights and water from the dehydrogenated products in a three-phase (vapor, aqueous and organic) decanter, the organic phase (SM, benzene, toluene, unreacted EB and styrene tar) is to be separated or recycled in styrene

*Corresponding author E-mail address: [email protected] (J. Sun).

1

ACCEPTED MANUSCRIPT distillation unit, the most popular option to manipulate reactor outputs commercially [6-7]. However, it is significantly energy intensive to treat the organic phase by conventional distillation [6-7]. The low thermodynamic efficiency is the most responsible aspect of conventional distillation column (CDiC), with high-grade energy at reboiler and a similar output of waste heat at condenser [8]. This gives opportunities to the suggestion of numerous energy efficient distillation pathways to lower CDiC energy consumption [9-12]. The options, with the aid of intermediate heat exchangers to recover heat in common, include double-effect distillation (DED) with an additional distillation column [1314], heat pump distillation (HPD) with a compressor [15-17], internally heat-integrated distillation column (HIDiC) with a pressured rectifying section [18-21], and dividing wall column (DWC) with an equivalent prefractionator [22-25]. Although HIDiC and DWC are very less commonly industrialized [11], DED and HPD are welcomed by patent licensers and engineering companies for reliability with rich design experience [26]. Recently, Jongmans et al. added a conceptual extractive distillation (ED) to the warehouse of separating EB/SM system by using sulfolane [1,27,28] or ionic liquid [29] as the extraction agent. ED was announced to be able to reduce energy requirement by ~40-45% compared to conventional distillation [27]. However, this pathway is lack of reports of successful industrialization in the field of styrene, possibly in fear of contamination of high purity SM product by external extraction agents [27]. This uncertainty undermines the consideration of ED herein. This work starts from energy, exergy and economic analyses (3E analyses) of current state-of-theart commercial styrene distillation schemes. Since these schemes are based on the CDiC, they are referred to conventional distillation schemes (CDSs) later in this work, whose sequences are studied for further amelioration. Then DED and HPD configurations are applied on the CDSs, to give birth to advanced distillation schemes (ADSs). Finally, 3E analyses of ADSs are carried out, with the aid of sensitivity analyses to investigate optimal operational parameters of columns. Moreover, an economic evaluation of the payback period (PBP) for additional investment is necessary to determine the feasibility of the ADSs. To clearly compare total energy consumption of CDSs and ADSs, all of the utilities depleted—as secondary energy consumption—are converted into primary standard oil consumption, in close link with economic expenditures. Together with operating costs, the capital investments, including distillation columns, condensers, reboilers and compressors, are calculated and included into total annualized cost (TAC) for economic evaluation. 2. Theory 2.1. Distillation sequences Currently most industrial multicomponent separations are carried out employing conventional distillation sequences (CDiSs) [30]. A CDiC is defined as a separator with one feed with two products, equipped with one condenser and one reboiler [31]. For a N-component mixture, if N product streams are required, a CDiS is necessarily composed by N-1 CDiCs [32]. The number of possible CDiSs is provided by Eq. (1) [33]: 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑆𝑒𝑞𝑢𝑒𝑛𝑐𝑒𝑠 =

[2(𝑁 ‒ 1)]! 𝑁!(𝑁 ‒ 1)!

(1)

The introduction of state-of-the-art commercial styrene distillation schemes starts by the synthesis of CDiSs, in hope of improving energy efficiency and economic benefit at the beginning. As shown in Fig. 1, this separation process consists of five product streams and therefore four CDiCs are necessary for the separation task. Herein, to reduce the enumeration of CDiSs, the SM and tar are lumped as the

2

ACCEPTED MANUSCRIPT heavies, requiring an additional column. Under this circumstance, a pseudo four products to be purified are benzene (a), toluene (b), EB (c), SM and tar (d). Fig. 2 enumerates all five possible CDiSs in consideration. The molecular structures and normal boiling points (NBPs) of these products are listed in Table 1. Notably, the NBP of EB is very close to that of SM, so EB/SM mixture is composed to a close-boiling system. Since SM will polymerize over a certain high temperature [5-6], the distillation operates under vacuum with 2-sec-butyl-4,6-dinitrophenol (DNBP), as polymerization inhibitor, to be added in the dehydrogenated products. Reasonably, it is favored by heating SM as fewer times as possible. In CDiSs 1 and 2, SM is boiled three times, but it is only heated twice in 3, 4 and 5, hence 1 and 2 are eliminated for the sake of safety and higher SM yield. Practically, the mass flow rate of benzene and toluene is much smaller than that of EB, so it consumes more steam to boil EB three times. Compared with only two times in 3 and 5, sequence 4 is removed as well. The CDiS analysis shows that only sequences 3 and 5 are acceptable in grassroots process design. The choices are consistent with commercialized state-of-the-art styrene distillation schemes, which will be referred to in detail in the following sections. 2.2. DED DED (Fig. 3) implements two columns instead of one in CDiC [11,13-14]. It uses the overhead vapor from high-pressure column (HPC) to drive the subsequent reboiler of the low-pressure column (LPC), combining the condenser of the former with the reboiler of the latter, eliminating a heat exchanger as well as the corresponding utilities. Instead of treating the entire crude feed in one column, the feed is split into two approximately same streams and entered into HPC and LPC, respectively. The pressure difference of adjacent columns provides temperature difference at HPC overhead and LPC bottom, creating heat transfer drive from the pressured stream to the heat-receiving column. Compared to CDiC, DED could upgrade half of low-grade heat and reuse it as hot utility, saving considerable energy. 2.3. HPD HPD facilitates the upgraded waste heat by compressing overhead vapor instead of enhancing pressure of entire column to achieve heat upgrade and transfer, safely avoiding elaborate automatic control scheme that DED required [9]. It is expected that HPD will bring about 20-50% of energy savings [34]. HPD is classified into different patterns, such as vapor compression, thermal and mechanical recompression [15], aiming at escalating and utilizing heat capacity of stream from the column top to drive its bottom. Vapor compression (VC) and mechanical vapor recompression (MVR) are usually available for commercial purposes [15]. The schemes of VC and MVR are shown in Fig. 4. In VC, working fluid evaporated by top vapor of column discharges heat to the reboiler through a compressor to provide required work input and an expansion valve to close the cycle. As a remarkable feature, all units involved are external to the distillation process. Consequently, the column does not require major modifications except for possible adjustments in heat exchangers for changing utilities. On the other hand, MVR compresses overhead vapor of column to heat the reboiler, preventing the intermediate medium from cooling below the boiling point of the top products, thus saving one more heat exchanger and enjoying lower investments than VC. The coefficient of performance (COP) [15] is usually used to evaluate the performance of heat

