Exergy analysis and flowsheet modification of PT. PUSRI II ...

ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERINGAsia-Pac. J. Chem. Eng. 2011; 6: 760–770Published online 14 March 2011 in Wiley Online Library(wileyonlinelibrary.com) DOI:10.1002/apj.565Special theme research articleExergy analysis and flowsheet modification of PT. PUSRI IIPalembang’s reformer unit for the production of synthesisgas from natural gasMuhammad Djoni Bustan, 1 * Shuhaimi Mahadzir 2 and Dian Kharismadewi 11 Chemical Engineering Graduate Program, University of Sriwijaya, Palembang 30139, Indonesia2 Chemical Engineering Department, Universiti Teknologi Petronas, Bandar Seri Iskandar, Trononh 31750, Perak, MalaysiaReceived 28 July 2010; Revised 12 January 2011; Accepted 23 January 2011ABSTRACT: Several processes which produce synthesis gas from natural gas have been analyzed by the exergy method.To minimize the exergy losses associated to chemical reaction, it is important to manage the exergy losses through theanalysis, reconsidering the chemical path. A generally applicable and schematic way of performing exergy analysisand dealing with their results are illustrated. It is integrated to the modified flowsheeting reformer. As petrochemicalindustry in Indonesia, PT. Pupuk Sriwijaya II Palembang uses natural gas as a raw material to produce synthesis gasthrough the primary and the secondary reformers in a series. In the modified flowsheeting reformer, partial oxidationprocess has been added before the primary reformer and the secondary reformer has been eliminated from the processpathways. The exergy method of analysis has been implemented for each process, in order to calculate the exergies ofthe material streams along with the energy and mass-balance calculations both in the modified flowsheeting reformerand in a series process of the primary and the secondary reformers. And, the results of the calculations of the modifiedflowsheeting reformer in the process proposed modification are then compared to the results of the calculations of theseries process of reformers in the existing conventional process. From the analysis, the overall exergy losses in a seriesprocess of the primary and secondary reformers can be reduced until 61.33%, by applying the modified flowsheetingreformer. © 2011 Curtin University of Technology and John Wiley & Sons, Ltd.KEYWORDS: exergy analysis; flowsheet modification; reformer unit; partial oxidation; synthesis gas; natural gasINTRODUCTIONIndonesia, as an agricultural country, still considers theagricultural sector as a significant contributor to itseconomy besides the growing industrial sector. Petrochemicalis one of the industries, which is still developingin this country. PT. Pupuk Sriwijaya (PUSRI) II isthe largest petrochemical plant in Palembang, Indonesiathat produces fertilizers for the large agriculturalsector.In the production of fertilizers, natural gas is usedas a raw material to generate synthesis gas as a basefor forming ammonia and urea fertilizer. Synthesisgas is produced in the primary and the secondaryreformer units. The two reformers require large amountof fuel and heat to achieve a synthesis gas conversionof only around 97.8%. In addition, exergy analysis*Correspondence to: Muhammad Djoni Bustan, Chemical EngineeringGraduate Program, University of Sriwijaya, Jalan Padang SelasaNo. 524 Bukit Besar, Palembang 30139, Indonesia.E-mail: djajashanta@yahoo.co.id© 2011 Curtin University of Technology and John Wiley & Sons, Ltd.Curtin University is a trademark of Curtin University of Technologyaround the reformers in PT. PUSRI II shows thatthe exergy losses are definitely high. In this study,modification of the process through flowsheet simulationof reformer is been explored, in which a partialoxidation process has been added before the primaryreformer. The secondary reformer has been eliminatedfrom the process pathways. The flowsheet simulationenables the optimum process condition to bedetermined and the process with the minimum exergylosses for the production synthesis gas production to beselected.EXERGY ANALYSIS METHODThe method of exergy analysis is based on the secondlaw of thermodynamics and the concept of irreversibleproduction of entropy. [1] The method has found earlydevelopment and application in the Soviet Union andEurope, primarily Germany and Poland. The earliestuse of the word exergy has been attributed to Rant in1956 by Bosjnakovic (1960), Trepp (1961) and Baehr

