You are on page 1of 11

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267041621

Aromatization of Propane: Techno-Economic
Analysis by Multiscale “Kinetics-to-Process”
Simulation
Article in Computers & Chemical Engineering · January 2015
DOI: 10.1016/j.compchemeng.2014.10.001

CITATIONS

READS

3

478

5 authors, including:
Michele Corbetta

Carlo Pirola

Politecnico di Milano

University of Milan

30 PUBLICATIONS 139 CITATIONS

115 PUBLICATIONS 1,365 CITATIONS

SEE PROFILE

SEE PROFILE

M.V. Tsodikov

Andrey V. Chistyakov

Institute of Petrochemical Synthesis

Institute of Petrochemical Synthesis Russian …

153 PUBLICATIONS 562 CITATIONS

38 PUBLICATIONS 133 CITATIONS

SEE PROFILE

SEE PROFILE

Some of the authors of this publication are also working on these related projects:
A novel manufacturing process for photocatalytically activate ceramic tiles by digital printing View
project
AG2S (Acid Gas to Syngas) technology, a novel process to reuse CO2 View project

All content following this page was uploaded by Michele Corbetta on 23 October 2014.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.

Via Golgi. Introduction The recent interest in third generation biorefineries results in an increased attention on correlated processes for the valorization of by-/co-products. Flavio Manenti a. Tsodikov c . Focusing the attention on oleaginous biomass sources (e. E-mail address: flavio.10. HRJ fuel is substantially composed by a pool of long chain paraffins free from aromatics compounds. saturation of double bonds. Andrey V. which is defined as the mono-alkyl esters of fatty acids (FAME) derived from triglycerides. Beside biofuels. The revised kinetic scheme is then embedded in a process simulation. 2014). 2005) for the threshold value of 25 vol% of aromatics in the Jet A-1 category. Similarly. starting either from waste animal fat (Bianchi et al. a general kinetic model of propane aromatization has been developed. Basing on the process simulation and on available price assessments. the biomass synthetic jet fuel production process involves removing oxygen molecules and other undesirable materials from triglycerides through hydrotreatment. Transesterification with an alcohol. 20133 Milano. 2010). © 2014 Elsevier Ltd. co-production of large amounts of propane is observed. 1. 19. Chistyakov c a Politecnico di Milano. Mark V.: +39 0 2 2399 3273. Piazza Leonardo da Vinci. Tel. 20133 Milano. Materiali e Ingegneria Chimica “Giulio Natta”. Italy c A. Based on these results. Italy Università degli Studi di Milano.V.. microalgae). The recent interest in this technological route derives from the development of new third generation biorefinery concepts.g. Manenti).manenti@polimi.compchemeng.. Kinetic studies of propane aromatization over H-ZSM-5 zeolite in a wide range of conversions are reported in the literature.elsevier. two main technological routes are adopted for the conversion of raw bio-oil to biofuels.com/locate/compchemeng Aromatization of propane: Techno-economic analysis by multiscale “kinetics-to-process” simulation Michele Corbetta a . such as methanol. The main difference between biodiesel and green diesel consists in their chemical characterization.1016/j. occurs under mild conditions in the presence of a base catalyst and co-produces low-value glycerol.. which is not true for biodiesel (Knothe.Computers and Chemical Engineering 71 (2014) 457–466 Contents lists available at ScienceDirect Computers and Chemical Engineering journal homepage: www. Leninskiy prospect. which are the perquisite for a zero-waste and economic sustainable plant. performed with the commercial code SimSci PRO/II by Schneider Electric.org/10. a techno-economic analysis has been performed to show limits as well as potentialities of the proposed layout.it (F. Carlo Pirola b .doi. 29. Recent works show the possibility to pre-treat raw oil (broadly composed by a triglyceride mixture) with a catalytic deacidification step in order to lower the free-fatty-acids (FFA) content and to avoid soap formation in the subsequent transesterification process. which can be upgraded by a catalytic conversion to aromatics on zeolites. Dipartimento di Chimica. As a consequence.. Russian Federation b a r t i c l e i n f o Article history: Received 17 April 2014 Received in revised form 26 September 2014 Accepted 4 October 2014 Available online 12 October 2014 Keywords: Propane aromatization Multiscale simulation Kinetic modelling Process development Excel unit operation module Hydroprocessed renewable jet (HRJ) fuel a b s t r a c t This paper addresses the techno-economic analysis of the propane aromatization process. The second technological route consists in a hydrodeoxygenation at high temperature and pressure in the presence of a catalyst. HRJ fuel meets the world jet fuel specifications (ExxonMobil Aviation. or raw sunflower oil (Pirola et al.Topchiev Institute of Petrochemical Synthesis. . in which. 32. 2009) ∗ Corresponding author. Dipartimento di Chimica. Hydroprocessing is employed in conventional refinery processes to firstly deoxygenate and remove undesirable materials including nitrogen.2014. and a subsequent selective hydrocracking/rearrangement of the atomic structure to yield lighter hydrocarbons suitable for aviation (Jet A-1: C8–C16) (Blakey et al. sulphur and residual metals and break down carbon chain lengths. 2011). All rights reserved. All rights reserved. Moscow 119991. The first one that had been developed is the transesterification to biodiesel. fax: +39 0 2 2399 3280. by adopting a novel kinetics-to-process approach. heteroatoms rejection and isomerization. which can be subjected to aqueous-phase-reforming to produce hydrogen.∗ .001 0098-1354/© 2014 Elsevier Ltd. with the production of drop-in renewable (green) diesel and hydrotreated renewable jet (HRJ) fuels. Green diesel is a drop-in fuel with a chemical structure similar to the petro-derived diesel irrespective of the feedstock and without oxygen content. These catalytic processes determine the cracking of triglycerides. http://dx. algal oil is subjected to catalytic hydrodeoxygenation processes for the production of (Hydrotreated Renewable Jet) HRJ fuels.