3

ACCEPTED MANUSCRIPT pump. It is defined as discharging heat in condenser over electrical power required for upgrade the energy. The higher the COP of a heat pump, the better the performance. 2.4. Energy analysis based on standard oil In order to compare the consumption of utilities with different energy qualities, all the utilities depleted, including electrical power, recycling cooling water (CW) and steams with different bands, are converted into standard oil consumption in reference to P. R. China national standard [35], provided in Table 2. In this study, the price of standard oil was consulted to Shandong Qilu Petrochemical Engineering Co. Ltd, and rated as 0.30US$/kg. It should be noted that this price is fluctuating with international market. 2.5. Exergy analysis of distillation As an important thermodynamic property, exergy (𝐸) is defined as the maximum quantity of mechanical work one could obtain from a given heat source [36-37]. Clearly, exergy can be used as the measurement of energy quality, and energy quality degeneration leads to exergy loss. For a continuous flow, the specific exergy is defined as: 𝐸 = (𝐻 ‒ 𝐻0) ‒ 𝑇0(𝑆 ‒ 𝑆0)

(2) where H and S are the enthalpy and entropy of a process stream under operational conditions. 𝑇0 is the ambient standard temperature (e.g. atmosphere, river and ocean). According to Sun et al. [38], exergy losses in a distillation column are mainly distributed in reboiler, condenser and trays. In the reboiler, steams with different band are used as hot utility. If the transfer of an infinitesimal quantity of heat 𝑑𝑄 is considered, and the temperature of the reboiler is 𝑇𝑅, then the exergy of this heat is: 𝑑𝐸𝑄 = (1 ‒

𝑇0 𝑇𝑅

(3)

)𝑑𝑄

The received infinitesimal exergy of the heated stream (𝑇𝑆) is: 𝑇0 𝑑𝐸𝑄' = (1 ‒ )𝑑𝑄 𝑇𝑆

(4)

The total exergy loss at reboiler is determined by integrating the difference between Eqs. (3) and (4) from the inlet temperature 𝑇𝑖𝑛 to the outlet temperature 𝑇𝑜𝑢𝑡: 𝐸𝑙𝑜𝑠𝑠,𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 =

∫

(

𝑇𝑜𝑢𝑡

𝑇0

𝑇𝑖𝑛

𝑇𝑆 𝑇𝑅

‒

𝑇0

)

𝑑𝑄

(5)

If constant temperature difference in reboiler is postulated, exergy loss is derived out as follow: 𝐸𝑙𝑜𝑠𝑠,𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 =

(

𝑇0

‒

𝑇0

𝑇𝑆 𝑇𝑅

)

𝑄

(6)

Inducing the derivative of 𝐸𝑙𝑜𝑠𝑠,𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 with respect to 𝑇𝑅: 𝑑𝐸𝑙𝑜𝑠𝑠,𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 𝑑𝑇𝑅

=

𝑇0𝑄 𝑇𝑅2

>0

(7)

Eq. (7) tells that the larger the driving force, the larger exergy loss in reboiler, and the growth rate of exergy loss is inversely proportional to the square of utility steam temperature. In other words, the choice of proper steam band has great impact on the exergy loss during heat transfer.

4

ACCEPTED MANUSCRIPT Likewise in the condenser, cooling utility (𝑇𝐶) is required and the exergy loss is: 𝐸𝑙𝑜𝑠𝑠,𝑐𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟 =

∫

(

𝑇𝑜𝑢𝑡

𝑇0

𝑇𝑖𝑛

𝑇𝐶 𝑇𝑆

)

𝑇0

‒

𝑑𝑄

(8)

If CW is chosen as the cold utility, its temperature will not maintain constant during heat transfer. In this case, the 𝑇𝐶 should be replaced by the logarithmic mean of inlet and outlet temperature difference (𝑇𝐶): 𝐸𝑙𝑜𝑠𝑠,𝑐𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟 =

(

𝑇0

‒

)

𝑇0

𝑇𝐶 𝑇𝑆

𝑄

(9)

where: 𝑇𝐶 =

𝑇𝐶,𝑜𝑢𝑡 ‒ 𝑇𝐶,𝑖𝑛

(10)

𝑙𝑛(𝑇𝐶,𝑜𝑢𝑡 𝑇𝐶,𝑖𝑛)

Suppose that the heat loads of the condenser and reboiler are equal, the exergy loss in condenser is usually larger than that in reboilers [38]. The exergy loss of the trays is due to the irreversible mass and heat transfer during distillation process [38]. The vapor rising from i+1 tray encounters with the liquid falling from the i-1 tray, and the higher temperature of the former will help less volatile components of the latter to condensate and release heat on the ith tray. Simultaneously, absorbing the released heat, the more volatile components of the latter will enrich in the vapor phase. Assuming this mass and heat transfer process is under adiabatic condition, where ∑𝐻𝑖𝑛 = ∑𝐻𝑜𝑢𝑡, then the exergy loss on trays will only determined by the entropy of streams and can be obtained: 𝐸𝑙𝑜𝑠𝑠,𝑡𝑟𝑎𝑦 =‒ 𝑇0(𝑆𝑉,𝑖 + 1 + 𝑆𝐿,𝑖 + 1 ‒ 𝑆𝑉,𝑖 ‒ 𝑆𝐿,𝑖)

(11)

For a distillation column, the minimum work required for separation is the exergy difference between the product streams and the feed streams [39]. This minimum work is accessible only if the column has infinite theoretical stages and infinite heat exchangers [13]. Eq. (12) reveals this relationship: 𝑊𝑚𝑖𝑛 =

∑𝑚 𝐸 ‒ ∑𝑚 𝐸 𝑗 𝑗

𝑜𝑢𝑡

(12)

𝑖 𝑖

𝑖𝑛

For a CDiC, it requires high-grade heat 𝑄𝑅 at reboiler and discharges low-grade heat 𝑄𝐶 at condenser. The total work 𝑊𝑡 supplied from the utility system is given by: 𝑇0 𝑇0 𝑊𝑡 = 𝑄𝑅(1 ‒ ) ‒ 𝑄𝐶(1 ‒ ) 𝑇𝑅 𝑇𝐶

(13)

When a heat pump is involved into the distillation process, the utility supplies of electrical power, ' ' cold utility and steam are changed to 𝑊𝑒, 𝑄𝐶 and 𝑄𝑅. This way the total work supplied from utilities

should be rewritten as follow: 𝑊𝑡 = 𝑊𝑒 + 𝑄𝑅' (1 ‒

𝑇0 𝑇𝑅

) ‒ 𝑄𝐶' (1 ‒

𝑇0

) 𝑇𝐶

(14)