Asia-Pacific Journal of Chemical Engineering THE PRODUCTION OF SYNTHESIS GAS FROM NATURAL GAS 761(1962). Exergy has been generally referred to as workcapacity or available work. The concept was initiallyintroduced by Gibbs and Helmholtz for thermodynamicanalysis of chemical systems. A review on the use ofexergy in systems analysis since its early beginning hasbeen provided in the early millennium year. [2] Bosnjakovic(1960) demonstrated the exergy analysis methodon a power plant using the enthalpy–entropy diagram topresent his results. He also made reference to the benefitof using exergy analysis in other related systems.Trepp (1961) examined exergy losses in refrigerationmachines and develop optimum low temperature cycledesign. Beahr (1962) provided an extensive discussionon exergy and presented his analytical results by comparingflowcharts of exergy and energy.Szargut (2005) mentioned that exergy analysis isbased upon the second law of thermodynamics whichstipulates that all macroscopic processes are irreversible.Every such irreversible process entails a nonrecoverableexergy lost, which is expressed as the productof the environment temperature and the entropygenerated. The quantity of entropy generated is the sumof the values of the entropy increase for all the bodiestaking part in the process. [3] The elementary irreversiblephenomena that generate entropy are mechanical orhydraulic friction, heat transfer with a finite temperaturegradient, diffusion with a finite gradient of concentration,and the mixing of substances with differentparameters and chemical composition.EXISTING CONVENTIONAL PROCESSIn the conventional process at PT. PUSRI II, naturalgas is converted into synthesis gas through steamreformers. The essential configuration of the primarysteam reformer is first established by BadischeAnilin and Soda Fabrik (BASF) at the end of1920s and the technology was used in 1931 by StandardOil of New Jersey to produce hydrogen fromoff-gases at its refineries. Since then, the processhas been improved such as by M.W. Kellog at PT.PUSRI II.In this work, process used by PT. PUSRI II Palembangwould be as reference process in modificationdesign and calculations analysis. This reference process,labeled as process A, uses a primary and a secondaryreformer in series as shown in Fig. 1. Natural gas, asfeed for the reforming process in the primary reformer,enters the feed preparation step where a heavy hydrocarbonand suspended particles are removed. Naturalgas feed to the primary reformer consists of 81.90%CH 4 by volume and other higher hydrocarbons with5.45% C 2 H 6 , 2.87% C 3 H 8 , 0.44% i-C 4 H 10 , 0.56% n-C 4 H 10 , 0.15% i-C 5 H 12 , 0.08% n-C 5 H 12 , 0.03% C 6+ ,0.20% CO 2 , 0.01% Ar, 1.63% N 2 , and 6.68% H 2 .In addition, CO 2 removal, dehydration, and desulfurizationare also applied to the feed gas. Natural gasfeed is then preheated to 430 ◦ C and pressured to33.87 atm before entering radiant section of the primaryNatural Gas Fuel AirFlue GasFEED PREPARATIONSteamPRIMARY REFORMERHeat (Flue Gas)FEED PREHEAT COILT = 703.15 KP = 33.87 atmExhaust tostackCONVECTION SECTION(Coils)Flue GasRADIANT SECTION(Tube Catalyst Ni)T = 1088.15 KP = 31.82 atmHeatAirFlue GasHeat toutilitySECONDARY REFORMER(Tube Catalyst Ni)T = 1251.15 KP = 30.72 atmSYNGAS PREPARATIONT = 644.15 KP = 35.50 atmFigure 1. Block diagram of the series reformers, process (A).© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

762 M. DJONI BUSTAN, S. MAHADZIR AND D. KHARISMADEWI Asia-Pacific Journal of Chemical Engineeringreformer. Energy from the hot flue gas is recoveredin the convection section to raise steam or for otheruses.In the radiant section, previously natural gas feedis mixed with saturated steam to give methane-tosteamratio of 1 : 4.25 then reacts in endothermicreaction with Ni-catalyst in a tubular reactor of theradiant section of the primary reformer at 815 ◦ C,31.82 atm. To increase and enrich the synthesis gasoutput from the primary reformer, the secondaryreformer is then applied with air supply and theoperating condition is 1251.15 ◦ K, 30.72 atm. Its