STM Investigation of CO Ordering on Pt(111): From an Isolated

Article pubs.acs.org/JPCC

STM Investigation of CO Ordering on Pt(111): From an Isolated Molecule to High-Coverage Superstructures Hyun Jin Yang,‡,§ Taketoshi Minato,§,† Maki Kawai,‡ and Yousoo Kim*,§ ‡

Department of Advanced Materials Science, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8651, Japan Surface and Interface Science Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan

§

ABSTRACT: Carbon monoxide (CO) adsorbed on Pt(111) has been extensively studied as a model catalyst. However, there remain some ambiguities in overlayer structures, particularly regarding bridge-site occupation. Here, we report real-space observations of CO on Pt(111) using scanning tunneling microscopy (STM) under ultrahigh vacuum at a cryogenic temperature, from a single CO adsorbed on the atop site to the gradual development of overlayer structures including atop-dominant (√3 × √3)R30° islands, c(4 × 2) domains, and 1 × 1 boundaries in c(4 × 2)-2CO domains. Bridge CO appear at the edge of (√3 × √3)R30° islands, significantly in the center of local c(√3 × 2)rect geometry which is equivalent to c(4 × 2)-2CO. In the c(4 × 2) domain including bridge vacancies, the height of the atop CO in the STM image is modulated according to the number of adjacent bridge CO, which implies the interadsorbate interaction between two different adsorption species. The real space observation presented here not only resolves ambiguities about overlayer structures but also describes an atop-bridge interadsorbate interaction.

1. INTRODUCTION Carbon monoxide (CO) on a platinum (Pt) surface is one of the most extensively studied systems in the field of surface science, not only for its practical importance in heterogeneous catalysis but also for the fundamental understanding of elementary processes at surfaces, including adsorption, diffusion, desorption, and reactions.1 For this reason, CO on Pt(111) has been investigated as a prototype model system with a wide range of experimental techniques to analyze electronic structure,2−6 vibrational properties,7−18 overlayer structures,7,9,10,19−25 and surface dynamics.7,17,26−35 As well as experiments, theoretical investigations have been undertaken for a long time, from the simple Blyholder model36 to density functional approaches.37−42 A brief summary of previous results concerning the adsorption geometry at a temperature range between 100 and 150 K is as follows: (1) atop sites are primarily occupied by CO, followed by bridge-site occupation at increased coverage of ∼0.2 monolayer (ML, 1.58 × 1015 molecules/cm2),7,29,43,44 (2) atop-dominant (√3 × √3)R30° structures appear up to coverage of ∼0.3 ML,9 and atop-bridge mixed c(4 × 2)7,45 appears at higher coverage (see Figure 1). Low energy electron diffraction (LEED) experiments have shown that (√3 × √3)R30°-derived structures appear at 0.17 ML, 0.25 ML, and 0.33 ML,7 and Tüshaus et al. suggested real space models of (4 × 4) and (8 × 8) islands with (√3 × √3)R30° local structure having all molecules in the atop position for 0.17 and 0.33 ML, respectively.9 However, they did not mention the location of bridge CO molecules and their contribution to the formation of the overlayer structure, although bridge CO must be involved in surface coverage © 2013 American Chemical Society

Figure 1. Schematic models for the overlayer structures of CO on Pt(111) and conventional notation of them.

higher than 0.2 ML. Indeed, both infrared absorption− reflection spectroscopy (IRAS)9,46,47 and electron energy loss spectroscopy (EELS)7,43,44,48 confirmed that the vibrational peak corresponding to the C−O internal stretch mode of bridge CO appears from 0.2 ML. So far, there has been no report that precisely describes where the bridge CO is located inside superstructures. The most reasonable space for bridgeCO occupation is between islands, but still there has been no direct evidence to prove it. Another issue to be discussed is the evolution of overlayer structures from (√3 × √3)R30° to c(4 × 2) in the coverage range from 0.3 to 0.5 ML. The line width of M−C stretch mode (ν(M−C)) of atop CO measured by IRAS is sensitive to the structural homogeneity of the adsorbate layer,18,46 providing precise information on structural changes. Ryberg observed a distinct minimum and a sharp increase in line width Received: April 29, 2013 Revised: July 19, 2013 Published: July 22, 2013 16429

dx.doi.org/10.1021/jp404231t | J. Phys. Chem. C 2013, 117, 16429−16437

The Journal of Physical Chemistry C

Article

surface atoms, was estimated by counting the number of CO molecules in the STM images.

at 0.3 ML, which revealed a change in the overlayer structure from (8 × 8) islands with (√3 × √3)R30° local structure to a c(4 × 2) domain.18 Interestingly, Ryberg observed that the line width showed a sharp maximum at 0.38 ML, while the LEED result indicated that the c(4 × 2) overlayer structure was kept between 0.33 and 0.5 ML,18 which was also observed by Malik and Trenary in a similar manner.46 Yet, no detailed explanation for this coverage-dependent line width change has been provided in terms of overlayer structure. Scanning tunneling microscopy (STM) is one of the most powerful tools to provide real space information with atomic resolution to study the overlayer structures of an adsorption system, which can resolve those two issues on overlayer structures. However, the number of STM studies regarding CO on Pt(111) is surprisingly small, and most of them are limited in terms of coverage range. Stroscio et al. observed two different types of isolated CO molecules at very low coverage and fabricated a small (√3 × √3)R30° island by molecular manipulation with an STM tip.24 Song et al.22 and Pederson et al.23 independently investigated overlayer structures in the 0.5 ML range, showing the c(4 × 2) structure with clearly resolved bridge species. The c(4 × 2) structure was also observed by Wintterlin et al., during the study of catalytic oxidation of CO with a preadsorbed oxygen layer.49 So far, neither detailed studies of overlayer structure evolution other than 0.5 ML nor spectroscopic studies with STM such as scanning tunneling spectroscopy (STS), inelastic electron tunneling spectroscopy with STM (STM-IETS), or action spectroscopy have been reported. In this Article we report the results of a molecularly resolved STM investigation on CO/Pt(111) from an isolated molecule to highly ordered overlayer structures, with addressing the unsolved issues on the overlayer structures in intermediaterange of coverage. At the low coverage limit, i.e. less than 0.01 ML, the adsorption site of a single molecule was determined with atomic resolution, and the electronic structure of the unoccupied state was studied by STS and action spectroscopy with STM. By varying the surface coverage up to ∼0.6 ML, the molecular ordering of atop COs to (√3 × √3)R30° islands, the appearance of bridge COs inside the overlayer structure, the phase transition of (√3 × √3)R30° to the fully covered c(4 × 2) domain (0.5 ML), and 1 × 1 boundary formation by further adsorption were characterized by STM imaging. In particular, at coverages between ∼0.4 and 0.5 ML, it was found that c(4 × 2) domains had developed with well-ordered bridge vacancies. Finally, we will discuss the intermolecular interaction of atop CO with adjacent bridge CO, leading to changes in the topographic height of atop COs in the STM images.

