High-Temperature Catalytic Reforming of n‑Hexane over Supportedand Core−Shell Pt Nanoparticle Catalysts: Role of Oxide−MetalInterface and Thermal StabilityKwangjin An,†,‡ Qiao Zhang,§ Selim Alayoglu,†,‡ Nathan Musselwhite,†,‡ Jae-Youn Shin,†

and Gabor A. Somorjai*,†,‡

†Department of Chemistry, University of California, Berkeley, California 94720, United States‡Chemical Sciences and Materials Sciences Divisions, Lawrence Berkeley National Laboratory, Berkeley, California 94720, UnitedStates§Institute of Functional Nano & Soft Materials (FUNSOM) and Jiangsu Key Laboratory for Carbon-Based Functional Materials &Devices, and Collaborative Innovation Center of Suzhou Nano Science and Technology, Soochow University, Suzhou 215123, China

*S Supporting Information

ABSTRACT: Designing catalysts with high thermal stability andresistance to deactivation while simultaneously maintaining theircatalytic activity and selectivity is of key importance in high-temperature reforming reactions. We prepared Pt nanoparticlecatalysts supported on either mesoporous SiO2 or TiO2.Sandwich-type Pt core@shell catalysts (SiO2@Pt@SiO2 andSiO2@Pt@TiO2) were also synthesized from Pt nanoparticlesdeposited on SiO2 spheres, which were encapsulated by eithermesoporous SiO2 or TiO2 shells. n-Hexane reforming was carriedout over these four catalysts at 240−500 °C with a hexane/H2ratio of 1:5 to investigate thermal stability and the role of thesupport. For the production of high-octane gasoline, branched C6isomers are more highly desired than other cyclic, aromatic, andcracking products. Over Pt/TiO2 catalyst, production of 2-methylpentane and 3-methylpentane via isomerization was increasedselectively up to 420 °C by charge transfer at Pt−TiO2 interfaces, as compared to Pt/SiO2. When thermal stability was comparedbetween supported catalysts and sandwich-type core@shell catalysts, the Pt/SiO2 catalyst suffered sintering above 400 °C,whereas the SiO2@Pt@SiO2 catalyst preserved the Pt nanoparticle size and shape up to 500 °C. The SiO2@Pt@TiO2 catalyst ledto Pt nanoparticle sintering due to incomplete protection of the TiO2 shells during the reaction at 500 °C. Interestingly, over thePt/TiO2 catalyst, the average size of Pt nanoparticles was maintained even after 500 °C without sintering. In situ ambientpressure X-ray photoelectron spectroscopy demonstrated that the Pt/TiO2 catalyst did not exhibit TiO2 overgrowth on the Ptsurface or deactivation by Pt sintering up to 600 °C. The extraordinarily high stability of the Pt/TiO2 catalyst promoted highreaction rates (2.0 μmol·g−1·s−1), which was 8 times greater than other catalysts and high isomer selectivity (53.0% of C6 isomersat 440 °C). By the strong metal−support interaction, the Pt/TiO2 was turned out as the best catalyst with great thermal stabilityas well as high reaction rate and product selectivity in high-temperature reforming reaction.

KEYWORDS: Pt, core@shell, selectivity, reforming, thermal stability, n-hexane

Catalytic reforming is an important industrial process,which converts the naphtha feedstock into high-octane

reformates as a major blending product for gasolines.1−4

Straight-run hydrocarbons or naphthas through hydrocrackingprocess are unsuitable for direct gasoline blending because oftheir low octane numbers. Therefore, catalytic reforming ofnaphtha and isomerization of light alkanes are prerequisite toimprove the quality of gasolines with high-octane numbers(>70).2 The catalytic reforming of n-hexane is an excellentmodel for this reaction. Hexane reforming has four distinctreaction pathways, which are isomerization, cracking, cycliza-tion, and dehydrogenation.5−8 The desired products in thisreaction are isomers and cyclization products, while aromatic

compounds are undesired due to their carcinogenic nature andthe cracking products, which can act as coke precursors anddeactivate the Pt catalyst.2 In order to achieve the desiredreaction rate, high temperatures (400−500 °C) must be appliedfor the reforming reaction without exhibiting deactivation bysintering or cracking of C6 isomers.

