686
Langmuir 1986, 2, 686-688
Direct Determination of Blocking of Surface Defects by Carbonaceous Deposits during Hydrocarbon Reactions on Platinum Single Crystals F. Zaerat and G. A. Somorjai* Materials and Molecular Research Division, Lawrence Berkeley Laboratory, and Department of Chemistry, University of California, Berkeley, California 94720 Received May 28, 1986. I n Final Form: July 18, 1986 A CO titration technique has been developed to quantitatively determine the presence of different kinds of catalytic sites on single-crystal surfaces. A direct correlation was found between platinum catalytic activity for conversion of hydrocarbons and the existence of bare defects on the metal surface. Isomerization and hydrogenolysis of either isopentane or butane are enhanced by the presence of steps or kinks on the catalyst surface. The same effect is not observed for methylcyclopentaneor n-hexane conversion, where carbonaceous deposits cover the surface defects within a few minutes of starting the reactions.
Many catalytic reactions are believed to occur at specific surface sites with well-defined geometries. This explanation is frequently invoked to explain changes in catalytic activity as a function of particle size on supported catalysts. The influence of metal particle size on the catalyst activity for hydrocarbon conversion was first reported by Gault,' and it is a topic that has attracted extensive research s i n ~ e . ~The , ~ results of such studies are sometimes contradictory, but it is generally accepted that many reactions are sensitive to the catalyst topography at the molecular level. Noticeable changes in either reaction rates or product selectivity have been reported for conversion of molecules like butane4 and n e ~ p e n t a n e . ~Similar ,~ changes have been seen by Anderson' and by Santacessarias for n-hexane reactions over ultrathin films and over platinum particles smaller than 10 A supported on silica. Dautzenberg and PlatteeuwgJoand Ponec and co-workers,ll on the other hand, did not find any significant structure sensitivity for the same system using catalysts with particle size from 15 to 80 A in diameter. The structure sensitivity of these reactions can be explained by the combined effects of two or three structural features. First, the surface of metal particles is composed of a combination of faces with different crystallographic orientations, including close-packed planes ((100) and (111) planes for fcc metals). There are also edges, corners, and other irregularities on the surface that may display different catalytic behavior than the flat terraces. The ratio of metal atoms on terraces to those at defects changes dramatically with particle size. A third contribution, due to the metal-support interface, has been used to explain some of the experimental results." In order to study the relative importance of these different sites to the overall performance of the catalyst in more detail, we have recently investigated the catalytic conversion of several hydrocarbons over a series of single-crystal surfaces with selected orientation^.'^-'^ It was found that isomerization and consecutive rearrangement of light alkanes (isobutane, n-butane, and neopentane) were maximized on platinum surface with high concentration of (100) microfacets.12 These reactions also show enhanced activity over stepped and kinked surfaces compared to flat metal surfaces. n-Hexane isomerization, by contrast, was structure-in~ensitive.~~ Only n-hexane aromatization was accelerated by the presence of (111) terraces. Present address: Brookhaven Natioiial Laboratory, NSLS Department, Bldg. 510E, Upton, NY 11973.
