Search for adsorption potential energy minima of NO on Pt(9 9 7) at 11 K

We observed four kinds of adsorbed NO molecules on Pt(9 9 7) at 11 K using infrared reflection absorption spectroscopy (IRAS). The peaks at 1690, 1484 and 1615 cm⁻¹ are assigned to the N-O stretching modes of the on-top site and the hollow site on the terrace and the bridge site at the step, respectively. The 1385 cm⁻¹ peak is observed below ∼70 K. We assign the 1385 cm⁻¹ peak to the hollow site of the (1 1 1) microfacet at the step or the lower-terrace hollow site nearest to the step. By heating, site-to-site hopping to the more stable site occurs and the relative stability of four adsorption sites can be determined.     

The dissimilar twins - a comparative, site-selective in situ study of CO adsorption and desorption on Pt(3 2 2) and Pt(3 5 5)

The adsorption and desorption of CO on stepped Pt(3 2 2) = Pt(S)-[5(1 1 1) × (1 0 0)] and Pt(3 5 5) = Pt(S)-[5(1 1 1) × (1 1 1)] were investigated using in situ high-resolution X-ray photoelectron spectroscopy at BESSY II, which allows to clearly distinguish between different step and terrace adsorption sites. For the two surfaces, with the same nominal terrace width of five atomic rows, but different step orientation, significant differences are observed. While for Pt(3 5 5) CO adsorption at steps only occurs at on-top sites, on Pt(3 2 2) both step on-top and bridge sites are occupied, albeit with a significantly lower coverage (0.07 vs. 0.13 ML at 200 K). On both surfaces terrace sites are only occupied when the step sites are almost saturated confirming the enhanced binding energy at step sites. CO adsorbed at the (1 1 1) steps on Pt(3 5 5) is more strongly bound than on the (1 0 0) steps on Pt(3 2 2), which is attributed to the different electronic and geometric structure of the steps. The relative occupation of terrace and step sites at a given coverage remains the same between 120 and 290 K on Pt(3 5 5) K, but shows major changes on Pt(3 2 2), between step on-top and bridge sites as well as terrace on-top and bridge sites. On Pt(3 5 5) a smaller CO terrace coverage is found (0.36 vs. 0.40 ML on Pt(3 2 2) at 200 K), mainly due to the lower occupation of terrace bridge sites. For Pt(3 2 2), an ordered adsorbate phase is deduced from a c(4 × 2)-like LEED pattern, which indicates adsorbate order beyond the extension of a single terrace. A model for this structure is proposed.     

Comparison of the adsorption of N2 on Ru(1 0 9) and Ru(0 0 1) - A detailed look at the role of atomic step and terrace sites

We present a direct side-by-side comparison of the adsorption and desorption of nitrogen on the atomically-stepped Ru(1 0 9) surface and the atomically-flat Ru(0 0 1) surface. Both infrared reflection absorption spectroscopy (IRAS) and temperature programmed desorption (TPD) are employed in this study, along with density functional theory (DFT). We find that the chemisorptive terminal binding of N₂ is stronger on the atomic step sites than on the terrace sites of Ru(1 0 9) as indicated by TPD and by a reduction of the singleton vibrational frequency, ν(N₂), by ∼9 cm⁻¹, comparing steps to terraces. In addition, we find that metal-metal compression effects on the terrace sites of Ru(1 0 9) cause stronger binding of N₂ than found on the Ru(0 0 1) surface, as indicated by a reduction of the terrace-N₂ singleton vibrational frequency by ∼11 cm⁻¹ when compared to the singleton N₂ mode on Ru(0 0 1). These spectroscopic results, comparing compressed terrace sites to Ru(0 0 1) sites and confirmed by TPD and DFT, indicate that N₂ bonds primarily as a σ-donor to Ru. Using equimolar ¹⁵N₂ and ¹⁴N₂, it is found that dynamic dipole coupling effects present at higher N₂ coverages may be partially eliminated by isotopically detuning neighbor oscillators. These experiments, considered together, indicate that the order of the bonding strength for terminal-N₂ on Ru is: atomic steps > atomic terraces > Ru(0 0 1). DFT calculations also show that 4-fold coordinated N₂ may be stabilized in several structures on the double-atom wide steps of Ru(1 0 9) and that this form of bonding produces substantial decreases in the N₂ vibrational frequency and increases in the binding energy, compared to terminally-bound N₂. These highly coordinated N₂ species are not observed by IRAS.     