Based on Eqs. (12) ~ (14), the total exergy loss of the distillation column can be calculated as follow: 𝐸𝑥,𝑙𝑜𝑠𝑠 = 𝑊𝑡 ‒ 𝑊𝑚𝑖𝑛

(15)

2.6. Economic analysis

5

ACCEPTED MANUSCRIPT To evaluate the performance of an industrial plant, economic analysis has to be involved. The cost must be annualized to study the economics in terms of TAC (US$/year). TAC is composed by the capital investment (𝐶𝑐𝑎𝑝𝑖𝑡𝑎𝑙) in all units and operational expenditure (𝐶𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙) on utilities, expressed as: 𝑇𝐴𝐶 = 𝑓 ∙ 𝐶𝑐𝑎𝑝𝑖𝑡𝑎𝑙 + 𝐶𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙

(16)

in which, f is annualization factor defined by Smith [32]. Given a fixed period n (year) at a fixed rate of interest I, f is presented as: 𝑛

I ∙ (1 + I)

(17) (1 + 𝐼)𝑛 ‒ 1 In this study, to simplify the mathematical model of TAC, only distillation columns, condensers, 𝑓=

reboilers and compressors are considered into the capital investment. The operating costs contain thermal utilities (steam, CW) and electrical services. The annualization factor f=0.163 is considered, which corresponds to a fixed rate of interest of 10% over a plant life span of 10 years. The annualized operation period of plant is postulated to be 8000 hours. The capital investment of distillation columns is estimated from column diameter and height. For the diameter and height estimation the correlations of Rathore et al. [40] are used. The diameter of the column 𝐷𝑐(m) is estimated based on: 𝐷𝑐 =

[(

( )( )( )]

𝑇𝐷 1 4 1 (𝐷)(𝑅𝑅 + 1)(22.4) 𝜋𝑉 273 𝑃 3600

)

12

(18)

in which, V (m/s), D (kmol/h), RR, 𝑇𝐷(K), P (atm) are average gas flow rate, overhead product molar flow rate, reflux ratio, overhead temperature and pressure, respectively. The results of diameter are rounded according to the national standard of P. R. China [41]. The estimation of average gas flow rate V (m/s) is:

()

𝑉 = 0.761

1 𝑃

12

(19)

Height of the column is calculated based on: 𝐻𝑐 = 0.61

𝑁𝑡

() 𝜂

(20)

+ 4.27

where 𝑁𝑡 is the number of theoretical stages, and 𝜂 is stage efficiency. This study postulates 𝜂 = 0.8 in all stages [42]. Based on diameter and height of column, the capital investment of column 𝐶𝑐𝑜𝑙𝑢𝑚𝑛 (US$) is estimated as [42]: 0.802 1.55 𝐶𝑐𝑜𝑙𝑢𝑚𝑛 = (101.9𝐷1.066 𝑐 𝐻 𝑐 ∙ 3.18 + 4.7𝐷 𝑐 𝐻𝑐)

( ) 803 274

(21)

The capital investment of condensers and reboilers considers the heat transfer area (A) of heat exchangers, which are calculated as: 𝐴𝑐𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟 = 𝑄𝑐 (𝑈𝑐Δ𝑇)

(22)

𝐴𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 = 𝑄𝑟 (𝑈𝑟Δ𝑇)

(23)

where 𝑄𝑐, 𝑈𝑐, 𝑄𝑟, 𝑈𝑟, Δ𝑇 are condenser duty, condenser overall heat transfer coefficient, reboiler duty, reboiler overall heat transfer coefficient and logarithmic mean temperature difference, respectively. In

6

ACCEPTED MANUSCRIPT this study, the overall heat transfer coefficients of condenser and reboiler are postulated to be 600𝑊/( 𝑚2 ∙ 𝐾). With the knowledge of the heat transfer area, the capital investment (US$) of condensers and reboilers is conducted based on the following equations [42]:

( ) ( )

0.65 𝐶𝑐𝑜𝑛𝑑𝑒𝑛𝑠𝑒𝑟 = 101.3 ∙ 𝐴𝑐𝑜𝑛𝑑𝑠𝑒𝑛𝑠𝑒𝑟 ∙ 3.29 ∙ 0.65 𝐶𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 = 101.3 ∙ 𝐴𝑟𝑒𝑏𝑜𝑖𝑙𝑒𝑟 ∙ 3.29 ∙

803 274

803 274

(24) (25)

The capital investment (US$) of compressors 𝐶𝑐𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑜𝑟 is estimated by using the correlations of Couper et al. [43]: 𝐶𝑐𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑜𝑟 = 9476.17𝑊𝑐0.62

(26) where 𝑊𝑐(kW) is the electrical power of compressor. In this study, all the compressions are considered isentropic, with the isentropic efficiency postulated to be 0.8. In this study, the straight PBP, not including tax and depreciation, is used to evaluate retrofit proposals. The definition of the straight PBP is given by [44]: 𝑇𝐼𝐶 𝑃𝐵𝑃 = 𝑂𝐸𝐶𝐷𝑆 ‒ 𝑂𝐸𝐴𝐷𝑆

(27)

where TIC is the total investment cost. 𝑂𝐸𝐶𝐷𝑆 and 𝑂𝐸𝐴𝐷𝑆 represent operational expenditure of the CDS and ADS, respectively. In this study, TIC only focuses on the costly investment of the distillation column and compressor since they are much more expensive than other auxiliaries. In the calculation of operational expenditures, the reboilers are heated by steams with suitable band, leaving minimum temperature difference Δ𝑇𝑚𝑖𝑛 = 20℃. The cold utility uses CW, and its inlet and outlet temperature are 25°Cand 40°C, respectively. The utilities for heating, cooling, and electrical power are converted into equivalent standard oil consumption, using data in Table 2, and then multiply with unit price of standard oil, obtaining 𝐶𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙. 3. Design of CDSs The purpose of this section is to introduce state-of-the-art commercial CDSs to refine styrene from the dehydrogenated crude feed (Table 3), obtaining benzene and toluene as byproducts and recycling EB as reactant back to the reactor. The crude feed data considered is from a real styrene plant running in Northern China [45]. Based on Peng-Robinson model [5] predicting thermodynamic properties, the purification demand in Table 4 must be satisfied [45]. CDiSs 3 and 5, as suitable grassroots process design options, are referred to CDS-I and CDS-II, respectively, facilitating 3E analyses and exploring better retrofit alternatives. Fig. 5 presents the CDS-I as well as the operating parameters for each column. This scheme designates to separate benzene and toluene (as distillates) from EB and heavy components (as bottoms) initially in toluene/EB column C1. Its overhead product isfurther refined in benzene/toluene column C4, while bottom products of C1 are fed into the EB column C2, where recycle EB is obtained overhead. The styrene mixtures, leaving from bottom of C2, are refined in styrene column C3 with a flash tank connected to further strip styrene from tar. The first three columns C1, C2 and C3 are run under vacuum to suppress styrene polymerization. The operational parameters, including operating pressure, number of theoretical stages, and feed location etc., are carried out with reference to in-situ styrene plants [45]. Energy consumptions and exergy losses are illustrated in Table 5, while column configurations and economic evaluations are shown in Table 6. CDS-II (Fig. 6) removes benzene, toluene and EB initially in EB/SM column C5. The recycle EB

