Effect of normal paraffins separation from naphtha on reaction kinetics for olefins and aromatics production

Computers and Chemical Engineering 74 (2015) 128–143

Contents lists available at ScienceDirect

Computers and Chemical Engineering journal homepage: www.elsevier.com/locate/compchemeng

Effect of normal paraffins separation from naphtha on reaction kinetics for olefins and aromatics production Truong Xuan Do a , Young-il Lim a,∗ , Jinsuk Lee b , Woojo Lee b a b

CoSPE, Department of Chemical Engineering, Hankyong National University, Jungangno 327, Anseong-si, Gyonggi-do 456-749, Republic of Korea Samsung Total Petrochemicals Co., Ltd., 411-1 Dokgot-Ri, Daesan-Eup, Seosan-Si, Chungnam 356-711, Republic of Korea

a r t i c l e

a b s t r a c t

i n f o

Article history: Received 31 May 2014 Received in revised form 22 December 2014 Accepted 1 January 2015 Available online 9 January 2015 Keywords: Simulated moving-bed (SMB) Reaction kinetics Naphtha Thermal cracking Catalytic reforming Petrochemical complex (PCC)

The objective of this study is to investigate the effect of the normal paraffins (n-paraffins) separation by the simulated moving-bed (SMB) on reaction kinetics of the naphtha thermal cracking (NTC) and catalytic reforming (NCR) for the olefins and aromatics production, respectively. First a process simulation of the SMB unit integrated to the petrochemical complex (PCC) was performed. Chemical reaction kinetics of NTC and NCR were proposed and validated. And a retrofit PCC with the SMB unit (rPCC) was compared to a conventional PCC (cPCC) in terms of products composition, flow rate, and energy consumption. It was found that olefins and aromatics yields of NTC and NCR can increase by 14 wt% (41 kt/yr) and 11 wt% (127 kt/yr) for the naphtha capacity of 2475 kt/yr, respectively. However, the total energy consumption of rPCC increased by about 67.8 MW (or 25%) because of the desorbent recovery in the SMB unit. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction The paradigm of separations is changing. Factors like energy consumption, process performance or process environmental cleanness have become more and more of a major issue (Sá Gomes et al., 2009). The simulated moving-bed (SMB) process has been successfully applied to petrochemical separations such as p-xylene separation from its C8 isomers, normal paraffins (n-paraffins) from branched and cyclic hydrocarbons, and olefins separation from paraffins (Sá Gomes et al., 2009). The MaxEne process, which is used to separate n-paraffins from naphtha, was developed by UOP to integrate refining and petrochemical facilities (UOP). MaxEne allows an increase of ethylene productivity with existing naphtha cracking units without making major changes to the existing

Abbreviations: ATC, annualized total cost; BTX, benzene, toluene, and xylene; CCR, continuous catalyst regenerative; cPCC, conventional petrochemical complex; CR, cyclic regenerative; H-HC, heavy hydrocarbons; L-HC, light hydrocarbons; NCR, naphtha catalytic reforming; NHT, naphtha hydro-treating; NTC, naphtha thermal cracking; n-paraffins, normal-paraffins; non n-paraffins, iso-paraffins, naphthenes, and aromatics; i-paraffins, iso-paraffins; PCC, petrochemical complex; PFD, process flow diagram; rPCC, retrofit petrochemical complex with SMB unit; SMB, simulated moving-bed; SR, semi-regenerative. ∗ Corresponding author. Tel.: +82 31 670 5207; fax: +82 31 670 5209. E-mail addresses: [email protected] (T.X. Do), [email protected] (Y.-i. Lim), [email protected] (J. Lee), [email protected] (W. Lee). http://dx.doi.org/10.1016/j.compchemeng.2015.01.002 0098-1354/© 2015 Elsevier Ltd. All rights reserved.

equipment. n-Paraffins are preferred to be fed into the naphtha cracker to raise the yield of light olefins (ethylene and propylene). The catalytic reforming yield increases significantly when n-paraffins are removed from the feed. Therefore, the separated n- and non n-paraffins can be effectively applied to raw materials for olefins and aromatics production, respectively (Chang et al., 2005; Liu et al., 2009). n-Butane was used as a desorbent in the simulated moving-bed (SMB) unit for the separation of n-paraffins from C5 –C10 hydrocarbons (Liu et al., 2009). It may be of interest to evaluate technical feasibility as well as economical profitability for the integration of the SMB unit into an existing petrochemical plant. Naphtha is a fraction of petroleum which typically constitutes 15–30 wt% of crude oil and boils between 30 ◦ C and 200 ◦ C. This complex mixture consists of hydrocarbon molecules with 5–12 carbon atoms, and it mainly includes paraffins, olefins, naphthenes, and aromatics (Rahimpour et al., 2013). Naphtha can be subdivided into 30–90 ◦ C light naphtha (C5 and C6 ), 90–150 ◦ C medium naphtha (C7 –C9 ), and 150–200 ◦ C heavy naphtha (C9 –C12 ). In commercial practice, the medium naphtha is the most-preferred feed for catalytic reforming (Rodríguez and Ancheyta, 2011), while the light naphtha is suitable for thermal cracking (Belohlav et al., 2003). Basic petrochemicals such as ethylene, propylene, butadiene and aromatics like benzene, toluene, and xylene (BTX) are the main products of the petrochemical industry (Ren et al., 2009). Currently, most of them are produced via conventional routes such as the

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Nomenclature cp d D Ea F −H i j k0 N Ns p Pbutane Q R r s T z

heat capacity (kJ/(kmol K)) reactor diameter (m) distillate flow rate (t/h) activation energy (kJ/mol) molar flow rate (kmol/s) heat of reaction (kJ/mol) component index reaction index  pre-exponential factor (kmol/m3 s Pa ˛i ) number of reactants theoretical number of stages partial pressure (Pa) n-butane purity (%) heat flow rate (kJ/s) gas constant, J/(mol K) or molar reflux ratio reaction rate (kmol/(m3 s)) stoichiometric factor temperature (K) reactor length (m)

Greek letters ˛i exponent of ith reactant tray efficiency 

reforming and cracking of naphtha, which are vital processes in petrochemical refineries (Ren et al., 2009). The significance of these industrial processes has induced researchers to investigate different aspects of both naphtha cracking and reforming, intensively (Rahimpour et al., 2013). Naphtha thermal cracking (NTC) of hydrocarbons is the most important source for the production of olefins which are the main feedstock of the polymers industries (Niaei et al., 2004). Steam has been traditionally used to partially remove coke which causes several problems along the reactor such as low heat transfer and high pressure drop. Naphtha is usually the mixture of manifold hydrocarbons, and each hydrocarbon has complex reactions in the cracking process. The cracking performance and products yield depend on the composition and the carbon number of n-paraffins, iso-paraffins (i-paraffins), naphthenes and aromatics (Liu et al., 2009). In the typical reaction condition, n-paraffins in naphtha contribute most to the ethylene in the products. i-Paraffins are the main sources of propylene. Naphthenes mainly produce butadiene, and aromatics can hardly produce olefins (Liu et al., 2009). The naphtha catalytic reforming (NCR) unit occupies a key position in refineries to obtain high octane gasoline and BTX components which are the basic substances of petrochemical industries (Gyngazova et al., 2011; Iranshahi et al., 2013; Rahimpour et al., 2013; Rodríguez and Ancheyta, 2011). Hydrogen is a valuable by-product of the naphtha catalytic reforming process, which in most refineries is used for the hydrocracking, hydrotreating, and other hydrogen-consuming processes. The NCR is accomplished mainly by converting of n-paraffins and naphthenes in naphtha to i-paraffins and aromatics over bifunctional catalysts. A large number of reactions take place during the catalytic reforming process such as alkylcyclohexane, dehydrogenation, dehydrocyclization to alkylcyclopentane, aromatization, isomerization, transalkylation, hydrodealkylation, hydrocracking, and coke formation (Rodríguez and Ancheyta, 2011). Industrial catalysts used in catalytic reforming units consist of -Al2 O3 support. Some metals such as Pt, Re, Ge, Ir, and Sn promote dehydrogenation reactions, and acid function such as chlorine increases isomerization and dehydrocyclization reactions (Rahimpour et al., 2013). All reactions are desirable except

