Simulação do Processo de Produção de Cumeno

See discussions stats and author profiles for this publication at httpswwwresearchgatenetpublication262774974 Economic Design and Optimization of ZeoliteBased Cumene Production Plant Article in Chemical Engineering Communications May 2014 DOI 101080009864452013806312 CITATIONS 9 READS 25198 4 authors including Hamid Reza Norourzi Amirkabir University of Technology 34 PUBLICATIONS 572 CITATIONS SEE PROFILE Bahram Haddadi TU Wien 54 PUBLICATIONS 263 CITATIONS SEE PROFILE Navid Mostoufi University of Tehran 328 PUBLICATIONS 6059 CITATIONS SEE PROFILE All content following this page was uploaded by Hamid Reza Norourzi on 17 October 2016 The user has requested enhancement of the downloaded file This article was downloaded by University of Saskatchewan Library On 04 June 2014 At 1134 Publisher Taylor Francis Informa Ltd Registered in England and Wales Registered Number 1072954 Registered office Mortimer House 3741 Mortimer Street London W1T 3JH UK Chemical Engineering Communications Publication details including instructions for authors and subscription information httpwwwtandfonlinecomloigcec20 ECONOMIC DESIGN AND OPTIMIZATION OF ZEOLITEBASED CUMENE PRODUCTION PLANT H R Norouzi a M A Hasani a B HaddadiSisakht a N Mostoufi a a Process Design and Simulation Research Centre Oil and Gas Centre of Excellence School of Chemical Engineering College of Engineering University of Tehran Tehran Iran Accepted author version posted online 14 Apr 2014Published online 29 May 2014 To cite this article H R Norouzi M A Hasani B HaddadiSisakht N Mostoufi 2014 ECONOMIC DESIGN AND OPTIMIZATION OF ZEOLITEBASED CUMENE PRODUCTION PLANT Chemical Engineering Communications 20110 12701293 DOI 101080009864452013806312 To link to this article httpdxdoiorg101080009864452013806312 PLEASE SCROLL DOWN FOR ARTICLE Taylor Francis makes every effort to ensure the accuracy of all the information the Content contained in the publications on our platform However Taylor Francis our agents and our licensors make no representations or warranties whatsoever as to the accuracy completeness or suitability for any purpose of the Content Any opinions and views expressed in this publication are the opinions and views of the authors and are not the views of or endorsed by Taylor Francis The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information Taylor and Francis shall not be liable for any losses actions claims proceedings demands costs expenses damages and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with in relation to or arising out of the use of the Content This article may be used for research teaching and private study purposes Any substantial or systematic reproduction redistribution reselling loan sublicensing systematic supply or distribution in any form to anyone is expressly forbidden Terms Conditions of access and use can be found at httpwwwtandfonlinecompageterms andconditions Economic Design and Optimization of ZeoliteBased Cumene Production Plant H R NOROUZI M A HASANI B HADDADISISAKHT AND N MOSTOUFI Process Design and Simulation Research Centre Oil and Gas Centre of Excellence School of Chemical Engineering College of Engineering University of Tehran Tehran Iran The aim of this work is intensification of an industrialscale production process of cumene to obtain higher profitability and reduce the energy requirements of the process In the first step two topological changes were made in the reactor configur ation and benzene separation column which led to a considerable reduction in utility requirements and equipment sizes in the process In the second step parametric optimization was carried out by a statistical method full factorial design to adjust the process parameters In the final step a pinch analysis was done to reduce utility requirements of the process A mathematical model was developed based on the statistical analysis which was then used to obtain optimized conditions of the process The capital investment of the process remained almost intact around 45 million and the utility requirements reduced from 244 to 149 millionyr These changes greatly enhanced the profitability of the process for which the net present value was increased from 082 to 453 million Keywords Cumene Economic optimization Factorial design Statistical method Zeolite catalyst Introduction Finding a configuration and operational conditions of a process in which a chemical is produced in a reliable and economical manner with low energy consumption low initial capital investment low or negligible environmental impact and high yield is the goal of plant design engineers Optimization of a process is conducted when the flow sheet of the base case and detailed mass and energy balances of the process are available Moreover the economic aspects of the plant including size of equipment capital investment and utility requirements must be evaluated Seider et al 2010 It requires a great amount of knowledge experience and hard work In this work intensification of the liquidphase alkylation of benzene with propylene was carried out to make the process more effective and thus more profitable Address correspondence to N Mostoufi Process Design and Simulation Research Centre Oil and Gas Centre of Excellence School of Chemical Engineering College of Engineering University of Tehran PO Box 111554563 Tehran Iran Email mostoufiutacir Color versions of one or more of the figures in the article can be found online at www tandfonlinecomgcec Chem Eng Comm 20112701293 2014 Copyright Taylor Francis Group LLC ISSN 00986445 print15635201 online DOI 101080009864452013806312 1270 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 The vast majority of cumene isopropylbenzene manufactured worldwide is used in the production of phenol and acetone Schmidt 2005 Reactions occur in the pres ence of different catalysts such as solid phosphoric acid SPA aluminum chloride AlCl3 and zeolites The first license for cumene production belongs to the corpor ation UOP in this process the reaction occurs in the gas phase in the presence of SPA catalyst Stefanidakis and Gwyn 1977 This catalyst had been originally used for converting light olefins into gasoline components Gary and Handwerk 1994 The side reactions occurring when using SPA catalyst produce polyisopropylbenzenes PIPB which cannot be converted back to cumene by the transalkylation reaction These heavy components must be removed in the following separation section The problem in this process is the low overall yield high temperature of the reactor and shortlife and unregenerable catalyst which has hazardous environmental effects Another license belongs to Monsanto and Kellogg which uses a mixture of AlCl3 and HCl as the catalyst in a homogeneous liquidphase reaction of benzene and propylene The transalkylation of PIPB is also possible in this process Thus this process has a high yield However corrosion of pipes and equipment are still a problem Canfield et al 1986 Liquidphase and zeolitebased catalytic processes were developed by CDTech Pohl and Ram 2005 Sy et al 1993 MobilBadger and UOP Hwang and Chen 2010 This