1−4

Industrial catalysts for catalytic reforming mainly consist ofthe Pt supported on alumina with small amounts of a second

Received: June 30, 2014Revised: July 24, 2014Published: July 31, 2014

Letter

pubs.acs.org/NanoLett

© 2014 American Chemical Society 4907 dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−4912

promoter metal such as Re, Sn, and Ir.2 Recent advances incolloid chemistry have enabled the design of metal nano-particles with tunable particle size, shape, and metalcomposition beyond conventional impregnated industrialcatalysts.9,10 Despite that vast research on catalytic activityand selectivity enhancements has been made depending on thestructure of nanoparticles, the thermal and chemical stabilitiesof nanoparticle catalysts received relatively little attention inhigh-temperature reactions.11,12 Recently, a core@shell-typePt@mesoporous silica was designed as a thermally stablecatalyst, in which silica shells encage Pt nanoparticle cores. Thecatalyst was found to be stable up to 750 °C in air andmaintained the size and structure of Pt nanoparticle coreshaving low melting points during CO oxidation at 330 °Cwithout aggregation.13,14 As a supported catalyst, a metal@porous oxide core@shell nanoparticle is expected to enhance acatalytic performance in high-temperature reactions, due to thepromoting role at the oxide−metal interface and the stabilityagainst sintering.In this study, we conducted the n-hexane reforming reaction

over supported Pt nanoparticle catalysts and sandwich-typecore@shell catalysts at temperatures ranging from 240 to 500°C. In order to investigate thermal stability and the role of thesupport, four different kinds of catalysts were prepared

(Scheme 1): Pt nanoparticles supported on mesoporous SiO2(Pt/SiO2) and TiO2 (Pt/TiO2), and sandwich-type Pt core@shell catalysts (SiO2@Pt@SiO2 and SiO2@Pt@TiO2). In ourprevious study, it was found that the TiO2 played a crucial roleto increase the production of C6 isomers and facilitated highturnover in n-hexane reforming by charge transfer at oxide−metal interfaces.8

Colloidal Pt nanoparticles stabilized with poly-(vinylpyrrolidone) (PVP) were loaded into a support ofmesoporous silica or titania by sonication.15,16 SBA-15 withhexagonal channels was used as a mesoporous silica support.17

The titania support, which contained meso- and macropores,was prepared by a Pluronic P123 surfactant and a polystyrenebead, respectively.8,18−20 Figure 1a,b shows that PVP-capped Ptnanoparticles with an average diameter of 2.7 nm (Figure S1,Supporting Information) were supported into the pores ofmesoporous SiO2 and TiO2, respectively. In order to synthesizecore@shell structures, silica spheres with an average diameter of200 nm were prepared as a core, then Pt nanoparticles weredeposited on the surface of the silica.21−24 When the outer shellwas coated with either silica or titania, sandwich-type SiO2@Pt@oxides core@shell catalysts were generated (Scheme 1 andFigure 1c,d). Upon calcinations at 550 °C, the outer oxidesbecame mesoporous shells, having BET surface areas of 24.4

Scheme 1. Preparation of Supported Pt Nanoparticle Catalysts and Sandwich-Type Pt Core@Shell Catalysts

Figure 1. TEM images of supported Pt nanoparticle catalysts (a,b) and sandwich catalysts (c,d): (a) Pt/SiO2, (b) Pt/TiO2, (c) SiO2@Pt@SiO2, and(d) SiO2@Pt@TiO2. (e) Scanning TEM high angle annular dark field (STEM-HAADF) image of SiO2@Pt@TiO2 catalysts and (f) theircorresponding energy dispersive spectroscopy (EDS) phase mapping (Pt is shown in red, Si in blue, and Ti in green, respectively).