The structures of physisorbed and chemisorbed formic acid on Si(1 1 1)-7 × 7

The reaction of formic acid on Si(1 1 1)-7 × 7 was investigated using X-ray photoelectron spectroscopy (XPS), ultraviolet photoelectron spectroscopy (UPS) and high-resolution electron energy loss spectroscopy (HREELS). The hydroxyl and carbonyl O 1s core levels of chemisorbed formic acid display chemical shifts of 2.4 and 0.2 eV respectively, compared with those of physisorbed molecules. The HREELS spectra of chemisorbed formic acid show the absence of stretching and bending modes of the O-H bond, the appearance of Si-H (2089 cm⁻¹) and the Si-O (680 cm⁻¹) stretching modes and the retained stretching mode of CO at 1703 cm⁻¹. Our results clearly suggest that formic acid dissociates to form monodentate formate species and H-atom on the adatom-rest atom pair of Si(1 1 1)-7 × 7.     

Sum frequency generation and density functional studies of CO-H interaction and hydrogen bulk dissolution on Pd(1 1 1)

CO-H interaction and H bulk dissolution on Pd(1 1 1) were studied by sum frequency generation (SFG) vibrational spectroscopy and density functional theory (DFT). The theoretical findings are particularly important to rationalize the experimentally observed mutual site blocking of CO and H and the effect of H dissolution on coadsorbate structures. Dissociative hydrogen adsorption on CO-precovered Pd(1 1 1) is impeded due to an activation barrier of ∼2.5 eV for a CO coverage of 0.75 ML, an effect which is maintained down to 0.33 ML CO. Preadsorbed hydrogen prevented CO adsorption at 100 K, while hydrogen was replaced from the surface by CO above 125 K. The temperature-dependent site blocking of hydrogen originates from the onset of hydrogen diffusion into the Pd bulk around 125 K, as shown by SFG and theoretical calculations using various approaches. When Pd(1 1 1) was exposed to 1:1 CO/H₂ mixtures at 100 K, on-top CO was absent in the SFG spectra although hydrogen occupies only threefold hollow sites on Pd(1 1 1). DFT attributes the absence of on-top CO to H atoms diffusing between hollow sites via bridge sites, thereby destabilizing neighboring on-top CO molecules. According to the calculations, the stretching frequency of bridge-bonded CO with a neighboring bridge-bonded hydrogen atom is redshifted by 16 cm⁻¹ when compared to bridging CO on the clean surface. Implications of the observed effects on hydrogenation reactions are discussed and compared to the C₂H₄-H coadsorption system.     

Collision-induced migration of CO on Pt(9 9 7) surface

Lateral displacement of adsorbates induced by collisions with energy-controlled rare gas atoms was examined in an ultra high vacuum chamber using Fourier-transform infrared (FTIR) spectroscopy and a supersonic molecular beam apparatus. A stepped Pt(9 9 7) surface was exposed to CO molecules and subsequently to energy-controlled Ne or Ar atoms. There was no change in the CO stretching mode region of the FTIR spectrum of the Pt(9 9 7) surface after Ne atoms having an average translational energy of 0.23 eV were collided with it. However, when Ne atoms having an average translational energy of 0.56 eV were collided with the surface, the intensity of the peak assigned to the CO stretching mode at terrace sites decreased, while that at step sites increased with increasing the exposure to the Ne atoms. This is the demonstration of collision-induced migration, showing that CO molecules adsorbed at the terrace sites migrate laterally to the step sites upon collision with high-energy Ne atoms. In addition, the experimental results demonstrate the existence of an additional energy barrier for jumps across the steps. This investigation demonstrates an advantage of using a molecular beam for studying adsorbate migration.     