3. RESULT AND DISCUSSION 3-1. At a Low Coverage Limit: Isolated Molecules. Individual CO molecules on a Pt(111) surface appear as a round-shaped protrusion in an STM image at bias voltages near the Fermi level, as shown in Figure 2a. The diameter of the protrusion is approximately 0.5 nm, and the height differs from 25 to 40 pm, according to the applied sample bias voltage.

Figure 2. Topographic STM images of CO on Pt(111) in low coverage limit. (a) Typical view of low coverage CO on Pt(111). Bright protrusions are CO molecules (image size 10 × 10 nm2, VS = 0.3 V, IT = 1 nA). (b) An isolated molecule (image size 3 × 3 nm2, VS = 2 V, IT = 1.6 nA) and (c) its adsorption site. Region A: VS = 2 mV, IT = 1.6 nA and Region B: VS = 20 mV, IT = 1.6 nA. See text.

STM images in b and c of Figure 2 show an identical CO molecule imaged under different scan conditions. In particular, Figure 2c was obtained with changing tunneling resistance to get both the atomic position of the substrate (region marked as A, Vsample = 2 mV and Itunnel = 1.6 nA) and the position of the molecule (region marked as B, Vsample = 20 mV and Itunnel = 1.6 nA) within a single image.50 The superimposed grid was made to indicate the position of the Pt atoms in region A. Atop adsorption of a CO molecule was confirmed by showing that the center-dip protrusion was exactly on the cross point of grids, as is widely accepted from the vibrational frequency of CO stretch mode in both IRAS and EELS.48 Stroscio and Eigler reported the coexistence of two different protrusions of CO, which was suggested as a difference in appearance according to the adsorption site, atop and bridge, involved in low temperature adsorption at 5 K.24 However, all the CO molecules in our observation showed an identical protrusion, implying that only atop species exist at this coverage. We anticipate the reason for this atop-site occupation to the higher adsorption temperature (∼ 50 K) than that used in their experiment, which was enough to evoke the thermal conversion of bridge CO to atop CO.32 Isolated CO molecules were immobile upon scanning up to ∼3.5 V, but higher bias scans caused lateral hopping of the molecules. Figure 3a exhibits the lateral hopping of a CO molecule, before dosing tunneling electrons (left), the disappearance of the target molecule during electron injection (4 V) (middle), and the subsequently repositioned target molecule (right). The topographic height of the individual CO molecules did not change before and after the hopping event, indicating that the atop adsorption geometry was retained. We measured the hopping probability per electron as a function of the sample bias voltage, as shown in Figure 3b. Two sets of reaction probability data for two different CO molecules with the same tip were in agreement with each other to show a rapid increase in lateral hopping probability at 4.5 V. Figure 3c shows the dI/dV spectra measured on a CO molecule and on the Pt

2. EXPERIMENTAL METHODS All the experiments were performed using a low-temperature scanning tunneling microscope (Omicron GmbH) in an ultrahigh vacuum (UHV) chamber (Base pressure (PBase) = 2.7 × 10−9 Pa). A Pt(111) single crystal surface was cleaned by several cycles of Ar ion sputtering (1 keV, 3 μA) and annealing (1100 K), followed by oxidation (1.0 × 10−5 Pa, 800 K) and flashing at 1070 K. All data, such as STM images, STS spectra, and action spectrum were collected at 4.7 K, using an electrochemically etched tungsten tip. CO molecules were adsorbed onto the Pt(111) surface through a dosing tube located near the surface at ∼50 K. The surface coverage of CO (θ), defined by the number ratio of adsorbate molecules to 16430



Article

bare surface, where the signal from CO stiffly increased at ∼4 V. In previous studies, two-photon photoemission (2PPE)5 and inverse photoemission3 reported the unoccupied 2π* state of CO on Pt(111) at 4.5 eV for atop CO species. Thus, the threshold near 4.5 V in the action spectrum corresponds to the tail of the 2π* state of atop CO on Pt(111) where the electrons tunnel into, resulting in lateral hopping of CO, which is similar to the pathway of energy transfer in CO hopping on Cu(110) by 2π* excitation, as shown by Bartels et al.51 Note that no hopping motion was observed near the vibrational energy level of CO internal stretch (260 meV for atop CO) during scanning as well as injecting tunneling electrons, in contrast to CO on Pd(110).52 3-2. Appearance of Bridge Species at Low Coverage (∼0.1 ML): Temperature Dependence and Potential Energy Surface. Figure 4a shows an STM image at 0.11 ML of surface coverage. Most of the CO molecules are irregularly scattered on the surface, but a few small rectangular units are also observed. In detail, the rectangular units (Figure 4b) are composed up of four 40 pm-high atop COs, bearing a 20 pm-high protrusion in the center. The dimension of the rectangular unit is the same as that of the c(√3 × 2)rect (equivalent to c(4 × 2)-2CO, see Figure 1) structure, and the center species is identified to be a bridge-adsorbed CO based on the high resolution STM images (Figure 4b) and an accompanying schematic model (Figure 4c). The inset image of Figure 4b was obtained after injecting tunneling electrons at 300 mV into the center of the rectangular unit in the leftbottom of Figure 4b as depicted by the yellow arrow. The electron injection induced the conversion of the 20-pm species to the 40 pm-species and the rearrangement of adjacent molecules, which clearly supports the existence of bridge species in the center of the rectangular unit. In our best observation, rectangular units without a bridge CO at the center were rarely found. Also the bridge-adsorbed CO did not

Figure 3. (a) STM topographic images with applying bias voltage on the top protrusion: before (left), with applying bias voltage (middle), after the hopping event (right). (Scale bar is 1 nm, image size 2.5 × 6 nm2, VS = 0.1 V, IT = 0.5 nA.) (b) Two sets of independent action spectrum from isolated CO molecules. (c) STS from a single CO molecule (blue), Pt surface (red) curve.