such as gallium. Applications of CAPE-OPEN standard in the field of PSE include. CORBA. propane aromatization process) consists of complex systems occurring at different levels.. one of the biggest concerns of alternative fuels has come from their low aromatic content. and different space times (Bhan et al. it should be noted that a global chemical process (e. A strategy to produce HRJ fuel that satisfies this requirement is to blend it with aromatics produced from bio-based propane aromatization. the use of the excel unit operation module allows directly post-processing and reporting simulation results in tables and graphs. HRJ fuels are produced along with propane. Going further on. 2008).. 2005).NET Framework).. For this reason the aromatic content of synthetic fuel blends is currently fixed at a minimum of 8% (Blakey et al. Blocks diagram of a possible third generation zero-waste biorefinery. regression of kinetic and thermodynamic models) could by directly embed within process simulators without the need for further modifications. Light olefins formed in the initial reaction steps give rise to fast oligomerization/cracking steps that control olefin distribution. Moreover. the bridging between PMC and PME is obtained by a proper Middleware (e. The necessity to include a custom process modelling component (PMC) derives from the choice of describing the catalytic reactor with a detailed surface kinetic mechanism. 2012). The lack of commercial software able to integrate all these multidisciplinary features (Charpentier. Section 3 shows results of the process simulation of the Cyclar process.. and alkane activation by carbonium ion-like transition states. This work proposes a new kinetic mechanism (micro and reactor scales) of propane aromatization. 2008).. In both cases. The novelty of this work relies on the use of a different standard and Middleware. 2000).. Particularly. 2006). This concept consists in a main hydroprocessing step of the algal oil. The same authors proposed additional investigations on the role of gallium modified zeolites (Bhan and Nicholas Delgass.. which has been extensively adopted both in the industrial and in the academic practice to model process plants that do not yet exist and to evaluate their economic viability (Lam et al. Corbetta et al. This is because Jet-A contains about 20% aromatics while HRJ consists mostly of iso-paraffin and cyclo-paraffin. The process simulation is performed with the PRO/II process modelling environment (PME). the modelling of thermodynamic properties and the modelling of new unit operations (Morales-Rodríguez et al.. Analysis of the experimental and kinetic modelling data has confirmed that propane and n-butane transformation over H-ZSM-5 occurs via both protolytic cracking of C C and C H bonds in paraffin molecules and hydrogen transfer between the propane and product olefins adsorbed on acid sites. On the other hand.g. 2.. During this catalytic conversion. Hui and co-workers studied the ignition and combustion of alternative jet fuels and they found that cetane numbers (CN) are higher than that of Jet-A.g.. as the seals would shrink. which is embedded in a multiscale process simulation (process scale). The rates of production of several species are reported in both papers. 2011). . Microsoft Excel is adopted to manage all communications between the PRO/II PME and the PMC (catalytic reactor unit). 2007) many efforts have been devoted to the definition of an efficient computer-aided framework to integrate PMCs in PMEs. / Computers and Chemical Engineering 71 (2014) 457–466 BTX AROMATIZATION BENZENE TOLUENE XYLENE Propane MICROALGAE SEPARATION Lipids HYDROPROCESSING BIO-JETFUEL Non-lipidic fraction DIGESTION/REFORMING Hydrogen Fig. In fact. homemade codes developed by engineers during R&D activities (e. Reaction kinetic studies of propane catalytic conversion to aromatics were also conducted by Bhan and co-workers on the same catalyst at atmospheric pressure and temperatures in the range of 520–550 ◦ C. The resulting integrated third generation biorefinery is illustrated in Fig. According to Blakey and co-workers. 1. Additionally. Process System Engineering (PSE) provides numerical tools and platforms devoted to the investigation of the so-called Chemical Supply-Chain (Grossmann and Westerberg. 1. could be added to enhance the dehydrogenation function. The advantage of this approach derives from its easiness and usability.g.. COM. In general. due to their high resistance to deactivation. Unfortunately the selectivity towards high-value aromatics is reduced by the significant co-production of fuel gas (methane and ethane). For instance. which are less reactive compared to n-paraffins (Hui et al. Concern has been raised that the ageing seals in the aircraft and engine would leak if the aromatic content is too low. 1994) and propane aromatization (Lukyanov et al. for instance. 2011). derived from the separation between the lipid and non-lipid fractions of the harvested microalgae. Nguyen and co-workers performed experimental and modelling studies of propane and n-butane aromatization over H-ZSM-5 zeolite at 500 ◦ C in a wide range of conversions (Nguyen et al. 2009) drives the research towards the integration of different tools targeting different scales.458 M. metal components. beyond the kinetic modelling and towards the plant scale. which has a lower economic value.. 2005). Section 2 addresses the reactor and kinetic modelling of propane aromatization on H-ZSM-5 catalysts. performed . as demonstrated in both the mechanistic studies of ethene and propene aromatization (Lukyanov et al. as outlined in Fig. zeolites with MFI pore structure are used. reactions of these alkoxide species by carbenium ion-like transition states. This co-product is then aromatized over H-ZSM-5 catalyst to produce aromatics for blending and for sale on the petrochemical market. They developed a kinetic model considering surface species as neutral alkoxides. each with its own unique multiscale structure (Li et al. In the literature (MoralesRodríguez and Gani. which spans from the science-oriented study at the molecular level up to the system-oriented study of process networks and beyond. the CAPE-OPEN standard has been widely adopted and integrated within PMEs. 1994).