7

ACCEPTED MANUSCRIPT is obtained at bottom of toluene/EB column C6, and C6 top products enter benzene/toluene column C8 to obtain benzene and toluene. Crude styrene product, leaving from bottom of C5, is refined in styrene column C7. It is noteworthy that CDS-II requires only two columns C5 and C7 running sub-ambient, instead of three in CDS-I. Likewise, the relevant results are shown in Tables 5 and 6. The results show that TAC and exergy loss in CDS-II are slightly less than that in CDS-I, and the purification of benzene, EB and SM as well as the purified SM yield of the former are more competitive than the latter. In addition, economic evaluations indicate operating costs are more responsible for TACs than capital investments. Consequently, diminishing TAC through cutting down energy consumptions is critical. Despite energy consumptions of condensers measured in energy quantity are approximately equal to that of the reboilers (Table 5), the CW consumption is less responsible for operating cost when relevant heat loads are measured in standard oil. In other words, cutting down heat duty in reboilers is the priority for saving operating expenditures. In Fig. 7, on the basis of standard oil, heat duties of reboilers in CDSs are distributed in two pie charts. Obviously, columns C2 and C5, occupying ~65% of energy depletion in each scheme, are energy dominator in CDS-I and CDS-II, respectively. It is revealed the fact that separating EB from SM is the most difficult and energy intensive step among the whole distillation processes. This fact is also predictable from the close-boiling nature of the EB/SM system [1,46]. On the other hand, caused by high reflux ratio as well as large number of stages, C2 and C5 are the most expensive columns amongst C1 to C8. As a result, in order to lower operational expenses, it will be favored by applying energy efficient distillation operations on C2 and C5. Particularly in this case study, sequence 5 only slightly performs better than 3. It is still unpredictable on final TAC results when energy efficient technologies are introduced into the CDSs, since their TAC difference is only ~2.5%. In the following design of ADSs, DED and HPD configurations are applied on CDSs to explore better styrene distillation scheme. 4. Design of ADSs 4.1. Flowsheet description of ADSs The purpose of designing ADSs is to reduce energy consumption in CDSs. As mentioned in Fig. 7, revamping on C2 and C5 is advantageous in reduction of operating cost as well as TAC. As a consequence, DED and HPD configurations are applied on C2 and C5, respectively. For DED, operating pressure of HPC should be chosen carefully to prevent styrene polymerization. While in HPD, because overhead vapor of C2 and C5 is mainly EB, it is suitable to choose MVR instead of VC, circumventing the procedure of introducing a new working fluid. In this study, only MVR type of HPD is considered. Advanced distillation configurations ADS-I-DED and ADS-I-HPD are based on the CDS-I. ADSI-DED (Fig. 8) is obtained by dividing C2 into a HPC C10 and a LPC C11, using C10 overhead vapor to drive C11 bottom. Herein, the top pressure of C10 is identical to that of C2 to prevent styrene polymerization. The C11 runs at 6.7kPa, creating sufficient temperature difference for heat integration. In ADS-I-HPD (Fig. 9), the overhead vapor of EB column C15 is compressed to 80kPa to drive the reboiler. Likewise, ADS-II-DED (Fig. 10) revamps C5 into a HPC C18 running at 36kPa and a LPC C19 running at 6.7kPa, with operating pressure the same to that of C10 and C11. In ADS-II-HPD (Fig. 11), the overhead vapor of EB/SM column C23 is compressed to 60kPa to drive the reboiler.

8

ACCEPTED MANUSCRIPT 4.2. Energy, exergy and economic analysis of ADSs The calculation results of ADSs are shown in Tables 7 to 10. Energy consumptions of CDSs and ADSs, based on energy quantity and quality, are illustrated in Fig. 12. It is concluded that, in terms of energy quantity, scheme family II, including CDS-II, ADS-II-DED and ADS-II-HPD, requires much less energy than scheme family I. However, when these energy consumptions are converted to primary energy depletion, the former only slightly outperforms its counterpart. Taking CDS-I and CDS-II as respective benchmark, ADS-I-DED, ADS-I-HPD, ADS-II-DED and ADS-II-HPD could lower operating cost by 32.96%, 44.41%, 31.62% and 43.79%, respectively. Consequently, HPD configuration outperforms DED in decrease of energy depletion. This is attributed to high heat pump COP in ADS-I-HPD (11.41) and ADS-II-HPD (9.57). Chemical/Petrochemical plants normally have an annual maintenance period of 10-14 days for major repair and replacement [47]. The proposed retrofit from CDSs to ADSs is preferred to take place within this period to avoid production loss. It is worth noting that the diameters of all columns in CDSI are same to the corresponding columns in ADSs. In practice, this means that the revamping of existing plant is possible by reusing the existing columns. Specifically, for ADS-I-DED, a new column C10 is required to facilitate the DED configuration, while for ADS-I-HPD, a new compressor is needed to fulfill the retrofit. The corresponding PBPs for ADS-I-DED and ADS-I-HPD are 581 and 2459 hours, respectively. As for scheme family II, the retrofit of CDS-II to ADS-II-HPD no longer requiresadditional column but a new compressor. The ADS-II-DED has 2 columns (C18 and C19) of diameter 4.4 and 6.0m to form DED configuration (Table 10). The CDS-II has only one corresponding column C5 with diameter of 6.8m. The existing column C5 can be used for retrofitting the C19, leaving C18 as a new column needed to purchase. The PBPs for ADS-II-DED and ADS-II-HPD are 624 and 2912 hours, respectively. Notably, the attractive PBPs make such retrofit scenarios should be certainly considered for revamping exiting styrene distillation plants. The total exergy loss in CDSs and ADSs are shown in Fig. 13. The result indicates that the total exergy losses in ADSs are always higher than those in CDSs. Compared with CDS-I and CDS-II, ADS-I-DED, ADS-I-HPD, ADS-II-DED and ADS-II-HPD increase exergy loss by 27.36%, 58.35%, 25.07%, 63.53%, respectively. Exergy loss in scheme family II is slightly less than that in I, indicating the former is more thermodynamic reversible than the latter. As shown in Fig. 14, non-retrofit columns maintain approximately same exergy loss. In other words, the higher exergy loss is due to the pressure varying in revamping columns. For example, C2 is responsible for 313kW of exergy loss. C10, running under the same pressure as C2 but the feed capacity halved, has 159kW of exergy loss, which is half of C2, indicating the extensive property of exergy. While for C11, the lower pressure (6.7kPa) brings about three times of exergy loss (480kW) compared to that in C10 (159kW). The higher exergy loss in C11 is attributed to: (1) low-pressure operation reduces the separation difficulty, leading a lower minimum work 𝑊𝑚𝑖𝑛 required; (2) lowpressure operation makes C11 a larger temperature difference between overhead and bottom (28.3°C) than that in C10 (14.9°C), increasing total work 𝑊𝑡 supplied. Likewise, for retrofit schemes using HPD, when heat pump is introduced, the exergy loss of C15 achieves to 851kW, which is much higher than that of the original C2 (313kW), because heat pump consumes electrical power with high exergy compared to the low-quality steam utility. Therefore, although energy efficient distillation could conclusively cut down considerable energy consumption, it is also responsible for more exergy loss as punishment. The TAC comparison of CDSs and ADSs is shown in Fig. 15. Because annualized capital costs