Figure 4. Topographic images of CO molecules with ∼0.1 ML coverage. (a) After CO exposure at ∼50 K (image size 10 × 10 nm2, VS = 0.1 V, IT = 3 nA). A few rectangular units were observed among the scattered molecules. (b) Zoomed-in image for rectangular units (image size 3 × 3 nm2, VS = 0.1 V, IT = 0.5 nA) and (c) corresponding ball model (atop CO: red, bridge CO: blue, Pt atom: gray). (e) After room-temperature annealing. All rectangular units disappeared (image size 10 × 10 nm2, VS = 0.1 V, IT = 0.5 nA). (f) A zoomed-in image (image size 3 × 3 nm2, VS = 0.1 V, IT = 1 nA), showing the minimum intermolecular distance of ∼4.6 Å and (c) corresponding ball model. 16431



Article

exist as an isolated form but only appeared at the center of a rectangular unit. It is known that the bridge site can be occupied by CO molecules even at low coverage if the adsorption temperature is low enough to suppress surface migration,32 and the relative population of bridge CO exponentially increases upon increasing total coverage.53 Indeed, bridge CO molecules were not observed at 0.005 ML, but we could find them at 0.1 ML, which is consistent with the previous report suggesting an increased probability of finding bridge CO at higher coverage.53 The adsorption temperature, 50 K, was higher than the threshold temperature (37−42 K) for bridge-to-atop hopping obtained by an IRAS study,32 resulting in atop adsorption for most CO molecules. Nevertheless, the bridge CO inside the c(√3 × 2)rect unit still survives. One plausible explanation is the high diffusion barrier of bridge-to-atop hopping inside the c(√3 × 2)rect unit, due to repulsive intermolecular interaction in the resultant (1 × 1) arrangement. Another possible explanation could be that the center bridge CO interacts with the surrounding atop COs to stabilize the total energy of the c(√3 × 2)rect unit. Most of the c(√3 × 2)rect units, including a bridge CO, disappeared after heating the surface to room temperature, as seen in Figure 4d. This is consistent with the results of an IRAS study,32 insisting the rearrangement of adsorbed CO from bridge to atop site by thermal excitation (100 K). All CO molecules appear identical in shape and height in the STM image, confirming the existence of only one adsorption species. In addition, triangular units with an intermolecular distance of 4.6 Å often appeared as shown in Figure 4e, corresponding to the partial (√3 × √3)R30° arrangement (Figure 4f). It is notable that 4.6 Å, the second nearest neighbor distance, is the minimum atop-to-atop distance up to 0.5 ML of coverage, implying that the 1 × 1 arrangement cannot exist due to repulsive interactions. 3-3. Evolution of a Hexagonal Superstructure and How Bridge Species Appear. A gradual increase in CO exposure brought forth the ordering of molecular layers. In order to form thermodynamically stable arrangements, the surface was annealed to room temperature prior to STM scanning at 4.7 K. Figure 5 shows the evolution of (√3 × √3)R30° islands from a randomly distributed molecular arrangement (Figure 5a) with increasing surface coverage. In Figure 5b−d, all molecules inside the islands show the (√3 × √3)R30° arrangement, and the average number of molecules in the island at each coverage was measured to be 3.09 ± 1.37 molecules (0.16 ML, Figure 5b), 5.47 ± 2.66 molecules (0.25 ML, Figure 5c), and 13.04 ± 2.31 molecules (0.315 ML, Figure 5d). Although small-sized (10 nm × 10 nm) STM images are presented in Figure 5, the homogeneity of island size and structures was confirmed by wide-area scanning at various positions. The formation of well-ordered (√3 × √3)R30° islands was also supported by the real-space model deduced from the LEED patterns.9 In our best observation, no large-area domains with the (√3 × √3)R30° arrangement were found, which implies the existence of intermolecular interactions limiting the size of islands. In order to take a closer look at the island structures in c and d of Figure 5, marked regions with yellow squares in the STM images are depicted with the corresponding schematic models, as shown in e and f of Figure 5, respectively. Although the internal arrangement of each island is (√3 × √3)R30°, neighboring islands are laterally shifted relative to each other by

Figure 5. Topographic STM images with increasing coverage. (a) θ = 0.11 ML (VS = 0.1 V, IT = 1 nA), (b) 0.16 ML (VS = 0.1 V, IT = 1 nA), (c) θ = 0.25 ML (VS = 0.1 V, IT = 0.5 nA), (d) θ = 0.315 ML (VS = 0.1 V, IT = 0.5 nA). The size of each image in (a)−(d) is 10 × 10 nm2. (e) and (f) are schematic models of marked regions (yellow square) in (c) and (d), respectively (Pt lattice: gray line, atop CO: red point, bridge CO: blue point).

a unit lattice of the Pt(111) surface (r0 = 2.7 Å) in both cases. This lateral shift results in the formation of an antiphase boundary between neighboring islands. However, despite the same magnitude of shift, the island structures in Figure 5e and f are different from each other in terms of the interisland distance and boundary structure. Here the interisland distance is defined as the shortest intermolecular distance between two neighboring islands, which corresponds to √3r0 in Figure 5c and 2r0 in Figure 5d, and is represented by orange lines in Figure 5e and light blue lines in Figure 5f. The interisland distances are remarkably homogeneous in both coverages. In Figure 5d, it is notable that 2r0 of the interisland distance leads to the formation of the (√3 × 2)rect arrangement between neighboring islands, consistent with the suggested (8 × 8) real space model from LEED observations.9 The (√3 × 2)rect arrangement is identical to the structure of the rectangular units observed in Figure 3b, which provides the possibility of locating a bridge CO (marked with light blue dots in Figure 5f) in the center of each rectangular unit (c(√3 × 2)rect). This would imply that bridge-adsorbed COs can coexist with atop-dominating (√3 × √3)R30° islands by forming a local boundary with a c(√3 × 2)rect structure. Note that some CO molecules at the rim of each island show a lower protrusion than COs inside the island, which will be discussed in the next section as the evidence of bridge occupation with c(√3 × 2)rect arrangement. If every available center of the rectangular units is occupied by a bridge CO, the coverage of this surface (Figure 5d) could reach 0.41 ML while it was previously estimated to be 0.315 ML when only atop CO molecules were counted. . 16432



