A Batch Reactor Study of the Effect of Deasphalting on Hydrotreating of Heavy Oil | Petroleum | Catalysis

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report



Views: 10 | Pages: 8

Extension: PDF | Download: 0

Catalysis Today 150 (2010) 264–271 Contents lists available at ScienceDirect Catalysis Today journal homepage: www.elsevier.com/locate/cattod A batch reactor study of the effect of deasphalting on hydrotreating of heavy oil ´ mano a,*, Fania Guerrero b, Jorge Ancheyta c, Fernando Trejo a, Jose ´ A.I. Dı ´az a Vicente Sa a ´n en Ciencia Aplicada y Tecnologı´a Avanzada, Unidad Legaria del Instituto Polite ´cnico Nacional (CICATA-IPN), Centro de Investigacio ´n, Me ´xico D.F. 11500, Mexico Legar
  A batch reactor study of the effect of deasphalting on hydrotreating of heavy oil Vicente Sa´mano a, *, Fania Guerrero b , Jorge Ancheyta c , Fernando Trejo a , Jose´A.I. Dı´az a a Centro de Investigacio´ n en Ciencia Aplicada y Tecnologı´ a Avanzada, Unidad Legaria del Instituto Polite´ cnico Nacional (CICATA-IPN),Legaria 694, Col. Irrigacio´ n, Me´  xico D.F. 11500, Mexico b Universidad Nacional Auto´ noma de Me´  xico (UNAM), Me´  xico D.F. 04360, Mexico c Instituto Mexicano del Petro´ leo, Eje Central La´  zaro Ca´ rdenas 152, Col. San Bartolo Atepehuacan, Me´  xico D.F. 07730, Mexico 1. Introduction Studies of asphaltenes precipitation and characterization havebeen traditionally carried out with the main objective of under-standing the asphaltene deposition problem: flocculation, aggre-gation and precipitation mechanisms from crude oils, and theinfluence of metallic particles present[1]. However, the decliningproduction of light crude oils together with the continuousgrowing up in the production of heavy and extra heavy crude oils,havecausedanincreaseintheasphaltenescontentoftheproducedoils which are sent to petroleum refineries[2].Ongoing advances in established technologies, together withrecent commercial applications of residue fluid catalytic cracking(RFCC), hydroprocessing, solvent deasphalting and gasification of pitch and coke, have markedly enhanced options for processingand economically using residues[3]. In addition, processing of heavy oils to obtain more gasoline and other liquid fuels isnowadays a necessity; hence the knowledge of the constituents of these higher boiling point feedstocks is of great importance. It hasbeen recognized that the problems associated to processing of heavyfeedstockscanbeequatedtothechemicalcharacterand theamount of complex and higher boiling components in thefeedstock. Refining heavy oils is not just a matter of applyingknow-how derived from refining conventional light crude oils butrequires knowledge of the chemical structure and chemicalbehavior of these more complex feedstocks[4]. Asphaltenes,being the most intrincate molecules present in petroleum, are alsothemostrefractoryanddifficulttoprocessportionofthecrudeoil.The main problems associated with hydroprocessing of heavy oilshaving high amount of asphaltenes are: precipitation on thecatalystsurfaceandblockingofthecatalystporemouth,theyactascoke precursors which ends up as catalyst deactivation, and limitthe maximum level of conversion due to sediment formation[5].Common fixed-bed catalytic hydroprocesses, which use atmo-spheric residue, vacuum residue, or heavy crude oils as feed, aremultiple reactor units with graded catalyst systems to achievedesired levels of impurity removals and conversion. The catalyticsystem is frequently integrated by a hydrodemetallization (HDM)catalyst for metal removal, a balanced hydrodemetallization/hydrodesulfurization (HDM/HDS) catalyst and a hydrodesulfur-ization (HDS) catalyst, at the front, middle and last sectionsrespectively[4,6,7].Thefirstcatalystisdesignedtohaveoptimizedtextural properties, shape, active metal loading, etc. in order toachieve high metal retention