Nano Letters Letter

dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−49124908

and 45.1 m2/g for SiO2@Pt@SiO2 and SiO2@Pt@TiO2,respectively. Transmission electron microscopy (TEM) imagesshown in Figure 1c demonstrated that the Pt nanoparticleswere decorated on the surface of the silica cores. The SiO2@Pt@TiO2 structures (Figure 1d) were fully covered with TiO2

shells, which made it impossible to observe the Pt nano-particles. In order to characterize the catalyst, high angleannular dark field (HAADF) imaging in the scanning TEM(STEM) mode was used. The X-ray line scan profile along theline indicated in the HAADF-STEM image (Figure 2e) and theelemental analysis by energy dispersive spectroscopy (EDS)phase mappings (Figure 2f) on the SiO2@Pt@TiO2 structures

demonstrated Pt nanoparticles were located in between theSiO2 inner spheres and TiO2 outer shells.n-Hexane reforming reactions were carried out over the

supported Pt nanoparticle and core@shell-type catalysts in alaboratory-scale flow reactor operated at atmospheric pressureand temperature ranges of 240−500 °C with a hexane/H2 ratioof 1:5. All products were analyzed by a gas chromatographequipped with a flame ionization detector. Catalysts (0.5−0.7g) diluted by quartz were loaded in the tubular type catalystbed. For the catalyst pretreatment, a reduction in a flow of 50vol % mixture of H2 and N2 at 20 sccm total flow rate wasconducted under ambient pressure at 260 °C for 2 h. Figure 2ashows the possible products of n-hexane reforming. Branched

Figure 2. (a) Schematic illustration showing possible products of n-hexane reforming. (b) Temperature-dependent selectivity over four differentcatalysts in n-hexane reforming at 420 °C at ambient pressure (hexane/H2 ratio = 1:5).

Figure 3. TEM images and corresponding schemes of (a) Pt/SiO2, (b) Pt/TiO2, (c) SiO2@Pt@SiO2, and (d) SiO2@Pt@TiO2 after reformingreaction at 500 °C.

Nano Letters Letter

dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−49124909

C6 isomers (including 2- and 3-methylpentane) were majorproducts at the temperature range of 240−380 °C over allcatalysts except the SiO2@Pt@SiO2. As the reaction temper-ature was increased (400−500 °C), the fractions of olefins,dehydrogenated cyclic compounds (cyclohexene, methylcyclo-hexene, and benzene), and cracking products were increasedgradually. Table S1, Supporting Information, summarizes thecatalytic activity and selectivity as a function of temperatureover four different catalysts in the n-hexane reforming reaction.In the previous study, we reported that isomer production wasgreatly increased over supported Pt nanoparticle catalysts in n-hexane reforming at 360 °C, when the SiO2 support wasreplaced by TiO2, Nb2O5, or Ta2O5.

8 The product selectivity at340 °C plotted in Figure 2b exhibited that the two TiO2-basedcatalysts produced C6 isomers preferentially due to the Pt−TiO2 interaction, whereas the two SiO2-based catalysts favoredolefins by dehydrogenation at this temperature. When thereaction temperature was further increased, the Pt/TiO2catalyst still maintained high reaction rates and high isomerselectivity up to 440 °C (see Supporting Information, FigureS2). For all catalysts, branched C6 isomers were decreased infavor of olefins, cracking, and other undesired products astemperature increased beyond 440 °C.In order to investigate thermal stability of Pt nanoparticle

catalysts, TEM was used for evaluating average diameters of thePt nanoparticles after the high-temperature reforming reaction.In Figure 3, it can be seen that the Pt in the Pt/SiO2 catalystsintered by diffusion and aggregation of Pt nanoparticles in thereforming reaction at 500 °C, while SiO2@Pt@SiO2 catalystpreserved the Pt size via a partial encapsulation by mesoporousSiO2 shells. In the absence of outer SiO2 shell, Pt nanoparticlecatalysts decorated on the surface of SiO2 spheres (SiO2@Pt)caused aggregation of Pt nanoparticles during the reforming at500 °C (Figure S3, Supporting Information). Because of thehigh thermal stability, the SiO2@Pt@SiO2 catalyst maintainedhigher activities than those of Pt/SiO2 catalyst over alltemperature ranges studied (Figure 4). However, the isomerselectivity of the SiO2@Pt@SiO2 catalyst was less than that ofPt/SiO2 catalyst (Figures 2b and S2, Supporting Information)because the SiO2 shell partially blocked the active surface site ofPt nanoparticles. After the high-temperature reforming