129

the hydrocracking because it converts valuable compounds to light gases. Reaction kinetics modeling of NTC and NCR for olefins and aromatics productions, respectively, is an attractive tool for feedstock selecting and mixing, production planning, optimal reactor controlling, reactor designing and revamping, and process modification (Belohlav et al., 2003). The large number of reactions and hundreds of components taking part in the actual reaction system make this a rather complex problem. An effective reaction kinetics model must properly represent all major types of reactions and at least account for the most important classes of chemical species present in the reaction mixture. Recently, various kinetics models for NTC (Belohlav et al., 2003; Jia et al., 2009; Keyvanloo et al., 2012; Niaei et al., 2004; Sadrameli and Green, 2005; Seifzadeh Haghighi et al., 2013) and NCR (Gyngazova et al., 2011; Hou et al., 2006; Iranshahi et al., 2013; Meidanshahi et al., 2011; Rahimpour et al., 2013; Rodríguez and Ancheyta, 2011; Zagoruiko et al., 2014) have been developed. There are several approaches for reaction kinetics modeling of the NTC process: empirical (Jia et al., 2009), semi-empirical (Sadrameli and Green, 2005), free radical (Keyvanloo et al., 2012; Niaei et al., 2004), and molecular mechanism (Belohlav et al., 2003; Seifzadeh Haghighi et al., 2013). Jia et al. (2009) evaluated the desirability of the Wiehe’s model (Wiehe, 1993) in describing the compositional changes of various oils in in situ processes. Sadrameli and Green (2005) applied an analytical semi-empirical model of pyrolysis to NTC. Niaei et al. (2004) used a rigorous kinetics model based on free-radical chain reactions for the decomposition of the naphtha feed. Keyvanloo et al. (2012) developed a semimechanistic kinetics model based on free radical chain reactions containing 4 pseudo components and 96 reactions. Belohlav et al. (2003) involved free radical reactions and a set of pure and formal molecular reactions as a kinetics model of NTC. Seifzadeh Haghighi et al. (2013) proposed a set of 20 molecular reactions for naphtha pyrolysis, which contained one primary decomposition and 19 secondary reactions. Taskar and Riggs (1997) modeled and optimized a semiregenerative catalytic naphtha reformer using a detailed kinetics scheme involving 35 pseudo components connected by a network of 36 reactions in the C5 –C10 range. Weifeng et al. (2007) applied an 18-lump reaction kinetics model for the naphtha continuous catalytic reforming. Gyngazova et al. (2011) developed an NCR kinetics model based on components aggregation into pseudo components according to their activity. Meidanshahi et al. (2011) used four dominant idealized reactions, which were dehydrogenation, dehydrocyclization, hydrocrackings of naphthenes and paraffins, and hydrodealkylation of toluene. Rodríguez and Ancheyta (2011) combined the simplicity of the lumped model with the complexity of the most advanced model to predict a detailed composition of the reformate. Iranshahi et al. (2013) applied a new kinetics model including 32 pseudo components with 84 reactions in a novel thermally coupled reactor in a continuous catalytic regenerative naphtha reforming process. It should be noticed that a compromise between the simplified and the rigorous reaction models is necessary to effectively represent real complex reactions within a reasonable calculation time. In this study, two chemical reaction models of NTC and NCR are proposed, which consist of 83 and 86 reactions, respectively. The kinetics models are solved conveniently in the Aspen Plus platform (AspenTech, USA, 2013) that contains the thermodynamic database of hydrocarbon components. The SMB unit is directly connected to these NTC and NCR reactors. The objective of this study is to investigate the effects of naphtha separation into n- and non n-paraffins on the NTC and NCR reaction kinetics in the retrofit petrochemical complex (rPCC) with the SMB unit. To achieve this objective, first a process model of the SMB unit

130

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Fig. 1. Conceptual flow diagram and mass flow rate.

is developed by using the commercial process simulator. Next, the reaction kinetics models of NTC and NCR are proposed to predict the product compositions. The energy consumption of SMB, NTC, and NCR is also concerned. The reaction kinetics models are validated with plant data. Finally, two cases of PCC with and without the SMB unit are compared. The paper is organized as follows: Section 2 describes two PCC plants with and without SMB unit as well as NTC and NCR reactors. Section 3 presents reaction kinetics models and the validation of NTC and NCR. Results of reaction kinetics are addressed in Section 4. The conclusion is followed in Section 5.

2. Process description Fig. 1 shows the conceptual flow diagram of two PCC configurations: cPCC (conventional petrochemical complex) and rPCC (retrofit petrochemical complex with the simulated moving-bed). The SMB unit separates naphtha into n- and non n-paraffins. As shown in Fig. 1(a), cPCC having a total capacity of 3800 kt/yr of naphtha includes two distillation columns to separate light and heavy hydrocarbons, a naphtha hydrotreating (NHT) area, a naphtha thermal cracking (NTC) reactor, a naphtha catalytic reformer (NCR), and olefins & aromatics separations areas. The inlets of the NTC reactor are light hydrocarbons (L-HC) which are the mixture of 1150 kt/yr leaving the distillation column 1 (Distil 1), 40 kt/yr from the reformer, 410 kt/yr exiting the extractive aromatics separation

area, and 600 kt/yr of raw naphtha. C5 –C10 hydrocarbons (C5 –C10 ) of 1890 kt/yr are pretreated by hydrogen in the NHT area to remove sulfur compounds before entering the naphtha catalytic reformer. The products of NTC and NCR are olefins and aromatics, respectively, which will be refined in successive downstream areas. In the rPCC plant, all naphtha (3800 kt/yr) is separated into LHC, C5 –C10 mixture and H-HC in the distillation columns (Distil 1 & 2 in Fig. 1(b)). The NHT area has to treat more of the C5 –C10 mixture (2477 kt/yr) than that in cPCC (1890 kt/yr). In the C5 –C10 mixture entering the NHT area, 2244 kt/yr comes from the Distil 2 and 234 kt/yr is brought independently to keep the same capacities of the NTC and NCR. After the separation in the SMB unit, the n-paraffins stream (587 kt/yr) goes into NTC, while the non n-paraffins flow (1888 kt/yr) enters the catalytic reformer. Less non-aromatics from the extractive aromatics separation of the rPCC (209 kt/yr) enters the NTC than those of the cPCC (410 kt/yr) because of a lower yield of non-aromatics in the NCR. The capacities of the downstream processes such as NTC, NCR and olefins & aromatics separation areas are almost the same in both configurations, as seen in Fig. 1(a) and (b). To investigate the effect of n-paraffins separation by the SMB unit on reaction kinetics of NTC and NCR, the conceptual flowsheets of cPCC and rPCC are simplified as illustrated in Fig. 2. Here, the feed capacities of the NTC and NCR reactors have the same amount in both cases, which is 2475 kt/yr. It is noted that the total feed capacity in Fig. 2 was specified as the feed flow rate of SMB in Fig. 1(b) to examine the influence of SMB separation.

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

131

Table 1 Naphtha composition fed toNCR of cPCC or SMB unit in rPCC. n-Paraffins (wt%) C5 C6 C7 C8 C9 C10 Total (wt%)

i-Paraffins (wt%)

Naphthenes (wt%)

Aromatics (wt%)

Total (wt%)

0.47 6.55 7.45 6.80 1.49 0.17

0.23 6.17 7.77 8.29 5.45 1.03

0.66 9.46 15.09 7.53 3.38 0.07

– 1.56 4.42 4.98 0.90 0.07

1.35 23.75 34.73 27.60 11.23 1.34

22.94

28.93

36.19

11.94

100.00

In the cPCC (see Fig. 2(a)), two naphtha feeds are considered: the feed composition to NTC is specified by the mixture of four streams entering to Mixer for NTC in Fig. 1(a), and the feed composition to NCR is given as the C5 –C10 mixture in Table 1. In the rPCC (see Fig. 2(b)), the SMB unit is added for the separation of n-paraffins. The feed composition to SMB is the same as that to NCR in Fig. 2(a). Hereafter, the process simulation study is based on the simplified flowsheets. 2.1. Separation of n- and non n-paraffins by SMB The SMB processes are widely used in sugar, petrochemical, and pharmaceutical industries. The SMB system consists of multiple columns connected to each other in series. The feed and desorbent are supplied continuously to the SMB unit. Simultaneously, the raffinate and extract products are withdrawn (Kawajiri and Biegler, 2008; Lim, 2012; Mun, 2013; Sutanto et al., 2012; Yao et al., 2013). Table 1 reports the naphtha composition fed to cPCC or the SMB unit in rPCC. It is assumed that sulfur compounds were removed previously in the NHT area from naphtha. The feedstock classified by hydrocarbon species comprises n-paraffins of 22.94 wt%, i-paraffins of 28.93 wt%, naphthenes of 36.19 wt%, and aromatics of 11.94 wt%. The carbon number is dominated by C6 , C7 , and C8 , which are 23.75 wt%, 34.73 wt%, and 27.60 wt%, respectively. A detailed SMB calculation is out of scope of this study. A size-exclusion adsorption with molecular-sieve 5A zeolite has been applied to separate n-paraffins from hydrocarbon mixtures in SMB (Bieser, 1977; Broughton, 1977; Kulprathipanja, 1991; Liu et al., 2014; Ragil et al., 2002). The SMB recovery of n-paraffins from C5 –C8 , C10 –C35 , and C10 –C16 mixtures was reported to 95%, 86%, and 90–95% in the patents of Ragil et al. (2002), Kulprathipanja (1991), and Broughton