synthesisgas reacts in exothermic reaction in the secondaryreformer with Ni-catalyst supported in a tubularreactor then enters the synthesis gas preparationsteps.Conversion of CH 4 in the primary reformer of the naturalgas feed only reaches to 51.69%, while all heavierhydrocarbons in the natural gas feed are completely converted.In the secondary reformer, conversion of CH 4 is95.48%. Consequently, the overall conversion of CH 4from the series reformers reaches to 97.82%.Large amount of energy is consumed to raise theprocess temperature from preparation step to 815 ◦ C inthe primary reformer and to 978 ◦ C in the secondaryreformer. The overall energy consumed from bothprocesses is 237 534.61 kJ h −1 .Material streams in processes A and B of this study(based on data of January 2009), could be seen fromTable 1, for our focus on a synthesis zone of each blockdiagram.The calculation of the material and energy balancesare given in Table 3 and are based on the previous studyabout natural gas conversion and will not be discussedhere, and only the calculation steps and results areshowed in this article. The exergy losses of processA in every step are shown in Table 4. It is shown thatthe exergy lost in process A is high. Clearly, althoughthe overall conversion for the reforming could reachup to 97.82%, the consumption of energy, fuel, andthe exergy losses in this process are definitely highbecause these parameters have not been optimized inthe design.PROPOSED PROCESS MODIFICATIONExergy analysis with a flowsheeting simulator for synthesisgas production from natural gas to becomemethanol as a final product has been learned by Hinderinket al. [4,12] In their method, they have implementeda set of external subroutines that have beenintegrated with a flowsheeting simulator, in order tocalculate exergies of material streams along with thetraditional energy and mass-balance calculations. Theyproposed four different pathways for the process: conventionalsteam reforming, conventional steam reformingand pre-reformer; combined reforming with a naturalgas bypass, and convective reforming parallel tonon-catalytic partial oxidation.Run from those types of pathways as reference byusing the operating condition and material streamssimilar to process A in our work, we have calculated itsconversion and exergy analysis of each process whichwould not be discussed and compared in this article.Material streams input in the synthesis part of thoseprocesses are similar to process A, which is shownin Table 1 for primary reformer input at temperature703.15 K and pressure 33.87 atm.The block diagram of each process is shown inFig. 2 for conventional steam reforming process (a) andTable 1. Material streams of processes A and B.CompoundsPrimary reformerinput (% kmole/h)Process ASecondary reformerinput (% kmole/h)Partial oxidation(% kmole/h)Process BSteam reformer(% kmole/h)Natural gasCH 4 18.37 5.38 59.00 5.69C 2 H 6 1.29 0.00 4.15 0.00C 3 H 8 0.81 0.00 2.59 0.00C 4 H 10 0.25 0.00 0.82 0.00CO 0.00 4.52 0.00 12.39CO 2 0.06 4.92 0.20 0.21H 2 0.86 31.42 2.77 10.24N 2 0.30 16.21 0.96 14.61Ar 0.00 0.19 0.01 0.00O 2 0.00 4.31 29.50 0.00H 2 O (steam) 78.05 33.04 0.00 56.87Methane : Steam input ratio 1 : 4.25 – – 1 : 10Methane : O 2 ratio – – 1 : 0.5 –N 2 : H 2 output ratio – 1 : 3 – 1 : 3© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering THE PRODUCTION OF SYNTHESIS GAS FROM NATURAL GAS 763Natural GasFuel AirFlue GasFEED PREPARATIONSteamHeat (Flue Gas)FEED PREHEAT COILPRIMARY REFORMERExhaust tostackCONVECTION SECTION(Coils)Flue GasRADIANT SECTION(Tube Catalyst Ni)HeatFlue GasSYNGAS PREPARATIONNatural Gas Fuel AirFlue GasFEED PREPARATIONSteamPRIMARY REFORMERHeat (Flue Gas)FEED PREHEAT COILNatural GasExhaustto stackCONVECTION SECTION(Coils)Flue GasRADIANT SECTION(Tube Catalyst Ni)NitrogenAirHeatOxygenAIRSEPARATORFlue GasHeat to utilityFlue GasPARTIAL OXIDATION(Tube Catalyst Rh)FuelAirSYNGAS PREPARATIONFigure 2. (a) Block diagram of the conventional steam reforming process (A). (b) Blockdiagram of the combined reforming with natural gas baypass process (B).combined reforming with a natural gas bypass process(b).From these two processes (Fig. 2(a) and (b)), it isknown that both processes give no satisfied conversionof methane and yield of hydrogen and carbon