reaction, however, the SiO2@Pt@TiO2 catalyst suffered severePt nanoparticle sintering, which was observed by HAADF-STEM and corresponding EDS elemental map (Figure S4,Supporting Information). Furthermore, the surface of theporous TiO2 shells became smooth at high temperatures(Figure 3d), though BET surface areas (43.0 m2/g) andcrystalline structures of the SiO2@Pt@TiO2 showed littlechange before and after the reaction (Figure S5, SupportingInformation). From these results, it was concluded that SiO2@Pt@TiO2 catalyst was not as thermally stable as the SiO2@Pt@SiO2 catalyst. Interestingly, no significant sintering was foundfor the Pt/TiO2 catalyst after the reaction (Figure 3b). TheTiO2 seemed to immobilize Pt nanoparticles against sintering,unlike the thermally unstable Pt/SiO2 catalyst.High thermal stability and specific Pt−TiO2 interactions of

the Pt/TiO2 catalyst allowed for eight times higher activity (2.0μmol·g−1·s−1 at 440 °C) than any other catalysts in this high-temperature reforming regime. In Figure 4, it can be seen thatthe reaction rates were maximized at 440 °C and decreasedbeyond this point. The activity drop was not attributed to theloss of available surface sites by sintering of Pt nanoparticles, asconfirmed through TEM. In order to characterize if there wasany diffusive encapsulation of TiO2 onto the surface of Ptnanoparticles during the high-temperature reaction, weconducted APXPS experiments with the Pt/TiO2 catalystunder ambient H2 atmospheres.

25−30 In Figure 5, normalized Pt4f spectra (730 eV photons) are identical up to 600 °C, whileXPS Ti 2p spectra maintained Ti3+ components under 0.1 TorrH2 in the 350−600 °C range. From the APXPS results, weconfirmed that % Pt on the surface (corresponding to ∼650 eVphotoelectons) of the Pt/TiO2 catalyst was preserved even atvery high temperatures (600 °C). Therefore, the activity dropof the Pt/TiO2 beyond 440 °C was not attributed to theencapsulation of Pt surface by TiO2. On the basis ofthermogravimetric analysis (TGA), the spent Pt/TiO2 catalysthad higher mass loss in the 300−800 °C temperature rangethan the fresh catalyst. We ascribed this mass loss to carbondeposited on the surface, indicating that the activity dropbeyond 450 °C during the reforming reaction was possibly dueto increased carbon coverage or even poisoning (Figure S6,Supporting Information).In summary, we conducted n-hexane reforming over

supported Pt nanoparticle catalysts and sandwich-type Ptcore@shell catalysts in the temperature range of 250−500 °C.Over Pt/TiO2 and SiO2@Pt@TiO2 catalysts, production of C6isomers was increased selectively up to 420 °C by chargetransfer at Pt−TiO2 interfaces, compared to Pt/SiO2-basedcatalysts.31−34 For thermal stability, the Pt/SiO2 catalystsuffered deactivation by Pt sintering above 400 °C. However,SiO2@Pt@SiO2 catalyst was thermally stable against sinteringof the Pt nanoparticles up to 500 °C, so its reaction rates werehigher than those of the Pt/SiO2 over all temperature rangesstudied. The SiO2@Pt@TiO2 catalyst produced isomersdominantly up to 400 °C; however, the mesoporous TiO2shells did not restrain Pt from sintering. In these reactionconditions, Pt/TiO2 showed the highest reaction rate, whichwas 8 times greater than any other catalyst at 440 °C andmaintained high isomer selectivity. The strong metal supportinteraction of the Pt/TiO2 catalyst affected not only thereaction rate and product selectivity but also endowed thermalstability against Pt nanoparticle sintering. In situ APXPSexhibited that the unique interactions between Pt and TiO2resulted in neither TiO2 encapsulation over the Pt surface nor

Figure 4. Catalytic activity over Pt nanoparticle catalysts in n-hexanereforming as a function of temperature.