(1977), respectively. In this study, the SMB recovery of n-paraffins from C5 –C10 hydrocarbons was set to 90% as the reference value. Fig. 3 shows the process flow diagram (PFD) of the SMB unit. It consists of an SMB, extract and raffinate distillation columns to recover the desorbent (n-butane), heat exchangers, and cooler. The specification of the main equipment such as the SMB system, heat exchangers, extract and raffinate distillation columns is described in Table 2. The desorbent ratio to the feed (D/F ratio) varied from 2 to 10 in SMB for n-paraffin separation (Bieser, 1977; Broughton, 1977; Liu et al., 2014; Ragil et al., 2002). The D/F ratio was assumed to be 2.5 in this study, as Ragil et al. (2002) proposed for C5 –C8 mixture. It was supposed to use an eight-zone SMB process with the rotary valve that has run to separate para-xylene (PX) from its isomers in the commercial-scale (Sutanto et al., 2012). Fixing the maximum superficial velocity of each zone at 0.85 m/min which was derived from the PX-SMB, the bed diameter (dc ) was estimated at 6.8 m and the height (hc ) was set to 1.1 m. The 8-zone SMB process was composed of 24 beds. For the extract and raffinate distillation columns, the theoretical number of stages (Ns ) was found, minimizing the annualized total cost (ATC). The ATC is the sum of the operating cost of rebolier and the total capital investment (TCI) of column annualized for 10 years. At a given number of stages, the heat duty of reboiler (Qr ) is minimized with the variation of the molar reflux ratio (R) and the constant distillate flow rate (D), satisfying the n-butane purity (Pbutane ) of 99% in mass. In other words, the heat duty minimization problem is nested inside the ATC minimization. The tray efficiency () was set to 0.7. Naphtha and n-butane are pumped, and fed into the SMB unit at 125 ◦ C and 25 bar. The naphtha feed leaving from the NHT process enters into the pump 1 at 154 ◦ C and 10.1 bar. The extract and raffinate streams leave the SMB and enter the distillation columns at 127 ◦ C and 130 ◦ C, respectively. The counter-current heat exchanger 1, which locates between the SMB unit and the raffinate distillation column, transfers heat from the hot naphtha flow to the cold raffinate stream. The heat exchanger 2 lays between the SMB unit and the extract distillation column to exchange heat of the hot naphtha (135 ◦ C) and the cold extract streams (125 ◦ C) in the counter-current way. The hot and cold outlet temperature approach of the two heat exchangers was set to 5 ◦ C. In the two distillation columns, almost all of the n-butane is recovered due to its lowest boiling temperature in the mixtures. The n-paraffins stream which is rich in n-paraffins is then transferred to NTC for olefins production. At the same time, the non n-paraffins stream which contains almost all of the i-paraffins, naphthenes and aromatics is used to produce aromatics through catalytic reforming reactions.

2.2. Naphtha thermal cracking (NTC)

Fig. 2. Simplified conceptual flow diagram and mass flow rate.

Thermal cracking is an endothermic process where large molecules are broken down into smaller ones. The typical naphtha composition fed into the thermal cracking reactor in cPCC is indicated in Table 3. It is dominated by paraffins, about 80 wt%,

132

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Fig. 3. PFD of simulated moving-bed (SMB) unit.

consisting of 36 wt% of n-paraffins and 44 wt% of i-paraffins. The remaining components contribute by around 20 wt%. Indeed, there are several cracking reactors operated at different temperatures according to the different feeds to the NTC cracker. However, the feed composition averaged over the four different feeds (see

Fig. 1(a)) was used and an operating condition being able to represent the NTC cracker was assigned in this study (see Table 5). The flow rates of the SMB inlet and outlet streams in rPCC are listed in Table 4. The bottom product of the extract distillation column (n-paraffins stream) consists of 86 wt% n-paraffins and 14 wt%

Table 2 Description of key equipments in SMB unit. Unit

Function

Process model

Remarks

Simulated moving-bed (SMB)

Separate 90% n-paraffins from naphtha stream

Separator

Extract distillation column

Separate desorbent from extract stream

Rigorous fractionation column

Raffinate distillation column

Separate desorbent from raffinate stream

Rigorous fractionation column

Heat exchanger 1

Counter-current heat exchange of hot naphtha feed with cold raffinate stream Counter-current heat exchange of hot naphtha feed with cold extract stream leaving SMB unit

Heat exchanger

TSMB = 125 ◦ C, PSMB = 25 bar, D/F ratio = 2.5, 24 beds, dc = 6.8 m, and hc = 1.1 m. Ns = 13, Pext = 25 bar, Text = 127 ◦ C, Dext = 173.7 t/h, Rext = 0.515, Qr = 23.19 MW, dc = 3.5 m, and hc = 13.0 m,  = 0.7. Ns = 11, Praf = 25 bar, Traf = 127 ◦ C, Draf = 553.8 t/h, Rraf = 0.465, Qr = 66.85 MW, dc = 5.9 m, and hc = 11.1 m,  = 0.7. Hot stream: Tinlet = 155 ◦ C, Toutlet = 135 ◦ C, Cold stream: Tinlet = 125 ◦ C, Toutlet = 130 ◦ C Hot stream: Tinlet = 135 ◦ C, Toutlet = 132 ◦ C, Cold stream: Tinlet = 125 ◦ C, Toutlet = 127 ◦ C

Heat exchanger 2

Heat exchanger

Table 3 Typical naphtha composition fed to NTC of cPCC. n-Paraffins (wt%)

i-Paraffins (wt%)

C3 C4 C5 C6 C7 C8 C9 C10

0.03 2.34 9.91 13.05 5.71 3.09 1.68 0.35

0.00 0.33 8.07 17.32 9.78 4.10 3.77 1.08

Naphthenes (wt%) 0.00 0.00 1.81 4.25 4.45 2.09 0.60 0.00

Aromatics (wt%) 0.00 0.00 0.00 0.90 1.66 2.34 1.18 0.13

Total (wt%) 0.03 2.66 19.79 35.52 21.59 11.61 7.24 1.55

Total (wt%)

36.15

44.44

13.20

6.21

100.00

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

133

Table 4 Inlet and outlet stream results of SMB unit. Stream

Feed

Desorbent

SMB extract

SMB raffinate

n-Paraffins

Non n-paraffins

Temperature (◦ C) Pressure (bar) Mass flow (t/h)a n-C4 H10 n-C5 H12 n-C6 H14 n-C7 H16 n-C8 H18 n-C9 H20

154.00 10.13

125.00 25.00

125.18 25.00

125.18 25.00

252.81 25.00

257.13 25.00

0.00 1.36 19.06 21.67 19.78 4.85

727.50 0.00 0.00 0.00 0.00 0.00

173.40 1.22 17.16 19.51 17.80 4.37

554.10 0.14 1.91 2.17 1.98 0.49

1.44 0.92 16.27 19.24 17.74 4.36

5.84 0.10 1.79 2.13 1.97 0.48

66.72

727.50

233.45

560.77

59.96

12.32

0.65 17.95 22.60 24.10 15.85 3.00

0.00 0.00 0.00 0.00 0.00 0.00

0.03 0.77 0.97 1.03 0.68 0.13

0.63 17.18 21.63 23.07 15.17 2.87

0.02 0.71 0.94 1.02 0.68 0.13

0.42 15.63 20.85 22.72 15.15 2.87

84.15

0.00

3.61

80.55

3.49

77.63

1.92 27.52 43.88 21.90 9.84

0.00 0.00 0.00 0.00 0.00

0.08 1.18 1.88 0.94 0.42

1.84 26.34 42.00 20.96 9.42

0.07 1.17 1.88 0.94 0.42

1.70 26.24 41.99 20.96 9.42

105.06

0.00

4.50

100.56

4.49

100.31

4.53 12.85 14.81 2.63 0.22

0.00 0.00 0.00 0.00 0.00

0.19 0.55 0.63 0.11 0.01

4.33 12.30 14.17 2.52 0.21

0.19 0.55 0.63 0.11 0.01

4.28 12.30 14.17 2.52 0.21

Normal paraffins subtotal i-C5 H12 i-C6 H14 i-C7 H16 i-C8 H18 i-C9 H20 i-C10 H22 Isoparaffins subtotal C5 H10 C6 H12 C7 H14 C8 H16 C9 H18 Naphthenes subtotal C6 H6 C7 H8 C8 H10 C9 H12 C10 H14 Sub aromatics total Total (t/h)

35.03

0.00

1.50

33.53

1.50

33.48

290.97

727.50

243.05

775.41

69.35

221.61

n-Cn H2n+2 : normal paraffin, n = 4–9; i-Cn H2n+2 : isoparaffin, n = 5–10; Cn H2n : naphthene, n = 5–9; Cn H2n−6 : aromatic, n = 6–10. a One year is counted as 8500 h.

of i-paraffins, naphthenes, and aromatics. Thus, the total mass flow rate of the n-paraffins stream is about 69.4 t/h (=587 kt/yr) that goes into the NTC reactor. The configuration of the NTC reactor is briefly reported in Table 5. The reactor contains 28 coils, where one coil has a length of 22 m and a diameter of 0.127 m. It was supposed that the NTC naphtha feed is evenly distributed to the 28 coils. There are two main sections in the NTC reactor: convection and radiation (Seifzadeh Haghighi et al., 2013). Steam at 127 ◦ C and 2.5 bar is injected to dilute the feed and to decrease partial pressure of hydrocarbons as well as the amount of coke deposits in tubes. The steam/naphtha ratio was set to be 0.45, which is the same in cPCC. Then, naphtha and steam enter the radiation zone at 2.4 bar, and leave at 2.1 bar, where the pressure drop was obtained from a typical operating condition of the plant data. The cracking reactions only take place in the radiation zone, where naphtha is broken down to ethylene, propylene, butadiene, and other products at high temperatures from 650 ◦ C to 805 ◦ C through the cracker. Finally the product is cooled

Table 5 Configuration of NTC reactor. Coil length (m) Inner coil diameter (m) Feed flow rate per one coil (t/h) Number of coils Total naphtha flow rate (t/h) Steam/naphtha mass ratio Resident time (s) Inlet–outlet temperature (◦ C) Inlet–outlet pressures (bar)

22 0.127 2.48 28 69.4 0.45 0.24 650–805 2.4–2.1

down quickly, compressed and transferred to the olefins separation area. The PFD of the NTC reactor and its key operating parameters such as temperature, pressure, and mass flow rate are shown in Fig. 4. To treat the n-paraffins stream of 69.4 t/h from SMB, the consumed energy is about 106 MW consisting of 55 MW of the NTC heater and 51 MW of the cracking reactor (see also Section 4.3). The rPCC can produce more olefins at the cost of energy consumption.