Article

islands accompanied with widening of c(√3 × 2)rect boundary. Figure 6c represents both rectangular superlattice and (√3 × √3)R30° islands. Inside the rectangular superlattice, atop COs are clearly recognizable, whereas individual bridge COs are not clearly resolved in the STM image due to their low protrusion (∼20 pm). Dip structures are also observed at the center of the rectangular units, which corresponds to the vacancies of bridge CO. The bridge vacancies provide contrast to recognize individual bridge CO molecules. A detailed STM image (Figure 6d) and the line profile (Figure 6f) along the blue solid line apparently indicate that bridge vacancies appear as dip structures at the position where bridge COs are supposed to occupy. As a consequence, the rectangular units are classified into two types depending on the existence of bridge CO; one is c(√3 × 2)rect with a bridge CO, and the other is (√3 × 2)rect without any bridge CO. Note that both c(√3 × 2)rect and (√3 × 2)rect can be commonly considered as c(4 × 2) in the conventional Wood notation, as depicted in b and c of Figure 1, respectively. The bridge vacancies lie along the [11̅0] direction of the Pt surface in Figure 6c. The surface coverage of the rectangular domain region in Figure 6a was estimated to be 0.375 ML. This number lies in the coverage range of (√3 × √3)R30° islands region between 0.315 and 0.41 ML, according to the number of boundary rectangular units filled with bridge COs as mentioned in section 3-3. Thus, the hexagonalrectangular phase transition might occur at the observed surface coverage of 0.375 ML, which is consistent with the LEED studies observing that the c(4 × 2) structure appeared from 0.35 ML.7,18 Interestingly, Figure 6c shows atop CO molecules with various heights. A magnified image of a rectangular domain (Figure 6d), a schematic model of Figure 6d (Figure 6e), and the corresponding line profile along the red dashed line (Figure 6f) indicate that the CO molecule marked with an arrow is apparently dimer and lower than the other atop CO molecules. Considering the arrangement of brighter and dimer atop molecules, the height difference reveals a strong correlation with the number of neighboring bridge CO molecules. Each atop CO has four available sites for adjacent bridge occupation in a c(4 × 2) arrangement. In Figure 6d, the brighter atop COs have only two neighboring bridge CO molecules, while the dimer atop COs are fully surrounded by four bridge COs. In order to examine this observation quantitatively, another STM image of the bridge-vacant c(4 × 2) domain was obtained as shown in Figure 7a, and the height of each atop CO was

3-4. Phase Transition of Overlayer Structure and Interadsorbate Interaction. An additional dose of CO molecules on the surface of Figure 5 induced the emergence of a rectangular superlattice (Figure 6) on the surface covered

Figure 6. (a) Coexistence of hexagonal and rectangular lattice of CO molecules, appearing in the middle of phase transition (image size 30 × 20 nm2, VS = 0.1 V, IT = 0.5 nA), (b) and (c) are zoomed-in images from marked regions in (a). (b) Irregularly mixed phase with small domain of (√3 × √3)R30° phase and c(√3 × 2)rect phase. (image size 5 × 5 nm2, VS = 0.1 V, IT = 1 nA). (c) Boundary region between the (√3 × √3)R30° phase and c(√3 × 2)rect phase (image size 5 × 5 nm2, VS = 0.1 V, IT = 0.5 nA). (d) Detailed view of the c(√3 × 2)rect domain (image size 3 × 3 nm2, VS = 10 mV, IT = 0.5 nA) and (e) its ball model (red: atop CO, blue: bridge CO, light gray: Pt, semitransparent gray: vacancy of bridge CO). (f) Height profiles from (d), for bridge CO molecules (blue solid line) and for atop CO molecules (red dashed line).

with hexagonally arranged (√3 × √3)R30° islands shown in Figure 5d. At the early stage of rectangular lattice formation, three regions are observed (Figure 6a); the (√3 × √3)R30° islands that already existed, an inhomogeneous mixture of (√3 × √3)R30° and c(4 × 2) domains (Figure 6b), and a welldeveloped rectangular superlattice (Figure 6c). The size of most domains in the mixture region (Figure 6b) does not exceed the size of (√3 × √3)R30° islands (Figure 5d), so that the mixture region can be understood as an intermediate of the phase transition from hexagonal to rectangular superlattices which results from the compression of the (√3 × √3)R30°

Figure 7. (a) A STM image of the bridge-vacant c(4 × 2) domain (image size 5 × 5 nm2, VS = 2 mV, IT = 0.5 nA) and (b) a line profile from the image. (c) Height distribution of atop CO molecules according to the number of neighboring bridge CO (NnbCO) was extracted from all atop CO molecules in (a) by measuring the line profile correlated with NnbCO. Black circles for NnbCO = 2, red squares for NnbCO = 3, and green triangles for NnbCO = 4. 16433



Article

have the negative end outward (Cδ+−Oδ‑) by simultaneously measuring the workfunction change and temperature-programmed desorption.56 Dipole reversal at different adsorption sites has also been suggested by a recent report using DFT calculation59 calibrated with singlet−triplet splitting energy,42 by estimating the amount of charge in each atom of CO. According to the authors, the opposite dipoles in the c(4 × 2) arrangement are enhanced due to electrostatic interaction, and this interaction provides the stabilization of the structure. In this point of view, various heights of atop CO in the bridgevacant c(4 × 2) domain can be understood as the different degrees of dipole−dipole interaction according to the number of neighboring bridge CO in the local environment. Additionally, the dimer atop CO with fully occupied bridge CO may result from the lateral broadening of the charge density distribution due to the attractive electrostatic interaction between two opposite direction of dipoles. Our observation provides the microscopic evidence for the preceding discussions about the interadsorbate interactions. These two descriptions, namely the opposite direction of charge transfer and the opposite direction of dipole moments between atop and bridge CO, can provide additional explanations to our observations. One aspect regarding the attractive interaction between atop and bridge COs is the existence of rectangular c(√3 × 2)rect units with lowtemperature adsorption, as shown in Figure 4b. As an extended discussion following section 3-2, the electrostatic interaction between the atop CO dipole and the bridge CO dipole may provide significant stabilization of c(√3 × 2)rect units over the (√3 × 2)rect structure without a bridge CO. Also the alignment of the bridge-vacancy along the [11̅0] direction can be understood with the electrostatic dipolar interaction. The atop COs in the (√3 × 2)rect unit will repulsively interact with each other due to the same direction of their dipole moments, resulting in preference for the [11̅0] direction, which provides a longer atop−atop distance (2r0) than the [1̅12] direction (√3r0). The topographic height variation according to NnbCO can be used for finding bridge COs in (√3 × √3)R30° islands in Figure 5d, such that some atop COs at the rim of the islands are dimmer than inner-island atop COs. Although our STM image (Figure 5d) does not resolve the individual bridge COs between (√3 × √3)R30° islands, all the atop COs at the rim consist of c(√3 × 2)rect units so as to consider that the bridge CO molecules exist at the corresponding position if the atop CO at the rim shows a lower topographic height in the STM image. Observation of the bridge-vacant c(4 × 2) domain and the interadsorbate interaction can provide a plausible explanation about the maximum line width of the M−C stretch mode at 0.38 ML of surface coverage. As mentioned in the Introduction section, the line width has been interpreted as a measure of the ordering of the overlayer structure, so that the maximum line width at 0.38 ML with the well developed c(4 × 2) pattern was explained as indicating the c(4 × 2) island in a very disordered matrix.18 However our observation revealed that the overlayer structure at 0.38 ML is a well ordered c(4 × 2) domain accompanied with bridge vacancies, without any disordered matrix. Moreover atop COs with various height according to the number of neighboring bridge CO indicates the atop COs are in locally inhomogeneous environment due to the interadsorbate interaction, possibly resulting in the broadening of the line width.

