Paul Pudney

Infrared spectroscopic comparison of the chemisorbed species from ethene, propene, but-1-ene and cis- and trans-but-2-ene on Pt(111) and on a platinum/silica catalyst

The chemisorption of ethene, propene, but-1-ene, cis- and trans-but-2-ene and but-2-yne has been studied on a Pt(111) surface using reflection absorption infrared spectroscopy (RAIRS). The results have been compared with transmission infrared spectra of the alkenes adsorbed on a finely divided impregnated Pt/SiO2 catalyst.On the Pt(111) surface ethene, propene and but-1-ene are adsorbed in the corresponding n-alkylidyne form CH3(CH2)nCPt3(n= 0, 1, 2) at room temperature. The cis- and trans-but-2-enes are adsorbed in the form of but-2-yne bonded to the surface in a di-σ/π fashion, as was confirmed by adsorption of but-2-yne itself.The room-temperature spectra of ethene, propene and but-1-ene on Pt/SiO2 all exhibited prominent absorptions from the n-alkylidyne, but other adsorbed species also contributed to the spectra, considered to be predominantly the appropriate di-σ species PtCH2CH(R)Pt (R = H, CH3, C2H5) together with smaller amounts of the relevant π-complexes. In the case of the cis- and trans-but-2-enes, virtually identical spectra were obtained on Pt/SiO2 but with many of the same prominent features as from but-1-ene. Clearly the finely divided catalyst had sites which led to double-bond isomerization. A smaller proportion of di-σ/π bonded but-2-yne was also present.The infrared spectra proved to be very effective at delineating the similarities and differences between the species chemisorbed on Pt(111) and on the Pt/SiO2 catalyst.

Structural studies of platinum/ZSM-5 catalysts

Abstract Pt/ZSM-5 catalysts have been prepared, characterised by extended X-ray absorption fine structure (EXAFS), and electron microscopy, and their activity in ethane hydrogenolysis has been measured. Calcination temperature is important in determining the size of the platinum particles formed. For a 0.5% Pt/H-ZSM-5 catalyst calcined at 720 K and reduced at 620 K or 790 K, small particles within the zeolite framework are observed, with a nearest neighbour coordination number of ca. 6 and an average diameter of about 8 A. Calcination of a similar catalyst at 790 K resulted in larger metal particles, with the average diameter in the range 12–15 A, estimates from EXAFS and electron microscopy being in agreement within experimental error. Contractions of up to 4% in the average PtPt interatomic distance were observed, but there is clear evidence that the larger particles at least retain the face-centred cubic structure. EXAFS indicates the presence of Pt0 bonding as a result of coordination to the zeolite framework, with bond lengths of 1.92−2.03 A. These catalysts have higher specific activity in ethane hydrogenolysis and alkane aromatization than materials where the platinum is in the form of large particles, ( d > 100 A), at the external surface of the zeolite, and reasons for this are discussed.

A simple molecular beam system for surface reactivity studies

A simple design for a thermal molecular beam system is described, which can be added as a bolt‐on to a conventional vacuum system. It provides a beam of well‐defined spatial distribution, which can be heated or cooled, and used for a variety of applications. These include the accurate and rapid determination of sticking probabilities and their coverage dependence, the gas temperature dependence of sticking, and reactive scattering measurements. The use of the beam is exemplified for the dissociative sticking of oxygen on Cu(110) and the reaction of a beam of methanol with a predeposited patch of oxygen. The initial O2 sticking probability is determined to be 0.21 (±0.01) independent of substrate temperature (300–600 K). Methanol is dehydrogenated by preadsorbed O atoms, but by different mechanisms at high and low substrate temperatures.

Activated dissociation of oxygen on Cu(110)

Abstract Oxygen adsorption on Cu(110) has been studied using a thermal molecular-beam system. The sticking probability (S) is independent of substrate temperature above 220 K but depends strongly on beam temperature, showing the adsorption to be directly activated from the gas phase by 3kj mol −1 . However, at low substrate temperatures there is also evidence of the involvement of an accommodated, molecularly chemisorbed species in the dissociation process. The adsorption scales with total energy of the incoming molecule and so eitheir the potential surface appears isotropic in space, or there is a rather tortuos entrance channel to reach the barrier, which exists between unionised (physisorbed) oxygen and O− 2 at the surface.