2.3. Naphtha catalytic reforming (NCR) According to the catalyst regeneration procedure, the catalytic reaction process could be categorized in three types, including: semi-regenerative (SR), cyclic regenerative (CR), and continuous catalyst regenerative (CCR) (Iranshahi et al., 2013). The CCR is the most modern type in which the catalyst is continuously removed from the last reactor, regenerated in a catalyst regenerator, and then transferred back to the first reactor. Regenerated catalysts enter the top of the first annular-type reactor, move down through the reactor by gravity and exit from the bottom of the fourth annular reactor in several days. The inlet naphtha feed is injected to the top of reactor, proceeds downward near the shell of reactor, and then flows radially in the catalyst bed. Although it is not easy to maintain the continuous flow of solid particles, the CCR technology has significant advantages such as the ease of catalyst regeneration, high conversion rate, and good selectivity (Iranshahi et al., 2013). During the reforming process, the catalyst is deactivated by the formation of coke on its surface, which denotes the carbonaceous residues formed from secondary reactions within the catalyst. By

134

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Fig. 4. PFD of one coil in naphtha thermal cracking (NTC) reactor.

burning-off the coke with oxygen and nitrogen mixtures the catalyst is regenerated (Ren et al., 2002). The PFD of NCR consists of four beds and four heaters, as illustrated in Fig. 5. The configuration of the NCR annular reactor is listed in Table 6. The length increases significantly from bed 1 (8.50 m) to bed 4 (15.39 m). The outer diameter is almost double of the inner diameter in the four annular-type beds. The outer diameter increases gradually from the first bed to the fourth one. This reformer configuration results in the increase of the resident time from 0.57 s in bed 1 to 0.99 s in bed 4. The inlet and outlet temperatures of each bed were obtained from the plant data. Even though the inlet temperatures were desired to be kept constant, the real inlet temperatures varied from 530 ◦ C to 550 ◦ C. The non n-paraffins stream from the SMB unit is mixed with hydrogen at a hydrogen/naphtha molar ratio of 3.26 and then heated to 532 ◦ C at about 5.1 bar. The pressure drop of each bed was also given in Table 6 according to the plant data. Because the reformer is operated adiabatically and the major reforming reactions are endothermic and very fast, a very sharp temperature drop occurs in NCR. For this reason, catalytic reformers are designed with a multiple-bed reactor with heaters between the reactors to keep the reaction temperature at operable levels (Gyngazova et al., 2011; Rahimpour et al., 2013; Rodríguez and Ancheyta, 2011). As the total reactor charge passes through the sequence of heating and reacting, the reactions become less and less endothermic. Finally the reforming products move into the stabilizer and then the separation area to refine aromatics and other petrochemicals. Inlet and outlet temperatures of the 4 NCR beds in rPCC were specified as the same value as those in cPCC. The total energy

consumption of the NCR reactor is the sum of the heat duties of 4 NCR beds and heaters (see Fig. 5 and also Table 6). 3. Reaction kinetics models and validation The SMB unit separates naphtha into n-paraffins and non n-paraffins. The n-paraffins stream goes into NTC to produce olefins, while non n-paraffins stream enters NCR to make aromatics, as mentioned previously. Two reaction kinetics models were developed to examine the changing effects of the naphtha composition on the yield of the NTC and NCR products. The NTC kinetics model consists of 83 reactions of 6 types comprising hydrogenation, aromatization, isomerization, formal reactions, paraffin cracking, and naphthene cracking. The NCR kinetics model is composed of 86 reactions of 10 types including dehydrogenation, alkylcyclohexane, dehydrocyclization to alkylcyclopentane, paraffin dehydrocyclization to aromatic, naphthene isomerization, paraffin isomerization, transalkylation, paraffin cracking, naphthene cracking, and hydrodealkylation. The reaction rate is expressed by the power law equation:



rj = k0,j exp

−Ea,j

 N

RT

pi ˛i

(1)

i=1

where rj (kmol/(m3 s)) is the reaction rate of jth reaction, k0,j is the pre-exponential factor of jth reaction, Ea,j (kJ/mol) is the activation energy of jth reaction, R (J/(mol K)) is the gas constant, T (K) is the reaction temperature, pi (Pa) is the partial pressure of ith reactant,

Fig. 5. PFD of naphtha catalytic reforming (NCR) reactor.

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

135

Table 6 Configuration of NCR reactor. Naphtha flow rate (t/h) Hydrogen/naphtha molar ratio

221.6 3.26

Bed

Bed 1

Bed 2

Bed 3

Bed 4

Resident time (s) Inlet temperature (◦ C) Outlet temperature (◦ C) Inlet/outlet pressures (bar) Length (m) Inner/outer diameters (m)

0.57 531.6 394.5 5.1/5.0 8.5 1.25/2.19

0.65 528.0 452.8 5.0/4.9 10.4 1.25/2.35

0.74 549.0 488.9 4.8/4.7 12.1 1.30/2.53

0.99 548.7 514.7 4.6/4.3 15.4 1.30/2.89

and ˛i is the exponent of ith reactant. N is the number of reactants in jth reaction, and j varies from 1 to 83 or from 1 to 86 in the NTC or NCR kinetics models, respectively. The activation energy (Ea ) was referred in literatures (Belohlav et al., 2003; Iranshahi et al., 2013), whereas the pre-exponential factors (k0 ) were tuned to fit typical plant data. Major components which were ethylene, propylene, and methane in NTC, and benzene, toluene, m-xylene, and methane in NCR were firstly focused. This step is called the reaction trend determination or raw k0 estimation. The k0 of these reactions, in which the major components are involved, may vary in the wide range. And then the yield of minor components was concerned. Normally, the k0 for the minor components was determined by changing k0 in the narrow range. Finally, a fine estimation was performed until the correlation coefficient between plant data and the reaction model results was acceptable. The detailed reactions and their kinetics parameters of NTC and NCR are listed in Tables A1 and A2 in Appendices. For the reaction kinetics, the

set of ordinary differential equations were solved by an implicit time integrator at a temperature and a residence time given in Tables 5 and 6. In the commercial plant, the coke thickness grew by 1.6 mm/month on the tube wall of the NTC. However, the effect of coke on reaction kinetics was limited, since the 28 coils of the NTC reactor were regularly cleaned one by one to remove coke. The coke formation was thus ignored in this study. A one-dimensional plug flow model was solved for both NTC and NCR. The steady-state mass balance is described in the plug-flow reactor as:

⎛

⎞

 dFi d2 sji rj ⎠ =⎝ 4 dz

(2)

j

where Fi (kmol/s) is the molar flow rate of ith component, z (m) is the axial direction of the reactor, sji is the stoichiometric factor of

(a)

(b)

Fig. 6. Comparison of NTC product compositions between calculated results and plant data of typical naphtha cracking.

136

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

(a)

(b)

Fig. 7. Correlation between calculated composition and plant data.

(a)

(b)

Fig. 8. Temperature profiles of NTC and NCR reactors versus reactor length.

ith component in jth reaction, and d (m) is the reactor diameter. The steady-state energy balance is:

 i

Fi cp,i

d2  dT rj (−H)j =Q+ 4 dz

(3)

j

where cp,i (kJ/(kmol K)) is the heat capacity of ith component, Q (kJ/s) is the heat flow rate across the reactor wall, and (−H)j (kJ/mol) is the heat of jth reaction. The heat of reaction (−H) is calculated on the basis of the standard heat of formation of reactants and products in the databank of the process simulator. The two kinetics models were validated with cPCC plant data. Fig. 6 shows NTC product compositions at three different NTC outlet temperatures for the typical naphtha feed given in Table 3. The calculated compositions of four major products (ethylene, propylene, aromatics, and n-paraffins) have the same trend as the plant data, as depicted in Fig. 6(a). Here, n-paraffins are composed of normal C1 –C4 of approximate 90 wt% and normal C5 –C10 of about 10 wt%. For minor products like butadiene, i-paraffins, naphthenes, and hydrogen, Fig. 6(b) compares the computed product compositions

(a)

to the plant data within the range from 0 wt% to 6 wt%. Except butadiene, the calculated compositions agree well with the plant data. For the naphtha feed given in Table 1, Table 7 reports compositions of 13 NCR products obtained from the plant data and the reaction kinetics model. The difference between them is limited ±2 wt%. Since the proposed NCR kinetics model has no reaction path for the olefin formation, the biggest error is found in the olefin composition. The calculated compositions of n-paraffins and i-paraffins are a little overestimated compared to the plant data. Fig. 7(a) and (b) illustrates the correlation of the compositions between the plant data and the model calculation for NTC and NCR, respectively. The correlation coefficients (R2 ) are 0.990 and 0.988 for the reaction kinetics models of NTC and NCR, respectively. The closer to 1 this factor is, the better the model is. It is expected that the two reaction models are acceptable to predict the product composition in changing the naphtha feed composition. Fig. 8(a) and (b) shows the axial temperature profiles of NTC and NCR, respectively. As mentioned previously in Sections 2.2 and 2.3, the inlet and outlet temperatures of the two reactors were specified according to the plant data. The temperature profiles were applied

(b)

Fig. 9. Mass fraction of NTC products with respect to reactor length in cPCC.