alkylcyclohexenes. which.C8 Fig. As a consequence. can activate paraffins with H-transfer reactions (Kazansky et al.5 times faster. 1996). PROPANE 2. in turn. 3. and finally Section 5 includes results and general comments.i Pi lib [−] . Once olefins are produced... or aromatize with other olefins.. 1995). 3. while the full set of reactions (79 reactions). etc. ∗ i = Kads. For instance.. along with kinetic parameters. plus adsorption and desorption steps. It is worth mentioning that other chemical species (i. alkyl-cyclohexadienes. Accordingly.1. Kads ). being the first route roughly 2. Reaction pathways.M.5 1 Protolytic Cracking H-Transfer The reactive system has been modelled considering the following 19 chemical species and lumps. Stoichiometries are generated according with the following rules. paraffins are activated on acid sites by protolytic cracking or dehydrogenation.C8 Aromatization (Cyclization + 3 H-Transfer) 3 PARAFFINS C2 . Logical blocks of the multiscale approach applied to the aromatization process. PARAFFINS C2 . Corbetta et al. The chemical species listed in Table 1 are involved in several reactive steps.C8 + OLEFINS C4 – C5 3 OLEFINS C2 . 1997). 2. paraffin C6 includes all the saturated chain isomers with six carbon atoms. is reported in Appendix 1. These intermediates species participate in the fast sub-steps of the aromatization of large olefins. with the commercial code SimSci PRO/II by Schneider Electric. The reaction pathway is illustrated in Fig. / Computers and Chemical Engineering 71 (2014) 457–466 459 Fig. which could be categorized in 5 reactions classes. Section 4 reports some economic considerations.e. they can rapidly oligomerize yielding heavier olefins (Buchanan et al.C8 AROMATICS (BTX) C6 . evolving the corresponding paraffins.) are observed just in traces and for this reason they are not taken into account in the model.Cracking OLEFINS C6 . each reaction is characterized by only one parameter (adsorption constant. Adsorption/desorption on acid sites (Zeolite HZSM-5) Olefins Aromatics Cn An Cn ↔ (Cn ) An ↔ (An )* ∗ Adsorption and desorption steps are considered at equilibrium because they are faster than the other reactive steps (Lukyanov et al. Hydrogen Paraffins Olefins Aromatics H2 CH4 –C8 H18 C2 H4 –C8 H16 B/T/X 1 8 7 3 2.C8 Protolytic Dehydrogenation RDS + + CH4 H2 Oligomerization . 2. Kinetic modelling Table 1 Chemical species and lumps.