Infrared reflection absorption spectral study for CO adsorption on Pd/Pt(1 1 1) bimetallic surfaces

Infrared reflection absorption spectroscopy (IRRAS) was used to investigate carbon monoxide (CO) adsorption on 0.15 nm-thick-0.6 nm-thick Pd-deposited Pt(1 1 1) bimetallic surfaces: Pd_x/Pt(1 1 1) (where x is the Pd thickness in nanometers) fabricated using molecular beam epitaxial method at substrate temperatures of 343 K, 473 K, and 673 K. Reflection high-energy electron diffraction (RHEED) measurements for Pd_0.15-0.6 nm/Pt(1 1 1) surfaces fabricated at 343 K showed that Pd grows epitaxially on a clean Pt(1 1 1), having an almost identical lattice constant of Pt(1 1 1). The 1.0 L CO exposure to the clean Pt(1 1 1) at room temperature yielded linearly bonded and bridge-bonded CO-Pt bands at 2093 and 1855 cm⁻¹. The CO-Pt band intensities for the CO-exposed Pd_x/Pt(1 1 1) surfaces decreased with increasing Pd thickness. For Pd_0.3 nm/Pt(1 1 1) deposited at 343 K, the 1933 cm⁻¹ band caused by bridge-bonded CO-Pd enhanced the spectral intensity. The linear-bonded CO-Pt band (2090 cm⁻¹) almost disappeared and the bridge-bonded CO-Pd band dominated the spectra for Pd_0.6 nm/Pt(1 1 1). With increasing substrate temperature during the Pd depositions, the relative band intensities of the CO-Pt/CO-Pd increased. For the Pd_0.3 nm/Pt(1 1 1) deposited at 673 K, the linear-bonded CO-Pt and bridge-bonded CO-Pd bands are located respectively at 2071 and 1928 cm⁻¹. The temperature-programmed desorption (TPD) spectrum for the 673 K-deposited Pd_0.3 nm/Pt(1 1 1) showed that a desorption signal for the adsorbed CO on the Pt sites decreased in intensity and shifted ca. 20 K to a lower temperature than those for the clean Pt(1 1 1). We discuss the CO adsorption behavior on well-defined Pd-deposited Pt(1 1 1) bimetallic surfaces.     

An IRAS study of CO bonding on Sn/Pt(1 1 1) surface alloys at maximal pressures of 10 Torr

The adsorption of CO on Pt(1 1 1), (2 × 2) and (√3 × √3)R30° Sn/Pt(1 1 1) surface alloys has been studied using temperature programmed desorption (TPD), low energy electron diffraction (LEED) and infrared reflection adsorption spectroscopy (IRAS). The presence of Sn in the surface layer of Pt(1 1 1) reduces the binding energy of CO by a few kcal/mol. IRAS data show two C-O stretching frequencies, ∼2100 and ∼1860 cm⁻¹, corresponding to atop and bridge bonded species, respectively. Bridge bonded stretching frequencies are only observed for Pt(1 1 1) and (2 × 2) Sn/Pt(1 1 1) alloy surfaces. A slight coverage dependence of the vibrational frequencies is observed for the three surfaces. High pressure IRAS experiments over a broad temperature range show no indication of bridge bonded CO on any of the three surfaces. Direct CO adsorption on Sn sites is not observed over the measured temperature and pressure ranges.     