dioxidegas compared to process A as reference and also highin exergy losses.On the basis of those processes results of calculationslearned before and its weaknesses and alsofrom weaknesses of process A as our reference in thisarticle, process modification is proposed by adding apartial oxidation process before the primary reformerand eliminating the secondary reformer. The modifiedprocess is designated as process B. Process Bhas been proposed to optimize the process design inthe production of synthesis gas from natural gas andthe improvement in exergy losses will be comparedin reference to process A. Process B is described© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

764 M. DJONI BUSTAN, S. MAHADZIR AND D. KHARISMADEWI Asia-Pacific Journal of Chemical EngineeringFuelAirNatural GasFEED PREPARATIONFuelAirFEED PREHEAT COILNaturalGasT = 703.15 KP = 33.87 atmOxygenAirAIRSEPARATORFlue GasPARTIAL OXIDATION(Tube Catalyst Rh)T = 1173.15 KP = 20 atmExhaust tostackCONVECTION SECTION(Coils)NitrogenSteamSteamSTEAM REFORMER(Tube Catalyst Ni)T = 773.15 KP = 20 atmSYNGASPREPARATIONT = 664.15 KP = 35.50 atmFigure 3. Block diagram of the modified flowsheeting reformer, process (B).below, while the block diagram for calculating exergiesof the material streams is shown in Fig. 3.The operating conditions are taken from the previousstudy. [5]Figure 3 illustrates the major process steps of themodified flowsheeting reformer in a block diagram. Thepartial oxidation process has been added before theprimary reformer and the secondary reformer eliminatedin process B. Natural gas with the same characteristicsand compositions with process A supplies as a feed forthe partial oxidation process, previously it will enterthe feed preparation steps. In this first step, heavyhydrocarbon and suspended particles separation, CO 2removal and dehydration process are applied to thenatural gas feed. Similar to the process A, natural gasfeed is then preheated to the temperature of 430 ◦ C inthe feed preheat coils and pressured to 33.87 atm, toenter the partial oxidation process. Natural gas feed isthen mixed with limited oxygen to give methane-tooxygenratio of 1 : 0.5 and reacts in exothermic reactionwith Rh-catalyst in a tubular reactor of the partialoxidation, [6,7] at the operating condition 900 ◦ C, 20 atm.To increase and enrich the synthesis gas outputfrom the partial oxidation, synthesis gas output is thenmixed with steam in methane-to-steam ratio of 1 : 10 inthe primary/steam reformer at the operating condition500 ◦ C, 20 atm. Its synthesis gas reacts in exothermicreaction with Ni-catalyst supported in a tubular reactorof the primary reformer. Nitrogen has been added inthe primary/steam reformer in this process because inthe process A as a reference condition to the next stepto produce ammonia, nitrogen is needed with ratio ofN 2 -to-H 2 1:3.Material streams in this modified flowsheetingreformer (process B) could be seen from Table 1 previewedbefore, for focus in a synthesis zone of Figure 3block diagram. Material streams of natural gas feedcompositions in process B are same as process A.EXERGY CALCULATIONS OF THE MATERIALSTREAMSThe overall reactions and their standard Gibbs energyof reactions presented for the conversion of naturalgas to synthesis gas for each process, is determined inTable 2. [8,9]In Table 2, the minus of standard Gibbs energy of theoverall reaction is denoted as available reaction exergy.It plays an important role in understanding the exergylosses associated to chemical reactions.In basic thermodynamics calculation of the materialstreams, the conversion and fraction equations wereperformed and adopted from Smith and Van Ness. [10]Determining the reactions that occur in catalytic partialoxidation process from natural gas with higher hydrocarbonsexist would be the first step to do in thermodynamicscalculation with certain pressure and rangeof temperature for trial the best operating condition.© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering THE PRODUCTION OF SYNTHESIS GAS FROM NATURAL GAS 765Table 2. The overall reactions.Process Reactions G 0 r 298K (kJ kmole −1 )AB1. In the primary reformer, from natural gas to synthesis gasCH 4 + H 2 O ⇔ CO + 3H 2 141 863C 2 H 6 + 2H 2 O ⇔ 2CO + 5H 2 214 661C 3 H 8 + 3H 2 O ⇔ 3CO + 7H 2 298 499C 4 H 10 + 4H 2 O ⇔ 4CO + 9H 2 349 042CO + H 2 O ⇔ CO 2 + H 2 −28 6182. In the secondary reformer, from synthesis gas to enriched synthesis gasCH 4 + H 2 O ⇔ 3H 2 + CO 141 8632H 2 + O 2 ⇔ 2H 2 O −457 144CO + O 2 ⇔ 2CO 2 −651 5491. In the catalytic partial oxidation, from natural gas to synthesis gas2CH 4 + O 2 ⇔ 2CO + 4H 2 −86 709C 2 H 6 + O 2 ⇔ 2CO + 3H 2 −242 4832C 3 H 8 + 3O 2 ⇔ 6CO + 8H 2 −387 217C 4 H 10 + 2O 2 ⇔ 4CO + 5H 2 −565 2462CO + O 2 ⇔ 2CO 2 −257 1902. In the steam reformer, from synthesis gas to enriched synthesis gasCH 4 + H 2 O ⇔ CO + 3H 2 141 863CO + H 2 O ⇔ CO 2 + H 2 −28 618To calculate the equilibrium mixture for multi-reactionequilibria, we need the equilibrium constant K for thereaction, and is shown in Eqn (1).−RT ln K = ∑ iυ i G 0 i ≡ G 0 (1)The final term G 0 is the conventional way of representingthe quantity ∑ i υ iGi 0 . It is called the standardGibbs energy change of reaction. The function of G 0is the difference between the Gibbs energies of theproducts and reactants (weighted by their stoichiometriccoefficients) when each species in its standard stateas a pure substance at the system temperature and at afixed pressure. Thus, the value of G 0 is fixed for agiven reaction once the temperature is established andis independent of equilibrium pressure and composition.And, G 0 /RT could be explained by Eqn (2).G 0RT = G 0 0 − H 00 + H 00RT 0 RT + 1 T ICPH − ICPS (2)Integral Differential Heat Capacity for EnthalpyCalculations (ICPH) and Integral Differential HeatCapacity for Entropy Calculations (ICPS) are symbolsfor computational purposes as described in Eqns (3)and (4).∫ TCP0 ICPH =T 0R dT = (A)T 0(τ − 1)+ B2 T 0 2 (τ 2 − 1) + C 3 T 0 3 (τ 3 − 1)+ D ( ) τ − 1(3)T 0 τICPS =+∫ TT 0C 0 PR[BT 0 +dT= A ln τT(CT0 2 + Dτ 2 T02) ( τ + 12)](τ − 1)(4)The evaluation of the integral ∫ C p dT is accomplishedby substitution for C p , followed by formal integration,for temperature limits of T 0 and T , whereτ = T T 0(5)Thus, G 0 /RT (= −ln K) as given by Eqn (2) isreadily calculated at any temperature from standardheat of reaction and the standard Gibbs energy changeof reaction at a reference temperature, T 0 (usually298.15 ◦ K) and from two functions which can becalculated by standard computational procedures.When two or more reactions proceed simultaneously,mole fractions y i of the species are related to ε or calledreaction coordinate, characterized the extent or degreeto which a reaction has taken place, shown by Eqn (6).y i =n i0 + ∑ jn 0 + ∑ jv i,j ε jv j ε j(i = 1, 2, . . . N ) (6)where j is the reaction index, and associates a separatereaction coordinate ε j with each reaction. The stoichiometricnumbers are doubly subscripted to identify their© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

766 M. DJONI BUSTAN, S. MAHADZIR AND D. KHARISMADEWI Asia-Pacific Journal of Chemical Engineeringassociation with both a species and a reaction. Thus,υ i,j designates the stoichiometric number of species iin reaction j .And, when the equilibrium state in a reacting systemdepends on two or more independent chemical reactions,the equilibrium composition can be found by adirect extension of the methods developed for singlereactions. If the equilibrium mixture is an ideal gas, itmay be written as Eqn (7), where P 0 = 1 atm.∏( )(y i ) v i,j P −vj=P 0 K j (7)iThe mole fractions in Eqn (7) may be solved byusing the mathematical program for the multireactionequilibria such Maple 11.Heat requirements of the process (Q) are expressedby the summation of heat input to the reaction processand the heat of reactions itself as shown inEqns (8) and (9), respectively.Q = Q input + Q reactions (8)(Q = nR) (Cp0∫ )TT 0R dT + Hro 0 + R Cp0T 0R dT∫ T(9)In Eqn (9), Hr00 is heat of reaction at referencetemperature T 0 .In calculating exergies of the material streams, thefollowing steps of calculating were performed andadopted from Ahern. [3] A primary use of the exergyanalysis is to show the location, type, and magnitude ofthe exergy losses, in a system. [11] As a result, the systemefficiency can be effectively increased by reducing theselosses. The exergy losses are generated throughout thesystem by irreversible production of entropy caused bythe non-ideal performance inherent in all real systemsand components.The