0743-7463/86/2402-0686$01.50/0
In this paper, we report a direct determination of the relative importance of terraces, steps, and kinks in catalyzing hydrocarbon conversion reactions. The high-pressure-UHV apparatus used in these studies is described in detail e1~ewhere.l~ It is equipped with standard surface science instrumentation (Auger electron spectroscopy,mass spectroscopy, low-energy electron diffraction) and a retractable cell to isolate the sample from vacuum and enclose it in a loop that acts as a well-mixed batch reactor for high-pressure reactions. Four platinum single crystals were employed, cut in the ( l l l ) ,(755), (332), and (10,8,7) orientations. A schematic representation of these surfaces is shown in Figure 1. They all have (111) terraces but different kinds of steps and kinks. Initial rates for isobutane, neopentane, n-hexane, and methylcyclopentane (MCP) conversion are presented in Figure 2 for the four platinum crystallographic faces. n-Hexane and MCP activities vary only slightly in the presence of the different irregularities. Isobutane and neopentane, on the other hand, react as much as 4 times faster when defects are present on the catalyst surface. A more detailed discussion of the results obtained for the lighter alkanes has been presented el~ewhere.'~J~ Only a small fraction of the surface area of the metal is actually active for catalysis. More than 50% of such surfaces is covered by carbonaceous deposits within the first few minutes of reaction and remains blocked during actual catalytic condition.16 These residues passivate the metal and inhibit further hydrocarbon conversion. Carbon monoxide adsorption-thermal desorption methods were developed to titrate active platinum sites before and after (1) Gault, F. G. Ann. Chim.(Paris) 1960, 5, 645. (2) Clarke, J. K. A.; Rooney, J. J. Adu. CataE. 1976, 25, 125. (3) Gault, F. G. Adu. Catal. 1981, 30, 1. (4) Anderson, J. R.; Avery, N. R. J. Catal. 1966,5, 446. (5) Foger, K.; Anderson, J. R. J. Catal. 1978, 54, 318. (6) Boudart. M.; Aldag, A. W.; Ptak, L. D.; Benson, J. E. J. Catal. 1968, 11, 35. (7) Anderson, J. R.; Shimoyama, Y. h o c . 5th Int. Congr. Catal., 1972 1973,695. (8) Santacessaria, E.; Gelosa, D.; Carra, S.J. Catal. 1975, 39, 403. (9) Dautzenberg, F. M.; Platteeuw, J. C. J. Catal. 1970, 19, 41. (10) Dautzenberg, F. M.; Platteeuw, J. C. J. Catal. 1972, 24, 264. (11) Lankhorst, P. 0.; DeJongste, H. C.; Ponec, V. Catalyst Deactiuation; Elsevier: Amsterdam, 1980. (12) Davis, S.M.; Zaera, F.;Somorjai, G. A. J. Am. Chem. SOC.1982, 104, 7452. (13) Davis, S.M.; Zaera, F.; Somorjai, G. A. J. Catal. 1984, 85, 206. (14) Zaera, F.; Godbey, D.; Somorjai, G. A. J. Catal., in press. (15) Blakely, D. W.; Kozak, E.; Sexton, B . A.; Somorjai, G . A. J. Vac. Sci. Technol. 1976, 13, 1901. (16) Davis, S. M.; Zaera, F.;Somorjai, G. A. J. Catal. 1982, 77, 439. (17) Kramer, R.; Zeugg, H. J. Catal. 1983, 80, 446.
0 1986 American Chemical Society
Langmuir, Vol. 2, No. 5, 1986 687
Letters
CO thermal desorption following reaclion studies on PI(S)-[6(111) xLlOOI]
Y)
c
--rlfier
irabvtane reocflon
Afler melkyicyclopenfant
" 0
373 473
573 673
Temperature ( K l
Figure 1. Idealized atomic surface structures for the flat (111), stepped (332) and (7551, and kinked (10.d.7) platinum singlecrystal surfacer.
I
I I I alkane skeletal rearrangement and
hydrogenolysis s t r ~ c t u r e
sensitivity 573 K , PH2:200Torr, PHC=PTor
11111
(332)
I5571
(10.8.7
Crystal Orientation Figure 2. Structure sensitivities for alkane conversion catdyzed over platinum single-crystal surfaces. reaction rate studies.I6 This technique provides information related not only to the amount of uncovered surface available for catalysis but also t o the nature of these sites. Examples of results obtained from these studies are shown
Figure 3. Comparison between CO thermal desorptionfrom (755) platinum surface clean and following hydrocarbon reactions. CO-adsorption temperature = 31w315 K. Heating rate = 80 K/s. CO exposure = 36 langmuirs.