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

(a)

137

(b)

Fig. 10. Mass fraction of NTC products with respect to reactor length in rPCC.

to both cPCC and rPCC. The temperature of the NTC reactor (or coil) linearly increases from 650 ◦ C to 805 ◦ C. The temperature difference across the four beds of the NCR reactor decreases from 137 ◦ C to 34 ◦ C, as also seen in Table 6. In this work, the naphtha flow was reheated to around 530–550 ◦ C before entering each bed. 4. Results of reaction kinetics The naphtha feed composition of cPCC was given in Table 1 for NCR and Table 3 for NTC, while that of rPCC was listed in Tables 1 and 4. The effects of the naphtha feed composition change on the NTC and NCR reactions are presented below. 4.1. Results of NTC reaction kinetics Figs. 9 and 10 show the fraction of the NTC products versus the reactor length in cPCC and rPCC based on dry mass, respectively. The major products of NTC are plotted in Figs. 9(a) and 10(a), while the minor products are in Figs. 9(b) and 10(b). The reactor inlet temperature of NTC was set to 650 ◦ C and the outlet was 805 ◦ C (see Section 2.2 and Fig. 4). Due to thermal cracking, the mass fractions of small molecules like C1 –C4 n-paraffins, ethylene, propylene, butadiene, and hydrogen rise gradually with the reactor length in the two cases. The mass fraction of C5 –C10 n-paraffins decreases with the reactor length from 34 wt% to 4 wt% in cPCC and from 90 wt% to 6 wt% in rPCC. i-Paraffins almost disappear at the cracker exit in both cases. Ethylene is more produced in rPCC (35 wt%) than in

cPCC (28 wt%), as expected. The production of butadiene is a little greater in rPCC than in cPCC. In Fig. 11, the final product compositions at the exit of rPCC are shown with respect to the exit temperature of NTC in a range between 795 ◦ C and 835 ◦ C. The feed inlet temperature was fixed at 650 ◦ C. Fig. 11(a) displays major products such as olefins and butadiene, whereas Fig. 11(b) minor products such as i-paraffins, naphthenes, and hydrogen. The exit temperature slightly affects the NTC product compositions. The ethylene yield can reach a maximum value of 35.5 wt% at 825 ◦ C. The propylene yield slightly reduces from 16 wt% to 14 wt% as the temperature rises from 795 ◦ C to 835 ◦ C. Butadiene increases from 4 wt% to 8 wt% in this temperature range. The total olefins yield (ethylene and propylene) varies in a range of 49–51 wt%. 4.2. Results of NCR reaction kinetics Figs. 12 and 13 present the mass fraction along the length of the NCR four beds in cPCC and rPCC, respectively. The nonaromatic products of NCR and the total aromatic fractions are plotted in Figs. 12(a) and 13(a), while the aromatic products are

(a)

Table 7 Comparison of calculated compositions and plant data of NCR in cPCC. Products

Hydrogenb n-Paraffins i-Paraffins Naphthenes Olefins Benzene Toluene meta-Xylene ortho-Xylene para-Xylene Ethyl-benzene Aromatic-C9 Aromatic-C10 Subtotal aromatics Total

Calculated compositions (wt%)

Plant data (wt%)

Difference (wt%)a

3.31 19.99 9.99 1.89 1.70 9.91 22.97 10.72 5.12 2.87 3.16 7.33 1.03

−0.31 +1.16 +1.55 −0.23 −1.70 +0.55 +0.39 −0.31 +0.05 −0.12 −0.17 −0.90 +0.03

62.64

63.11

−0.48

100.00

100.00

0.00

3.00 21.15 11.54 1.67 0.00 10.46 23.36 10.42 5.17 2.75 2.99 6.42 1.06

Difference = calculated composition (wt%) − plant data (wt%). Hydrogen is a net value calculated by the subtraction of the recycled flow rate from the total flow rate of hydrogen.

(b)

a

b

Fig. 11. Effect of NTC temperature on products mass fraction in rPCC.

138

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Fig. 12. Mass fraction of NCR products versus reactor length in cPCC.

Fig. 13. Mass fraction of NCR products versus reactor length in rPCC.

in Figs. 12(b) and 13(b). The total aromatic fraction increases gradually with the reactor length for both cases. Since the mass fraction of n-paraffins keeps high over the whole beds in cPCC, the aromatic production is less than that of rPCC. As mentioned earlier, the inlet stream to rPCC is dominated by non n-paraffins like i-paraffins, naphthenes, and aromatics which promote the production of aromatics. Hydrogen linearly increases from 6 wt% to about 9 wt% with the reactor length in both cases, resulting in a net hydrogen composition of around 3 wt% (see Table 8). In Figs. 12(b) and 13(b), it is noticeable that the mass fraction of toluene increases up to 21.8 wt% and 24.2 wt% in cPCC and rPCC, respectively. Owing to the higher compositions of i-paraffins and naphthenes in the feedstock, a higher yield of aromatics is gained in rPCC.

4.3. Comparison of cPCC and rPCC Table 8 lists product compositions (wt%) and flow rates (t/h) of the two investigated cases, cPCC and rPCC. 9 major products are reported in NTC and 13 products are in NCR. In the NTC, the total composition of ethylene, propylene, and butadiene in cPCC is about 48.9 wt% and that in rPCC is about 55.9 wt%. In the NCR, the total composition of aromatics in cPCC is 62.6 wt% and that in rPCC is 69.4 wt%. The difference of compositions and flow rates between rPCC and cPCC is also indicated in Table 8. Significant increases of ethylene and toluene are observed in NTC and NCR of rPCC, respectively. To clearly explain the difference of flow rates between cPCC and rPCC, Fig. 14 illustrates the production rates of 8 main products (ethylene, propylene, butadiene, hydrogen, benzene, toluene, xylene, and other aromatics). The rPCC shows a significant improvement of olefins and aromatics productions in NTC and NCR, respectively. In summary, the olefins production rate increases by 14 wt% and the aromatics production by 11 wt% in rPCC. More

olefins (41 kt/yr) and aromatics (127 kt/yr) are produced in rPCC having naphtha capacity of 2475 kt/yr. In the rPCC, the energy consumption has to increase because of the SMB unit. Table 9 compares the energy consumption of the two cases. The two distillation columns of the SMB unit require 90.0 MW to treat C5 –C10 naphtha of 291 t/h. The heat duties of the NTC cracker and the four NCR beds are indicated, which correspond to Q in Eq. (3). Since the NTC of rPCC produces more olefins than that of cPCC, the cracker heat duty of rPCC increases by about 2.9 MW. However, due to the higher temperature (253 ◦ C) of the n-paraffins stream in the rPCC than that in the cPCC (154 ◦ C), the heat duty of the NTC heater reduces by 6.9 MW in the rPCC. Totally, the energy consumption of the NTC area decreases by 4.0 MW in the rPCC. The energy consumption of NCR is also saved by 18.2 MW according to the smaller heat duty of the heater 1 (23.5 MW). It results from the higher temperature of the NCR feed (257 ◦ C) in rPCC than that

Fig. 14. Comparison of main product flow rates between cPCC and rPCC.

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

139

Table 8 Products compositions and flow rates of cPCC and rPCC. Area

Products

cPCC

rPCC

Compositions (wt%)

Flow rate (t/h)

Compositions (wt%)

Differencea (wt%)

Flow rate (t/h)

Differenceb (t/h)

NTC

Ethylene Propylene Butadiene Aromatics n-Paraffins i-Paraffins Naphthenes Hydrogen Others Subtotal

28.3 17.5 3.1 13.1 22.9 1.2 2.3 0.6 11.2 100.0

19.6 12.1 2.1 9.1 15.9 0.8 1.6 0.4 7.8 69.4

35.0 16.1 4.8 9.2 26.5 0.2 0.9 0.6 6.7 100.0

6.7 −1.4 1.7 −3.8 3.6 −1.0 −1.3 0.0 −4.5 0.0

24.3 11.2 3.3 6.4 18.4 0.1 0.6 0.4 4.7 69.4

+4.6 −0.9 +1.2 −2.7 +2.5 −0.7 −0.9 0.0 −3.1 0.0

NCR

Hydrogenc n-paraffins i-paraffins Olefins Naphthenes Benzene Toluene meta-Xylene ortho-Xylene para-Xylene Ethyl-benzene Aromatic-C9 Aromatic-C10 Subtotal aromatics Subtotal

3.0 21.2 11.5 1.7 0.0 10.5 23.4 10.4 5.2 2.8 3.0 6.4 1.1 62.6 100.0

6.7 46.9 25.6 3.7 0.0 23.2 51.8 23.1 11.5 6.1 6.6 14.2 2.3 138.8 221.6

3.1 15.3 10.3 1.9 0.0 11.4 26.0 11.4 5.3 3.2 3.5 7.3 1.3 69.4 100.0

+0.1 −5.8 −1.3 +0.3 +0.0 +0.9 +2.6 +1.0 +0.1 +0.4 +0.5 +0.9 +0.2 +6.8 0.0

6.9 33.9 22.7 4.3 0.0 25.2 57.5 25.3 11.7 7.1 7.7 16.3 2.9 153.8 221.6

+0.2 −13.0 −2.8 +0.6 0.0 +2.1 +5.8 +2.3 +0.3 +1.0 +1.1 +2.0 +0.5 +15.0 0.0

a b c

Difference = composition (wt%) of rPCC − composition (wt%) of cPCC. Difference = flow rate (t/h) of rPCC − flow rate (t/h) of cPCC. Hydrogen is a net value calculated by the subtraction of the recycled flow rate from the total flow rate of hydrogen.