exergy lost results from heat loss during combustioncould be calculated by using Eqn (10). This wasassumed to be (1 - overall efficiency) and considered asa direct exergy loss as it reduced the available work inthe fuel by that amount.Ex a = Q release of fuel × (1 − overall efficiency)× mole of fuel (10)The exergy of fuel is Ex 1,7 the heat release fromnatural gas fuel to the furnace.Exergy lost resulting from transfer of chemical energyto heat energy is calculated using Eqn (11). During thecombustion of the fuel, the remainder of the chemicalenergy is transferred into thermal energy,Ex 2,8 = (Q release of fuel × mole of fuel) − Ex a (11)A transfer of useful chemical energy into thermalenergy is an irreversible process expressed by Eqn (12),Ex b = m 9,4 (H out − H in ) T 0T g(12)The exergy loss resulting from a heat transfer betweennatural gas fuel and feed of process shows in Eqn (13).[Ex c = T 0 (m 9,4 ) (S out − S in ) − (H ]out − H in )T g(13)The total exergy loss of fuel natural gas to naturalgas feed/synthesis gas in the process is Ex b + Ex c .Furthermore, the exergy loss during the outflow ofsynthesis gas of certain process in piping from one stepprocess to the other is expressed in Eqn (14).Ex 3,5 = (H out − H in ) − T 0 (S out − S in ) (14)In Figs 4 and 5, a synthesis gas in conditions 4 and6 has a work capacity, soEx 4,6 = (H 4,6 − H 9,4 ) − T 0 (S 4,6 − S 9,4 ) (15)Moreover, the exergy lost in process is expressed byEqn (16) below.Ex d = Ex 3,5 − Ex 4,6 (16)The exergy loss of friction from the partial oxidationto the steam reformer or from the primaryreformer to the secondary reformer or from the steamreformer/secondary reformer to the synthesis gas preparationis(P2Ex e = m 3,5 − P )1(17)ρ 2 ρ 1A heat transfer in piping is m(H 3,5 − H 4,6 ).Figure 4 describes the material and energy calculationssteps in the proposed process modification. Thecomposition moles input used is similar to the existingconventional process A. It is divided into two bigblocks, named 1 and 2.Block 1 is the step for calculating material conversionand moles fraction of natural gas that reacts in a process.Furthermore, block 2 is the step for calculating theenergy streams in a process that is continuing from stepsin block 1.Mass and heat input of the material streams from theseries reformers of process A and modified reformer© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering THE PRODUCTION OF SYNTHESIS GAS FROM NATURAL GAS 767Estimate the reactions occur in a process,variations temperature and fixed pressureconditionCalculate ∆G°, ∆H° in standard reference° °temperature of 298 °KCalculate ICPH and ICPS for all temperaturevariations and reactionscondition at PT. PUSRI II reformer furnace. Natural gasfuel compositions consist of CH 4 64.61%, C 2 H 6 7.97%,C 3 H 8 7.08%, i-C 4 H 10 1.25%, n-C 4 H 10 1.76%, i-C 5 H 121.01%, n-C 5 H 12 0.56%, and CO 2 15.76% moles.The exergy calculation of the material streams ateach point of flows, based on the block diagram in theprocess descriptions, is described in Figs 5 and 6 forthe related processes.Calculate ∆G° for all temperature variationsand reactionsCalculate K in all temperature variations andreactionsCalculate ξ (conversion) in each reaction withMaple 11 programCalculate y i (mole fraction) for all componentsin reactionsBlock 1Calculate Q input , Q reactionsCalculate Q total = Q input + Q reactionsBlock 1Figure 4. Block flow of calculation of material and energystreams.process B for the same initial feed of a raw natural gasis shown in Table 3Assumption has been taken for the conventionalcombustion temperature, which is 1927 ◦ C and theambient temperature is 25 ◦ C. The overall efficiency ofthe furnace is 88% with heat release of the natural gasfuel is 15 189 Btu lb −1 based on the actual operatingRESULTS AND DISCUSSIONThe calculation of exergy at discrete points in a systemwill indicate the total exergy change, reversible andirreversible in the process and in the equipment betweenany two points. These exergy values are explicit valuesrelative to the reference condition of the surroundingenvironment. On the basis of the calculation of exergyperformed, the results of the exergy analysis are shownin Tables 4 and 5.The overall reactions of natural gas feed occurs inthe primary