9

ACCEPTED MANUSCRIPT are much less than annualized operating costs, additional capital investment in ADSs becomes less responsible for TAC. Compared with corresponding CDS, ADS-I-DED, ADS-I-HPD, ADS-II-DED and ADS-II-HPD can save TAC by 35.62%, 41.47%, 34.09% and 40.79%, respectively. In this case, the TAC requirement for unit styrene production in CDS-I, CDS-II, ADS-I-DED, ADS-II-DED, ADSI-HPD and ADS-II-HPD are 0.0290US$/kg, 0.0282US$/kg, 0.0187US$/kg, 0.0186US$/kg, 0.0169US$/kg and 0.0167US$/kg, respectively (Fig. 15). It is concluded that the ADS-II-HPD is the most competitive scheme amongst all these proposed styrene distillation schemes. Because the TAC will vary under various economic models and international market, more detailed economic evaluation should be considered during engineering design period. 4.3. Sensitivity analysis and optimization In this study, sensitivity analysis is used to determine the optimal operational parameters of columns in styrene distillation process [38]. Herein, the effect on revamping columns is considered of the operating pressure and the number of theoretical stages, which are responsible for the significant increase of exergy loss and potentially to be diminished. Fig. 16 shows the effect of pressure on exergy loss distribution on trays of C2. In this case, other operational parameters are maintained while the C2 top pressure is changed from 36kPa, the C2’s original pressure, to 6.7kPa, the C11’s operating pressure. As a result, with the decrease in C2’s pressure, the exergy loss on each tray is increased. Exergy loss on each tray in rectifying section is higher than that in stripping section, touching local maxima at feed stage. In other words, stripping trays are more efficient than rectifying cousins and feed mixing with inner column streams causes more thermodynamic irreversibility. Fig. 17 shows the monotonous relationship of exergy loss on trays in C2 and column top-bottom temperature difference versus operational pressure. The exergy loss on trays increases from 349kW to 973kW when C2 top pressure is descended from 36kPa to 6.7kPa. This is attributed to the ascending pressure decreases the top-bottom temperature difference. For example, when top pressure is 36kPa, the temperature difference is 14.9°C, while when the former decreases to 6.7kPa, the latter increases to 28.3°C. Consequently, it is concluded that the exergy loss increase is largely dependent on top-bottom temperature difference. In other words, increasing pressure benefits decreasing tray exergy losses, explaining the higher exergy loss in LPCs C11 and C19. Because the operating pressure of HPCs C2, C10 and C18 is restricted by the styrene stability, in the aim of minimum exergy losses, 36kPa is an optimum. This is naturally induced that, in order to minimize exergy losses, maintain styrene stability and construct DED setup, 6.7kPa of overhead pressure in C11 and C19 is also an optimum. Likewise, the effect of number of theoretical stages has influence on exergy loss. In order to lower exergy loss on each tray, more stages are required in a column [38]. Taking C11 (a representative of LPC in DED) as an example, because rectifying trays loss more exergy than stripping ones, an increment of 10 theoretical stages is specified on rectifying section of C11 stepwise. As shown in Fig. 18, exergy loss on each tray in rectifying section decreases gradually. However, the total exergy loss on trays increases from 480kW to 483kW when the number of theoretical trays varies from 92 to 142 (Fig. 19). On the other hand, the increase in number of theoretical trays brings in only a slight decrease in reboiler energy consumption (Fig. 19), at the cost of higher capital investment as well as TAC. Therefore, the best strategy is to maintain the current number of theoretical stages. This conclusion can be extended to other LPCs. For C15 and C23, as columns using MVR setup, their operational parameters are identical to C2

10

ACCEPTED MANUSCRIPT and C5. Based on Eqs. (14) and (15), higher exergy in electrical power is the main reason for their higher exergy loss (Fig. 14). In other words, varying column configuration has less impact on decreasing exergy losses. However, it is noteworthy that higher exergy loss is not necessary correspond to high TACs. In order to lower retrofit difficulty, the best strategy is to maintain the current pressure and number of theoretical stages. 5. Conclusions Significant energy savings as well as TAC reductions are obtained in the styrene distillation schemes by the aid of DED and HPD. Compared to the corresponding CDS, ADSs with DED and HPD can achieve energy savings up to 30% and 40%, respectively. Due to a synergistic effect of considerably reducing energy costs and highly reusing most of the equipment, the PBPs of retrofit designs are very attractive. Specifically, the proposed ADS-I-DED, ADS-I-HPD, ADS-II-DED and ADS-II-HPD can save TAC by 35.62%, 41.47%, 34.09% and 40.79% compared to the CDSs, and decrease unit styrene yield cost to 0.0187US$/kg, 0.0186US$/kg, 0.0169US$/kg and 0.0167US$/kg, respectively. The results show both DED and HPD are attractive options in industrial implementation, although the latter is slightly better than the former. This study also shows that although the ADSs can reduce the energy requirements greatly, they can also cause higher exergy losses than the CDSs. This study puts forward several retrofit strategies for styrene manufacture. The proposed schemes are suitable for either grassroots process design of a new plant or retrofit of an existing one. Moreover, the revamping procedure in this paper is general to be extended to other distillation systems. Acknowledgement The authors are grateful for the industrial consultant of Shandong Qilu Petrochemical Engineering Co. Ltd, Zibo, Shandong, China. Nomenclature Acronyms ABS