which could be due to coupled chemicalphysical phenomena.6 ARO 0. comprising all species with at least 5 carbon atoms.n Pn  kKIN. B-Scission/oligomerization (Cn )∗ ↔ (Cn−m ) + (Cm )∗ b.9 C3 0.2.. (1) reports the material balance for the ith species. The nonlinear regression has been accomplished minimizing the objective function of Eq.j  (3) RT Regression of kinetic parameters (Buzzi-Ferraris and Manenti. which is well reproduced by the kinetic model.k (yi. BzzMath library includes special numerical methods able to simultaneously handle the so-called narrow-valley problem.1 0.8 C1-C4 0. b)) (4) i=1 k=1 Model predictions are compared with experimental data in Fig. Activation of paraffins on acid sites (Zeolite HZSM-5) rj = kKIN.. The objective function is a weighed (ωi. Comparisons of model simulations and experimental data (Nguyen et al. leading to a reduction of the computational time.% in B/T/X (Giannetto et al. C5+ products. which typically arise from the estimation of kinetic and thermodynamic parameters. 2012).6 C5+ 0. Panel (b) shows the model .  Nreact 2.9 C3 0. Propane is progressively converted to two main products. 4. which are aromatics and fuel gas (essentially methane and ethane).it/. while C2 –C4 product composition is almost constant in the range of 7 wt. Eq. 2. 2006). are found in small amounts.k − gk (xi .4 0.7 C2=-C4= 0.chem. Note that the model nicely fits experimental data of species mass fractions as a function of WHSV. Finally.j rj MWi − Eact. Nreact i. (4). b) is the model prediction of the k-th dependent variable. 1995).5 0. using experimental data from the available literature (Nguyen et al.2 0. 4. and the possible presence of bad-quality measures with the identification of outliers.3 0. kinetic constants (kKIN.polimi.460 M. Experimental aromatics distribution has an average trend of 26/51/23 wt. Panel (b): comparison between the full model (solid line) and the power law model without adsorption.2 0. 1994). the possible multicollinearities. Panel (a): full model.j (a) Protolytic cracking a.. Such optimizers are based on the object-oriented programming and parallel computing.3 0.7 Mass fracon [-] Mass fracon [-] 1 C5+ 0. in which gk ( xi . (2) reflect the saturation effect of olefins and aromatics adsorbed on surface active sites.k ) least sum of square errors.%. / Computers and Chemical Engineering 71 (2014) 457–466 1 0. Aromatization (Cm )∗ + 3 Cn → 3 Cn + (Am )∗ The mathematical model of the catalytic system consists in a PFR isothermal pseudo-homogeneous reactor model with reaction rates evaluated according with the Eley–Rideal (E–R) surface kinetic mechanism.j = Aj · exp (Cn )∗ → (Cn )∗ + H2  (2) n=1 (b) Protolytic dehydrogenation dωi = ri MWi = d  Nads 1+ (Cn )∗ → (Cn−m )∗ + Cm Pm m=1 (1)   N exp Ny S(b) = 2 ωi.1 0 0 0 10 20 1/WHSV [h] 30 0 40 10 20 1/WHSV [h] (a) 30 40 (b) Fig. H-Transfer (Cn = )∗ → Cm + Cn + (Cm = )∗ c. Corbetta et al. E–R reaction rates of Eq.4 0. 2009) has been performed by means of the set of very robust optimizers belonging to the BzzMath library (Buzzi-Ferraris and Manenti. Reactivity of insaturated on acid sites (Zeolite HZSM-5) j=1 KADS. as a function of independent variables (xi ) and model parameters (b). The library is available at http://super.8 C1-C4 0.j ) are considered with an Arrhenius expression.3.5 ARO C2=-C4= 0. in which ω represents its mass fraction and  is the reciprocal of the weight hourly space velocity (WHSV = total mass flow/catalyst mass load).

The material balances and the global performances of the plant are reported in Table 2. The BP/UOP Cyclar aromatization process has been selected and simulated by means of the commercial code PRO/II (Fig.5 4.. while the last one separates toluene and xylene. Propane and heavies are recycled. the sensitivity of product distribution on aromatization rates appears to be negligible. in order to find out the role of the different reaction classes on the catalytic conversion process (Fig.21 1.97 0. From Fig.1 0 Specification 8.820 0. with an asymptotic value of about 80%. The first tower separates benzene from the top. 2013) and in surface reactive CFD simulations (Corbetta et al. neglecting the denominator of Eq. which overestimate the conversion of propane to aromatics and fuel gas.396 0.900 0.2 325 0. where the cut between ethane and propane is realized.e.8 0. Multiscale process simulation Fig. 5) could be useful to understand the surface reaction mechanism. mainly decreasing the Once the kinetic scheme has been validated against experimental data.. Finally. From this parametric study emerges that protolytic cracking/dehydrogenation is the limiting step. results considering a power-law model (i. corresponding to the saturation effect).9 Surface coverage [-] 0. highlighting the role of the different adsorbed species on the saturation of the catalyst. oligomerization and cracking reaction classes do not affect the conversion of propane but slightly change the product distribution.010 100 29.74 0.00E-03 0. On the other hand.40 1 1.4 0. Its reaction rate is the smallest one (about 1000 times smaller than other reactions).790 1.3 2. The sensitivity analysis has been carried on by multiplying by ten all the reaction rates belonging to the same class.00E-03 0.00E-02 Table 3 Reactor specifications. exploiting the excel unit operation module. the reactor is a multiple inter-heated adiabatic annular moving-bed catalytic reactor.00E-03 0 10 20 1/WHSV [h] 30 40 θC3=* [◦ C] [bar] [t] [−] [−] Reactor temperature Reactor pressure Catalyst load Propane conversion (mass) BTX yield (mass) 500 1. 5. The propane feed enters the catalytic reactor after a mixing step with the unconverted and recycled propane.00E-03 0. The reactor effluent enters a feed-effluent-heat-exchanger (FEHE) and it is compressed and sent to a separator.00E+00 3. it has been corroborated with a process simulation. 6).7 6..2 115 −214 215 1. Increasing H-transfer rates results in a higher conversion of propane to light paraffins.69 0. (2). This unit has been simulated with a custom model.00 0.99 0.32 . while in Table 3 and Table 4 reactor and distillation columns specifications are summarized respectively.20 9.373 θC4=* θC5=* θC6=* amount of light olefins (C2 –C4 ). / Computers and Chemical Engineering 71 (2014) 457–466 461 Table 2 Global material balances. 0. The Cyclar process is composed by a catalytic section and a downstream section.6 θfree θARO* θC2=* 0. Corbetta et al. and it determines the highest sensitivity to product distribution. Number of trays Condenser pressure Condenser temperature Condenser duty Reboiler duty Reflux ratio [−] [bar] [◦ C] [kW] [kW] [−] De-ethanizer Stabilizer Benz-distill TX-splitter 30 20 −74 −491 57 0. Industrially.6 9 10 −78 −23 134 1 17 1. a liquid stream mainly composed of aromatics is fed to the last two purification columns. This is due both to the fact that aromatics adsorption constant is five times greater than the olefins one.96 0. From the bottom of the stabilizer. Moreover. the gas phase is further compressed and fed to a de-ethanizer.91 3. while in the partial condenser a liquid stream rich in ethane (with methane) and a gas stream rich in methane (with hydrogen) are withdrawn. Stream Mass flowrate [t/h] Molar flowrate [kmol/h] Mass fraction of the key component [−] Propane (feed) Fuel gas C2 H6 – CH4 (70–30% mol) CH4 – H2 – C2 H6 (60–30–10% mol) Benzene Toluene Xylene Purge (propane – benzene) 4. the process simulator (PME) calls a Microsoft Excel file (Middleware) with a Visual Basic macro that executes a reactor model routine (PMC) with the detailed kinetic model previously reported. 2014). 7).78 0.7 134 6. Model prediction of surface coverage by olefins and aromatics.2 0. In this way.410 0. In the downstream section.484 0. as outlined in recent spectrokinetic analysis (Visconti et al. Table 4 Distillation columns specifications. surface coverage of active sites (Fig.M. Once the kinetic scheme has been validated. where light hydrocarbons are stripped from the top and recycled to the catalytic section. and to the low partial pressures of olefins. 0. 5.910 0.2 49 −555 572 Total 30 1. we can notice that aromatics are largely responsible for the coverage. a sensitivity analysis on kinetic constants has been performed. The liquid stream effluent from the flash is pumped to a stabilizer.