reformer is endothermic reaction, that consistsof higher hydrocarbons reactions. From Table 4, itcan be seen that the largest amount of the exergy lostis in the primary reformer that accounts for 95.17%of the total exergy lost in process A. The presenceof high exergy lost in the primary reformer may indicatea poor match between the quality of the energysupplied to the process and the quality required bythe process to perform a given task. This mismatchresults in large irreversibility. Heat released by fuel ofnatural gas can reach temperature to 1927 ◦ C throughthe conventional combustion furnace, while the maximumtemperature required by the process only reaches815 ◦ C.In the secondary reformer, the exergy loss is inthe minimum percentage in which the temperatureoperation needed reaches 978 ◦ C for the exothermicreaction occurred with large amounts of moles inputand heat used. Although the exergy values alwaysSyngas PRSyngas PRAir3 4Fuel In Flue Gas Out Fuel InFlue Gas OutPRIMARYSECONDARYREFORMERREFORMER12 78(PR)(SR)95Steam Natural Gas Feed Syngas SRTo SyngasPreparation (SP)6Figure 5. Exergy analysis diagram of the series reformers, process (A).© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

768 M. DJONI BUSTAN, S. MAHADZIR AND D. KHARISMADEWI Asia-Pacific Journal of Chemical EngineeringTable 3. Mass and heat input of the material streams.ProcessTotal moles input(kmole h −1 )Total heat input(kJ h −1 )A 1. In the primary reformer, from natural gas to synthesis gas 5857.06 99 392 276.412. In the secondary reformer, from synthesis gas to enriched synthesis gas 9657.16 265 845 258.21B 1. In the catalytic partial oxidation, from natural gas to synthesis gas 1823.88 34 765 753.392. In the steam reformer, from synthesis gas to enriched synthesis gas 7628.99 158 957 291.79Syngas POXSyngas POXSteamNitrogen3 4Fuel In Flue Gas Out Fuel InFlue Gas Out1PARTIALOXIDATION(POX)2 7STEAMREFORMER(SR)89Oxygen Natural Gas Feed Syngas SR5To SyngasPreparation (SP)6Figure 6. Exergy analysis diagram of the modified flowsheeting reformer,process (B).show positive to the reference surrounding environment,in the secondary reformer it shows negative value.The zero reference for enthalpy and entropy valuesin the literature often results in positive and negativevalues of exergy calculation for the system. Totalexergy losses of the entire processes in process A are2 821 615.23 MJ h −1 , its amounts will reduce throughthe modified flowsheeting reformer which is shown inTable 5.As shown in Table 5, the total exergy lost for theentire process B is 1 091 015.74 MJ h −1 . The flowsheetmodification of process A is done through replacementof the secondary reformer with a partial oxidationprocess upstream of the primary reformer. Process Bshows a significant potential reduction in the totalexergy lost in primary reformer (steam reformer) byup to 78.11% relative to the conventional process A.Furthermore, the operating conditions in the primaryreformer of process B is reduced to 500 ◦ C, 20 atmthereby reducing the consumption of natural gas fuelin the reformer. Heat released from the exothermicpartial oxidation process may be recovered to provideheating for the endothermic reaction in steam reformerof process B.So, the overall exergy losses in process A can bereduced through applying a modification such in processB, as much as% exergy savings( )exergy losses in process A − exergy losses in process B=exergy losses in process A× 100%( )2821615.23 MJ/h − 1091015.74 MJ/h=× 100%2821615.23 MJ/h% exergy savings = 61.33%CONCLUSIONSThis article presents a path for calculating exergy lossesfor each step in the main parts of synthesis gas productionfrom natural gas. The path is integrated withthe modified flowsheeting reformer of process B andcompared to the series reformers of process A. Exergyanalysis is found to be useful for the process analysis.It provides an objective measurement for selectingand judging processes regard to their energy (or exergy)utilization. The overall exergy losses (%) in process Acan be reduced until 61.33% through modification toprocess B. The result is achieved for the same initialfeed of a raw natural gas. Furthermore, the exergy loss© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