catalyst is regenerable and transalkylation reactions also occur over this catalyst Among different the zeolitebased processes the technology developed by CDTech uses reactive distillation while the others use two fixedbed reactors for performing alkylation and transalkylation reactions Many researchers have investigated the optimization of the abovementioned processes Luyben 2010 studied the gasphase process on the SPA catalyst and econ omically optimized the process with a simple model Total annualized cost TAC which accounts for initial investment with a threeyear payback time operating costs and raw material costs was the objective function to be minimized in that study Pathak et al 2011 studied the vaporphase and reactive distillation RD process of CDTech for cumene They also used TAC as the objective function and concluded that the reactive distillation process is 47 cheaper than the conventional gasphase process Lei et al 2009 considered a zeolitebased process that uses two fixedbed reactors They started with side stream draw from a distillation column to reduce energy consumption They further modified the alkylation reactor and replaced it with a reactive distillation column and concluded that the energy consumption and total investment can be reduced in this way However they did not include any econ omic evaluation in the optimization procedure Norouzi and Fatemi 2012 studied gasphase production of cumene on SPA catalyst and developed an economic model that comprised initial investment operating costs labor costs raw material costs revenue plant life and inflation among other factors The net present value NPV was chosen as the objective function and a statistical procedure response surface method was chosen to adjust operational conditions and to find the global optimum of the process to maximize the NPV In the present study liquidphase alkylation of benzene on a zeolite catalyst was investigated This process corresponds to the process proposed by UOP and Mobil Badger Hwang and Chen 2010 which uses two fixedbed reactors and a separation train to recover the product A full economic model was developed for optimization This model includes grass root capital investment operating and raw material costs revenue plant life taxes and inflation and was adjusted for the year 2011 The ZeoliteBased Cumene Production Plant 1271 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 optimization procedure was divided into two parts First some changes were made in the topology of the process to reduce the utility requirements and to enhance the NPV of the process Second full factorial design which is a statistical method was used to optimize the operational conditions of the process to obtain a higher NPV Process Description Cumene is produced by alkylation of benzene with propylene on a proper catalyst like zeolitebased catalysts such as β Y ZSM12 and MCM22 Corma et al 2000 Dimian and Bildea 2008 Han et al 2001 Perego et al 1996 The reactions occur in the liquid phase at a temperature range that yields complete conversion of propylene and under a proper pressure to maintain the reactants in the liquid phase throughout the reactor Different licenses that use liquidphase reactors consider a series of two reactors In the first reactor known as the alkylation reactor the following reactions take place C3H6 Propylene C6H6 benzene k1 C9H12 cumene C3H6 Propylene C9H12 cumene k2 C9H12 DIPB The first reaction yields the main product of the process The produced cumene may further react with propylene to produce DIPB or higher propylbenzenes Here for simplicity it was assumed that the only byproduct is DIPB The reaction rate constants are listed in Table I Dimian and Bildea 2008 It was shown that the alkylation reactions follow the EleyRideal kinetic model in which the adsorption of propylene is the ratedetermining step Corma et al 2000 In the range of operating conditions of this process this kinetic law reduces to a firstorder reaction as shown in Table I Dimian and Bildea 2008 The reactor is an adiabatic fixed bed of catalyst pellets with the inlet temperature range of 150 to 200C The pressure is maintained high enough to ensure that the boiling point temperature of the solution is at least 20C higher than the temperature anywhere else in the reactor Therefore the pressure has to be maintained between 25 and 35 MPa depending on the composition of the mixture and the temperature of the reactor The benzene to propylene mole ratio BP is maintained at more than 4 in the reactor Perego and Ingallina 2002 An excess amount of benzene is required in the reactor for the following reasons to absorb heat generated by exothermic reactions in the rector to enhance the selectivity of the cumene reaction over the DIPB reaction and to suppress the reaction between propylene molecules which produces higher linear hydrocarbons The boiling temperature of these hydrocarbons is close to that of benzene thus separating them from benzene to avoid accumulation becomes problematic In the second adiabatic reactor known as the transalkylation reactor the produced DIPB is converted back to cumene in a reaction with benzene in the liquid phase C6H6 benzene C12H18 DIPB k3k4 2 C9H12 cumene 3 The inlet temperature for this reactor is about 220240C and the benzene to DIPB mole ratio is between 5 and 10 The reaction rate constants of these reactions are listed in Table I Pathak et al 2011 Utilization of the second reactor enhances the overall yield of the alkylation process and hence the profitability of the whole plant A typical process flow diagram of cumene production on a zeolite catalyst is shown in Figure 1 Fresh and pure benzene at the molar rate of 100 kmolh is fed to the vessel V1 mixed with the recycled benzene coming from the separation section and pumped to 36 MPa The recycled benzene is utilized to maintain the benzene to propylene mole ratio at the desired value The refinerygrade propylene at the molar flow rate of 105 kmolh 5 kmolh propane as impurity is pumped to 36 MPa and mixed with the benzene stream The mixed feed stream is heated to 170C by the hot outlet product stream of the alkylation reactor in the heat exchanger E1 The previously mentioned reactions occur Equations 1 and 2 in the alkylation reactor R1 Both reactions are highly exothermic thus the temperature of the reactants increases gradually along the reactor length The inlet temperature and BP must be maintained in a way that no vapor is generated in the reactor Complete conversion of propylene is desired due to two important facts first the unreacted propylene would escape from the process in the propane column C1 with propane which is Figure 1 Process flow diagram of liquid phase cumene production for the base case Table II Stream table of process flow diagram shown in Figure 1 Stream Fresh benzene Fresh propylene Mixed feed Alkylation effluents Fuel gas C2feed Benzene ovlhd C2 bottoms Cumene DIPB Trans feed Trans effluents Temperature T C 250 250 1700 2575 567 1133 817 1613 1518 2134 2400 2403 Pressure MPa 010 115 354 354 150 013 010 012 010 011 350 350 Vapor fraction 000 000 000 000 100 