acrylonitrile-butadiene-styrene

ADS

advanced distillation scheme

CDiC

conventionaldistillation column

CDiS

conventional distillation sequence

CDS

conventional distillation scheme

COP

coefficient of performance

CW

coolingwater

DED

double-effect distillation

DNBP

2-sec-butyl-4,6-dinitrophenol

DWC

dividingwall column

EB

ethylbenzene

ED

extractive distillation

HIDiC

heat-integrated distillation column

HPC

high-pressure column

HPD

heat pump distillation

11

ACCEPTED MANUSCRIPT LPC

low-pressure column

MVR

mechanical vapor recompression

NBP

normal boiling point

PBP

payback period

PS

polystyrene

SAN

styrene-acrylonitrile

SBC

styrenic block copolymers

SBL

styrene-butadiene latex

SBR

styrene-butadiene rubber

SM

styrene monomer

TAC

total annualized cost

VC

vapor compression

3E analyses

energy, exergy and economic analyses

Roman letters A

heat transfer area

C

cost

D

overhead product molar flow rate

Dc

diameter of the column

E

exergy

f

annualization factor

H

enthalpy

Hc

height of the column

m

mass flow rate

Nt

number of theoretical stages

OE

operationalexpenditure

P

pressure

Q

heat

RR

reflux ratio

S

entropy

T

temperature

TIC

total investment cost

U

overall heat transfer coefficient

V

average gas flow rate

W

work or power

Greek letters η

stage efficiency

References [1] Jongmans MTG. Energy efficient trace removal by extractive distillation. PhD Thesis. Eindhoven University of Technology. 2012.

12

ACCEPTED MANUSCRIPT [2] Nederlof C. Catalytic dehydrogenations of ethylbenzene to styrene. PhD Thesis. Delft University of Technology. 2012. [3] Liao SJ, Chen T, Miao CX, Yang WM, Xie ZK, Chen QL. Effect of TiO2 on the structure and catalytic behavior of iron–potassium oxide catalyst for dehydrogenation of ethylbenzene to styrene. Catalysis Communications. 2008; 9:1817-1821. [4] Ba H, Podila S, Liu Y, Mu X, Nhut JM, Papaefthimiou V, Zafeiratos S, Granger P, Pham-Huu C. Nanodiamond decorated few-layer graphene composite as an efficient metal-free dehydrogenation catalyst for styrene production. Catalysis Today.2015; 249:167-175. [5] Luyben WL. Design and control of the styrene process. Industrial & Engineering Chemistry Research. 2011; 50: 1231-1246. [6] Welch VA. Cascade reboiling of ethylbenzene/styrene columns. US Patent. US 6171449 B1. 2001. [7] Parra-Santiago JJ, Guerrero-Fajardo CA, Sodre JR. Distillation process optimization for styrene production from a styrene-benzene-toluene system in a Petlyuk column. Chemical Engineering and Processing: Process Intensification.2015; 98: 106-111. [8] Kiss AA. Advanced Distillation Technologies: Design, Control and Applications. John Wiley & Sons Ltd. 2013. [9] Jana AK. Heat integrated distillation operation. Applied Energy. 2010; 87: 1477-1494. [10] Halvorsen IJ, Skogestad S. Energy efficient distillation. Journal of Natural Gas Science and Engineering. 2011; 3: 571-580. [11] Cui C, Yin H, Yang J, Wei D, Sun J, Guo C. Selecting suitable energy-saving distillation schemes: Making quick decisions. Chemical Engineering and Processing: Process Intensification.2016; 107: 138-150. [12] Cui C, Sun J. Coupling design of interunit heat integration in an industrial crude distillation plant using pinch analysis. Applied Thermal Engineering. 2017; 117: 145-154. [13] Sun J, Wang F, Ma T, Gao H, Wu P, Liu L. Energy and exergy analysis of a five-column methanol distillation scheme. Energy. 2012; 45: 696-703. [14] Cui C, Li X, Sui H, Sun J. Optimization of coal-based methanol distillation scheme using process superstructure method to maximize energy efficiency. Energy. 2017; 119: 110-120. [15] Kiss AA, Landaeta SJF, Ferreira CAI. Towards energy efficient distillation technologies – Making the right choice. Energy.2012; 47:531-542. [16] Chua KJ, Chou SK, Yang WM. Advances in heat pump systems: A review. Applied Energy. 2010; 87: 3611-3624. [17] Jana AK. Advances in heat pump assisted distillation column: A review. Energy Conversion and Management. 2014; 77: 287-297. [18] Nakaiwa M, Huang K, Endo A, Ohmori T, Akiya T, Takamatsu T. Internally heat-integrated distillation columns: A review. Chemical Engineering Research and Design.2003; 81: 162-177. [19] Suphanit B. Design of internally heat-integrated distillation column (HIDiC): Uniform heat transfer area versus uniform heat distribution. Energy. 2010; 35: 1505-1514. [20] Suphanit B. Optimal heat distribution in the internally heat-integrated distillation column (HIDiC). Energy. 2011; 36: 4171-4181. [21] Harwardt A, Marquardt W. Heat-integrated distillation columns: Vapor recompression or internal heat integration? AIChE Journal. 2012; 58: 3740-3750. [22] Asprion N, Kaibel G. Dividing wall columns: Fundamentals and recent advances. Chemical Engineering and Processing: Process Intensification. 2010; 49: 139-146.