5 0.7 C2=-C4= Mass fracon [-] Mass fracon [-] 1 C5+ 0.8 C1-C4 0.6 ARO 0. optimizing some process parameters (feed trays.4 0. Sensitivity analysis on reaction class kinetic constants.6 ARO 0.8 C1-C4 0.3 0. The simulation has been carried on with a SRK thermodynamic package.3 0.5 0.9 C3 0. . Cyclar process layout adopted in the simulation.2 0 ARO 0. stabilizer.1 0 0 0 10 20 1/WHSV [h] 30 40 0 10 20 1/WHSV [h] 30 40 Fig.4 0. / Computers and Chemical Engineering 71 (2014) 457–466 1 0.9 C3 0. number of stages. The resulting liquid molar fraction profiles inside the 4 columns (de-ethanizer.4 0.6 C5+ 0.3 C5+ 0.7 C2=-C4= 0.) by minimizing heat duties.462 M. 7.1 10 20 1/WHSV [h] 30 0 40 0 10 20 1/WHSV [h] 30 40 1 1 0. Bdistiller. 8.5 0. etc. Corbetta et al. 4.7 C2=-C4= Mass fracon [-] Mass fracon [-] 0.1 0. some economic consideration is discussed below. First of all.1 0.2 0.8 C1-C4 0.2 0. 6.4 0.2 0 C5+ 0. prices of the involved chemicals have been collected from the Platts Fig.5 0.3 0.9 C3 0.7 C2=-C4= 0. and TX-splitter) are reported in Fig.6 ARO 0. Economic considerations On the basis of the detailed process simulation.8 C1-C4 0.9 C3 0.

6 TOLUENE 0. and a premier source of benchmark price assessments for those commodity markets (Platts.9 5 Tray number 7 9 TX-SPLITTER 0.9 0.8 Liquid molar fracon [-] 0.8 DE-ETHANIZER 0. . Bulk chemicals prices show a significant fluctuation.5 ETHANE 0.3 0.2 0. Corbetta et al.9 0. Fig. 9. where the Platts Global Petrochemical Index (PGPI) is reported for Toluene in the period May 2013–July 2014. a leading global provider of energy. 8.8 Liquid molar fracon [-] 463 0.4 0.5 PROPANE 0.2 0. The typical fluctuation of petrochemical prices is highlighted in Fig. metals and agriculture information. database (Table 5).6 0.4 PROPANE 0. Liquid composition profiles of the four distillation columns (purity specifications are highlighted with circles).4 BENZENE TOLUENE 0. petrochemicals. In this study.2 0.1 0.3 Liquid molar fracon [-] Liquid molar fracon [-] STABILIZER PROPANE 0.1 0 0 1 6 1 11 16 Tray number 21 1 26 3 1 BENZ-DISTILLER 0.5 0. 2014).M.6 0.3 0. 9. according with the process yields.5 XYLENE 0. annual average values of 2013 have been adopted.8 0.7 0. supported by experimental data.7 0.4 TOLUENE 0.1 0. and.3 0.7 0.7 0.1 0 0 1 6 11 Tray number 16 1 6 11 16 Tray number 21 26 Fig. Platts global petrochemical index (PGPI) for Toluene in the period June 2013–July 2014 (Platts. 2014).6 0. / Computers and Chemical Engineering 71 (2014) 457–466 1 0.2 ETHANE BENZENE 0.