072 000 000 000 000 000 000 Mole flow kmolh 1000 1050 6050 5050 510 53237 42500 10737 9990 747 3247 3247 Mass flow kgh 78114 44285 445474 445474 2282 475504 343274 132230 120112 12118 32312 32312 Comp mole flow kmolh Benzene 100 100 47476 37841 0099 3983 3982 0144 0144 000 2342 200 Propylene 5 5 1000 0002 0002 00 00 00 00 00 00 00 Propane 0 0 25237 500 500 00 00 00 00 0003 1580 8427 Cumene 0001 11794 1264 2683 9954 9953 7466 7467 4043 DIPBs 3649 7692 0001 0225 0225 economically unfavorable second benzene and propylene are reacted with a 11 mole ratio and the inlet flow rate of benzene to the process should be reduced in the fresh benzene stream which in turn reduces the overall productivity of the plant The reactor effluent is passed through the heat exchangers E1 and E2 and then is fed to the propane column in which propane is completely removed from other components The condenser E3 is partial and its pressure is 15 MPa The overhead stream has fuel value and can be used in the fuel cycle of the process The bottom liquid leaving the propane column is mixed with the product stream of transalkyla tion reactor R2 trans effluents and is fed to the benzene column C2 in which benzene is collected in the overhead and recycled back to both reactors In this column the concentration of benzene in the bottom stream must be negligible to ensure high purity of cumene in the next column Bottoms of the benzene column are sent to the cumene column C3 where cumene product is obtained overhead with a minimum purity of 995 The bottoms of this column containing mostly DIPB are mixed with the recycled benzene from the benzene column transalkylation recycle and fed to the transalkylation reactor R2 In this reactor DIPB is converted back to cumene The effluent from this reactor is then sent to the benzene column C2 A stream table of the base case described above is supplied in Table II The number of trays and reflux ratios of columns were determined according to the following procedure The minimum reflux ratio and minimum number of trays were determined using the shortcut method Seider et al 2010 The actual number of trays and feed stage were determined based on the actual reflux ratio which was chosen to be 125 times the minimum reflux ratio Since the shortcut method does not provide the exact solution of the column an exact simulation of the column was made using the rigorous distillation method After simulating the column with the number of trays and the feed stage the reflux ratio was adjusted to achieve the desired purification Economic Model As the main goal of this study is the economic optimization of the cumene production process the consideration of a comprehensive economic model that takes into account all effective parameters is essential This economic model requires infor mation about the fixed capital investment the annual utility requirements the raw material costs the operating labor costs and the revenue of the process The methods for evaluating all these factors are introduced in the following Capital Investment There are methods for estimating the capital cost of a process at the preliminary design stage such as the Lang factor the sixtenths rule and the bare module cost Turton et al 1998 The bare module cost is a more detailed cost estimation method that considers the construction materials the operating pressure and the special configuration of equipment in cost estimation This technique relates all costs to the purchased cost of equipment evaluated for some base conditions The bare module cost of equipment is determined by the following relation C0 BM ¼ CPF0 BM ð4Þ ZeoliteBased Cumene Production Plant 1275 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 To estimate the bare module cost of all major equipment in the cumene process it is necessary to know the size the construction material the configuration and the operating pressure of all equipment Detailed information about the construction materials and the operating conditions of the major equipment in the process base case is listed in Table III The bare module correlations for estimating the equipment costs belong to past data on purchased equipment and they need to be updated according to changes in economic conditions such as inflation The Chemical Engineering Plant Cost Index CEPCI was used to account for the effect of inflation CBM02 CBM01 CEPCI2CEPCI1 where subscripts 1 and 2 refer to the base time and the present time of cost estimation respectively The correlations used for estimating equipment costs belong to 1996 with a CEPCI of 382 The CEPCI for the year 2011 was 5857 and all costs were adjusted according to this value Utility Cost The fuel price fluctuates more than equipment costs and in fact it does not follow a general trend like the CEPCI The cost of utilities in a plant is greatly influenced by the price of fuel In general the utility cost is linked to two separate variable costs inflation CEPCI and fuel cost To reflect this dual dependence the following twofactor equation was used Ulrich and Vasudevan 2006 Cu aCEPCI bCf The values of a and b for different utilities are presented by Ulrich and Vasudevan 2006 The price of fuel is based on the average price of natural gas in 2011 in the US obtained from the US Department of Energy 2011 The industrial price for natural gas is 491000 ft3 Assuming 381 MJstdm3 heating value for natural gas Perry et al 1999 the price of natural gas was estimated to be 454GJ According to the price of natural gas as the reference fuel costs of different utilities were calculated and are listed in Table IV This table also includes the values of a and b for different utilities Available utilities differ from one process to another Here low pressure LPS medium pressure MPS and high pressure HPS steams at 05 10 and 41 MPa gauge respectively and cooling water entering at 30C and leaving at 45C were considered as well as electricity for driving pumps Raw Material and Product Prices Prices of raw materials benzene and propylene and products cumene were obtained from ICIS 2011 The prices of benzene propylene and cumene were considered as 0947 0970 and 1050kg respectively for the year 2011 Heat value of the fuel produced in the fuel gas stream was estimated according to the price of natural gas 454GJ The heating values for burning propane propylene benzene cumene and DIPB were calculated to be 918 868 1454 2271 and 3098kmol 0208 0206 0186 0189 and 0191kg respectively Table III Operating conditions and sizing data for main equipment in the process Uoveral Wm2 K1b Material of constructionc Pressure Utility type Tube side Shell side Heat Exchangersa E1 Heat ex 700 CSCS 355 354 None E2 Cooler 700 CSCS 353 4 CW E3 Condenser 450 CSCS 15 4 CW E4 Reboiler 750 CSCS 152 43 HPS E5 Condenser 450 CSCS 1 4 CW E6 Reboiler 750 CSCS 12 11 MPS E7 Condenser 450 CSCS 1 4 CW E8 Reboiler 900 CSCS 11 43 HPS E9 Heater 800 CSCS 35 43 HPS Residence time mind Configuration materiale LD Vessels V801 10 Horizontal CS 3 V802 10 Horizontal CS 3 V803 10 Horizontal CS 3 V804 10 Horizontal CS 3 Towersf C1 Reflux ratio 5 pressure 15 MPa vessel material carbon steel 12 