measured. Figure 7b represents one of the line profiles used for measuring the height of atop COs, with each atop CO marked with the number of neighboring bridge CO (NnbCO) using black circles (NnbCO = 2), red squares (NnbCO = 3), and green triangles (NnbCO = 4). There are 98 atop CO molecules in Figure 7a: 63 atop COs with two neighboring bridge COs (NnbCO = 2), 8 atop COs with three neighboring bridge COs (NnbCO = 3), and 27 atop COs with four neighboring bridge COs (NnbCO = 4). The average height of atop COs according to NnbCO are 26.1 ± 1.2, 21.6 ± 0.9, and 17.8 ± 1.0 pm for NnbCO = 2, 3, and 4, respectively (Figure 7c). Therefore, our observation clearly indicates that the protrusion of atop CO is significantly dimmed as the number of adjacent bridge CO increases. One of the possible explanations accounting for the height modulation of atop CO would be a change in real topographic height, by altering the M−C or C−O bond length. However, the relative position of each atom in atop CO at 0.3 ML of surface coverage coincides well within error bars of ±0.1 Å with that at 0.5 ML of surface coverage, according to the analysis of LEED intensity.19,20 Another possibility might be tilting of atop CO resulting from a repulsive intermolecular interaction that forces the atop CO to lean toward the bridge vacancy. However, this explanation does not seem reasonable, as the atop CO at this coverage are known to stand upright from electron-stimulated desorption ion angular distribution (ESDIAD),21 near edge X-ray absorption fine structure (NEXAFS),54 and LEED19 studies. Another possible explanation would be a change in the charge density distribution of atop CO by the surrounding bridge CO, considering that the topographic STM image provides information not only about geometry but also about the charge density distribution based on the local density of state.55 The charge density distribution of each adsorption species is dependent on (1) the amount and direction of charge transfer between the metal and CO and (2) the internal charge distribution of the CO molecule itself. On one hand, regarding the charge transfer between CO and the Pt surface, several groups have reported a workfunction decrease upon CO adsorption up to certain coverage and a subsequent increase by additional exposure, revealing that the minimum workfunction appears at the onset coverage for bridge occupation.7,45,56,57 This workfunction change has been interpreted as showing the opposite direction of charge transfer between atop CO and bridge CO, i.e. the direction of atop CO-to-metal and metal-tobridge CO charge transfer, respectively. At surface coverage exceeding the workfunction minimum, bridge COs depolarize the metal surface, which had been polarized by charge transfer from the atop COs. This depolarizing scheme may apply to the height modulation of atop COs in the bridge-vacant c(4 × 2) domain on the microscopic scale, such that the charge density of atop CO decreases upon adjacent bridge-CO adsorption due to charge redistribution through the metal substrate. The typical Blyholder model36 and further modified models58 of CO adsorption also support this explanation, since bridge CO have been considered to accept the charges from the Pt substrate from its lower C−O stretch frequency, which is evidence of stronger back-bonding to the 2π* orbital,36 while the atop CO have been considered to provide charges to the Pt substrate. On the other hand, regarding the internal charge distribution of each CO adsorbate, Norton et al. insisted that the dipole moment of each adsorption species must be opposite, as atop CO have the positive end outward (Cδ‑−Oδ+), while bridge CO 16434



Article

Figure 8. Gradual filling of the bridge vacancy of the incomplete c(√3 × 2)rect domain in accordance with overall coverage (a−d), and the schematic model (e−h). (a) ∼0.375 ML (VS = 0.1 V, IT = 1 nA), (b) ∼0.417 ML (VS = 0.1 V, IT = 0.5 nA), (c) ∼0.438 ML (VS = 0.1 V, IT = 1 nA), (d) fully covered c(√3 × 2)rect with θ = 0.5 ML (VS = 0.1 V, IT = 0.5 nA). The size of each image in (a)−(d) is 5 × 5 nm2.