CO interaction with Au atoms adsorbed on terrace, edge and corner sites of the MgO(1 0 0) surface. Electronic structure and vibrational analysis from DFT

The interaction of CO with Au atoms adsorbed on terrace and low-coordinates sites (edge and corner) of the MgO(1 0 0) surface was studied using the density functional theory (DFT) in combination with embedded cluster models. Surface anionic (O²⁻) and neutral oxygen vacancy (F_s) sites were considered. In all the cases, the CO stretching frequencies are shifted with respect to free CO with values between −232 and −358 cm⁻¹. In particular, the values for Au on F_s at edge and corner are shifted to higher stretching frequencies by 100 and 59 cm⁻¹, respectively, with respect to the value on a perfect MgO(1 0 0) surface. This result is in agreement with recent scanning tunneling microscopy and infrared spectroscopy experiments where a corresponding shift of 70 cm⁻¹ was observed by comparing the measurements on perfect and O-deficient MgO(1 0 0) surfaces. However, these results are different than expected because Au atoms on F_s centers are negatively charged and, therefore, according to the generally accepted scheme the CO frequency should be red-shifted with respect to the adsorption on anionic five-coordinated site where the Au atom is essentially neutral. The following picture emerges from the present results: the single occupied HOMO(α) of Au atom on F_s at low-coordinated sites consists in two lobes extended sideward the Au atom. For symmetry reasons, this MO overlaps efficiently with the 2π^∗ MO of CO. This bonding contribution to the Au-CO link is counteracted by a Pauli repulsion between the 5σ MO of CO and more internal orbitals (the HOMO-1(α) and the HOMO(β)) centered on Au. In consequence, CO is forced to vibrate against a region with a high electron density. This is the so-called “wall effect” which by itself contributes to higher CO frequency values.     

Signatures of epitaxial graphene grown on Si-terminated 6H-SiC (0 0 0 1)

Epitaxial graphene layers thermally grown on Si-terminated 6H-SiC (0 0 0 1) have been probed using Auger electron spectroscopy, Raman microspectroscopy, and scanning tunneling microscopy (STM). The average multilayer graphene thickness is determined by attenuation of the Si (L₂₃VV) and C (KVV) Auger electron signals. Systematic changes in the Raman spectra are observed as the film thickness increases from one to three layers. The most striking observation is a large increase in the intensity of the Raman 2D-band (overtone of the D-band and also known as the G′-band) for samples with a mean thickness of more than ∼1.5 graphene layers. Correlating this information with STM images, we show that the first graphene layer imaged by STM produces very little 2D intensity, but the second imaged layer shows a single-Lorentzian 2D peak near 2750 cm⁻¹, similar to spectra acquired from single-layer micromechanically cleaved graphene (CG). The 4-10 cm⁻¹ higher frequency shift of the G peak relative to CG can be associated with charge exchange with the underlying SiC substrate and the formation of finite size domains of graphene. The much greater (41-50 cm⁻¹) blue shift observed for the 2D-band may be correlated with these domains and compressive strain.     

Molecular dynamics of nonadiabatic processes at surfaces: Chemisorption of H/Al(1 1 1)

Time-dependent density functional theory for the electronic degrees of freedom has been combined with Ehrenfest dynamics for the nuclei to simulate electron-hole pair excitation due to electronic friction during the chemisorption of hydrogen atoms on an Al(1 1 1) surface. The H-atoms are assumed to be spin-unpolarized in the simulations. Trajectories starting with a hydrogen atom at rest above either the on-top or the fcc-hollow site evolve in qualitatively very different ways: at the fcc-hollow position the H-atom acquires sufficient kinetic energy in the chemisorption well to penetrate into the Al-substrate, thereby increasing the coupling of the motion of the H-atom to the substrate electrons. The electronic excitation spectra, however, are roughly characterized by an exponential decay with similar fictitious temperature parameters of the order of 10³ K for both kinds of trajectories. The energy dissipation into electron-hole pairs and the nonadiabatic contribution to the force acting on the hydrogen atom have been calculated along the trajectories.     