sieve trays tray spacing 61 cm 80 active area C2 Reflux ratio 025 pressure 01 MPa vessel material carbon steel 18 sieve trays tray spacing 61 cm 80 active area C3 Reflux ratio 07 pressure 01 MPa vessel material carbon steel 14 sieve trays tray spacing 61 cm 80 active area Reactors Alkylation reactor Length 7 m diameter 13 m vessel material carbon steel Transalkylation reactor Length 13 m diameter 07 m vessel material carbon steel Pumps For all pumps 60 hydraulic efficiency was considered construction material carbon steel a 20 overdesign b From Gas Processors Suppliers Association 1998 c From Stefanidakis and Gwyn 1997 d From Branan 1995 e From Turton et al 1998 f From Seider et al 2010 Table IV Utility cost estimation based on Ulrich and Vasudevan 2006 for the year 2011 Utility type GJ a b Electricity 20 130 104 0010 LP steam 5 MPa gauge 821 347 106 0004 MP steam 10 MPa gauge 954 347 106 0004 HP steam 42 MPa gauge 1204 347 106 0004 CW 30 to 45C 13 700 104 0003 Other Costs Other costs of production are operating labor maintenance cost and fixed operating cost The techniques used to estimate these costs are described by Turton et al 1998 The fixed operating costs are independent of changes in the production rate and include local taxes insurance and depreciation which are charged at a constant rate even when the plant is not in operation In the calculation of the fixed operating costs an average tax of 30 was assumed and the straightline depreciation method for a period of 10 years was used Table V Details of capital investment and utility requirements of different cases in the optimization procedure Base case 4Bed reactor Twocolumn case Optimized case Optimized case after pinch analysis Capital investment million Total 4481 388 426 422 450 Heat ex 1534 1405 1701 1685 1965 Towers 0954 0738 0795 0783 0783 Reactors 0483 0503 0503 0511 0511 Vessels 0637 0498 0503 0502 0502 Pumps 0837 0735 0758 0739 0739 Utility requirements million yr Total 2440 190 175 168 149 Steam 2053 1597 1466 1414 125 CW 0305 0253 0232 0215 0193 Electricity 0082 0048 0052 0051 0051 Catalyst 004 004 004 0032 Revenue million yr Total 105348 10544 10540 10542 10542 Cumene 104954 10505 1050 10523 10523 Fuel prod 0394 0391 0392 0393 0393 Production cost kg 1025 1018 1017 1017 1015 COM million yr 10253 10187 1017 1017 1015 NPV million 082 295 356 416 453 Present value ratio 118 176 184 199 201 Payback time years 33 18 19 19 19 Economic Status of the Base Case The calculation sequence for economic evaluation of the process is as follows Materials and energy balances were obtained for all operating units in the process until a final solution was obtained The results were heat duties work duties stream flow rates etc The results of materials and energy balances were applied in design equations with assumptions presented in Table III and all equipment was sized The results of materials and energy balances as well as sizes of all equipment were used in the economic model to determine different economic parameters such as utility costs equipment costs and NPV The process described above was considered to be the base case from which the optimization started Before proceeding to the optimization of the base case its economic status was evaluated to have a criterion for comparison According to the material balance of this process and assuming a stream factor of 095 the annual production rate of cumene is 99960 metric ton The grass root capital investment of the process obtained from the economic model was estimated to be 448 millionyr for the year 2011 The utility requirements of the process based on the energy balances were estimated to be 244 millionyr Almost 841 of utility requirements belongs to the steam followed by the cooling water and the electricity which are 125 and 34 respectively The net present value and cost of manufacturing COM of the base case are 082 million and 10253 millionyr respectively Details of capital investment utility requirements revenue and economic parameters of the base case are presented in Table V By inspecting the process flow diagram of the process two loops of materials and a onethrough flow Main Flow are distinguishable The first loop is the Figure 2 Cost breakdown of utilities requirements in each material processing flow in the plant Main flow Loop1 and Loop2 ZeoliteBased Cumene Production Plant 1279 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 benzene recycle to the alkylation reactor and the second loop is the benzene recycle to the transalkylation reactor Materials flowing in the first loop Loop1 are passed through the following utilitydemanding units P1 E2 E3 E4 E5 and E6 Also materials flowing in the second loop Loop2 are passed through P6 E9 E5 E6 and E8 The cost breakdown of each type of utility which is required for processing the materials flowing in each loop is shown in Figure 2 The main flow in this figure refers to the stream flow in which materials enter the process once and leave it as the final product As can be seen in this figure the contribution of Loop1 to the utility requirements in the whole process is much greater than that of Loop2 and Main Flow This shows that the benzene recycle to the alkylation reactor Loop1 can greatly affect the economics of the process Thus it is an important variable that should be adjusted in the process to obtain better profitability However adjustment of the second benzene recycle Loop2 seems to be less important In addition change in important parameters of the process such as reactor inlet temperature and volume to achieve higher conversion and selectivity and separation conditions can also change the utility requirement of the main flow in the process Economic Optimization Procedure There are two kinds of optimization procedures namely topologyconfiguration changes and parametricoperating condition changes Seider et al 2010 Turton et al 1998 Topological changes are carried out prior to the parametric because they have a greater influence on the profitability of the process When the topology of the process is fixed parametric optimization can be easily carried out by a proper strategy Based on the discussion of the utility cost breakdown of the process in the previous section Loop1 and Main Flow are the most important parts of the process that should be considered first in optimization The optimization procedure starts with topological changes in the process in which modifications are made to reduce the utility requirements of Loop1 and Main Flow in the process The procedure ends with parametric changes in which the most profitable operating conditions would be determined It is worth mentioning that since the overall yield of this process is 100 all the entering materials are converted into the desired product there is no need of extra space to enhance it by changing either topology or operating conditions Thus the economic condition of the process can be enhanced only through reducing the utility requirements and capital investment of the process Topological Changes Reactor Topology As discussed before to control the reactor temperature and to prevent production of linear