Nano Letters Letter

dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−49124910

deactivation of the Pt surface. The supported Pt/TiO2 catalystwas found to be the best choice for high-temperaturereforming, yielding selective isomer production as well ashigh reaction rates.

■ ASSOCIATED CONTENT*S Supporting InformationExperiment details; catalytic activity and selectivity details for n-hexane reforming (Figure S2 and Table S1); TEM, XRD, andTGA of catalysts (Figures S1 and S3−S6). This material isavailable free of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*(G.A.S.) E-mail: somorjai@berkeley.edu.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by the Director, Office of BasicEnergy Sciences, Materials Science and Engineering Divisionand the Division of Chemical Sciences, Geological andBiosciences of the U.S. Department of Energy under ContractNo. DE-AC02-05CH11231. The user project at the AdvancedLight Source and the Molecular Foundry at the LawrenceBerkeley National Laboratory was supported by the Director,Office of Science, Office of Basic Energy Sciences, U.S.Department of Energy, under Contract DE-AC02-05CH11231. The nanoparticle synthesis was funded byChevron Corporation. Q.Z. thanks Soochow University forstart-up funds, the 1000 Talented Program, and the NaturalScience Foundation of Jiangsu Province for funding support.We thank Professors A. Paul Alivisatos, Peidong Yang, andOmar Yaghi for use of the TEM and XRD instruments andProfessor Katz and Alexander Okrut for TGA measurement.

■ REFERENCES(1) Alfke, G.; Irion, W. W.; Neuwirth, O. S. Oil Refining. In Ullman’sEncyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, Germany,2012.(2) Hagen, J. Industrial Catalysis: A Practical Approach, 2nd ed.;Wiley-VCH: Weinheim, Germany, 2006.(3) Geus, J. W.; van Ween, J. A. R. Catalysis: An Integrated Approachto Homogeneous, Heterogeneous and Industrial Catalysis; Moulijn, J. A.,van Leeuwen, P. W. N. M., van Santen, R. A., Eds.; Elsevier:Amsterdam, The Netherlands, 1993.