The high-resolution infrared spectrum of isoxazole vapor between 800 and 1700 cm

The Fourier transform gas-phase IR spectrum of isoxazole, C₃H₃NO, between 550 and 1700 cm⁻¹ was measured with a resolution of ca. 0.003 cm⁻¹. Ten fundamental bands in the region 800-1700 cm⁻¹ have been analyzed by the Watson Hamiltonian model to yield upper state spectroscopic constants. A number of local resonances have been identified in the bands and explained qualitatively, and the unobserved ν₁₄(A″) fundamental band has been located at 897.5(5) cm⁻¹ from its perturbation effects on the neighboring fundamentals.     

Adsorption of carbon monoxide on Pd(3 1 1) and (2 1 1) surfaces

Adsorption of carbon monoxide on Pd(3 1 1) and (2 1 1) stepped surfaces has been investigated by the extended London-Eyring-Polyani-Sato (LEPS) method constructed using a 5-parameter Morse potential. The calculated results show that there exist common characteristics of CO adsorption on the two surfaces. At low coverage, CO occupies threefold hollow site of the (1 1 1) terrace and is tilted with respect to the surface normal. Among the threefold hollow sites on the (1 1 1) terrace, the nearer the site is to the step, the greater is the influence of the step. The twofold bridge site on the (1 0 0) step is also a stable adsorption site at high coverage. Because of the different lengths of the (1 1 1) terraces, the (3 1 1) and (2 1 1) stepped surfaces have different characteristics. A number of new sites are exposed on the boundary regions, including the fourfold hollow site (H₄) of the (3 1 1) surface and the fivefold hollow site (H₅) of the (2 1 1) surface. At high coverage, CO resides in the H₅ site of the (2 1 1) surface, but the H₄ site of the (3 1 1) surface is not a stable adsorption site. This study further shows that the on-top site on the (1 0 0) step of Pd(3 1 1) is a stable adsorption site, but the same type of site on Pd(2 1 1) is not.     

Infrared reflection absorption study of carbon monoxide adsorption on Cu(1 0 0)-c(2 × 2)-Pd surfaces formed by palladium vacuum-depositions at various temperatures

Using infrared reflection absorption spectroscopy (IRRAS) and temperature programmed desorption (TPD), we investigated carbon monoxide (CO) adsorption and desorption behaviors on atomic checkerboard structures of Cu and Pd formed by Pd vacuum deposition at various temperatures of Cu(1 0 0). The 0.15-nm-thick Pd deposition onto a clean Cu(1 0 0) surface at room temperature (RT) showed a clear c(2 × 2) low-energy electron diffraction (LEED) pattern, i.e. Cu(1 0 0)-c(2 × 2)-Pd. The RT-CO exposure to the c(2 × 2) surfaces resulted in IRRAS absorption caused by CO adsorbed on the on-top sites of Pd. The LEED patterns of the Pd-deposited Cu(1 0 0) at higher substrate temperatures revealed less-contrasted c(2 × 2) patterns. The IRRAS intensities of the linearly bonded CO bands on 373-K-, 473-K-, and 673-K-deposited c(2 × 2) surfaces are, respectively, 25%, 22%, and 10% less intense than those on the RT-deposited surface, indicating that Pd coverages at the outermost c(2 × 2) surfaces decrease with increasing deposition temperature. In the initial stage of the 90-K-CO exposure to the RT surface, the band attributable to CO bonded to the Pd emerged at 2067 cm⁻¹ and shifted to higher frequencies with increasing CO exposure. At saturation coverage, the band was located at 2093 cm⁻¹. In contrast, two distinct bands around 2090 cm⁻¹ were apparent on the spectrum of the 473-K-deposited surface: the CO saturation spectrum was dominated by an apparent single absorption at 2090 cm⁻¹ for the 673-K-deposited surface. The TPD spectra of the surfaces showed peaks at around 200 and 300 K, which were ascribable respectively to Cu-CO and Pd-CO. Taking into account the TPD and IRRAS results, we discuss the adsorption-desorption behaviors of CO on the ordered checkerboard structures.     