hydrocarbons reaction between propylene molecules the BP should not fall below a certain threshold Thus benzene is recycled back to the alkylation reactor to achieve such conditions However this recycling requires larger equipment and more utility usage The calculations above showed that this recycle Loop1 includes a large portion in the utility requirement Thus the reduction of this stream can greatly influ ence both capital investment smaller equipment and utility requirements There are different configurations proposed for the alkylation reactor in the literature Hwang and Chen 2010 Schmidt 2005 In one configuration both propylene and benzene are premixed heated and enter the reactor In another configuration the propylene 1280 H R Norouzi et al Downloaded by University of Saskatchewan Library at 1134 04 June 2014 stream is divided into several parts eg four and injected stagewise along the reactor The reactor may have interstage cooling to keep the temperature below the bubble point When propylene is reduced by onefourth the required mole flow of benzene in the reactor inlet to obtain the same BP is also reduced On the other hand since the total amount of benzene in the reactor is reduced the temperature rise in the reactor is enhanced and interstage cooling is needed to cool down reacting components For the first change a stagewise propylene injection was employed to test whether or not it would change the total profitability of the process The configur ation illustrated in Figure 3 was used instead of the original one The new configur ation consists of four catalytic beds and is called the fourbed reactor hereafter To keep the temperature of the reactor below the bubble point an intercooler was used after the second bed In the fourbed reactor the flow rate of benzene entering the reactor was reduced from 500 to 225 kmolh However the BP was increased from 5 to 9 which is favorable to the process A minimum value of 225 kmolh benzene is required to maintain the temperature below the boiling point throughout the reactor Temperature and BP profiles along the reactor for the base case and the fourbed reactor are shown in Figure 4a and 4b respectively In Figure 4a for the base case the temperature increases along the reactor as the reactions proceed In the fourbed reactor the temperature increases gradually along the bed and drops at the end of each bed where the fresh and cold propylene is injected into the bed The fresh propylene cannot absorb all the heat generated by the reaction thus an interstage cooler is required to reduce the reactor temperature The boiling points in the inlet and outlet of each bed are also illustrated in this figure Figure 4a demonstrates that the temperature is well below the bubble point of the mixture Figure 3 Alkylation reactor with four catalytic beds and an interstage cooler and side feeding of propylene ZeoliteBased Cumene Production Plant 1281 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 throughout the reactor BP is shown along the reactor in Figure 4b this ratio increases along the reactor with the consumption of propylene As can be seen in this figure this value is always greater than the minimum allowable BP in the reactor The overall conversion of propylene in both reactor configurations is nearly complete The economic condition of the process is enhanced considerably with the above mentioned modification to the reactor topology As expected the utility consumption of the process is reduced from 244 to 19 millionyr Moreover due to the Figure 4 Temperature a and BP ratio b profiles along the reactor for the base case and the fourbed reactor 1282 H R Norouzi et al Downloaded by University of Saskatchewan Library at 1134 04 June 2014 reduction in the amount of materials processed in Loop1 the size of equipment P1 P3 E1 E2 E4 E5 E6 C1 C2 etc is reduced This leads to less total investment for the process from 448 to 388 million even though an interstage cooler is added in the fourbed reactor configuration All these effects are reflected in the profitability of the process the NPV of the process is increased from 082 to 295 million and payback time is reduced from 33 to 18 years More details about the capital investment utility requirements revenue and economic parameters of the new plant are presented in Table V Benzene Column Topology The topology of the flow sheet can still be changed in order to lower the utility requirement of the process and enhance its economic condition Hot and cold utilities constitute the majority of the utility requirement more than 185 out of 19 millionyr Among the columns in this process the topology of the benzene column C2 can be modified to reduce the utility requirement A large amount of materials enters this column 25000 kgh which requires large equipment and large utility usage The condenser and reboiler duties are 1023 and 387 GJh respectively which impose 450000yr utility cost on the process A reduction in the utility requirement of this tower can enhance the economic condition of the process The old and new configurations for this column are shown in Figure 5a and 5b The liquid stream entering the column C2Feed is split into two distinct streams which are fed to columns C2 and C2 in the new configuration The split fraction is defined as SF Flow rate of split 1Flow rate of C 2 Feed 7 Figure 5 a Original benzene column and b the new configuration for benzene column Operating pressures of C2 and C2 are 10 and 80 bar respectively These values are selected in a way that the reboiler temperature of C2 becomes lower by at least 10C than the condenser temperature of C2 Moreover since the HP steam is available at 254C the reboiler temperature of C2 should not exceed 244C Since the new configuration of the benzene column does not change the operating conditions and the flow rates of the streams in other parts of the entire process and indeed the capital investment and operating cost of the rest of the process profitability analysis of the new separation configuration should be focused on the benzene column alone It was assumed that both proposed configurations have the same operating lives as the plant life ie 10 years However the operating costs and the initial investment of these configurations are essentially different Many criteria are available to evaluate the profitability of the new configuration over the old one Here the net present value of the benzene separation section was chosen as the criterion of analysis The net present value returns all the investments fixed capital and operating costs paid during the plant life converted back to the present time Thus it reflects the effect of time including plant life and construction periods on the value of money The net present value for the benzene separation section assuming 2 years for startup 10 years plant life and 10 interest rate can be estimated by the following equation Turton et al 1998 NPV FCITM Oprt Cost PA 01 10PF 01 1 8 In which FCITM is the total module cost Optr Cost is the annual operating cost of the section including utility operating labor maintenance etc The expression in the first