3-5. Formation of a Complete c(4 × 2)-2CO Structure and Further Adsorption. The height difference of atop CO according to the number of neighboring bridge CO (NnbCO) consistently appears with increased surface coverage, as shown in Figure 8. As more CO molecules were dosed, the number of brighter atop species gradually decreased, resulting in a complete c(√3 × 2)rect structure with homogeneous height of atop COs (Figure 8d). The unit cells in Figure 8 are simple combinations of the c(√3 × 2)rect unit and (√3 × 2)rect unit, and we can find the position of bridge CO vacancies from the relative height of atop COs in the rectangular unit. The number ratio of bridge vacancies over all the available bridge sites was modeled to be 1/2 for 0.375 ML (Figure 8a and e), 1/ 3 for 0.4175 ML (Figure 8b and f), 1/4 for 0.4375 ML (Figure 8c and g), and eventually zero for 0.5 ML (Figure 8d and h). The minimum surface coverage to form rectangular domains was estimated to be 0.375 ML as shown in Figure 8a. Thus we infer the minimum number ratio of bridge CO over all available bridge sites is 1/2 for stabilizing the c(4 × 2) domain. As the surface coverage increased, i.e. filling the bridge vacancy, the arrangement of bridge CO vacancies along the [11̅0] direction of the Pt crystal lattice became irregular. Figure 9 describes the adsorption structure when the coverage was higher than 0.5 ML. As suggested in previous studies as the ″fault-line model″,60,61,10 a higher dose of CO to the c(√3 × 2)rect domain causes the compression of the c(√3 × 2)rect structure to form a 1 × 1 domain boundary, resolved at the individual molecular level (Figure 9a and b). The intermolecular interaction between neighboring atop CO molecules in the 1 × 1 domain appears repulsive, deduced from the intermolecular distance (3.0 Å) and the angle (close to 90°), significantly larger than the Pt−Pt distance (r0, 2.7 Å) and smaller than the Pt−Pt angle (120°), respectively. Consequently, the CO molecules at the 1 × 1 boundary may be tilted with respect to the surface normal direction by repulsive interactions, as suggested in an ESDIAD study.21 It is generally known that achieving coverage greater than 0.5 ML is difficult in a UHV environment, and we found that this is true especially when the coverage is gradually increased with annealing at room temperature. To increase the coverage up to approximately 0.55 ML, additional exposure to CO at low temperature is mandatory. Figure 9d shows a reduction in the

Figure 9. (a) Structure of a ″fault line″, a 1 × 1 boundary between c(√3 × 2)rect domains which starts to appear with θ > 0.5 ML (image size 10 × 10 nm2, VS = 0.1 V, IT = 0.5 nA), (b) zoomed-in view of the domain structure (image size 5 × 5 nm2, VS = 0.1 V, IT = 0.5 nA), (c) schematic model of (b), and (d) higher number density of the fault line boundary with increased coverage (∼0.56 ML) (image size 10 × 10 nm2, VS = 20 mV, IT = 0.5 nA).

c(√3 × 2)rect domain size as well as an increase in the relative area of 1 × 1 boundaries, which is consistent with the previously predicted c(√3 × 5)rect structure.7,60,61

4. CONCLUSION In conclusion, we have presented the adsorption behavior of CO on Pt(111) from a single-molecule limit to high coverage superstructures. An adsorption site of a single CO molecule was determined to be atop in atomically resolved STM images. STS and action spectroscopy showed the existence of DOS around 4.5 eV, which corresponds to the 2π* state of atop CO. The adsorption geometry and the formation of overlayer structures 16435



Article

(8) Baró, A. M.; Ibach, H. New Study of CO Adsorption at Low Temperature (90 K) on Pt (111) by EELS. J. Chem. Phys. 1979, 71, 4812−4816. (9) Tüshaus, M.; Schweizer, E.; Hollins, P.; Bradshaw, A. M. Yet Another Vibrational Study of the Adsorption System Pt{111}-CO. J. Electron Spectrosc. Relat. Phenom. 1987, 44, 305−316. (10) Schweizer, E.; Persson, B. N. J.; Tüshaus, M.; Hoge, D.; Bradshaw, A. M. The Potential Energy Surface, Vibrational Phase Relaxation and the Order-Disorder Transition in the Adsorption System Pt{111}-CO. Surf. Sci. 1989, 213, 49−89. (11) Poelsema, B.; Palmer, R. L.; Comsa, G. A Thermal He Scattering Study of CO Adsorption on Pt(111). Surf. Sci. 1984, 136, 1−14. (12) Persson, B. N. J.; Ryberg, R. Vibrational Line Shapes of LowFrequency Adsorbate Modes: CO on Pt(111). Phys. Rev. B 1989, 40, 10273−10281. (13) Lahee, A. M.; Toennies, J. P.; Wöll, C. Low Energy Adsorbate Vibrational Modes Observed with Inelastic Helium Atom Scattering: CO on Pt(111). Surf. Sci. 1986, 177, 371−388. (14) Graham, A. P. The Low-Frequency Vibrational Modes of c(4 × 2) CO on Pt(111). J. Chem. Phys. 1998, 109, 9583. (15) Hähner, G.; Toennies, J. P.; Wöll, C. Normal Modes of CO Adsorbed on Metal Surfaces. Appl. Phys. A: Mater. Sci. Process. 1990, 51, 208−215. (16) Engström, U.; Ryberg, R. Coupling to Dipole-Forbidden Modes: CO on Pt(111) Studied by Infrared Spectroscopy. Phys. Rev. Lett. 1997, 78, 1944. (17) Beckerle, J. D.; Casassa, M. P.; Cavanagh, R. R.; Heilweil, E. J.; Stephenson, J. C. Ultrafast Infrared Response of Adsorbates on Metal Surfaces: Vibrational Lifetime of CO/Pt(111). Phys. Rev. Lett. 1990, 64, 2090−2093. (18) Ryberg, R. Phase Transitions in a Chemisorbed Overlayer Studied by Infrared Spectroscopy: CO on Pt(111). Phys. Rev. B 1989, 40, 865−868. (19) Ogletree, D. F.; Van Hove, M. A.; Somorjai, G. A. LEED Intensity Analysis of the Structures of Clean Pt(111) and of CO Adsorbed on Pt(111) in the c(4 × 2) Arrangement. Surf. Sci. 1986, 173, 351−365. (20) Blackman, G. S.; Xu, M.-L.; Ogletree, D. F.; Van Hove, M. A.; Somorjai, G. A. Mix of Molecular Adsorption Sites Detected for Disordered CO on Pt(111) by Diffuse Low-Energy Electron Diffraction. Phys. Rev. Lett. 1988, 61, 2352−2355. (21) Kiskinova, M.; Szab, A.; Yates, J. T., Jr. Compressed CO Overlayers on Pt(111)  Evidence for Tilted CO Species at High Coverages by Digital ESDIAD. Surf. Sci. 1988, 205, 215−229. (22) Song, M.-B.; Yoshimi, K.; Ito, M. STM Observations of BridgeBonded CO on Pt(111) and Asymmetric On-Top CO on Pt(100). Chem. Phys. Lett. 1996, 263, 585−590. (23) Pedersen, M. Ø.; Bocquet, M.-L.; Sautet, P.; Lægsgaard, E.; Stensgaard, I.; Besenbacher, F. CO on Pt(111): Binding Site Assignment from the Interplay Between Measured and Calculated STM Images. Chem. Phys. Lett. 1999, 299, 403−409. (24) Stroscio, J. A.; Eigler, D. M. Atomic and Molecular Manipulation with the Scanning Tunneling Microscope. Science 1991, 254, 1319−1326. (25) Vestergaard, E. K.; Thostrup, P.; An, T.; Lægsgaard, E.; Stensgaard, I.; Hammer, B.; Besenbacher, F. Comment on “High Pressure Adsorbate Structures Studied by Scanning Tunneling Microscopy: CO on Pt(111) in Equilibrium with the Gas Phase”. Phys. Rev. Lett. 2002, 88, 259601. (26) Brako, R.; Šokčević, D. Adsorbate Interactions of CO Chemisorbed on Pt(111). Surf. Sci. 1998, 401, L388−L394. (27) German, E. D.; Sheintuch, M.; Kuznetsov, A. M. Diffusion on Metal Surfaces: Formalism and Application to CO Diffusion. J. Phys. Chem. C 2008, 112, 15510−15516. (28) Inoue, K.; Watanabe, K.; Matsumoto, Y. Instantaneous Vibrational Frequencies of Diffusing and Desorbing Adsorbates: CO/Pt(111). J. Chem. Phys. 2012, 137, 024704−024704−6.