Growth and thermal evolution of submonolayer Pt films on Ru(0 0 0 1) studied by STM

The growth of submonolayer Pt on Ru(0 0 0 1) has been studied with scanning tunneling microscopy. We focus on the island evolution depending on Pt coverage θ_Pt, growth temperature T_G and post-growth annealing temperature T_A. Dendritic trigonal Pt islands with atomically rough borders are observed at room temperature and moderate deposition rates of about 5 × 10⁻⁴ ML/s. Two types of orientation, rotated by 180° and strongly influenced by minute amounts of oxygen are observed which is ascribed to nucleation starting at either hcp or fcc hollow sites. The preference for fcc sites changes to hcp in the presence of about one percent of oxygen. At lower growth temperatures Pt islands show a more fractal shape. Generally, atomically rough island borders smooth down at elevated growth temperatures higher than 300 K, or equivalent annealing temperatures. Dendritic Pt islands, for example, transform into compact, almost hexagonal islands, indicating similar step energies of A- and B-type of steps. Depending on the Pt coverage the thermal evolution differs somewhat: While regular islands on Ru(0 0 0 1) are formed at low coverages, vacancy islands are observed close to completion of the Pt layer.     

The catalytic reduction of NO by H2 on Ru(0 0 0 1): Observation of NHads species

Adsorption of NO and the reaction between NO and H₂ were investigated on the Ru(0 0 0 1) surface by X-ray photoelectron spectroscopy (XPS). Surface composition was measured after NO adsorption and after the selective catalytic reduction of nitric oxide with hydrogen in steady-state conditions at 320 K and 390 K in a 30:1 mixture of H₂ and NO (total pressure = 10⁻⁴ mbar). After steady-state NO reduction, molecularly adsorbed NO in both the linear on-top and threefold coordinations, NH_ads and N_ads species were identified by XPS. The coverage of the NH_ads and N_ads species was higher after the reaction at 390 K than the corresponding values at 320 K. Strong destabilisation of N_ads by O_ads was detected. A possible reaction mechanism is discussed.     

Self-regulated Ni cluster formation on the TiO2(1 1 0) terrace studied using scanning tunneling microscopy

We have studied the growth mode and morphology of Ni clusters on a TiO₂(1 1 0) surface with a wide terrace using scanning tunneling microscopy (STM) at a low coverage (less than 3 atoms nm⁻²). The Ni clusters are formed on the terrace at the low coverage of 0.2 atoms nm⁻². Their average dimensions are constant in three directions up to 1 atoms nm⁻². The Ni clusters have an oval shape with average sizes of 1.8 nm (along [0 0 1]) × 1.4 nm (along (in the [1 1 0] directions). Above the coverage of 1.0 atoms nm⁻², an increase in the cluster height occurs, retaining an almost constant lateral size. It is proposed that the interaction of the Ni cluster and the support surface regulates the Ni cluster size.     

Carbon monoxide adsorption on Co deposited Pt(1 0 0)-hex: IRRAS and LEED investigations