Asia-Pacific Journal of Chemical Engineering THE PRODUCTION OF SYNTHESIS GAS FROM NATURAL GAS 769Table 4. The exergy analysis of the series reformers (process A).Station orcomponentExergy(MJ h −1 )Change inexergy (MJ h −1 )Heat transfer(MJ h −1 )Chemicalchange (MJ h −1 )Heat transfer(MJ h −1 )Exergy lossesFriction(MJ h −1 )Heat rejection(MJ h −1 )Total(MJ h −1 )Total exergyloss (%)1 49.94Furnace primary 5.99 5.99 5.99 −0.00(Ex a)2 43.95Primary reformer −2 685 274.29 43.95 −1 306 860.27 −1 378 414.01 −2 685 274.29 95.17(Ex b + Ex c) (Ex b) (Ex c)3 −1187.94Piping 898.13 −156 418.00 −156 418.00 0.01 −156 417.99 5.54(PR to SR) (Ex d )4 −2086.07Secondary reformer 22 955.61 117.55 1 222 160.24 −1 199 204.63 22 955.61 −0.00(Ex b + Ex c) (Ex b) (Ex c)5 931.45Piping −0.30 −2900.54 −2900.54 −0.05 −2900.59 0.10(SR to SP) (Ex d )6 931.757 133.58Furnace secondary 16.03(Ex a) 16.03 16.03 −0.008 117.55−2 821 615.23 100.00© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj

770 M. DJONI BUSTAN, S. MAHADZIR AND D. KHARISMADEWI Asia-Pacific Journal of Chemical EngineeringTable 5. The exergy analysis of modified flowsheeting reformer (process B).Station orcomponentExergy(MJ h −1 )Change inexergy(MJ h −1 )Heattransfer(MJ h −1 )Chemicalchange(MJ h −1 )Heattransfer(MJ h −1 )Exergy LossesFriction(MJ h −1 )Heatrejection(MJ h −1 )Total(MJ h −1 )Totalexergyloss (%)1 17.47Furnace POX 2.10 2.10 2.10 −0.00(Ex a )2 15.37Partial −500 907.94 15.37 270 586.09 −771 494.03 −500 907.94 45.91oxidation (Ex b + Ex c ) (Ex b ) (Ex c )3 1369.34Piping −137.05 −2504.92 −2504.92 −0.01 −2504.92 0.23(POX to SR) (Ex d )4 1506.40Steam −587 614.55 70.29 −295 897.19 −291 717.36 −587 614.55 53.86reformer (Ex b + Ex c ) (Ex b ) (Ex c )5 −209.17Piping (SR to 0.00 0.00 0.00 −0.01 −0.01 0.00SP) (Ex d )6 −209.177 79.87Furnace SR 9.58 9.58 9.58 −0.00(Ex a )8 70.29−1 091 015.74 100.00in the primary reformer itself can be reduced to 78.11%relative to process A.NOMENCLATURET 0T 0 , TyThe ambient temperature, ◦ KReference temperature; Process temperature,◦ KMole fractionυ Stoichiometric numberεReaction coordinateρ 2 , ρ 1 Density of components in the materialstreams at point 2 and 1, kg cm −3H Total enthalpy of material streams, kJkmole −1S Total entropy of material streams, kJkmole −1G 0 , G 0 0 Gibbs energy, standard Gibbs energy ofpure component at 298.15 ◦ K, J mole −1H 0 , H 0 0 Enthalpy, standard enthalpy of pure componentat 298.15 ◦ K, J mole −10C p Heat capacityEx 1,2,...,9 The exergy in point 1 or 2 or . . . to 9,kJ h −1K Equilibrium constantm Mole input to the process, molen i,0 Mole of pure component; Mole total, moleP 2 , P 1 Pressure of process at point 2 and 1, atmP 0 , P Reference pressure; Process pressure, atmQ Heat, kJ h −1R Universal gas constant, J mol −1 K −1T g The conventional combustion furnacetemperature, ◦ KREFERENCES[1] I. Dincer, M. Rosen. Exergy: Energy, Environment, andSustainable Development, Elsevier: Oxford, United Kingdom,2007.[2] E. Sciubba, G. Wall. Int. J. Thermodyn., 2007; 10(1), 1–26.[3] J.E. Ahern. The Exergy Method of Energy Systems Analysis,John Wiley & Sons Inc: USA, 1980.[4] A.P. Hinderink, F.P.J.M. Kerkhof, A.B.K. Lie, J.D.S. Arons,H.J.V.D. Kooi. Chem. Eng. Sci., 1996; 51(20), 4693–4700.[5] M.D. Bustan, S. Haryati, I.G. Mendera. Studi Analisis PerbandinganAntara Konvensional dan Modifikasi FlowsheetingSistem Produksi Syngas Terhadap Konversi dan PenurunanKonsumsi Panas Pada Reformer PT. PUSRI II Palembang,Thesis, Sriwijaya University, Indonesia, 2009.[6] M.V. Twigg. Catalyst Handbook, Wolfe Publishing Ltd:England 2 nd edn., 1989.[7] S. Bharadwaj, L. Schmidt. Catalytic Partial Oxidation ofNatural Gas to Syngas Fuel. Proc. Tech, 1995.[8] M.R. Felder, R.W. Rousseau. Elementary Principles of ChemicalProcesses, John Wiley & Sons Inc: USA, 2005.[9] O. Levenspiel. Chemical Reaction Engineering, Wiley Eastern:New Delhi, 1999.[10] J.M. Smith, H.C. Van Ness. Introduction to ChemicalEngineering Thermodynamics, McGraw-Hill: New York,1987.[11] W.D. Seider, J.D. Seader, D.R. Lewin. Process DesignPrinciples; Synthesis, Analysis and Evaluation, John Wiley &Sons Inc: USA, 1999.[12] A.P. Hinderink, F.P.J.M. Kerkhof, A.B.K. Lie, J.D.S. Arons,H.J.V.D. Kooi. Chem. Eng. Sci., 1996; 51(20), 4701–4715.© 2011 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. 2011; 6: 760–770DOI: 10.1002/apj