Molecular-beam studies of methanol partial oxidation on Cu(110)

A thermal molecular-beam system has been used to examine the oxidative dehydrogenation of methanol on the Cu(110) surface. The initial sticking probability for oxygen is 0.21 (±0.01) at room temperature and shows a near linear dependence of sticking coefficient on atomic coverage due to an island growth mechanism with dissociation on an oxygen dilute inter-island phase. Beam temperature variations show this adsorption to be activated. The reaction of methanol in the beam with a predeposited patch of oxygen depends strongly on surface temperature and oxygen coverage. There is a change in stoichiometry in the reaction from 2CH3OH + O(a)→ 2H2CO + H2+ H2O at 330 K to CH3OH + O(a)→ H2CO + H2O above ca. 450 K. Oxygen promotes the methanol adsorption and reaction at low coverages, but shows poisoning effects at half a monolayer [saturation, p(2 × 1) structure] where the rate of reaction is very much reduced, and there is an induction time before products are seen in the gas phase (ca. 30 min at 333 K under these conditions). This is explained by stabilisation of the methoxy species when the adjacent (110) trough sites are blocked by either adsorbed oxygen or hydroxyl groups; a kinetic model is being developed to describe these complex kinetics, based on a slow production of vacant sites in these (110) troughs. This is shown to describe the kinetics quite well in a semi-quantitative manner. Use of CD3OD in the beam shows a marked isotope effect, whereas CH3OD does not, again indicating that it is methoxy decomposition which limits the product evolution rate.

In situ studies of supported rhodium catalysts

Rhodium catalysts have been prepared, supported on γ-alumina, vanadium(III) oxide, chromia and molybdena. The activity and selectivity of these materials in the conversion of synthesis gas, (H2/CO: 1/1, 523 K and 5 bar pressure), to methanol, ethanol and hydrocarbons has been studied. All catalysts showed similar activity except for chromia, which was almost inactive at the temperature chosen. Selectivity to higher oxygenates followed the trend: V2O3 > MoO3 > Al2O3 > Cr2O3. The alumina supported catalysts had the highest selectivity to methanol. Catalysts have been characterised in situ by extended X-ray absorption fine structure (EXAFS) and the alumina-supported material has been examined in the greatest detail. On reduction at 473 K, rhodium particles with diameter 30 A diameter, on the chromia support.The relation between catalyst structure and performance is discussed.

Oxygen adsorption on Ag powder

Oxygen adsorption on a Ag powder sample is different from that on a supported Ag catalyst. The powder shows the presence of a very high temperature desorption state (at 900 K) which diminishes adsorption into the state which is common to all types of silver samples (desorption peak at 560 K). This behaviour, together with the increased sticking probability which is observed on the powder, indicates that the powder has a very much rougher surface structure than a catalyst, which consists of mostly (111) planes. The powder probably consists of a high proportion of (111) planes with a high density of steps and defects since the high-temperature desorption state has been seen on a stepped single-crystal plane [Ag(331)].

A molecular beam study of the oxidative dehydrogenation of alcohols on Cu(110)

The adsorption and reaction of ethanol, propan-1-ol, and propan-2-ol have been studied on clean and oxygen-covered copper(110) by using a thermal molecular beam. The preadsorption of oxygen greatly enhances the sticking probability of all the alcohols, although a saturation coverage (p (2 × 1) LEED pattern) poisons the oxidative dehydrogenation reaction. The reaction mechanism has been worked out in detail with the help of isotopic labeling experiments and is common to all alcohols examined. There is a change of stoichiometry observed with temperature, at low temperatures (at ambient or just above) 2R1R2CHOH + O(a) → 2R1R2CO + H2O + H2, changing to R1R2CHOH + O(a) → R1R2CO + H2O at higher temperature, the exact temperature depending on the alcohol involved. The rate of reaction follows the stability of the intermediate alkoxy species except for 2-propoxy which shows a slower rate than expected. This effect and the poisoning observed for saturation coverages of preadsorbed oxygen are due to the stabilization of the alkoxy species by coadsorbed surface hydroxyl groups or oxygen atoms. Overall the reaction rate is limited by the decomposition of the alkoxy species.

Deuterated benzoate species on copper (110), a vibrational spectroscopic study

We have studied the effect of deuteration on the adsorption of benzoic acid on a copper (110) surface with RAIRS and HREELS. The resulting benzoate species has been characterized by the assignment of the spectra to modes of A 1 symmetry under the C 2v point group, which shows the carboxylate group and aromatic ring are perpendicular to the surface. Comparison to the non-deuterated species shows some unexpectedly large intensity changes and frequency shifts for several aromatic ring breathing modes due to coupling with modes of the carboxylate group

Precursor state effects in catalysis: Oxidative dehydrogenation of propan-2-ol on Cu(110)

The oxidative dehydrogenation of propan-2-ol has been investigated using a thermal molecular beam system. The main products after predosing the surface with oxygen atoms are the dehydrogenated molecule (acetone), water and hydrogen. The effect of oxygen is to enhance the sticking of propan-2-ol from approximately 0.05 on the clean surface to 0.76 (±0.02). By varying the oxygen predose we have shown unambiguously that precursor state effects dominate the catalytic reaction, since at oxygen coverages as low as 0.008 monolayers the propanol sticking probability remains at 0.76. This is due to the long lifetime and diffusivity of the reactant on the surface enabling it to‘seek out’ the active sites (those with oxygen atoms) on the surface.