process modelling.909 186.941 1. A minimum of about 60 wt. by increasing the selectivity towards aromatics.000 142. The paper proposes a comprehensive study for the propane aromatization process specifically addressed to the integration within a third generation (microalgae) biorefinery concept. The kinetic scheme is reported below (Table A1) along with pre-exponential factors and activation energies. Table A1 Kinetic scheme of propane aromatization over H-ZSM-5 catalyst.000 142. the proposed solution does not imply the consumption of fossil fuels.000 142. chemical reactors.45 0.729 93.8881 186. 2002). once methane and ethane are produced by cracking reactions.000 142.% in aromatics’ yield is necessary for economic sustainability. EP2 = NP  D P.909 186. to have a positive economic balance.000 142.3 105. The kinetic scheme has been tuned with available experimental data by means of nonlinear regression.000 142. By doing some rough consideration on prices.909 186.165 −2641.000 .909 186.909 186.3 Platts (2014) Platts (2014) Platts (2014) Doty (2004) Platts (2014) Platts (2014) Platts (2014) the global second order economic potential results in a negative value. In order to assess this technological route.909 186.000 142. which are also active in methane aromatization (Shu et al.000 142. Consequently. This approach relies on the use of Microsoft Excel as Middleware between the PRO/II PME and the PMC of custom unit operations. is here provided.56 475.000 142. along with kinetic constants.000 142. Catalysts with a higher selectivity towards aromatics (up to 80%) are reported in the literature (Wang et al. providing an aprioristic reliable assessment of the techno-economic benefits for novel and innovative biomass conversion pathways.i ni < 0 (5) i=1 This is due to the fact that large amounts of fuel gas are coproduced along with the main BTX material stream.000 142. Propane Ethane Methane Hydrogen Benzene Toluene Xylene 0.909 142. and the full mechanism.599 0.909 186.000 142.000 142. In fact.polimi.909 186..000 142.38 101.000 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Protolytic C C Cracking C3 H8 > C2 H4 + CH4 C4 H10 > C2 H4 + C2 H6 C4 H10 > C3 H6 + CH4 C5 H12 > C2 H4 + C3 H8 C5 H12 > C3 H6 + C2 H6 C5 H12 > C4 H8 + CH4 C6 H14 > C2 H4 + C4 H10 C6 H14 > C3 H6 + C3 H8 C6 H14 > C4 H8 + C2 H6 C6 H14 > C5 H10 + CH4 C7 H16 > C2 H4 + C5 H12 C7 H16 > C3 H6 + C4 H10 C7 H16 > C4 H8 + C3 H8 C7 H16 > C5 H10 + C2 H6 C7 H16 > C6 H12 + CH4 71.000 142. and chemical Appendix 1.000 142. Conclusions A completely new methodology to face the relevant issue of economic and industrial feasibility of process layouts in the area of biorefineries has been developed and applied to a specific case study. Pre-exponential factor [mol/g/h/Pa] or [mol/g/h/Pa2 ] # Reaction 1 2 3 4 5 6 7 Protolytic C H Cracking C2 H6 > C2 H4 + H2 C3 H8 > C3 H6 + H2 C4 H10 > C4 H8 + H2 C5 H12 > C5 H10 + H2 C6 H14 > C6 H12 + H2 C7 H16 > C7 H14 + H2 C8 H18 > C8 H16 + H2 Activation energy [J/mol] 21. 5. thus requiring further efforts on more selective catalysts development. The kinetic scheme is also available at the section download of the website http://super.909 186.729 93.302 93.3 461.909 186.729 93. relying on detailed surface kinetics.909 186.6 17. Moreover.000 142. 1999). This evidence suggests that new catalysts should be developed in order to turn the aromatization process into economic sustainability. Activation energies are adapted from Bhan et al.chem. they are not anymore catalytically active on H-ZSM-5 zeolites.982 0.022 0.3 856. The methodology allowed to integrate and exploit at best all the information relating to the modelling of chemical kinetics.729 93. Corbetta et al.% in aromatics.6 D /kg.j nj − j=1 NR  D R.000 142. the reactor should yield at least 60 wt. Basing on the process simulation.909 186. Chemical compound Unit cost/price [D /kg] Economic flux [D /h] Refs. a multiscale process simulation has been proposed.000 142.000 142. propane price is about 0. some economic considerations have been argued. thus reducing the amount of fuel gas. surface kinetics.302 21. (2005).04 D /kg (global average prices in 2013).464 M..it in CHEMKIN format.909 186. BTX 1 D /kg and fuel gas 0.729 142. Such integration allows to overcome main concerns that have been arisen by scientists and technicians about the absence of aromatics in HRJ fuels. resulting in a flexible and easy-to-use tool. / Computers and Chemical Engineering 71 (2014) 457–466 Table 5 Economic parameters.058 1.