(4) Somorjai, G. A. Introduction to Surface Chemistry and Catalysis;Wiley: New York, 1994.(5) Davis, S. M.; Zaera, F.; Somorjai, G. A. J. Catal. 1984, 85, 206−223.(6) Musselwhite, N.; Alayoglu, S.; Melaet, G.; Pushkarev, V. V.;Lindeman, A. E.; An, K.; Somorjai, G. A. Catal. Lett. 2013, 143, 907−911.(7) Alayoglu, S.; Pushkarev, V. V.; Musselwhite, N.; An, K.;Beaumont, S. K.; Somorjai, G. A. Top. Catal. 2012, 55, 723−730.(8) An, K.; Alayoglu, S.; Musselwhite, N.; Na, K.; Somorjai, G. A. J.Am. Chem. Soc. 2014, 136, 6830−6833.(9) An, K.; Somorjai, G. A. ChemCatChem 2012, 4, 1512−1524.(10) Park, J. Y. Current Trends of Surface Science and Catalysis;Springer: New York, 2014.(11) Hansen, T. W.; Delariva, A. T.; Challa, S. R.; Datye, A. K. Acc.Chem. Res. 2013, 46, 1720−1730.(12) Campbell, C. T. Acc. Chem. Res. 2013, 46, 1712−1719.(13) Joo, S. H.; Park, J. Y.; Tsung, C. K.; Yamada, Y.; Yang, P. D.;Somorjai, G. A. Nat. Mater. 2009, 8, 126−131.(14) Nakagawa, K.; Tanimoto, Y.; Okayama, T.; Sotowa, K. I.;Sugiyama, S.; Takenaka, S.; Kishida, M. Catal. Lett. 2010, 136, 71−76.(15) Rioux, R. M.; Song, H.; Hoefelmeyer, J. D.; Yang, P.; Somorjai,G. A. J. Phys. Chem. B 2005, 109, 2192−2202.(16) Pushkarev, V. V.; An, K. J.; Alayoglu, S.; Beaumont, S. K.;Somorjai, G. A. J. Catal. 2012, 292, 64−72.(17) Zhao, D. Y.; Feng, J. L.; Huo, Q. S.; Melosh, N.; Fredrickson, G.H.; Chmelka, B. F.; Stucky, G. D. Science 1998, 279, 548−552.(18) Dacquin, J. P.; Dhainaut, J.; Duprez, D.; Royer, S.; Lee, A. F.;Wilson, K. J. Am. Chem. Soc. 2009, 131, 12896−12897.(19) Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281,538−540.(20) Stein, A. Microporous Mesoporous Mater. 2001, 44, 227−239.(21) Zhang, Q.; Lima, D. Q.; Lee, I.; Zaera, F.; Chi, M. F.; Yin, Y. D.Angew. Chem., Int. Ed. 2011, 50, 7088−7092.(22) Lee, I.; Joo, J. B.; Yin, Y. D.; Zaera, F. Angew. Chem., Int. Ed.2011, 50, 10208−10211.(23) Lee, J. W.; Othman, M. R.; Eom, Y.; Lee, T. G.; Kim, W. S.;Kim, J. Microporous Mesoporous Mater. 2008, 116, 561−568.(24) Zhang, Q.; Lee, I.; Joo, J. B.; Zaera, F.; Yin, Y. D. Acc. Chem. Res.2013, 46, 1816−1824.(25) Somorjai, G. A.; Beaumont, S. K.; Alayoglu, S. Angew. Chem., Int.Ed. 2011, 50, 10116−10129.(26) Hashimoto, S.; Tanaka, A. Surf. Interface Anal. 2002, 34, 262−265.(27) Grass, M. E.; Karlsson, P. G.; Aksoy, F.; Lundqvist, M.;Wannberg, B.; Mun, B. S.; Hussain, Z.; Liu, Z. Rev. Sci. Instrum. 2010,81, 053106.(28) Melaet, G.; Ralston, W. T.; Li, C. S.; Alayoglu, S.; An, K.;Musselwhite, N.; Kalkan, B.; Somorjai, G. A. J. Am. Chem. Soc. 2014,136, 2260−2263.(29) Thomas, J. M.; Gal, P. L. Adv. Catal. 2004, 48, 171−227.

Figure 5. Ambient pressure XPS spectra of Pt/TiO2 catalysts at (a) Ti 2p and (b) Pt 4f core levels in 0.1 Torr (0.13 mbar) H2 or O2 at giventemperatures (the spectra were normalized by correction for the relevant photon energy photoelectron cross sections and removal of a Shirleybackground).

Nano Letters Letter

dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−49124911

(30) Boyes, E. D.; Gai, P. L. Ultramicroscopy 1997, 67, 219−232.(31) Boffa, A. B.; Lin, C.; Bell, A. T.; Somorjai, G. A. Catal. Lett.1994, 27, 243−249.(32) Tauster, S. J.; Fung, S. C.; Garten, R. L. J. Am. Chem. Soc. 1978,100, 170−175.(33) Baker, L. R.; Kennedy, G.; Van Spronsen, M.; Hervier, A.; Cai,X. J.; Chen, S. Y.; Wang, L. W.; Somorjai, G. A. J. Am. Chem. Soc. 2012,134, 14208−14216.(34) An, K.; Alayoglu, S.; Musselwhite, N.; Plamthottam, S.; Melaet,G.; Lindeman, A. E.; Somorjai, G. A. J. Am. Chem. Soc. 2013, 135,16689−16696.

Nano Letters Letter

dx.doi.org/10.1021/nl502434m | Nano Lett. 2014, 14, 4907−49124912