Table 9 Comparison of energy consumption between cPCC and rPCC. Area

Capacity (t/h)

Equipment

SMB

291.0

Extract distillation column Raffinate distillation column Subtotal

NTC

69.4

Heater Cracker Subtotal

221.6

Heater 1 Bed 1 Heater 2 Bed 2 Heater 3 Bed 3 Heater 4 Bed 4 Subtotal

NCR

Total heat duty a

cPCC

rPCC

Heat duty (MW)

Heat duty (MW)

0 0 0

Differencea (MW)

23.2 66.9 90.0

+23.2 +66.9 +90.0

62.0 47.7 109.7

55.0 50.6 105.7

−6.9 +2.9 −4.0

107.7 −10.0 33.9 −1.2 25.3 −7.4 15.9 2.2 166.5

84.2 −5.72 34.5 2.0 25.6 −6.0 16.1 −2.5 148.3

−23.5 +4.2 +0.6 +3.2 +0.3 +1.4 +0.2 −4.6 −18.2

276.2

318.9

+67.8

Difference = heat duty (MW) of rPCC − heat duty (MW) of cPCC.

in cPCC (154 ◦ C). Hence, 67.8 MW is totally consumed more in rPCC than in cPCC, which is about 25% increase of the total heat duty having 276.2 MW in cPCC. However, more rigorous study on the energy consumption is necessary because the heat network is complexly integrated in the real plant.

5. Conclusions This study investigated the effect of the naphtha separation into n-paraffins and non n-paraffins by using the SMB (simulated moving-bed) on the reaction kinetics of NTC (naphtha thermal cracking) and NCR (naphtha catalytic reforming) for the industrial

olefins and aromatics production. The SMB unit includes one SMB adsorption process, two distillation columns for desorbent recovery, and heat exchangers. A simple separation ratio of naphtha was used in the SMB process. Two conceptualized petrochemical complexes (PCC) were concerned: conventional PCC without the SMB unit (cPCC) and retrofit PCC with the SMB unit (rPCC). Simplified reaction kinetics of NTC and NCR were developed and validated with typical plant data obtained from a commercial petrochemical complex. The results from the two reaction kinetics models had a good agreement with the plant data. The olefins and aromatics production rates increase by 14 wt% and 11 wt%, respectively, in rPCC, since n-paraffins are cracked more in NTC and non n-paraffins produce more aromatics in NCR. Although

140

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

more olefins and aromatics are produced in rPCC, the energy consumption increases due to the SMB unit. The rPCC needs a total of 67.8 MW more than cPCC, which means a 25% increase of the heat duty for the C5 –C10 naphtha of 2475 kt/yr. The separation of naphtha into n-paraffins and non n-paraffins showed a significant improvement of olefins and aromatics production. It can be a promising way to increase the economic profitability of petrochemical plants. However, it is necessary to perform a techno-economic analysis for rPCC. Not only will cost energy consumption increase, but it will also pay capital investment costs for the SMB unit, the naphtha hydrotreating area, and the downtime costs for integrating the SMB unit to an existing plant. A rigorous calculation with adsorption isotherms on size exclusion 5A zeolite will help to accurately obtain the compositions of extract and raffinate in the SMB process. A kinetic model of coke formation is also necessary to consider the effect of coke formation in the NTC. Furthermore, technical problems of this naphtha separation on others areas of PCC have to be analyzed

before commercially applying the naphtha separation process to the petrochemical industry. Acknowledgements This work (grant no. C0002706) was financially supported by Korea Small and Medium Business Administration (http://www.smba.go.kr) in 2012 in the framework of Cooperative R&D between Industry, Academy, and Research Institute. The authors thank to GNG MC for the collaborations. We also appreciate the editing contribution of Patrick Bresnahan. Appendices. Kinetics parameters of NTC (naphtha thermal cracking) and NCR (naphtha catalytic reforming). Tables A1 and A2.

Table A1 Naphtha thermal cracking reactions and kinetics parameters.



Reactions

k0 (kmol/m3 s Pa

Hydrogenation

C2 H4 + H2 → C2 H6 C3 H6 + H2 → C3 H8 1-C4 H8 + H2 → n-C4 H10 2-C4 H8 + H2 → n-C4 H10 i-C4 H8 + H2 → i-C4 H10 C2 H2 + H2 → C2 H4 C3 H4 + H2 → C3 H6 C4 H6 + H2 → 1-C4 H8

2.13E+06 1.64E+12 8.22E+12 8.05E+12 2.05E+14 5.75E+08 7.44E+09 6.76E+10

191 340 340 340 340 278 278 278

9 10 11 12 13 14 15

Aromatization

C4 H6 + C2 H4 → C6 H6 + 2H2 C4 H6 + C3 H6 → (CH3 )-C6 H5 + 2H2 C4 H6 + 1-C4 H8 → (C2 H5 )-C6 H5 + 2H2 2C4 H6 → (C2 H3 )-C6 H5 + 2H2 C4 H6 + 2-C4 H8 → OX + 2H2 C4 H6 + C6 H6 → (C4 H3 )-C6 H5 + 2H2 C5 H8 + C4 H6 → (C3 H7 )-C6 H5 + H2

1.06E+03 6.52E+02 4.20E+01 1.11E+02 4.66E+01 8.09E+01 2.42E+01

116

16 17

Isomerization

1-C4 H8 → 2-C4 H8 2-C4 H8 → 1-C4 H8

1.37E+00 6.64E−01

42 32

C2 H6 → C2 H4 + H2 C2 H4 → C2 H2 + H2 C3 H6 → C3 H4 + H2 C3 H8 → C2 H4 + CH4 C3 H8 → C3 H6 + H2 n-C4 H10 → C2 H4 + C2 H6 n-C4 H10 → C3 H6 + CH4 n-C4 H10 → 2 C2 H4 + H2 n-C4 H10 → 1-C4 H8 + H2 i-C4 H10 → C3 H6 + CH4 i-C4 H10 → i-C4 H8 + H2 1-C4 H8 → C4 H6 + H2 1-C4 H8 → 2C2 H4 i-C4 H8 → C3 H4 + CH4 C5 H8 + H2 → C3 H6 + C2 H4 (1,4-C6 H10 ) + H2 → 2 C3 H6 (CH3 )-C6 H5 + H2 → C6 H6 + CH4 (C2 H5 )-C6 H5 + H2 → (CH3 )-C6 H5 + CH4 (C3 H7 )-C6 H5 + H2 → OX + CH4 (C4 H9 )-C6 H5 + H2 → (C3 H7 )-C6 H5 + CH4 C5 H8 + H2 → C3 H6 + C2 H4 (2,2-C5 H10 ) → C4 H6 + CH4 C4 H6 + H2 → 2 C2 H4 C3 H6 + H2 → C2 H4 + CH4 i-C4 H8 + H2 → C3 H6 + CH4 1-C4 H8 + H2 → C3 H6 + CH4 2-C4 H8 + H2 → C3 H6 + CH4 C2 H4 + CH4 → C3 H6 + H2 C3 H6 + CH4 → C2 H4 + C2 H6 C3 H6 + CH4 → 1-C4 H8 + H2 1-C4 H8 + CH4 → C2 H4 + C3 H8 2-C4 H8 + CH4 → C2 H6 + C3 H6 i-C4 H8 + CH4 → C2 H4 + C3 H8

4.52E+11 4.03E+07 2.18E+07 8.64E+06 1.30E+07 1.20E+08 3.50E+06 2.13E+10 3.17E+06 5.57E+06 6.37E+08 5.05E+07 4.85E+08 3.81E+06 5.35E+00 1.65E−01 1.32E−02 9.37E−01 1.93E−05 2.87E+00 1.16E−01 1.42E+02 3.27E−01 1.09E+09 2.00E+05 7.24E+07 2.07E+04 4.51E+08 1.55E+07 1.10E+08 1.09E+07 2.82E+07 3.23E+07

340 207 207 237 227 237 227 305 207 237 218 195 237 207 35 35 35 35 35 35 35 35 63 211 211 211 211 204 204 204 204 204 204

1 2 3 4 5 6 7 8

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Formal reactions

˛i

)

Ea (kJ/mol)

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

141

Table A1 (Continued)



Reactions

k0 (kmol/m3 s Pa

51 52 53 54 55 56 57 58 59 60 61

C4 H6 + CH4 → C2 H4 + C3 H6 C2 H4 + 2H2 → 2CH4 C3 H6 + C2 H2 → C5 H6 + H2 C2 H6 + C2 H4 → C3 H6 + CH4 C2 H4 + C2 H2 → C4 H6 C3 H6 + C2 H4 → C5 H8 + H2 2 C3 H6 → (1,4-C6 H10 ) + H2 2 C3 H6 → C5 H6 + CH4 + H2 C3 H6 + C4 H6 → C5 H6 + C2 H4 + H2 C5 H8 → C5 H6 + H2 (2,2-C5 H10 ) → C2 H4 + C3 H6