parenthesis of Equation 8 returns the sum of 10 years annual operating cost back to the year of startup equal to 6145 and the expression in the second parentheses returns this sum to the present time equal to 08264 Turton et al 1998 NPV FCITM 508123 Utility 018 FCITM 273 Labour 9 There are different factors that affect the NPV of the benzene column the split fraction of the C2 Feed SF the operating pressure of columns and the condensation rate of C2 The operating pressure of the columns changes the relative volatility of the components and the condensation rate changes the vapor and liquid flows in both columns and the size and utility requirements of both columns Of these three the operating pressures of both columns are kept fixed According to the requirements described above the temperatures of both columns also cannot be changed noticeably However the other two factors SF and condensation rate were tested to find an optimum condition for this column The NPV of the new benzene column configuration versus condensation rate at different split fractions is shown in Figure 6 The NPV of the original benzene column is 39 million The shaded rectangle of this figure shows the operating conditions at which improved economic conditions of the benzene column were obtained It can be seen in this figure that better economic conditions are obtained for split fractions 05 and 06 The proper selection of SF mainly depends on the feed composition and the place where the product streams are drawn from these columns When the feed stream is to be distributed between these two columns it is better to set a greater value of the flow rate to the top column C2 in which the pressure is lower and relative volatility is high Moreover this column requires cheap cooling water within the condenser while the bottom column needs expensive steam for evaporation Better economics would be obtained if the condensation rate is reduced at constant SF Reducing the condensation rate decreases the utility requirements of both the reboiler of C2þ and the condenser of C2 However a minimum condensation rate exists at which the two columns can together perform the desired task at each SF and the con densation rate cannot be reduced below this minimum For the current case the condensation rate of 90 kmolh and SF ¼ 06 were chosen for the new configuration of the benzene column The operating conditions and connections between these two columns are shown in Figure 5b The required condenser and reboiler duties are reduced by 54 and 25 respectively in the new configuration However it should be noted that the utility type of the reboiler chan ged from MPS to HPS which is 20 more expensive Details of capital investment utility requirements revenue and economic parameters of the new plant with new benzene column configuration are presented in Table V This table illustrates that the capital investment is increased while the utility requirement is reduced from 19 to 175 million The final outcome of this change is the enhancement of the NPV of the process and the reduction of COM Operational Conditions Optimization After performing topological changes to the process it is necessary to adjust the oper ating conditions of the new process at which better profitability is obtained The selec ted operating conditions should have a significant effect on process performance and hence on the economic status of the process Here four parameters were chosen as the manipulated variables inlet temperature of the alkylation reactor T molar flow rate of the recycle to the alkylation reactor BA alkylation reactor length L and molar flow of the recycle to the transalkylation reactor BT Statistical methods were used to find a model that describes the behavior of the process After that this model was used to find the optimum operating conditions By using a statistical method it is possible to determine the level of importance of each variable and to find the probable Figure 6 NPV of the new benzene column configuration vs condensation rate at different split fractions shaded rectangle shows the operating conditions at which better economic condition of benzene column is obtained ZeoliteBased Cumene Production Plant 1285 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 interactions between them Omidbakhsh et al 2010 Vining 1998 Zivorad 2004 Moreover the statistical analysis regression can give a mathematical model as a function of manipulated variables In this study the full factorial design was used to determine the level of significance of each manipulated variable on the objective function Based on the full factorial design 24 evaluations should be available to determine the significant effects of the variables and their binary interactions Zivorad 2004 Table VI shows the full factorial design of the four variables mentioned above The process was simulated at each operating condition and the NPV capital investment and utility requirements were calculated in each case and reported in this table In performing these simulations some constraints were imposed on the process to accomplish the aim of this process which is manufacturing cumene with a molar purity greater than 995 These constraints are listed below Concentration of benzene in cumene column C3 feed must be negligible to guarantee that highly purified cumene is obtained in the overhead of C3 Temperature of C2 bottoms must not exceed 242C The reboiler of this column is derived by HP steam Concentration of benzene recycle does not fall under 94 mol Temperature of liquid mixture through the alkylation reactor must be below the boiling point If this constraint is not satisfied in each case additional intercoolers have to be added to the reactor to keep the temperature in the range Analysis of variance ANOVA was carried out on the NPV values to determine the significance of singlefactor and twoway interactions A significance level of 10 Table VI Full factorial design table and NPV of the process as response No BA kmolh T C BT kmolh L m Capital investment million Utility million yr NPV million 1 95 160 20 5 4100 1670 404 2 150 160 20 5 4190 1510 418 3 95 180 20 5 4220 1690 395 4 150 180 20 5 4200 1780 365 5 95 160 40 5 4170 1730 384 6 150 160 40 5 4140 1550 363 7 95 180 40 5 4220 1800 274 8 150 180 40 5 4260 1890 123 9 95 160 20 8 4140 1720 388 10 150 160 20 8 4240 1790 343 11 95 180 20 8 4380 1670 400 12 150 180 20 8 4350 1850 311 13 95 160 40 8 4280 1740 400 14 150 160 40 8 4350 1860 304 15 95 180 40 8 4240 1730 388 16 150 180 40 8 4350 1870 306 17 1225 170 30 65 4140 1780 372 was considered for determining the significant variables The results are shown in Table VII The test values of Fischer and the pvalues were considered as the decision criteria Those factors with pvalues less than 01 were considered to have a significant effect on the NPV and the variables with pvalues greater than 01 were reported as nonsignificant parameters In addition case no 17 which is the center point in the full factorial design was evaluated to test whether or not there is a curvature secondorder functionality rather than linear in the respond function The