in Figure 3. The topmost trace represents a standard thermal desorption spectrum obtained after CO-saturation of a (755) stepped platinum surface (also called Pt(s) S(l11) X (100)). The major peak at 430 K is due to desorption from the (111) terraces, while the high-temperature shoulder is related to desorption from the steps. The two distinctive adsorption states allow us to monitor the structural changes that occur as a result of the carbon deposition during hydrocarbon reactions. Figure 3 also shows that after isohutane and neopentane reactions, a fraction of the platinum atoms on both terraces and steps is still uncovered by carbon, roughly in proportion to their concentration ratio on the clean platinum surface. After reactions with n-hexane and methylcyclopentane, on the other hand, only a fraction of the terraces remain catalytically active as the high-temperature CO-desorption peak, due t o desorption from defects, disappears, indicating their passivation by deposited carbon. There is a good correspondence between changes in reaction rates and selectivity of site blocking for these hydrocarbon conversion processes. Isobutane and neopentane isomerization and hydrogenolysis proved to be faster over surfaces with steps and kinks. The increase in reactivity can be associated with the atoms at defect sites because they are not covered by hydrocarbon residues and are therefore available for catalysis. Methylcyclopentane and n-hexane reaction rates were similar for the four crystals studied here. This result is to he expected because steps and kinks become covered hy organic fragments and therefore deactivated by carbonaceous deposits within the first minutes of reaction, so the catalysis occurs only on (111)terraces on all surfaces. Our experiments have been carried nut under highly reducing conditions, with hydrogen to hydrocarbon raticw between 101and 301. Since these ratios are higher than those used in commercial processes, we expect that our results are applicable to supported catalysts. In conclusion, a CO-titration technique was developed to directly identify catalytic sites on platinum singlecrystal
688
Langmuir 1986, 2, 688-688
surfaces. The importance of irregularities on the metal surface for isobutane and neopentane conversion was established by this method. n-Hexane and methylcyclopentane reactions, on the other hand, are not affected by the presence of such defects, as these heavier alkanes deposit carbonaceous fragments at the surface irregularities and passivate them during both catalytic reactions and for CO adsorption. As a result the structure sensitivity that is due to defects a t the surface (steps and kinks) should
be absent for the reactions of heavier alkanes while very much present for lighter alkanes.
Acknowledgment. This work was supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, Materials Sciences Division of the U S . Department of Energy, under Contract DE-AC03-76SF00098. Registry No. MCP, 96-37-7; Pt, 7440-06-4; CO, 630-08-0; isopentane,78-78-4; butane, 106-97-8;n-hexane, 110-54-3.
Nev-s and Announcements It is with deep regret that we note the death of Stephen Brunauer, perhaps best noted as coauthor of the BET equation. An obituary will follow.
Announcement Nominations for the Victor K. LaMer Award for graduate research in colloid and surface chemistry, sponsored by the American Chemical Society Division of Colloid and Surface Chemistry, must be submitted by March 1, 1987, to Dr. P. Stair, Department of Chemistry, Northwestern University, Evanston, IL 60201, phone 312-491-5266. New Journals Notice of two new private journals has been received. Journal of Adhesion Science and Technology; W. J. van Ooij (Department of Materials Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061) and K. L. Mittal (IBM Corporate Technical Institute, Thornwood, NY 10594) Eds. Published by VNU Science Press: Utrecht, The Netherlands. First issue, end of 1986. Published quarterly; annual subscription DM 358. Liquid Crystals; Prof. G. R. Luckhurst (Department of Chemistry, The University, Southampton, SO9 5NH, U.K.) and Prof. E. T. Samulski (Institute of Materials Science, U-136, The University of Connecticut, Storrs, CT 06268) Eds. Published by Taylor and Francis: Philadelphia. First issue, March 1986. Published bimonthly; annual subscription $160. Meetings in 1986 November 19-21. Foam: Basic Properties and Applications in Porous Media. 