4.36E+01 2.50E+08 2.71E+01 2.04E+08 5.34E+04 2.39E+04 2.23E+07 1.69E+02 5.35E−01 3.20E−02 5.00E+07

62 63 64 65

3.00E+08 4.02E+08 2.05E+08 2.50E+08

68 69 70 71 72 73 74

3n-C4 H10 → 2CH4 + 2 C2 H4 + C3 H8 + C3 H6 n-C5 H12 → 0.5n-C4 H10 + 0.5 C2 H6 + C2 H4 5n-C6 H14 → n-C4 H10 + n-C5 H12 + CH4 + 4 C2 H4 + 2 C3 H6 + C2 H6 + 1-C4 H8 + H2 15n-C7 H16 → 2 (1-C4 H8 ) + 5n-C5 H12 + 7CH4 + 15 C2 H4 + 5 C3 H6 + n-C4 H10 + C4 H6 + 2 (2-C4 H8 ) + 2 C2 H6 + H2 15n-C8 H18 → 2 (1-C4 H8 ) + 5n-C5 H12 + 9CH4 + 21 C2 H4 + 5 C3 H6 + (2,2-C5 H10 ) + 2iC4 H8 + (2-C4 H8 ) + C4 H6 + 2H2 n-C9 H20 → 0.2 (1-C4 H8 ) + 0.4n-C5 H12 + 0.6CH4 + 1.7 C2 H4 + 0.4 C3 H6 + 0.2 (2,2-C5 H10 ) i-C4 H10 → CH4 + 1.5 C2 H4 i-C5 H12 → 0.5i-C4 H10 + 0.5CH4 + 0.5 C2 H4 + 0.5 C3 H6 5i-C6 H14 → i-C4 H10 + 2i-C5 H12 + 2CH4 + 2 C2 H4 + 2 C3 H6 + i-C4 H8 15i-C7 H16 → i-C4 H10 + 6i-C5 H12 + 8CH4 + 18 C2 H4 + 5 C3 H6 + 2i-C4 H8 + C4 H6 + H2 15i-C8 H18 → 2i-C4 H8 + 6i-C5 H12 + 9CH4 + 26 C2 H4 + 4 C3 H6 + (2,2-C5 H10 ) + C4 H6 + H2 i-C9 H20 → 0.2 (1-C4 H8 ) + 0.4i-C5 H12 + 0.6CH4 + 1.7 C2 H4 + 0.4 C3 H6 + 0.2 (2,2-C5 H10 ) i-C10 H22 → (2,2-C5 H10 ) + 0.5n-C4 H10 + 0.5 C2 H6 + C2 H4

75 76 77 78 79 80 81 82 83

NA6-C6 → 0.4 C4 H6 + 0.4 (2,2-C5 H10 ) + 0.4CH4 + 0.4 C2 H4 + 0.4 C3 H6 15 NA6-C7 → 7 C4 H6 + 7 (2,2-C5 H10 ) + 7CH4 + 7 C2 H4 + 7C3 H6 15 NA6-C8 → 8 C4 H6 + 8 (2,2-C5 H10 ) + 8CH4 + 8 C2 H4 + 8 C3 H6 NA6-C9 → 0.6 C4 H6 + 0.6 (2,2-C5 H10 ) + 0.6CH4 + 0.6 C2 H4 + 0.6 C3 H6 NA5-C5 → 0.5 C4 H6 + 0.5CH4 + 0.5 C2 H4 + 0.5 C3 H6 NA5-C6 → 0.4 C4 H6 + 0.4 (2,2-C5 H10 ) + 0.4CH4 + 0.4 C2 H4 + 0.4 C3 H6 15 NA5-C7 → 7 C4 H6 + 7 (2,2-C5 H10 ) + 7CH4 + 7 C2 H4 + 7 C3 H6 15 NA5-C8 → 8 C4 H6 + 8 (2,2-C5 H10 ) + 8CH4 + 8 C2 H4 + 8 C3 H6 NA5-C9 + 0.6H2 → 0.6 C4 H6 + 0.6 (2,2-C5 H10 ) + 0.6CH4 + 0.6 C2 H4 + 0.6 C3 H6

5.21E+07 7.88E+08 2.83E+08 4.03E+08 2.63E+08 5.90E+08 2.51E+08 2.06E+08 4.03E+08

66 Paraffin cracking 67

Naphthene cracking

˛i

)

Ea (kJ/mol) 63 211 116 211 211 204 204 204 35 35 340

8.10E+08 287.75 1.01E+10 4.50E+08 1.50E+09 3.60E+08 6.01E+08 8.05E+08 1.51E+08 1.51E+09

287.75

OX: ortho-Xylene; NA6-Cn : cyclohexane where n = 6, 7, 8 or 9; NA5-Cn : cyclopentane where n = 5, 6, 7, 8 or 9; ˛i = 1, i = 1 − N, N: number of reactants.

Table A2 Naphtha catalytic reforming reactions and kinetics parameters.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25



Reactions

k0 (kmol/m3 s Pa

˛i

Dehydrogenation

NA6-C6 → C6 H6 + 3H2 NA6-C7 → (CH3 )-C6 H5 + 3H2 NA6-C8 → MX + 3H2 NA6-C8 → OX + 3H2 NA6-C8 → PX + 3H2 NA6-C8 → (C2 H5 )-C6 H5 + 3H2 NA6-C9 → (C3 H7 )-C6 H5 + 3H2

3.69E+08 1.01E+08 4.58E+09 1.57E+09 9.70E+08 1.04E+09 1.25E+10

162.12

Alkylcyclohexane

NA6-C6 + H2 → n-C6 H14 NA6-C7 + H2 → n-C7 H16 NA6-C8 + H2 → n-C8 H18 NA6-C9 + H2 → n-C9 H20 NA6-C6 + H2 → i-C6 H14 NA6-C7 + H2 → i-C7 H16 NA6-C8 + H2 → i-C8 H18 NA6-C9 + H2 → i-C9 H20

3.83E+07 4.34E+09 1.20E+09 8.41E+09 2.81E+09 1.52E+09 1.20E+10 8.41E+10

275.28

Dehydrocyclization to alkylcyclopentane

n-C5 H12 → NA5-C5 + H2 n-C6 H14 → NA5-C6 + H2 n-C7 H16 → NA5-C7 + H2 n-C8 H18 → NA5-C8 + H2 n-C9 H20 → NA5-C9 + H2 i-C5 H12 → NA5-C5 + H2 i-C6 H14 → NA5-C6 + H2 i-C7 H16 → NA5-C7 + H2 i-C8 H18 → NA5-C8 + H2 i-C9 H20 → NA5-C9 + H2

2.26E+10 7.44E+10 3.30E+11 3.96E+10 2.06E+11 9.02E+07 2.55E+08 2.04E+09 7.06E+09 2.06E+09

275.28

)

Ea (kJ/mol)

142

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143

Table A2 (Continued)



Reactions

k0 (kmol/m3 s Pa

˛i

1.02E+04 2.56E+04 1.30E+06 1.00E+05 6.40E+04 6.21E+04 2.81E+05 4.98E+06 1.06E+07 4.74E+07 4.68E+07 8.59E+06 2.85E+06 2.70E+07 9.03E+06

156.80

)

Ea (kJ/mol)

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Paraffin dehydrocyclization to aromatic

n-C6 H14 → C6 H6 + 4H2 n-C7 H16 → (CH3 )-C6 H5 + 4H2 n-C8 H18 → MX + 4H2 n-C8 H18 → OX + 4H2 n-C8 H18 → PX + 4H2 n-C8 H18 → (C2 H5 )-C6 H5 + 4H2 n-C9 H20 → (C3 H7 )-C6 H5 + 4H2 i-C6 H14 → C6 H6 + 4H2 i-C7 H16 → (CH3 )-C6 H5 + 4H2 i-C8 H18 → MX + 4H2 i-C8 H18 → OX + 4H2 i-C8 H18 → PX + 4H2 i-C8 H18 → (C2 H5 )-C6 H5 + 4H2 i-C9 H20 → (C3 H7 )-C6 H5 + 4H2 i-C10 H22 → (C4 H9 )-C6 H5 + 4H2

41 42 43 44

Naphthene Isomerization

NA5-C6 → NA6-C6 NA5-C7 → NA6-C7 NA5-C8 → NA6-C8 NA5-C9 → NA6-C9

3.65E+05 3.65E+05 1.57E+05 2.23E+05

197.96

Paraffin Isomerization

n-C4 H10 → i-C4 H10 n-C5 H12 → i-C5 H12 n-C6 H14 → i-C6 H14 n-C7 H16 → i-C7 H16 n-C8 H18 → i-C8 H18 n-C9 H20 → i-C9 H20

4.50E+10 2.91E+10 2.60E+10 2.31E+10 2.24E+10 2.52E+10

216.16

51 52 53 54 55 56

Aromatic Isomerization

OX → PX OX → MX MX → (C2 H5 )-C6 H5 OX → (C2 H5 )-C6 H5 (C2 H5 )-C6 H5 → PX MX → PX

8.88E+06 4.37E+07 4.50E+07 5.53E+07 8.67E+07 4.24E+06

176.59

57 58

Transalkylation

2(CH3 )-C6 H5 → C6 H6 + PX 2OX → (CH3 )-C6 H5 + (C3 H7 )-C6 H5

1.75E+04 5.28E+02

176.59

Paraffin cracking

C2 H6 + H2 → 2CH4 C3 H8 +H2 → CH4 + C2 H6 3n-C4 H10 + 3H2 → 2CH4 + 2 C2 H6 + 2 C3 H8 n-C5 H12 + H2 → 0.5n-C4 H10 + 0.5CH4 + 0.5 C2 H6 + 0.5 C3 H8 n-C6 H14 + H2 → 0.4n-C4 H10 + 0.4n-C5 H12 + 0.4CH4 + 0.4 C2 H6 + 0.4 C3 H8 15n-C7 H16 + 20H2 → 7n-C4 H10 + 7n-C5 H12 + 7CH4 + 7 C2 H6 + 7 C3 H8 15n-C8 H18 + 25H2 → 8n-C4 H10 + 8n-C5 H12 + 8CH4 + 8 C2 H6 + 8 C3 H8 n-C9 H20 + 2H2 → 0.6n-C4 H10 + 0.6n-C5 H12 + 0.6CH4 + 0.6 C2 H6 + 0.6 C3 H8 i-C4 H10 + 1.5H2 → 1.5CH4 + 0.5 C2 H6 + 0.5 C3 H8 i-C5 H12 + H2 → 0.5i-C4 H10 + 0.5CH4 + 0.5 C2 H6 + 0.5 C3 H8 i-C6 H14 + H2 → 0.4n-C4 H10 + 0.4n-C5 H12 + 0.4CH4 + 0.4 C2 H6 + 0.4 C3 H8 15i-C7 H16 + 20H2 → 7n-C4 H10 + 7n-C5 H12 + 7CH4 + 7 C2 H6 + 7 C3 H8 15i-C8 H18 + 25H2 → 8n-C4 H10 + 8n-C5 H12 + 8CH4 + 8 C2 H6 + 8 C3 H8 i-C9 H20 + 2H2 → 0.6n-C4 H10 + 0.6n-C5 H12 + 0.6CH4 + 0.6 C2 H6 + 0.6 C3 H8