calculated pvalue for the curvature was found to be 0514 see Table VII which shows that the response NPV is linear Hence a linear model twolevel factorial design is adequate to obtain a mathematical model for the NPV The Pareto chart of standardized effects for significant variables is shown in Figure 7 nonsignificant variables are not illustrated in this figure The vertical dashed lines indicate the significance limit of each term which is considered to be 10 Among the four abovementioned variables only the recycle to the transalkylation reactor BT has a nonsignificant effect on the NPV of the process As discussed before this recycle Loop2 has the least contribution to the utility requirements of the process hence any changes in this variable do not have a pronounced effect on the NPV of the process However the recycle to the alkylation reactor BA Loop1 has a significant and negative effect on the NPV Although attempts were made to reduce the contribution of Loop1 to the utility requirement of the process it still possesses a significant effect based on the statistical analysis Moreover the inlet temperature of the reactor T and reactor length L both have a positive effect on the NPV of the process Reducing either the reactor length or the inlet temperature reduces the conversion of propylene in the reactor In this state a part of the fed propylene will leave the process unreacted in the overhead of the propane column Therefore the NPV of the process is reduced as a result of a decrease in the production rate of cumene One of the most interesting capabilities of the statistical analysis used in this study is determining the binary twoway interactions between the variables The results in Figure 7 show that T L has a negative effect on the NPV This indicates that the NPV is enhanced when one variable is increased while the other is decreased Combined variables BA T and BA L have a positive effect Table VII ANOVA analysis of NPV for determining significant variables Variable Mean sum of squares Ftest pvalue Significance BA 2658 145 0004 Yes T 784 428 0069 Yes BT 0865 031 0604 No L 1756 958 0013 Yes BA T 1152 629 0033 Yes BA BT 0051 002 0899 No BA L 1292 705 0023 Yes T BT 0941 033 0589 No T L 8702 475 0057 Yes BT L 0504 018 0690 No Curvature 1395 049 0514 No Figure 7 Pareto chart of standardized effect for significant variables Dashed lines indicate statistically significant bond for each term on the NPV In other words when the benzene recycle to the alkylation reactor BA is increased the inlet temperature or the reactor length should be increased in order to reach higher reaction rate or longer residence time Hence the statistical analysis matches the expected trends of the process and therefore the mathematical model obtained from this analysis reflects the influence of different variables on the NPV reliably The following mathematical model was obtained for the NPV of the process NPV 1901 07132B A 00115T 63877L 0003086B AT 0021788 B AL 004917TL R2 095 10 Optimized Process Optimization was performed based on Equation 10 by the Simplex method Vining 1998 and the optimal conditions were determined to be inlet temperature Figure 8 Process flow diagram of liquidphase cumene production for the optimized case Table VIII Stream table of process flow diagram optimized case shown in Figure 8 Stream Fresh benzene Fresh propylene Mixed feed Alkylation effluents Fuel gas C2feed Benzene ovhd C2 bottoms Cumene DIPB Temperature C 25 25 160 2525 442 2425 1176 1580 1518 2133 2400 Pressure MPa 03 115 354 354 15 152 08 011 01 011 2400 Vapor fraction 0 0 0 0 1 0007 0 0 0 0 0 Mole flow kmolh Mass flow comp mole flow kmolh Benzene 100 100 22125 20001 501 22790 12000 10790 10000 790 3290 Propylene 5 5 25000 93128 0003 11312 11304 0074 0000 23550 19994 Propane trace trace 1250 0007 0007 trace 695 99861 99852 0010 8571 Cumene trace trace 5508 5000 5000 trace 10681 796 7887 7888 4332 of 160C reactor length of 6 m and benzene recycle to the alkylation reactor of 95 kmolh The results indicate that the NPV of the process becomes 482 million at this optimized condition A simulation was done at the optimized conditions and the results of different economic parameters of the process are reported in the last column of Table V It can be seen in this table that both capital investment and utility requirements of the optimized case are less than those of the twocolumn case resulting in an increase in the NPV of the process The NPV of the optimized process obtained from the simulation is 416 million which is greater than that of the twocolumn case which is 356 million And the utility cost of the process is reduced from 175 to 168 millionyr The process flow diagram of the optimized case is illustrated in Figure 8 and the corresponding streams of this process are shown in Table VIII Utility requirements of the optimized process are 207 and 141 GJh for cold and hot utilities respectively Pinch analysis was carried out for the optimized case to find out if further process enhancement is possible The composite curve of the optimized case is illustrated in Figure 9 Hot and cold pinch temperatures are located at 222 and 212C respectively The results show that minimum cold and hot utility requirements are 175 and 112 GJh respectively if perfect energy targeting is done on the process Smith 2005 This shows that more energy can be recovered through the process As can be seen in the flow sheet there are two heat exchangers that cross the pinch point Smith 2005 The first exchanger is the intercooler for which the inlet and outlet temperatures are 243 and 160C and the second exchanger is E1 for which the inlet and outlet temperatures are 253 and 194C These heat exchangers should be avoided and extra exchangers were added to the process In this way cooling is done in two stages first the hot stream is cooled to 222C by a cold stream above the pinch point E8 and then it is further cooled to the desired final temperature by a cold stream or utility These changes were made to the process and simulation was carried out again The results showed reductions in both hot and cold utilities The hot utility requirement was decreased by 11 16 GJh and the cold utility requirement was decreased by 135 28 GJh The economic state of the process is listed in Figure 9 Composite curve of optimized case and pinch point Table V The total annual utility cost was reduced from 168 to 149 millionyr and the total investment was increased from 422 to 450 million due to the addition of extra exchangers The NPV of the process in this condition was further enhanced and increased to 453 Norouzi and Fatemi 2012 studied the vapor phase production process of cumene on SPA catalyst with the same annual production capacity and the same economic model as this study Their optimized process required 102 million total investment and 383 million annual utility requirements These values are consider ably greater than those obtained for the optimized case of the zeolitebased process NPV of 479 million was obtained for the optimized case of the SPA process whereas it is 453 million for the optimized case of the zeolitebased