42nd Annual Southwest Regional Meeting of the American Chemical Society, Houston, TX. Dr. J. K. Borchardt, Shell Development Co., Houston, T X 77001. December 15-17. 4th Miami International Symposium on Multi-Phase Transport and Particulate Phenomena, Miami Beach, FL. Dr. T. Nejat Veziroglu, Clean Energy Research Institute, Coral Gables, FL 33124. Meetings in 1987 February 4-6. Formula. 1st International Forum on Physic0 Chemistry of Interfaces and Formulation of Chemical Specialties. Nice, France. Scientific Committee of Formula, SociBtB Francaise of Chimie, DBpartement Congrh, 75005 Paris, France. April 5-10. American Chemical Society National Meeting, Division of Colloid and Surface Chemistry. Denver, CO. General papers: Dr. R. Mackay, SMCCR-RCS, Chemical Division, Research Directorate, U S . Army Chemical Research and Development Center, Aberdeen Proving Ground, MD 21010. “Liquid Crystals and Ordered Fluids”; W. J. Benton, Department of Chemical Engineering, Rice University, Houston, TX 77251, or Dr. Anselm Griffin, Department of Chemistry, University of Southern Mississippi, Southern
Station, Hattiesburg, MS 39406. “Surface Chemistry in Biology, Medicine and Dentistry: Bioelectrochemistry”; Dr. M. Blank, Columbia University, Department of Physiology, New York, NY 10032. “General Papers in Catalysis and Related Topics”; Dr. I. I. Wachs, Exxon Research and Development Co., Clinton Township, Annandale, NJ 08801, or Dr. E. Stuva, Department of Chemical Engineering, University of Washington, Seattle, WA 98105. “Photochemical and Electrochemical Surface Science: Photoelectrochemistry”; Professor N. R. Armstrong, Department of Chemistry, University of Arizona, Tucson, AZ 85721. “Molecular Processes at Solid Surfaces: Spectroscopy of Intermediates and Adsorbate Interactions”; Dr. Neal D. Shinn, Sandia National Laboratories, Division 1134, Albuquerque, NM 87185. “Kendall Award Symposium”; to be announced. April 26-29. IUPAC-Symposium on Characterization of Porous Solids. Bad Soden, West Germany. Dechema, Attn. C. Birkenberg, D-6000, Frankfurt/M 97, West Germany. May 17-21. Advantages of Surfactant Mixtures. American Oil Chemists’ Society Annual Meeting. New Orleans, LA. Dr. J. F. Scamehorn, University of Oklahoma, Norman, OK 73019. June 8-12. The 1987 International Congress on Membranes and Membrane Processes. Tokyo, Japan. Prof. Dr. Shoji Kimura, Secretary-General, ICOM ’87, Institute of Industrial Science, University of Tokyo, Roppongi, Minato-ku, Tokyo 106, Japan. June 17-19. Conference on Chemically Modified Surfaces. Fort Collins, CO. W. T. Collins, Dow Corning Corp., Midland, MI 48686. June 21-24. 61st Colloid and Surface Science Symposium. Ann Arbor, MI. Prof. E. Gulari, 3168 Dow Building, Chemical Engineering Department, Univeristy of Michigan, Ann Arbor, MI 48109. June 22-25. Second International Conference on the Structure of Surfaces. Amsterdam, The Netherlands. Louise ROOS,Conference Secretary ICSOS-11, FOMInstitute for Atomic and Molecular Physics, Kruislaan 407, 1098 SJ Amsterdam, The Netherlands. July 13-18. 31st IUPAC Congress. Sofia, Bulgaria. Dr. D. Exerowa, Bulgarian Academy of Sciences, Institute of Physical Chemistry, 1040 Sofia, Bulgaria. August 12-19. World Congress of Theoretical Organic Chemists. Section B (Aug 17-19) includes Catalysis and Solid State Chemistry, Budapest, Hungary. E. A. Lang, WATOC Congress, Hungarian Chemical Society, H-1061 Budapest, Anker koz 1, Hungary. September 7-9. IACIS-Symposium. Eindhoven, The Netherlands. Prof. Dr. G . J. Fleer, Agricultural University, Laboratory for Physical and Colloid Chemistry, De Dreijen 6, 6703 Wageningen, The Netherlands.