2.74E+07 2.24E+08 2.73E+08 1.17E+08 1.88E+09 2.81E+09 4.00E+09 6.06E+09 5.01E+07 1.13E+08 4.00E+08 1.13E+08 1.07E+08 1.11E+08

287.75

Naphthene cracking

NA6-C6 + 2H2 → 0.4n-C4 H10 + 0.4n-C5 H12 + 0.4CH4 + 0.4 C2 H6 + 0.4 C3 H8 15NA6-C7 + 35H2 → 7n-C4 H10 + 7n-C5 H12 + 7CH4 + 7 C2 H6 + 7 C3 H8 15 NA6-C8 + 40H2 → 8n-C4 H10 + 8n-C5 H12 + 8CH4 + 8 C2 H6 + 8 C3 H8 NA6-C9 + 3H2 → 0.6n-C4 H10 + 0.6n-C5 H12 + 0.6CH4 + 0.6 C2 H6 + 0.6 C3 H8 NA5-C5 + 2H2 → 0.5n-C4 H10 + 0.5CH4 + 0.5 C2 H6 + 0.5 C3 H8 NA5-C6 + 2H2 → 0.4n-C4 H10 + 0.4n-C5 H12 + 0.4CH4 + 0.4 C2 H6 + 0.4 C3 H8 15 NA5-C7 + 35H2 → 7n-C4 H10 + 7n-C5 H12 + 7CH4 + 7 C2 H6 + 7 C3 H8 15 NA5-C8 + 40H2 → 8n-C4 H10 + 8n-C5 H12 + 8CH4 + 8 C2 H6 + 8 C3 H8 NA5-C9 + 3H2 → 0.6n-C4 H10 + 0.6n-C5 H12 + 0.6CH4 + 0.6 C2 H6 + 0.6 C3 H8

8.02E+10 4.59E+09 1.16E+09 2.89E+09 2.96E+10 1.11E+12 1.00E+11 9.06E+10 5.49E+10

287.75

Hydrodealkylation

(CH3 )-C6 H5 + H2 → C6 H6 + CH4 PX + H2 → (CH3 )-C6 H5 + CH4 (C2 H5 )-C6 H5 + H2 → C6 H6 + C2 H6 (C3 H7 )-C6 H5 + H2 → OX + CH4 (C3 H7 )-C6 H5 + H2 → (CH3 )-C6 H5 + C2 H6

1.00E+03 1.73E+03 1.25E+02 2.05E+02 1.00E+02

148.99

45 46 47 48 49 50

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

OX: ortho-Xylene; PX: para-Xylene; MX: meta-Xylene; NA6-Cn : cyclohexane where n = 6, 7, 8 or 9; NA5-Cn : cyclopentane where n = 5, 6, 7, 8 or 9; ˛i = 1, i = 1 − N, N: number of reactants.

References Belohlav Z, Zamostny P, Herink T. The kinetic model of thermal cracking for olefins production. Chem Eng Process Process Intensif 2003;42:461–73. Bieser HJ. Process for separating normal paraffins. United States patent US4006197 A; 1977.

Broughton DB. Process for separating normal and isoparaffins. United States patent US4036745 A; 1977. Chang B-M, Choi S, Choi A-S, Kang S-C, Kim S-W, Ku M-S, Lee S-J. Process for separating normal paraffins from hydrocarbons and applications for the separated hydrocarbons. United States patent US6870073 B2; 2005.

T.X. Do et al. / Computers and Chemical Engineering 74 (2015) 128–143 Gyngazova MS, Kravtsov AV, Ivanchina ED, Korolenko MV, Chekantsev NV. Reactor modeling and simulation of moving-bed catalytic reforming process. Chem Eng J 2011;176–177:134–43. Hou W, Su H, Hu Y, Chu J. Modeling, simulation and optimization of a whole industrial catalytic naphtha reforming process on Aspen plus platform. Chin J Chem Eng 2006;14:584–91. Iranshahi D, Rafiei R, Jafari M, Amiri S, Karimi M, Rahimpour MR. Applying new kinetic and deactivation models in simulation of a novel thermally coupled reactor in continuous catalytic regenerative naphtha process. Chem Eng J 2013;229:153–76. Jia N, Moore RG, Mehta SA, Ursenbach MG. Kinetic modeling of thermal cracking reactions. Fuel 2009;88:1376–82. Kawajiri Y, Biegler LT. Large scale optimization strategies for zone configuration of simulated moving beds. Comput Chem Eng 2008;32:135–44. Keyvanloo K, Sedighi M, Towfighi J. Genetic algorithm model development for prediction of main products in thermal cracking of naphtha: comparison with kinetic modeling. Chem Eng J 2012;209:255–62. Kulprathipanja S. Adsorptive separation process for the purification of heavy normal paraffins with non-normal hydrocarbon pre-pulse stream. United States patent US4992618 A; 1991. Lim Y-I. Optimal flushing flow rates in para-xylene simulated moving-bed considering geometric factor of dead volume. Adsorption 2012;18:469–82. Liu J-C, Shen B-X, Wang D-Q, Dong J-H. Separating group compositions in naphtha by adsorption and solvent extraction to improve olefin yields of steam cracking process. J Petrol Sci Eng 2009;66:156–60. Liu J-T, Ni T-Y, Liu J-C, Shen B-X. Adsorption separation of normal paraffins from naphtha by simulated moving bed based on molucule-scale management. Petrol Process Petrochem 2014;45:71–6. Meidanshahi V, Bahmanpour AM, Iranshahi D, Rahimpour MR. Theoretical investigation of aromatics production enhancement in thermal coupling of naphtha reforming and hydrodealkylation of toluene. Chem Eng Process Process Intensif 2011;50:893–903. Mun S. Improving performance of a tandem simulated moving bed process for sugar separation by making a difference in the adsorbents and the column lengths of the two subordinate simulated moving bed units. J Chromatogr A 2013;1277:48–57. Niaei A, Towfighi J, Sadrameli SM, Karimzadeh R. The combined simulation of heat transfer and pyrolysis reactions in industrial cracking furnaces. Appl Therm Eng 2004;24:2251–65.

143

Ragil K, Bailly M, Jullian S, Clause O. Process for chromatographic separation of a C5–C8 feed or an intermediate feed into three effluents, respectively rich in straight chain, mono-branched and multi-branched paraffins. United States patent US6353144 B1; 2002. Rahimpour MR, Jafari M, Iranshahi D. Progress in catalytic naphtha reforming process: a review. Appl Energy 2013;109:79–93. Ren T, Daniëls B, Patel MK, Blok K. Petrochemicals from oil, natural gas, coal and biomass: production costs in 2030–2050. Resour Conserv Recycl 2009;53:653–63. Ren XH, Bertmer M, Stapf S, Demco DE, Blümich B, Kern C, et al. Deactivation and regeneration of a naphtha reforming catalyst. Appl Catal A-Gen 2002;228: 39–52. Rodríguez MA, Ancheyta J. Detailed description of kinetic and reactor modeling for naphtha catalytic reforming. Fuel 2011;90:3492–508. Sá Gomes P, Lamia N, Rodrigues AE. Design of a gas phase simulated moving bed for propane/propylene separation. Chem Eng Sci 2009;64:1336–57. Sadrameli SM, Green AES. Systematics and modeling representations of naphtha thermal cracking for olefin production. J Anal Appl Pyrolysis 2005;73: 305–13. Seifzadeh Haghighi S, Rahimpour MR, Raeissi S, Dehghani O. Investigation of ethylene production in naphtha thermal cracking plant in presence of steam and carbon dioxide. Chem Eng J 2013;228:1158–67. Sutanto PS, Lim Y-I, Lee J. Bed-line flushing and optimization in simulated movingbed recovery of para-xylene. Sep Purif Technol 2012;96:168–81. Taskar U, Riggs JB. Modeling and optimization of a semiregenerative catalytic naphtha reformer. AlChE J 1997;43:740–53. UOP. http://www.uop.com/maxene-process-flow-scheme/ [accessed 20.05.14]. Weifeng HOU, Hongye SU, Shengjing MU, Jian CHU. Multiobjective optimization of the industrial naphtha catalytic reforming process. Chin J Chem Eng 2007;15:75–80. Wiehe IA. A phase-separation kinetic model for coke formation. Ind Eng Chem Res 1993;32:2447–54. Yao C, Tang S, Yao H-M, Tade MO. Continuous prediction technique for fast determination of cyclic steady state in simulated moving bed process. Comput Chem Eng 2013;58:298–304. Zagoruiko AN, Belyi AS, Smolikov MD, Noskov AS. Unsteady-state kinetic simulation of naphtha reforming and coke combustion processes in the fixed and moving catalyst beds. Catal Today 2014;220–222:168–77.