process The higher value of the NPV for the SPA process should be compared to 102 million total capital investment which yields 142 present value ratio while the present value ratio is 201 for the zeolitebased process The investment in a process with a higher present value ratio and lower capital investment is more tempting Conclusion The zeolitebased production of cumene from propylene and benzene was studied for intensification of an industrialscale production process to obtain higher profitability and reduce the energy requirements of the process A comprehensive model that considers all influential factors was established for the profitability analysis Using established economic models the total investment and yearly utility requirements of the base process were calculated to be 448 and 244 million respectively which yielded an NPV of 082 million A multistage procedure was followed to optimize the base process First topological changes were made to the process The reactor with one catalytic bed was replaced with a reactor with four catalytic beds with an interstage cooling in which propylene is added through side streams In addition the benzene column was replaced with two columns with operating pressures of 1 and 8 bar These topo logical changes led to a considerable reduction in utility requirements and equipment sizes in the process and thus to better profitability After that the parametric opti mization was carried out by a statistical method full factorial design to adjust process parameters at which the profitability is maximized The analysis of variance determined the influential parameters on the NPV of the process and showed that the linear model is adequate to obtain a mathematical model for the NPV A mathematical model was developed and utilized to get the optimized conditions of the process Comparisons between the base case and optimized case revealed that the capital investment and the utility requirements of the process were reduced from 4481 to 422 million and from 244 to 168 millionyr respectively These greatly enhanced the economic condition of the process for which the net present value was increased from 082 to 416 million Finally pinch analysis was done to find the heat flow across the pinch point Extra exchangers were added to the process to change hot and cold stream contact pattern a substantial reduction in utility cost was obtained and the NPV increased to 453 This work can be further improved by considering polyalkylation reactions in the process to gain a better picture of the process In addition dynamic simulation can be another option through which different startup scenarios can be tested and fault detection and troubleshooting of the process can be performed ZeoliteBased Cumene Production Plant 1291 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 Nomenclature a utility cost coefficient b utility cost coefficient CBM0 bare module equipment cost Cf price of fuel GJ CP purchased cost for the base condition Cu cost of utilities GJ CEPCI Chemical Engineering Plant Cost Index FBM0 bare module cost factor FCI total module cost k1 firstorder reaction constant in Equation 1 1s k2 firstorder reaction constant in Equation 2 1s k3 forward reaction constant in Equation 3 kmolm3s k4 backward reaction constant in Equation 3 kmolm3s NPV net present value million Oprt Cost operating cost y References Branan C R 1995 Rules of Thumb for Chemical Engineers Gulf Publishing Houston Tex Canfield R C Cox R C and McCarthy D M 1986 MonsantoLummus crest process produces lowest cost cumen in Proceedings of the AIChE 1988 Spring Meeting New Orleans LA American Institute of Chemical Engineers New York Corma A MartinezSoria V and Schnoeveld E 2000 Alkylation of benzene with shortchain olefins over MCM22 zeolite Catalytic behaviour and kinetic mechanism J Catal 192 163173 Dimian A C and Bildea C S 2008 Chemical Process Design ComputerAided Case Studies WileyVCH Hoboken NJ Gary J H and Handwerk G H 1994 Petroleum Refining Technology and Economics Macel Dekker New York Gas Processors Suppliers Association 1998 Engineering Data Book 11th ed Gas Processors Suppliers Association Tulsa Okla Han M Li X and Lin S 2001 Intrinsic kinetics of the alkylation of benzene with propylele over beta zeolite catalyst Kinet Catal 42 533538 Hwang S Y and Chen S S 2010 Cumene review in KirkOthmer Encyclopedia of Chemical Technology vol 8 147157 John Wiley Sons Hoboken NJ ICIS 2011 Chemical prices and chemical industry trends Accessed April 2011 from http wwwicispricingcom Lei Z Dai C Wang Y and Chen B 2009 Process optimization on alkylation of benzene with propylene Energy Fuels 23 31593166 Luyben W L 2010 Design and control of the cumene process Ind Eng Chem Res 49 719734 Norouzi H R and Fatemi S 2012 Economic optimization of the cumene production process using response surface methodology Chem Eng Commun 199 119 Omidbakhsh N Elkamel A Duever T A and Reilly P M 2010 Combining design of experiments techniques connectionist models and optimization for the efficient design of new product formulations Chem Prod Process Model 51 Pathak A S Agarwal S Gera V and Kaistha N 2011 Design and control of a vaporphase conventional process and reactive distillation process for cumene production Ind Eng Chem Res 50 33123326 Perego C and Ingallina P 2002 Recent advances in the industrial alkylation of aromatics New catalysts and new processes Catal Today 73 322 Perego C Amarilli S Millini R Bellussi G Girotti G and Terzoni G 1996 Experimental and computational study of beta ZSM12 Y mordenite and ERB1 in cumene synthesis Microporous Mater 6 395404 Perry J H Green D W and Maloney J O 1999 Perrys Chemical Engineers Handbook 7th ed McGrawHill New York Pohl S and Ram S 2005 ABB Lummus global cumene production via CD cumene technology in Handbook of Petrochemicals Production Processes ed R A Meyers McGrawHill New York Schmidt R J 2005 Industrial catalytic processesPhenol production Appl Catal A Gen 280 89103 Seider W D Seader J D Lewin D R and Widagdo S 2010 Product and Process Design Principles John Wiley Sons Hoboken NJ Smith R 2005 Chemical Process Design and Integration John Wiley Sons Chichester West Sussex England Stefanidakis G and Gwyn J E 1977 Alkylation Encyclopedia of Chemical Processing and Design ed J J McKetta and W A Cunningham vol 2 357 Marcel Dekker New York Sy A Smith L Chen J and Dautzenberg F M 1993 Catalytic distillation route for cumene paper presented at DeWitt Petrochemical Review Houston Texas Turton R Bailie R C Whiting W B and Shaeiwitz J A 1998 Analysis Synthesis and Design of Chemical Processes PrenticeHall Upper Saddle River NJ Ulrich G D and Vasudevan P T 2006 How to estimate utility costs Chem Eng 1134 6669 US Department of Energy 2011 United States natural gas industrial price Accessed April 2011 from httpwwweiagovdnavnghistn3035us3mhtm Vining G G 1998 Statistical Methods for Engineers Duxbury Press Pacific Grove Calif Zivorad R L 2004 Design of Experiments in Chemical Engineering WileyVCH Weinheim ZeoliteBased Cumene Production Plant 1293 Downloaded by University of Saskatchewan Library at 1134 04 June 2014 View publication stats