000 30.891e−007 2.029116 0.000 86.000 86.64(5):1061–74. Haag WO.000 125.000 30.891e−007 2.909 186. Visconti CG.000 125. Manenti F.000 125.909 186.000 30. Catal Lett 2009.03774 17. Kinetic models analysis. Perspective on multiscale methodology for product design and engineering.001217 2.056518 0.M. Comput Chem Eng 2014.85e−007 8.001217 2.891e−007 2. Bhan A.000 86.891e−007 2. Rye L.000 30.000 30. Nicholas Delgass W. J Catal 1996.22e−005 2.000 30.000 86. Catal Rev 2008.85e−009 1. Manenti F.54e−007 0. Blakey S.000 30.056518 0.056518 0.33(2):2863–85.000 Buchanan JS.000 30.21e−006 8. Aviation gas turbine alternative fuels: a review.000 125.6416 34.000 125.000 30.000 86.000 142. .2455 37.000 30.000 142. Blau G.001713 0.000 86.000 125.395628 0.000 30.134:179–85.33(5):936–46. Manenti F.21e−006 8.85e−007 0.909 186.891e−007 41 42 43 44 45 46 47 48 49 Oligomerization C2 H4 + C2 H4 > C4 H8 C2 H4 + C3 H6 > C5 H10 C2 H4 + C4 H8 > C6 H12 C2 H4 + C5 H10 > C7 H14 C2 H4 + C6 H12 > C8 H16 C3 H6 + C3 H6 > C6 H12 C3 H6 + C4 H8 > C7 H14 C3 H6 + C5 H10 > C8 H16 C4 H8 + C4 H8 > C8 H16 0.158(1):279–87. Chem Eng Sci 2009.000 30.21e−006 8. Buzzi-Ferraris G.891e−007 2.85e−007 0.001217 2.85e−007 0.50(1):19–51.000 142.719 239.000 142.891e−007 2.21e−006 References Bhan A. Caruthers JM.85e−007 0. Hsu SH.3587 165. Santiesteban JG.891e−007 2. BzzMath: library overview and recent advances in numerical methods.272315 0. Ragaini V.000 142.000 125.001217 2. Proc Combust Inst 2011.000 125.000 30.000 125. Corbetta et al.000 29 30 31 32 33 34 35 36 37 38 39 40 H-Transfer C3 H8 + C2 H4 > C3 H6 + C2 H6 C3 H8 + C4 H8 > C3 H6 + C4 H10 C3 H8 + C5 H10 > C3 H6 + C5 H12 C3 H8 + C6 H12 > C3 H6 + C6 H14 C3 H8 + C7 H14 > C3 H6 + C7 H16 C3 H8 + C8 H16 > C3 H6 + C8 H18 C4 H10 + C2 H4 > C4 H8 + C2 H6 C4 H10 + C3 H6 > C4 H8 + C3 H8 C4 H10 + C5 H10 > C4 H8 + C5 H12 C4 H10 + C6 H12 > C4 H8 + C6 H14 C4 H10 + C7 H14 > C4 H8 + C7 H16 C4 H10 + C8 H16 > C4 H8 + C8 H18 2.000 125. Microkinetic modeling of propane aromatization over HZSM-5. Low temperature de-acidification process of animal fat as a pre-step to biodiesel production.000 30.21e−006 8.001 239. Boffito DC.000 125.60:76–85.000 86.000 86.000 125.000 30. J Catal 2005.2455 34. Comput Chem Eng 2009. Propane aromatization over HZSM-5 and Ga/HZSM-5 catalysts.000 86.001217 2. Corbetta M.891e−007 2.000 30.000 125.719 8.000 86.000486 8.85e−007 0. Comput Aided Chem Eng 2012.21e−008 3.891e−007 2.395628 50 51 52 53 54 55 56 57 58 Cracking C4 H8 > C2 H4 + C2 H4 C5 H10 > C2 H4 + C3 H6 C6 H12 > C2 H4 + C4 H8 C7 H14 > C2 H4 + C5 H10 C8 H16 > C2 H4 + C6 H12 C6 H12 > C3 H6 + C3 H6 C7 H14 > C3 H6 + C4 H8 C8 H16 > C3 H6 + C5 H10 C8 H16 > C4 H8 + C4 H8 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 Aromatization C6 H12 + 3 C2 H4 > B + 3 C2 H6 C7 H14 + 3 C2 H4 > T + 3 C2 H6 C8 H16 + 3 C2 H4 > X + 3 C2 H6 C6 H12 + 3 C3 H6 > B + 3 C3 H8 C7 H14 + 3 C3 H6 > T + 3 C3 H8 C8 H16 + 3 C3 H6 > X + 3 C3 H8 C6 H12 + 3 C4 H8 > B + 3 C4 H10 C7 H14 + 3 C4 H8 > T + 3 C4 H10 C8 H16 + 3 C4 H8 > X + 3 C4 H10 C6 H12 + 3 C5 H10 > B + 3 C5 H12 C7 H14 + 3 C5 H10 > T + 3 C5 H12 C8 H16 + 3 C5 H10 > X + 3 C5 H12 C6 H12 + 3 C6 H12 > B + 3 C6 H14 C7 H14 + 3 C6 H12 > T + 3 C6 H14 C8 H16 + 3 C6 H12 > X + 3 C6 H14 C6 H12 + 3 C7 H14 > B + 3 C7 H16 C7 H14 + 3 C7 H14 > T + 3 C7 H16 C8 H16 + 3 C7 H14 > X + 3 C7 H16 C6 H12 + 3 C8 H16 > B + 3 C8 H18 C7 H14 + 3 C8 H16 > T + 3 C8 H18 C8 H16 + 3 C8 H16 > X + 3 C8 H18 1.909 142. CATalytic – Post Processor (CAT – PP): a new methodology for the CFD-based simulation of highly diluted reactive heterogeneous systems.000 30. 86.000 125.000 125.000 86.000 30.000 125. Charpentier JC.235(1):35–51.000 30.2455 34. Mechanistic considerations in acidcatalyzed cracking of olefins. Wilson CW.891e−007 2. / Computers and Chemical Engineering 71 (2014) 457–466 465 Table A1 (Continued) # Reaction Pre-exponential factor [mol/g/h/Pa] or [mol/g/h/Pa2 ] Activation energy [J/mol] 23 24 25 26 27 28 C8 H18 > C2 H4 + C6 H14 C8 H18 > C3 H6 + C5 H12 C8 H18 > C4 H8 + C4 H10 C8 H18 > C5 H10 + C3 H8 C8 H18 > C6 H12 + C2 H6 C8 H18 > C7 H14 + CH4 186.000 30. Bianchi CL.909 186.909 186.000 125. Pirola C.30(2):1312–6.891e−007 2.000 30. Venkatasubramanian V. Delgass WN. Buzzi-Ferraris G.061656 0.000 125.

Ind Eng Chem Res 1994. Biodiesel and renewable diesel: a comparison. Zhang J. Boffito DC. Propane aromatization over Mo/HZSM-5 catalysts. . Manenti F. .56:311–6.61(17):5881–94. Monque R. Guisnet MR. Technical Report. Corbetta et al. Chem Eng Sci 2006. Research challenges in process systems engineering. Frash MV. Catal Lett 1997.37:553–8. Multiscale compromise and multi-level correlation in complex systems. Spectrokinetic analysis of the NOx storage over a Pt–Ba/Al2O3 lean NOx trap catalyst.466 M. Guisnet MR. Kumar K. Pirola C.48(1–2):61–7.24:207–12. Daturi M. Catal Rev Sci Eng 1994. AIChE J 2000. Wang X. Fuel 2012. Wang J. Vazhnova T. / Computers and Chemical Engineering 71 (2014) 457–466 Doty FD. Morales-Rodríguez R. Kazansky VB.. Zhang Z. Baudouin O. Experimental studies on the combustion characteristics of alternative jet fuels. 2014 http://www. Déchelotte S. Kinetic modeling of propane aromatization reaction over HZSM-5 and GaHZSM-5.86(7):823–33. Ge W. Kinetic modeling of ethene and propene aromatization over HZSM-5 and GaHZSM-5. Software tools overview: process integration.83(6):574–82. Bifunctional behavior of Mo/HZSM-5 catalysts in methane aromatization. Van Santen RA.platts. Manenti F. Corbetta M.com/ Shu J.38(10):3860–7. A realistic look at hydrogen price projections. In: Doty Scientific Inc. Platts. Gani R.46(9):1700–3. 2005. Chem Eng Trans 2014. ExxonMobil Aviation: World Jet Fuel Specifications with Avgas Supplement 2005 Edition. Computer-aided multiscale modelling for chemical process engineering. Gnep NS. Westerberg AW. Heterogeneously catalyzed free fatty acids esterification in (monophasic liquid)/solid Packed Bed Reactors (PBR). McGraw Hill Financial. Lam HL. Adnot A. Kravanja Z. Knothe G. Comput Aided Chem Eng 2007. Morales-Rodríguez R. Galli F. Kang M.6(5):696–712. Giannetto G. A quantum-chemical study of hydride transfer in catalytic transformations of paraffins on zeolites. Lietti L. Chem Eng Res Des 2005. modelling and optimisation for energy saving and pollution reduction. Top Catal 2013. Combined experimental and kinetic modelling studies of the pathways of propane and n-butane aromatization over H-ZSM-5 catalyst. Galiasso R. Vacher A. Lukyanov DB. Leatherhead. UK: ExxonMobil Aviation International Ltd.33(2):223–34. Visconti C. Grandjean BP. Nguyen LH.98:176–82. Sung CJ. Transformation of LPG into aromatic hydrocarbons and hydrogen over zeolite catalysts.36(3):364–73. Edwards T. Pierucci S. et al. Kolaczkowski ST. View publication stats Lukyanov DB. Bianchi CL. 2004. Kwauk M. Gnep NS.11(1):43–50. Ind Eng Chem Res 1999. Corbetta M. ExxonMobil Aviation International Ltd. Lukyanov DB. Ind Eng Chem Res 1995. J Nat Gas Chem 2002. Li J. Varbanov PS. Pathways through adsorbed nonclassical carbonium ions. Klemeˇs JJ. Use of CAPEOPEN standards in the interoperability between modelling tools (MoT) and process simulators (Simulis® Thermodynamics and ProSimPlus). Prog Energy Combust Sci 2010. Gani R. Hui X.36(2):271–304. Gardner D. Asia Pac J Chem Eng 2011.34(2):516–23. Chem Eng Res Des 2008. Grossmann IE.