were observed in molecular resolution from coverage of 0.1 ML to approximately 0.56 ML, to show the evolution of (√3 × √3)R30° islands and phase transition to a c(4 × 2) rectangular lattice, and further 1 × 1 boundary formation in the c(4 × 2) domain. CO adsorption on the bridge site in the range of our observations involves the rectangular unit structure, c(√3 × 2)rect, from the appearance of bridge CO at 0.1 ML by lowtemperature adsorption to the boundary structure in (√3 × √3)R30° islands and bridge-vacant c(4 × 2) domains. The bridge-vacant c(4 × 2) domain exhibits height modulation of atop CO by adjacent bridge CO, indicating the existence of interadsorbate interactions between the atop and bridge CO. As the origin of the interaction, the directions of charge transfer and the dipole according to the adsorption site were discussed. The present study may clear up the ambiguities in the overlayer structure of CO on Pt(111) as well as provide insight into interadsorbate interactions.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address †

Office of Society-Academia Collaboration for Innovation, Kyoto University, Gokasho, Uji, Kyoto 611-0011, Japan.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was partially supported by a Grant-in-Aid for Scientific Research on Priority Areas “Electron Transport through a Linked Molecule in Nano-scale” and a Grant-in-Aid for Scientific Research(S) “Single Molecule Spectroscopy using Probe Microscope” from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), Japan and partially supported by the Global COE Program (Chemistry Innovation through Cooperation of Science and Engineering), MEXT, Japan. The authors sincerely thank Michael Trenary and Jaehoon Jung for fruitful discussions and reading of the manuscript. H.J.Y. acknowledges the Junior Research Associate (JRA) program of RIKEN for support.

■

REFERENCES

(1) Ertl, G. Reactions at Solid Surfaces, 2nd ed.; Wiley: 2009. (2) Apai, G.; Baetzold, R. C.; Jupiter, P. J.; Viescas, A. J.; Lindau, I. Influence of Acceptor and Donor Adsorbates (CO, K, NH3) on Pt Surface Core-Level Shifts. Surf. Sci. 1983, 134, 122−134. (3) Dose, V.; Rogozik, J.; Bradshaw, A. M.; Prince, K. C. Inverse Photoemission from CO Co-Adsorbed with K on Pt(111). Surf. Sci. 1987, 179, 90−100. (4) Trenary, M.; Tang, S. L.; Simonson, R. J.; McFeely, F. R. An Angle-Resolved Photoemission Study of CO Chemisorbed on the Pt(111) Surface. Surf. Sci. 1983, 124, 555−562. (5) Anazawa, T.; Kinoshita, I.; Matsumoto, Y. Two-Photon Photoemission Study of CO/Pt(111). J. Electron Spectrosc. Relat. Phenom. 1998, 88−91, 585−590. (6) Bondino, F.; Comelli, G.; Esch, F.; Locatelli, A.; Baraldi, A.; Lizzit, S.; Paolucci, G.; Rosei, R. Structural Determination of Molecules Adsorbed in Different Sites by Means of Chemical Shift Photoelectron Diffraction: c(4 × 2)-CO on Pt(111). Surf. Sci. 2000, 459, L467−L474. (7) Steininger, H.; Lehwald, S.; Ibach, H. On the Adsorption of CO on Pt(111). Surf. Sci. 1982, 123, 264−282. 16436