Infrared reflection absorption spectroscopy (IRRAS) was used to investigate carbon monoxide (CO) adsorption on Pt(1 0 0) surfaces deposited with Co layers with different thicknesses. Pt(1 0 0) surfaces cleaned in ultrahigh vacuum showed surface reconstruction, i.e., Pt(1 0 0)-hex: two absorption bands ascribable to adsorbed CO on the 1 × 1 surface and hex domains emerge at 2086 and 2074 cm⁻¹, respectively, after 1.0 L CO exposure. Deposition of a 0.3-nm-thick-Co layer on Pt(1 0 0)-hex at 333 K changes the low-energy electron diffraction (LEED) pattern from hex to p(1 × 1), indicating that the deposited Co lifts the reconstruction. The IRRAS spectrum for 1.0-L-CO-exposed Co_0.3 nm/Pt(1 0 0)-hex fabricated at 333 K yields a single absorption band at 2059 cm⁻¹. For Co_0.3 nm/Pt(1 0 0)-hex fabricated at 693 K, the LEED pattern shows a less-contrasted hex and the pattern remains nearly unchanged even after CO exposure of 11 L, although only 1.0 L CO exposure to Pt(1 0 0)-hex lifts the surface reconstruction. A Co_0.3 nm/Pt(1 0 0)-hex surface fabricated at 753 K exhibits an absorption band at 2077 cm⁻¹, which is considered to originate from CO adsorbed on the Pt-enriched surface alloy. Co_0.3 nm/Pt(1 0 0)-hex surfaces fabricated above 773 K show a clear hex-reconstructed LEED pattern, and the frequencies of the adsorbed CO bands are comparable to those of Pt(1 0 0)-hex, indicating that the deposited Co atoms are diffused near the surface region. The outermost surface of the 3.0-nm-thick-Co-deposited Pt(1 0 0)-hex is composed of Pt-Co alloy domains even at a deposition temperature of 873 K. Based on the LEED and IRRAS results, the outermost surface structures of Co_x/Pt(1 0 0)-hex are discussed.     

The surface structure of Fe3O4(1 1 1) films as studied by CO adsorption

We have studied adsorption of CO on Fe₃O₄(1 1 1) films grown on a Pt(1 1 1) substrate by temperature programmed desorption (TPD), infrared reflection absorption spectroscopy (IRAS) and high resolution electron energy loss spectroscopy (HREELS). Three adsorption states are observed, from which CO desorbs at ∼110, 180, and 230 K. CO adsorbed in these states exhibits stretching frequencies at ∼2115-2140, 2080 and 2207 cm⁻¹, respectively. The adsorption results are discussed in terms of different structural models previously reported. We suggest that the Fe₃O₄(1 1 1) surface is terminated by 1/2 ML of iron, with an outermost 1/4 ML consisting of octahedral Fe²⁺ cations situated above an 1/4 ML of tetrahedral Fe³⁺ ions, in agreement with previous theoretical calculations. The most strongly bound CO is assigned to adsorption to Fe³⁺ cations present on the step edges.     

Probing the properties of the (1 1 1) and (1 0 0) surfaces of LaB6 through infrared spectroscopy of adsorbed CO

The adsorption of carbon monoxide on the LaB₆(1 0 0) and LaB₆(1 1 1) surfaces was studied experimentally with the techniques of reflection absorption infrared spectroscopy and X-ray photoelectron spectroscopy. The interaction of CO with the two surfaces was also studied with density functional theory. Both surfaces adsorb CO molecularly at low temperatures but in markedly different forms. On the LaB₆(1 1 1) surface CO initially adsorbs at 90 K in a form that yields a CO stretching mode at 1502-1512 cm⁻¹. With gentle annealing to 120 K, the CO switches to a bonding environment characterized by multiple CO stretch values from 1980 to 2080 cm⁻¹, assigned to one, two, or three CO molecules terminally bonded to the B atoms of a triangular B₃ unit at the (1 1 1) surface. In contrast, on the LaB₆(1 0 0) surface only a single CO stretch is observed at 2094 cm⁻¹, which is assigned to an atop CO molecule bonded to a La atom. The maximum intensity of the CO stretch vibration on the (1 0 0) surface is higher than on the (1 1 1) surface by a factor of 5. This difference is related to the different orientations of the CO molecules on the two surfaces and to reduced screening of the CO dynamic dipole moment on the (1 0 0) surface, where the bonding occurs further from the surface plane. On LaB₆(1 0 0), XPS measurements indicate that CO dissociates on the surface at temperatures above 400 K.