capacity to protect the followingcatalysts against premature deactivation and allow the process forlong-term operations[8]. Since most of the amount of metalspresent in heavy crudes, mainly V and Ni, are concentrated inasphaltenes, removing this high-molecular weight material willconsequently reduce the metal content in the feed to hydro-processing. Therefore, deasphalting a heavy feed before itshydrotreating seems to be very convenient from operational andeconomical points of view.Solvent deasphalting (SDA) is a molecular weight-basedseparation process member of the family of carbon rejection Catalysis Today 150 (2010) 264–271 A R T I C L E I N F O  Article history: Available online 20 October 2009 Keywords: HydrotreatingHeavy oilAsphalteneDeasphaltingSpent catalysts A B S T R A C T The effect of deasphalting of heavy oil with different degrees of asphaltenes precipitation on catalytichydrotreating is reported in this work. Deasphalted oils were obtained in a pressurized vessel using n-heptane and n-pentane as solvents. Various samples with different amounts of asphaltenes wereprepared by varying precipitation conditions. Hydrotreating of deasphalted oils was conducted with acommercial NiMo catalyst in a batch reactor at the following reaction conditions: hydrogen pressure of 100 kg/cm 2 , temperature of 400 8 C, stirring rate of 750 rpm and reaction time of 4 h. The heavy oil, thedeasphalted oils and the hydrotreated products were characterized by sulfur, metals (Ni, V), asphaltenecontents,andAPIgravity.MetalsandcarboncontentsaswellastexturalpropertiesandX-raydiffractionwere also determined on fresh, spent and regenerated catalysts. ß 2009 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +52 55 9175 6619; fax: +52 55 9175 8429. E-mail address: vicente_samano@yahoo.com.mx(V. Sa´mano). Contents lists available atScienceDirect Catalysis Today journal homepage: www.elsevier.com/locate/cattod 0920-5861/$ – see front matter ß 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.cattod.2009.09.004  technologies, which has been used for more than fifty years toseparate heavy fractions of crude oil beyond the range of economical commercial distillation. SDA process produces alow-contaminant deasphalted oil (DAO) rich in paraffinic-typemolecules, and a pitch product rich in aromatic compounds andasphaltenes containing of course the majority of the feedimpurities. The application of SDA process has been reported forproduction of lube oil feedstocks from vacuum residue usingpropane as solvent, for preparation of feedstocks for catalyticcracking, hydrocracking and hydrodesulfurization units, as well asfor the production of specialty asphalts[9–11].The increasing production of heavy crude oils has forcedpetroleum refiners to deal with heavier feedstocks, which possesshigh amounts of asphaltenes. When sending these heavy feeds tohydroprocessing units the catalysts will be exposed to severedeactivation, in such cases deasphalting these heavy materialswould be more appropriate to prepare suitable hydroprocessingfeeds. Hydrotreating is a better choice for processing vacuumresidua with relatively low content of heavy metals. Thecombination of solvent deasphalting with asphalt gasificationwouldbeafeasibleupgradingwaytoconvertheavyfeedstockwithhighmetalcontent[12–14].Theaimofthisstudyisthentoanalyzethe effect of different degrees of deasphalting of heavy oils oncatalytic hydroprocessing behavior. 2. Experimental  2.1. Materials A vacuum residue (538 8 C+ VR) obtained from Maya heavycrude oil was employed for hydrotreating experiments and forpreparing different deasphalted oils. The main properties of Mayacrude and its VR are given inTable 1. The catalyst used in allhydrotreating experiments was a commercial sample with thefollowing main properties: 0.58% Ni, 2.2% Mo, 197 m 2 /g specificsurfacearea,0.85 cc/gporevolumeand173 A˚meanporediameter.n-Heptane, n-pentane and nitrogen were used for asphaltenesprecipitation while hydrogen was utilized for hydrotreating tests.  