Article

(29) Kinne, M.; Fuhrmann, T.; Whelan, C. M.; Zhu, J. F.; Pantförder, J.; Probst, M.; Held, G.; Denecke, R.; Steinrück, H.-P. Kinetic Parameters of CO Adsorbed on Pt(111) Studied by in Situ High Resolution X-ray Photoelectron Spectroscopy. J. Chem. Phys. 2002, 117, 10852. (30) Kuhl, D. E.; Lin, K. C.; Chung, C.; Luo, J. S.; Wang, H.; Tobin, R. G. Coverage-, Temperature-, and Frequency-Dependent Broadband Infrared Reflectance Change Induced by CO Adsorption on Pt(111). Chem. Phys. 1996, 205, 1−10. (31) McEwen, J. S.; Payne, S. H.; Kreuzer, H. J.; Kinne, M.; Denecke, R.; Steinruck, H. P. Adsorption and Desorption of CO on Pt(111): a Comprehensive Analysis. Surf. Sci. 2003, 545, 47−69. (32) Nekrylova, J. V.; Harrison, I. Single Hop Diffusion of CO from Bridge to Top Sites on Pt(111). J. Chem. Phys. 1994, 101, 1730−1733. (33) Reutt-Robey, J. E.; Doren, D. J.; Chabal, Y. J.; Christman, S. B. Microscopic CO Diffusion on a Pt(111) Surface by Time-Resolved Infrared Spectroscopy. Phys. Rev. Lett. 1988, 61, 2778−2781. (34) Poelsema, B.; Verheij, L. K.; Comsa, G. He-Scattering Investigation of CO Migration on Pt(111). Phys. Rev. Lett. 1982, 49, 1731−1735. (35) Yeo, Y. Y.; Vattuone, L.; King, D. A. Calorimetric Heats for CO and Oxygen Adsorption and for the Catalytic CO Oxidation Reaction on Pt(111). J. Chem. Phys. 1997, 106, 392−401. (36) Blyholder, G. Molecular Orbital View of Chemisorbed Carbon Monoxide. J. Phys. Chem. 1964, 68, 2772−2777. (37) Hammer, B.; Morikawa, Y.; Nørskov, J. K. CO Chemisorption at Metal Surfaces and Overlayers. Phys. Rev. Lett. 1996, 76, 2141− 2144. (38) Feibelman, P. J.; Hammer, B.; Nørskov, J. K.; Wagner, F.; Scheffler, M.; Stumpf, R.; Watwe, R.; Dumesic, J. The CO/Pt(111) Puzzle. J. Phys. Chem. B 2000, 105, 4018−4025. (39) Schimka, L.; Harl, J.; Stroppa, A.; Grüneis, A.; Marsman, M.; Mittendorfer, F.; Kresse, G. Accurate Surface and Adsorption Energies from Many-Body Perturbation Theory. Nat. Mater. 2010, 9, 741−744. (40) Wang, Y.; de Gironcoli, S.; Hush, N. S.; Reimers, J. R. Successful a Priori Modeling of CO Adsorption on Pt(111) Using Periodic Hybrid Density Functional Theory. J. Am. Chem. Soc. 2007, 129, 10402−10407. (41) Dabo, I.; Wieckowski, A.; Marzari, N. Vibrational Recognition of Adsorption Sites for CO on Platinum and Platinum−Ruthenium Surfaces. J. Am. Chem. Soc. 2007, 129, 11045−11052. (42) Mason, S. E.; Grinberg, I.; Rappe, A. M. First-Principles Extrapolation Method for Accurate CO Adsorption Energies on Metal Surfaces. Phys. Rev. B 2004, 69, 161401. (43) Cudok, A.; Froitzheim, H.; Schulze, M. Low-temperature Adsorption Kinetics of CO on Pt(111) Derived from Nonequilibrium Time-Resolved Electron-Energy-Loss Spectroscopy Measurements. Phys. Rev. B 1993, 47, 13682−13686. (44) Mieher, W. D.; Whitman, L. J.; Ho, W. A Time Resolved Electron Energy Loss Spectroscopy Study of CO on Pt(111): Adsorption Site Occupations Versus Coverage and Temperature. J. Chem. Phys. 1989, 91, 3228−3239. (45) Ertl, G.; Neumann, M.; Streit, K. M. Chemisorption of CO on the Pt(111) Surface. Surf. Sci. 1977, 64, 393−410. (46) Malik, I. J.; Trenary, M. Infrared Reflection-absorption Study of the Adsorbate-Substrate Stretch of CO on Pt(111). Surf. Sci. Lett. 1989, 214, L237−L245. (47) Hayden, B. E.; Bradshaw, A. M. The Adsorption of CO on Pt(111) Studied by Infrared Reflection-Absortion Spectroscopy. Surf. Sci. 1983, 125, 787−802. (48) Froitzheim, H.; Hopster, H.; Ibach, H.; Lehwald, S. Adsorption Sites of CO on Pt (111). Appl. Phys. 1977, 13, 147−151. (49) Wintterlin, J.; Völkening, S.; Janssens, T. V. W.; Zambelli, T.; Ertl, G. Atomic and Macroscopic Reaction Rates of a SurfaceCatalyzed Reaction. Science 1997, 278, 1931−1934. (50) Doering, M.; Buisset, J.; Rust, H.-P.; Briner, B. G.; Bradshaw, A. M. Adsorption Site Determination with Scanning Tunnelling Microscopy. Faraday Discuss. 1996, 105, 163−175.

(51) Bartels, L.; Meyer, G.; Rieder, K.-H.; Velic, D.; Knoesel, E.; Hotzel, A.; Wolf, M.; Ertl, G. Dynamics of Electron-Induced Manipulation of Individual CO Molecules on Cu(111). Phys. Rev. Lett. 1998, 80, 2004−2007. (52) Komeda, T.; Kim, Y.; Kawai, M.; Persson, B. N. J.; Ueba, H. Lateral Hopping of Molecules Induced by Excitation of Internal Vibration Mode. Science 2002, 295, 2055−2058. (53) Nekrylova, J. V.; Harrison, I. Site Resolved Adsorption Dynamics of CO on Pt(111). Chem. Phys. 1996, 205, 37−46. (54) Sette, F.; Stöhr, J.; Kollin, E. B.; Dwyer, D. J.; Gland, J. L.; Robbins, J. L.; Johnson, A. L. Na-Induced Bonding and Bond-Length Changes for CO on Pt(111): A Near-Edge X-ray-Absorption FineStructure Study. Phys. Rev. Lett. 1985, 54, 935−938. (55) Stroscio, J. A.; Kaiser, W. J. Scanning Tunneling Microscopy; Academic Press: 1993; Vol. 27. (56) Norton, P. R.; Goodale, J. W.; Selkirk, E. B. Adsorption of CO on Pt(111) Studied by Photoemission, Thermal Desorption Spectroscopy and High Resolution Dynamic Measurements of Work Function. Surf. Sci. 1979, 83, 189−227. (57) Poelsema, B.; Palmer, R. L.; Comsa, G. Helium Scattering and Work Function Investigation of Co Adsorption on Pt(111) and Vicinal Surfaces. Surf. Sci. 1982, 123, 152−164. (58) Föhlisch, A.; Nyberg, M.; Hasselström, J.; Karis, O.; Pettersson, L. G. M.; Nilsson, A. How Carbon Monoxide Adsorbs in Different Sites. Phys. Rev. Lett. 2000, 85, 3309−3312. (59) Deshlahra, P.; Conway, J.; Wolf, E. E.; Schneider, W. F. Influence of Dipole−Dipole Interactions on Coverage-Dependent Adsorption: CO and NO on Pt(111). Langmuir 2012, 28, 8408−8417. (60) Persson, B. N. J.; Tüshaus, M.; Bradshaw, A. M. On the Nature of Dense CO Adlayers. J. Chem. Phys. 1990, 92, 5034−5046. (61) Biberrian, J. P.; Van Hove, M. A. A New Model for CO Ordering at High Coverages on Low Index Metal Surfaces: A Correlation between LEED, HREELS and IRS: II. CO Adsorbed on Fcc (111) and Hcp (0001) Surfaces. Surf. Sci. 1984, 138, 361−389.

16437


STM Investigation of CO Ordering on Pt(111): From an Isolated

Recommend Documents