13

ACCEPTED MANUSCRIPT [23] Dejanovic I, Matijasevic L, Olujic Z. Dividing wall column – A breakthrough towards sustainable distilling. Chemical Engineering and Processing: Process Intensification.2010; 49: 559-580. [24] Yildirim O, Kiss AA, Kenig EY. Dividing wall columns in chemical process industry: A review on current activities. Separation and Purification Technology.2011; 80: 403-417. [25] StaakD, Grutzner T, Schwegler B, Roederer D. Dividing wall column for industrial multi purpose use. Chemical Engineering and Processing: Process Intensification.2010; 49: 559-580. [26] Sun J, Dai L, Shi M, Gao H, Cao X, Liu G. Further optimization of a parallel double-effect organosilicon distillation scheme through exergy analysis. Energy. 2014; 69: 370-377. [27] Jongmans MTG, Hermens E, Raijmakers M, Maassen JIW, Schuur B, de Haan AB. Conceptual process design of extractive distillation processes for ethylbenzene/styrene separation. Chemical Engineering Research and Design. 2012; 90: 2086-2100. [28] Jongmans MTG, Maassen JIW, Luijks AJ, Schuur B, de Haan AB. Isobaric low-pressure vaporliquid equilibrium data for ethylbenzene+ styrene + sulfolane and the three constituent binary systems. Journal of Chemical & Engineering Data.2011; 56: 3510-3517. [29] Jongmans MTG, Schuur B, de Haan AB. Ionic liquid screening for ethylbenzene/styrene separation by extractive distillation. Industrial & Engineering Chemistry Research.2011; 50: 1080010810. [30] Errico M, Rong BG. Modified simple column configurations for quaternary distillations. Computers & Chemical Engineering. 2012; 36: 160-173. [31] Errico M, Rong BG, Torres-Ortega CE, Segovia-Hernandez JG. The importance of the sequential synthesis methodology in the optimal distillation sequences design. Computers & Chemical Engineering. 2014; 62: 1-9. [32] Smith R. Chemical Process Design and Integration. 2nd edition. John Wiley &Sons, Ltd. 2005. [33] Thompson RW, King CJ. Systematic synthesis of separation schemes. AIChE Journal. 1972; 18: 941-948. [34] van de Bor DM, Ferreira CAI, Kiss AA. Low grade waste heat recovery using heat pumps and power cycles. Energy. 2015; 89: 864-873. [35] Standard for calculation of energy consumption in petrochemical engineering design. P. R. China National Standard: GB/T50441-2007. [36]

Smith

JM,

Van

Ness

HC,

Abbott

MM.

Introduction

to

Chemical

Engineering

Thermodynamics.5thedition. McGraw-Hill. 1996. [37] Ammar Y, Joyce S, Norman R, Wang Y, Roskilly AP. Low grade thermal energy sources and uses from the process industry in the UK. Applied Energy. 2012; 89:3-20. [38] Sun J, Wang F, Ma T, Gao H, Liu Y, Cai F. Exergy analysis of a parallel double-effect organosilicon monomer distillation scheme. Energy. 2012; 47: 498-504. [39] Ognisty TP. Analyze distillation columns with thermodynamics. Chemical Engineering Progress.1995; 91: 40-46. [40] Rathore RNS, Vanwormer KA, Powers GJ. Synthesis of distillation systems with energy integration. AIChE Journal.1974; 20:940-950. [41] P. R. China National Standard: JB-1153-73. [42] An W. Synthesis of complex distillation systems based on stochastic optimization. PhD Thesis. Tianjin University. 2003. [43] Couper JR, Penney WR, Fair JR,Walas SM. Chemical Process Equipment: Selection and Design. 3rd edition. Elsevier. 2012.

14

ACCEPTED MANUSCRIPT [44] Feng X, Berntsson T. Critical COP for an economically feasible industrial heat-pump application. Applied Thermal Engineering.1997; 17: 93-101. [45] Engineering Package – Styrene Distillation Unit. Shandong Qilu Petrochemical Engineering Co. Ltd. 2015. (In Chinese) [46] Cui C, Li X, Sui H, Sun J. Quick decision-making for close-boiling distillation schemes. Industrial & Engineering Chemistry Research. 2017; DOI: 10.1021/acs.iecr.7b00935 [47] Long NVD, Lee M. Optimal retrofit of a side stream column to a dividing wall column for energy efficiency maximization. Chemical Engineering Research and Design. 2013; 91: 2291-2298.

15

ACCEPTED MANUSCRIPT

Fig. 1. Styrene production process using EB dehydrogenation technology

Fig. 2. CDiSs for four-product separation

Fig. 3. CDiC and DED

1

ACCEPTED MANUSCRIPT

Fig. 4. VC and MVR

Fig. 5. Conventional design of CDS-I

Fig. 6. Conventional design of CDS-II

2

ACCEPTED MANUSCRIPT

Fig. 7. Energy consumption (based on standard oil) distribution of CDS-I and CDS-II

Fig. 8. Advanced design of ADS-I-DED

Fig. 9. Advanced design of ADS-I-HPD

3

ACCEPTED MANUSCRIPT

Fig. 10. Advanced design of ADS-II-DED

Fig. 11. Advanced design of ADS-II-HPD

Fig. 12. Energy consumption in quantity and standard oil

4

ACCEPTED MANUSCRIPT

Fig. 13. Total exergy loss of CDSs and ADSs

Fig. 14. Exergy loss distribution of CDSs and ADSs

Fig. 15. Total annualized cost and TAC requirement for unit styrene of CDSs and ADSs

5

ACCEPTED MANUSCRIPT

Fig. 16. The effect of pressure on exergy loss distribution on trays of column C2

Fig. 17. The effect of pressure on total exergy loss on trays and top-bottom temperature difference of column C2

6

ACCEPTED MANUSCRIPT

Fig. 18. The effect of number of theoretical stages on exergy loss distribution on trays of column C11

Fig. 19. The effect of number of theoretical stages on reboiler energy consumption and exergy loss of column C11

7

ACCEPTED MANUSCRIPT Highlights 1.

Energy efficient distillation technologies are applied on EB/SM column

2.

Innovative retrofit scenarios with attractive payback periods are proposed

3.

Significant reduction in energy consumption as well as total annualized cost

4.

Optimization of advanced distillation schemes through sensitivity analysis

ACCEPTED MANUSCRIPT Table 1. Molecular structures and NBPs of each component Component

Molecular structure

NBP (°C)

Benzene

80.1

Toluene

110.6

EB

136.2

SM

145.7

Styrene tar

-

>145.7

Table 2. The conversion of utilities into standard oil Utility type

Remark

Utility unit

Equivalent standard oil (kg)

Electrical power

-----

kWh

0.26

Recycling CW

-----

ton

0.10

Band 10.0MpaG steam

7.0MpaG ≤ P

ton

92

Band 5.0MpaG steam

4.5MpaG ≤ P < 7.0𝑀𝑃𝑎𝐺

ton

90

Band 3.5MpaG steam

3.0MpaG ≤ P < 4.5𝑀𝑃𝑎𝐺

ton

88

Band 2.5MpaG steam

2.0MpaG ≤ P < 3.0𝑀𝑃𝑎𝐺

ton

85

Band 1.5MpaG steam

1.2MpaG ≤ P < 2.0𝑀𝑃𝑎𝐺

ton

80

Band 1.0MpaG steam

0.8MpaG ≤ P < 1.2𝑀𝑃𝑎𝐺

ton

76

Band 0.7MpaG steam

0.6MpaG ≤ P < 0.8𝑀𝑃𝑎𝐺

ton

72

Band 0.3MpaG steam

0.3MpaG ≤ P < 0.6𝑀𝑃𝑎𝐺

ton

66

Band <0.3MpaG steam

-----

ton

55

Table 3. Dehydrogenated crude feed data Component

Mass flow rate (kg/h)