2.2. Preparation of DAO VRwasloadedinapressurizedsystem(Parrreactormodel4522and controller model 4841) equipped with stirring system,temperature and pressure controls, and vent. To obtain differentdegreesofdeasphalting,precipitationconditionswerevariedinthefollowing ranges:solvent-to-oil (S/O) ratiobetween 2 and 10 mL/g,750–1000rpm stirring rate, 15–30 min stirring time, atmospheric-25kg/cm 2 of nitrogen pressure and temperature from ambient to100 8 C.Typically,200 gofVRwereputintothereactorand1400mL ofn-heptanewereaddedforthecaseofasolvent-to-oilratioof7 mL/g. The reactor was pressurized with nitrogen and stirring beganwhen the temperature was adjusted to the desired value. Stirringwaskeptforthedesiredtime,stoppedandthenthe reactorcontentwassettledfor60 min,andwithdrawnbymeansofavalvelocatedatthebottomofthereactor.Theremainingsampleadheredtoreactorwalls was removed with n-heptane washing and collected in abeaker. The reactor content was filtered by using a vacuum systemand a Whatman 3 filter paper with 6 m m pore size to retainasphaltenes. The deasphalted oil was finally weighed and char-acterized. A schematic representation of this experimental meth-odology is shown inFig. 1.  2.3. Hydrotreating experiments Hydrotreatingtestswerecarriedoutinabatchreactor(1 L,Parrreactor model 4573 and controller model 4842). The followingconstant hydrotreating conditions were utilized: 100 kg/cm 2 hydrogen pressure, 400 8 C reaction temperature, 750 rpm stirringrate, and 4 h reaction time. Temperature, pressure and stirring arecontrolled automatically using a digital controller. In eachexperiment, 5 g of catalyst and 200 g of sample were loaded tothe reactor. Prior to the activity tests the catalyst was sulfided exsitu in a fixed-bed glass reactor at the following conditions:atmospheric pressure, 400 8 C temperature, 40 mL/min H 2 , and 3 h.Hydrogen is passed through a container having carbon disulfide,and then the H 2 /CS 2 mixture is passed through the glass reactor.The sulfided catalyst was transferred into the batch reactor innitrogen atmosphere so that contact with air was avoided. Afterclosing the reactor, it was purged several times with hydrogen toassure there was no air left inside the reactor. Heating was thenstartedfrom roomtemperature to 400 8 C. The reaction began untilallconditionswereestablishedandstirringratewasinitiated.Afterreaction,thereactorwasdepressurizedandhydrotreatedproductswere filtered to separate the catalyst from liquids.Fig. 1shows theschematic representation of this procedure.  2.4. Characterization of oils and catalyst  Oil samples were characterized by sulfur content by X-rayfluorescence (HORIBA model SLFA-2100/2800), Ni and V contentsby atomic absorption (Thermoelectron model Solaar AA), specificgravity by means of picnometers and then transformed to APIgravity (ASTM D-70). Asphaltene is defined as that materialinsoluble in n-heptane (ASTM D-3279).Metals content on catalysts was analyzed by atomic absorption(ThermoelectronmodelSolaar AA),carboncontentwitha LecoSC-444 instrument using ASTM C-1408 method with direct combus-tion-infrared detection, and textural properties (specific surface  Table 1 Properties of Maya crude oil, Maya VR (538 8 C+) and DAO’s.Maya VR DAO-7 DAO-4 DAO-0Yield of DAO, wt% – – 69.6 64.5 40.2API gravity 22.0 1.46 10.01 17.87 11.27Total S, wt% 3.60 5.32 4.62 4.46 4.29nC 7 insol., wt% 9.51 27.86 7.10 4.18 0.45Ni, wppm 53 140 59 44 30V, wppm 260 698 314 206 60Total Ni+V, wppm 313 838 373 250 90Hydrotreated productsAPI gravity – 10.78 17.76 20.86 22.96Total S, wt% – 3.06 2.61 2.20 1.79nC 7 insol., wt% – 16.24 6.18 5.88 1.67Ni, wppm – 144 84 36 36V, wppm – 534 125 117 30Total Ni+V, ppm – 678 209 153 66 V. Sa´ mano et al./Catalysis Today 150 (2010) 264–271 265  area, pore volume and pore size distribution) by nitrogenadsorption at 77 K (Quantachrome Nova 4000). Spent catalystwas washed with toluene under reflux to remove adsorbedresidual oil whereas solids obtained (spent catalyst + coke) weredried at 100 8 C before characterization to analyze texturalproperties. Dried catalyst was regenerated by oxidation underair atmosphere at 500 8 C during 4 h and the amount of coke wasdetermined as the weightdifference amongsolids after and beforecalcination. Both the spent and regenerated catalysts wereanalyzed by X-ray diffraction in a diffractometer model SiemensD-500 with Cu K a radiation ( l = 1.5418 A˚) operating at 40 kV and30 mA in the 5–70 8 scan range of 2 u  , with a scan rate of 0.02 8 /s. 3. Results and discussion  3.1. Preparation of DAO The effect of conditions on asphaltenes precipitation has beenstudied in previous works in a pressurized system[15]. It is well Fig. 1. Experimental methodology. Fig. 2. Effect of solvent-to-oil ratio and solvent on asphaltenes content in DAO. ( * )n-Heptane and ( * ) n-pentane. V. Sa´ mano et al./Catalysis Today 150 (2010) 264–271 266  known in the literature that the lower the precipitation time andsolvent-to-oil ratio, the lower the asphaltenes precipitation; andwhen temperature is increased and pressure is decreasedasphaltenes content tends to diminish. The opposite behavior, of course,isobservedforasphaltenescontentinDAO.Thatis,whenS/O ratio, time and pressure are decreased or temperature isincreased, DAO asphaltenes content tends to increase. To examinethe effect of precipitation conditions on the particular feed (VR from Maya crude oil) of the present work, experiments varyingsome parameterswere done.Fig. 2illustrates the effect of solvent-to-oil ratio on asphaltenes content in DAO. In the range of 2–10 mL/g solvent-to-oil ratio asphaltenes content in DAO variedfrom about2 to 8 wt%. Earlier studieshave also reportedthe use of S/O ratio of 10:1[16]. Neither higher than 10 nor lower than 2solvent-to-oil ratios were tested since either very low quantity of DAO was obtained and a number of precipitations would berequired or when filtering the sample different problems wereexperienced. By varying other precipitation conditions within theaforementionedrangesitwasnotpossibletoobtainneitherhighernorloweramountsofasphaltenesinDAO.Instead,thesolventwaschanged to n-pentane and concentration of asphaltenes in DAOwas reduced to a value very close to zero. As summary, we wereforced to use different solvents because when using n-heptaneonly,itwasnotpossibletoobtainDAOfreeof asphaltenes.Theuseof n-pentane allowed us to meet the requirements for a feedstockwith almost no amount of asphaltenes.The option of blending DAO either with virgin VR or withasphaltenes was also considered to prepare DAO with differentasphaltenes content. However, since asphaltenes properties mayvaryduringprecipitationanditseffectonhydrotreatingcanalsobedifferent, this alternative was discarded. On the other hand,dilution of VR with light petroleum fractions, e.g. straight-run gasoil has been already studied in our group and reported elsewhere[17].Basedontheseresults,thefollowingfeedswerepreparedattheindicated conditions, whose main properties are presented inTable 1.  DAO-0, with 0.45 wt% asphaltenes, precipitated with n-pentaneat 5 mL/g S/O ratio, 1000 rpm stirring rate, 30 min stirring time,25 kg/cm 2 pressure and 100 8 C temperature.  DAO-4, with 4.18 wt% asphaltenes, precipitated with n-heptaneat 7 mL/g S/O ratio, 1000 rpm stirring rate, 30 min stirring time,25 kg/cm 2 pressure and 100 8 C temperature.  