Component

Mass flow rate (kg/h)

Benzene

156.929

Alpha-methylstyrene

265.821

Toluene

597.781

Isopropylbenzene

0.641

EB

15327.930

N-propylbenzene

4.664

SM

26123.890

C10H14

42.024

DNBP

162.576

C13H12

71.360

P-xylene

1.861

C14H14

207.551

M-xylene

4.206

C18H14

393.408

O-xylene

0.303

Temperature

83.4°C

Pressure

24kPa

Flow rate

43419.7kg/h

1

ACCEPTED MANUSCRIPT Table 4. Purification demand on products Product

Purification demand (wt%)

SM

99.80

Recycle EB

98.00

Toluene

99.00

Table 5. Energy consumptions and exergy losses of CDS-I and CDS-II Conventional distillation scheme

CDS-I

CDS-II

Total condenser duty (kW)

25,164

23,193

Standard oil (10 kg/s)

37.10

34.18

Total reboiler duty (kW)

25,216

23,796

Standard oil (10 kg/s)

641.91

628.21

Total standard oil (10 kg/s)

679.01

662.39

Total exergy loss (kW)

1,022

980

-3

-3

-3

Table 6. Column configurations and economic evaluations of CDS-I and CDS-II Conventional distillation scheme

CDS-I

CDS-II

Column

C1

C2

C3

C4

C5

C6

C7

C8

Diameter (m)

3.0

6.0

4.6

0.3

6.8

1.4

4.4

0.3

43.92

74.42

31.72

30.96

74.42

43.92

31.72

30.96

Condenser area (m2)

234.55

380.90

217.88

1.66

506.87

28.84

197.84

1.41

Reboiler area (m )

311.24

1,328.59

454.60

6.87

1,374.93

180.41

421.24

6.43

Column cost (US$)

66,941

219,793

81,944

4,194

252,341

29,252

78,058

4,194

Condenser cost (US$)

33,917

46,483

32,330

1,358

55,970

8,685

30,365

1,221

Reboiler cost (US$)

40,765

104,707

52,147

3,418

107,066

28,598

49,627

3,274

Height (m)

2

Total capital cost (US$)

687,998

648,651

Total operational cost (US$/year)

5,866,646

5,723,050

Total annualized cost (US$/year)

5,978,790

5,828,780

Table 7. Energy consumptions and exergy losses of advanced distillation scheme I Advanced distillation scheme

ADS-I-DED

ADS-I-HPD

Total condenser duty (kW)

17,001

11,591

Standard oil (10-3kg/s)

25.06

17.10

Total reboiler duty (kW)

16,933

10,589

Standard oil (10-3kg/s)

430.12

267.79

Total electrical power (kW)

0

1,282

Standard oil (10 kg/s)

0

92.59

Total standard oil (10 kg/s)

455.18

377.48

Total exergy loss (kW)

1,291

1,619

-3

-3

Table 8. Column configurations and economic evaluations ofadvanced distillation scheme I Advanced distillation scheme

ADS-I-DED

ADS-I-HPD

Column

C9

C10

C11

C12

C13

C14

C15

C16

C17

Diameter (m)

3.0

4.4

6.0

4.6

0.3

3.0

6.0

4.6

0.3

2

ACCEPTED MANUSCRIPT Height (m)

43.92

74.42

74.42

31.72

30.96

43.92

74.42

31.72

30.96

Condenser area (m )

234.55

2.38

567.3

196.66

1.66

234.55

51.26

217.87

1.66

Reboiler area (m )

311.24

669.29

0

423.7

6.87

311.24

109.7

454.6

6.87

Column cost (US$)

66,941

156,265

219,793

78,058

4,194

66,941

219,793

81,944

4,194

Condenser cost (US$)

33,917

1,716

60,221

30,247

1,358

33,919

12,622

32,329

1,358

Reboiler cost (US$)

40,765

67,053

0

49,815

3,418

40,765

20,697

52,147

3,418

0

0

0

0

0

0

800,810

0

0

2

2

Compressor cost (US$) Total capital cost (US$)

813,761

1,370,938

Total operational cost (US$/year)

3,716,237

3,261,427

Total annualized cost (US$/year)

3,848,880

3,484,890

Table 9. Energy consumptions and exergy losses of advanced distillation scheme II Advanced distillation scheme

ADS-II-DED

ADS-II-HPD

Total condenser duty (kW)

15,196

9,555

Standard oil (10-3kg/s)

22.40

14.08

Total reboiler duty (kW)

15,800

8,577

Standard oil (10 kg/s)

430.56

243.96

Total electrical power (kW)

0

1,582

Standard oil (10 kg/s)

0

114.26

Total standard oil (10 kg/s)

452.96

372.30

Total exergy loss (kW)

1,226

1,603

-3

-3

-3

Table 10. Column configurations and economic evaluations ofadvanced distillation scheme II Advanced distillation scheme

ADS-II-DED

ADS-II-HPD

Column

C18

C19

C20

C21

C22

C23

C24

C25

C26

Diameter (m)

4.4

6.0

1.4

4.4

0.3

6.8

1.4

4.4

0.3

Height (m)

74.42

74.42

43.92

31.72

30.96

74.42

43.92

31.72

30.96

Condenser area (m2)

0.20

662.4

28.09

197.84

1.41

74.28

32.47

197.84

1.35

Reboiler area (m2)

702.71

0.00

181.97

425.52

6.43

113.83

173.26

421.24

6.43

Column cost (US$)

156,265

219,793

29,252

78,058

4,194

252,341

33,805

78,058

4,194

343

66,604

8,537

30,365

1,221

16,063

9,381

30,365

1,187

69,211

0

28,759

49,954

3,274

21,200

27,856

49,627

3,274

0

0

0

0

0

912,307

0

0

0

Condenser cost (US$) Reboiler cost (US$) Compressor cost (US$) Total capital cost (US$)

745,830

1,439,658

Total operational cost (US$/year)

3,720,038

3,216,672

Total annualized cost (US$/year)

3,841,608

3,451,336

3