DAO-7,with7.1 wt%asphaltenes,precipitatedwithn-heptaneat3 mL/g S/O ratio, 1000 rpm stirring rate, 30 min stirring time,25 kg/cm 2 pressure and 100 8 C temperature.APIgravitywassubstantiallyincreasedwhendifferentportionsof asphaltenes were separated from the VR feed. The DAO-0, withthelowestamountofasphaltene,didnotfollowtheexpectedtrendwhen reducing asphaltene content, which might be due to the useof n-pentane instead of n-heptane as in the cases of DAO-4 andDAO-7.API gravity of DAO-0 is higher compared with DAO-4 and DAO-7 due to the presence of microwaxes that cannot be properlydissolved by n-pentane since it takes longertime to dissolve them.In addition, when usingn-pentane as solvent,a fewer yield of DAOis obtained as observed inTable 1; however, the sulfurconcentration is still higher in DAO. In this way, microwaxes,sulfur compounds along with metals could be responsible for theincrease of API of DAO-0 because its concentrations are morenotorious in a smaller amount of recovered DAO-0 (40.2 wt%)compared with DAO-4 (64.5 wt%) and DAO-7 (69.6 wt%). Inaddition, asphaltenes precipitated with n-pentane are lighter thanthose obtained with n-heptane as reported in previous work inwhichaggregatemolecularweightsbyVaporPressureOsmometryof asphaltenes precipitated from Maya crude had values of 3680and 5190 g/mol when using n-pentane and n-heptane, respec-tively[18].SulfurcontentinDAO’swasslightlyreducedcomparedwiththesrcinal VR, and not much effect is observed when changing theconcentration of asphaltenes. On the contrary, metals (Ni and V)notoriously vary at different asphaltenes content in DAO. Thisbehavior in both S and metals indicates that the amount of asphaltenic sulfur is lower than that of the asphaltenic metals,sincemostofthesulfurconcentratedinDAOandmostofthemetalwas found to be asphaltenic in nature (i.e. it was concentrated inasphaltenes). For the case of Maya crude, it has been reported intheliteraturethatabout30%ofthesulfurand60%ofthemetalsareasphaltenic[18], which is in line with our findings.The increase in API gravity of DAO’s compared with that of VR feed, corroborates that apart from reduced impurities contentdeasphalted oil became rich in paraffinic-type molecules. Thechange in API gravity roughly indicates that DAO’s are lessaromatic at low amount of asphaltenes. The low aromaticity of DAO will favor hydrocracking of heavy molecules into smallerones.  3.2. Hydrotreating of DAO Hydrotreating experiments were carried out in a batch reactor.Sincethecatalystissubjecttodeactivationbycokeandmetals,andinitial concentration of asphaltenes (and those of the otherimpurities) is changing during time, samples at different timeswere not obtained such as when experiments are performed incontinuous flow reactors, in which initial concentrations arealways the same. Experiments may be conducted by loading andunloading fresh catalyst and performing the reaction at differenttimesforeachtest[19]however,itwouldtakemuchtimetodoso.Instead, only one sample was taken at the end of reaction at 4 h.Sincemuchofthesulfurandpartofthemetalswerefoundtobenon-asphaltenic in nature, their removals are expected to be highas shown inFig. 3, in which different amounts of asphaltenes inDAO are plotted against sulfur and metals conversions afterhydrotreating. S and metals conversions were calculated as ( M  1 – M  2 )/ M  1 ,where M  1 isgramsoftheimpurityinVRand M  2 thegramsoftheimpurityinhydrotreatedDAO.Thelocalizationof  M  1 and M  2 can be observed inFig. 1. The yield of DAO during deasphalting of VR is necessary to do this calculation, so that conversions reportedinFig. 3are on VR basis, and not on HDT reactor feed. This mannerto compare conversions is more suitable since reactor feeds havedifferent concentrations of S and metals. For DAO-0 sulfurconversion is high ( $ 86%) along with metals conversion ( $ 97%). Fig. 3. Effect of asphaltenes in DAO on impurities removal and API gravity of hydrotreated products. ( * ) Sulfur, ( * ) Ni + V and ( ~ ) API gravity. V. Sa´ mano et al./Catalysis Today 150 (2010) 264–271 267
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks