Mechanistic and Kinetic Measurements of Elementary Surface Reactions Using Temperature-Programmed X-ray Photoelectron Spectroscopy.

This paper describes a method by which surface-reaction kinetics can be measured by slowly and precisely ramping up the surface temperature at a constant rate while simultaneously collecting X-ray photoelectron spectra (XPS). This approach results in the collection of a large amount of data over relatively small temperature steps to produce quasi-continuous kinetic data. The method is illustrated for the desorption and reaction of diethyl disulfide (DEDS) on a Au(111) substrate in ultrahigh vacuum, where the results can be compared with previous conventional temperature-programmed desorption (TPD) data from Au(111). Experiments were carried out using a double-pass cylindrical-mirror analyzer with a channeltron detector to demonstrate how this approach can be implemented in a routine, multitechnique vacuum chamber. The approach will be even more effective in a more modern, specialized XPS apparatus with high-transmission hemispherical analyzers with multichannel array detectors, which will enable the spectra of several elements to be measured simultaneously. The results yielded an activation energy for multilayer desorption of DEDS of 41 ± 1 kJ/mol, with a pre-exponential factor of 8 ± 7 × 1012 s-1, an activation energy of 53 ± 6 kJ/mol and pre-exponential factor of 9 ± 8 × 1013 s-1 for monolayer desorption and an activation energy of 90 ± 6 kJ/mol with a prefactor of 1.0 ± 0.3 × 1015 s-1 for the reaction of adsorbed ethyl thiolate species to adsorbed DEDS. While these results were collected for a system for which the kinetic data could have been obtained using conventional TPD, this method can be more usefully applied to those surface reaction processes that do not rely on the formation of desorption products. This system, having been previously studied by TPD, facilitates a comparison with results obtained by conventional methods.

Exploring mechanochemical reactions at the nanoscale: theory versus experiment.

Mechanochemical reaction pathways are conventionally obtained from force-displaced stationary points on the potential energy surface of the reaction. This work tests a postulate that the steepest-descent pathway (SDP) from the transition state to reactants can be reasonably accurately used instead to investigate mechanochemical reaction kinetics. This method is much simpler because the SDP and the associated reactant and transition-state structures can be obtained relatively routinely. Experiment and theory are compared for the normal-stress-induced decomposition of methyl thiolate species on Cu(100). The mechanochemical reaction rate was calculated by compressing the initial- and transition-state structures by a stiff copper counter-slab to obtain the plots of energy versus slab displacement for both structures. The reaction rate was also measured experimentally under compression using a nanomechanochemical reactor comprising an atomic-force-microscopy (AFM) instrument tip compressing a methyl thiolate overlayer on Cu(100) (the same system for which the calculations were carried out). The rate was measured from the indent created on a defect-free region of the methyl thiolate overlayer, which also enabled the contact area to be measured. Knowing the force applied by the AFM tip yields the reaction rate as a function of the contact stress. The result agrees well with the theoretical prediction without the use of adjustable parameters. This confirms that the postulate is correct and will facilitate the calculation of the rates of more complex mechanochemical reactions. An advantage of this approach, in addition to the results agreeing with the experiment, is that it provides insights into the effects that control mechanochemical reactivity that will assist in the targeted design of new mechanochemical syntheses.

Surface chemistry at the solid-solid interface: mechanically induced reaction pathways of C8 carboxylic acid monolayers on copper.

Mechano- or tribochemical processes are often induced by the large pressures, of the order of 1 GPa, exerted at contacting asperities at the solid-solid interface. These tribochemical process are not very well understood because of the difficulties of probing surface-chemical reaction pathways occurring at buried interfaces. Here, strategies for following surface reaction pathways in detail are illustrated for the tribochemical decomposition of 7-octenoic and octanoic acid adsorbed on copper. The chemistry was measured in ultrahigh vacuum by sliding either a tungsten carbide ball or a silicon atomic force microscope (AFM) tip over the surface to test a previous proposal that the nature of the terminal group in the carboxylic acid, vinyl versus alkyl, could influence its binding to the counterface, and therefore the reaction rate. The carboxylic acids bind strongly to the copper substrate as carboxylates to expose the hydrocarbon terminus. The tribochemical reaction rate was found to be independent of the nature of the hydrocarbon terminus, although the pull-off and friction forces measured by the AFM were different. The tribochemical reaction is initiated in the same way as the thermal reaction, by the carboxylate group tilting to eliminate carbon dioxide and deposit alkyl species onto the surface. This reaction occurs thermally at ∼640 K, but tribochemically at room temperature, producing significant differences in the rates and selectivities of the subsequent decomposition pathways of the adsorbed products.

Adsorption and reaction pathways of 7-octenoic acid on copper.

The surface structure and reaction pathways of 7-octenoic acid are studied on a clean copper substrate in ultrahigh vacuum using a combination of reflection-absorption infrared spectroscopy, X-ray photoelectron spectroscopy, temperature-programmed desorption and scanning-tunneling microscopy, supplemented by first-principles density functional theory calculations. 7-Octenoic acid adsorbs molecularly on copper below ∼260 K in a flat-lying configuration at low coverages, becoming more upright as the coverage increases. It deprotonates following adsorption at ∼300 K to form an η2-7-octenoate species. This also lies flat at low coverages, but forms a more vertical self-assembled monolayer as the coverage increases. Heating causes the 7-octenoate species to start to tilt, which produces a small amount of carbon dioxide at ∼550 K and some hydrogen in a peak at ∼615 K ascribed to the reaction of these tilted species. The majority of the decarbonylation occurs at ∼650 K when CO2 and hydrogen evolve simultaneously. Approximately half of the carbon is deposited on the surface as oligomeric species that undergo further dehydrogenation to evolve more hydrogen at ∼740 K. This leaves a carbonaceous layer on the surface, which contains hexagonal motifs connoting the onset of graphitization of the surface.

Influence of the terminal group on the thermal decomposition reactions of carboxylic acids on copper: nature of the carbonaceous film.

The effect of the terminal groups on the nature of the films formed by the thermal decomposition of carboxylic acids on copper is studied in ultrahigh vacuum using temperature-programmed desorption (TPD), scanning tunneling microscopy (STM) and Auger electron spectroscopy (AES). The influence of the presence of vinyl or alkynyl terminal groups and chain length is studied using heptanoic, octanoic, 6-heptenoic, 7-octenoic, 6-heptynoic and 7-octynoic acids. The carboxylic acids form strongly bound carboxylates following adsorption on copper at room temperature, and thermally decompose between ∼500 and 650 K. Previous work has shown that this occurs by the carboxylate plane tilting towards the surface to eliminate carbon dioxide and deposit a hydrocarbon fragment. The fragment can react to evolve hydrogen or form oligomeric species on the surface, where the amount of carbon increases for carboxylic acids that contain terminal functional groups that can anchor to the surface. These results will be used to compare with the carbonaceous films formed by the mechanochemical decomposition of carboxylic acids on copper, which occurs at room temperature. This is expected to lead to less carbon being deposited on the surface than during thermal decomposition.

Chemisorptive enantioselectivity of chiral epoxides on tartaric-acid modified Pd(111): three-point bonding.

The chemisorption of two chiral molecules, propylene oxide and glycidol, is studied on tartaric-acid modified Pd(111) surfaces by using temperature-programmed desorption to measure adsorbate coverage. It is found that R-glycidol shows preferential enantioselective chemisorption on (S,S)-tartaric acid modified Pd(111) surfaces, while propylene oxide does not adsorb enantioselectively. The enantioselectivity of glycidol depends on the tartaric acid coverage, and is exhibited for low tartaric acid coverages indicating that the bitartrate phase is responsible for the chiral recognition. The lack of enantioselectivity when using propylene oxide as a chiral probe implies that the enantiospecific interaction between glycidol and bitartate species is due to hydrogen-bonding interactions of the -OH group of glycidol. Scanning tunneling microscopy images were collected for tartaric acid adsorbed on Pd(111) under the same experimental conditions as used for enantioselective experiments. When tartaric acid is dosed at room temperature and immediately cooled to 100 K for imaging, individual bitartrate molecules were found. Density functional theory (DFT) calculations show that bitartrate binds to Pd(111) through its carboxylate groups and the -OH groups are oriented along the long axis of the bitartrate molecule. An enantiospecific interaction is found between glycidol and bitartate species where R-glycidol binds more strongly than S-glycidol to (S,S)-bitartate species by simultaneously forming hydrogen bonds with both the hydroxyl and carboxylate groups, thereby providing three-point bonding.

Surface chemistry of isopropoxy tetramethyl dioxaborolane on Cu(111).

The surface chemistry of isopropoxy tetramethyl dioxaborolane (ITDB), tetramethyl dioxaborolane (TDB), and 2-propanol is studied on a clean Cu(111) single crystal using temperature-programmed desorption (TPD). 2-Propanol is found to have two competing reactions on the copper surface. Dehydration results in water and propene formation, and dehydrogenation results in the formation of acetone and hydrogen. ITDB directly adsorbed on the surface reacts completely and does not molecularly desorb. TDB and 2-propanol decompose desorbing mainly 2,3-dimethyl 2-butene and acetone, respectively. Both of those products desorb above room temperature and are present in TPDs of ITDB. An additional acetone desorption peak was observed for ITDB at higher temperatures than acetone desorption from 2-propanol. This higher temperature peak at ∼391 K was attributed to two acetone molecules forming from the tetramethyl end group resulting from a stronger bound surface species in ITDB compared to TDB despite their identical end groups. The copper surface seems to be reactive enough toward ITDB at room temperature that a potential boron-containing tribofilm could be produced for copper-copper sliding contacts. Despite their similarities, ITDB and TDB have different surface species present at room temperature, so their tribological properties will be investigated in the future.

Critical stresses in mechanochemical reactions.

The rates of mechanochemical reactions are generally found to increase exponentially with applied stress. However, a buckling theory analysis of the effect of a normal stress on an adsorbate that is oriented perpendicularly to the surface that reacts by tilting suggests that a critical value of the stress should be required to initiate a mechanochemical reaction. This concept is verified by using density functional theory calculations to simulate the effect of compressing a homologous series of alkyl thiolate species on copper by a hydrogen-terminated copper counter-face. This predicts that a critical stress is indeed needed to initiate methyl thiolate decomposition, which has a perpendicular C-CH3 bond. In contrast, no critical stress is found for ethyl thiolate with an almost horizontal C-CH3 bond, while a critical stress is required to isomerize propyl thiolate from a trans to a cis configuration. These predictions are tested by measuring the mechanochemical reaction rates of these alkyl thiolates on a Cu(100) substrate by sliding an atomic force microscope tip over the surface and finding a critical stress of ∼0.43 GPa for methyl thiolate, ∼0.33 GPa for propyl thiolate, but no evidence of a critical stress for ethyl thiolate, in accord with the predictions. These results provide insights not only into mechanochemical reaction mechanisms on surfaces, but also on the origin of critical phenomena in stress-induced processes in general. It also suggests novel approaches to designing robust surface films that can resist wear and damage.

Efficient transport of gold atoms with a scanning tunneling microscopy tip and a linker molecule.

A thiophene-containing molecule attached to a scanning tunneling microscopy (STM) tip is used to transport gold atoms on a Au(111) surface. The molecule contains eight thiophene rings and therefore has sulfur atoms that are known to bind to gold atoms. Using a gold-coated tip, the molecules previously deposited on the surface bind to the lower-coordination gold atoms of the tip. When that tip is used to scan the surface, the still free thiophene rings (not all of the sulfur atoms bind to the tip) can attach to gold atoms from the surface and drag them along the scanning direction, depositing them either at the position where the tip changes its scanning direction or where the tip encounters an "up step", whichever event occurs first.

Insights into the Mechanism of the Mechanochemical Formation of Metastable Phases.

The mechanochemical reaction kinetics of sulfur with copper to form a metastable copper sulfide phase at room temperature is investigated in ultrahigh vacuum by modifying the properties of the copper during cleaning in vacuum. The measured kinetics is in agreement with a theory first proposed by Karthikeyan and Rigney that predicts that the rate depends linearly both on the contact time and on the strain-rate sensitivity of the substrate. The mechanism for this process was investigated using thin samples of copper fabricated using a focused-ion-beam and by measuring the crystal structure and elemental composition of the copper subsurface region by electron microscopy after reaction. The measured sulfur depth distributions produced by shear-induced surface-to-bulk transport were in good agreement with values calculated using rate constants that also model the reaction kinetics. Sulfur was found both in crystalline regions and also concentrated along grain boundaries, implying that formation of metastable phases is facilitated by both the presence of dislocations and by grain boundaries.

Coverage effects on the palladium-catalyzed synthesis of vinyl acetate: comparison between theory and experiment.

The high adsorbate coverages that form on the surfaces of many heterogeneous catalysts under steady-state conditions can significantly lower the activation energies for reactions that involve the coupling of two adsorbed intermediates while increasing those which result in adsorbate bond-breaking reactions. The influence of the surface coverage on the kinetics of metal-catalyzed reactions is often ignored in theoretical and even in some ultrahigh vacuum experimental studies. Herein, first principle density functional theoretical calculations are combined with experimental surface titration studies carried out over well-defined Pd(111) surfaces to explicitly examine the influence of coverage on the acetoxylation of ethylene to form vinyl acetate over Pd. The activation energies calculated for elementary steps in the Samanos and Moiseev pathways for vinyl acetate synthesis carried out on acetate-saturated palladium surfaces reveal that the reaction proceeds via the Samanos mechanism which is consistent with experimental results carried out on acetate-saturated Pd(111) surfaces. The rate-limiting step involves a beta-hydride elimination from the adsorbed acetoxyethyl intermediate, which proceeds with an apparent calculated activation barrier of 53 kJ/mol which is in very good agreement with the experimental barrier of 55 +/- 4 kJ/mol determined from kinetic measurements.

The surface chemistry of dimethyl disulfide on copper.

The surface chemistry of dimethyl disulfide (DMDS) is studied on a Cu(111) single crystal and a polished copper foil in ultrahigh vacuum as a basis for understanding its tribological chemistry using a combination of temperature-programmed desorption (TPD), reflection-absorption infrared spectroscopy (RAIRS), and X-ray photoelectron spectroscopy (XPS). Low-energy electron diffraction reveals that the polished foil becomes ordered on heating in vacuo and displays identical surface chemistry to that found on the Cu(111) surface. Dimethyl disulfide reacts with the copper surface at 80 K to form thiolate species. Heating the surface to ∼230 K causes a small portion of the thiolate species to decompose to form methyl groups adsorbed on the surface. Further heating results in methane and C(2) hydrocarbon desorption at ∼426 K, due to a reaction of adsorbed methyl species, to completely remove carbon from the surface and to deposit atomic sulfur.

One-dimensional supramolecular surface structures: 1,4-diisocyanobenzene on Au(111) surfaces.

One-dimensional supramolecular structures formed by adsorbing low coverages of 1,4-diisocyanobenzene on Au(111) at room temperature are obtained and imaged by scanning tunneling microscopy (STM) under ultrahigh vacuum (UHV) conditions. The structures originate from step edges or surface defects and arrange predominantly in a straight fashion on the substrate terraces along the <110> directions. They are proposed to consist of alternating units of 1,4-diisocyanobenzene molecules and gold atoms with a unit cell in registry with the substrate corresponding to four times the lattice interatomic distance. Their long 1-D chains and high thermal stability offer the potential to use them as conductors in nanoelectronic applications.

Measuring and modelling mechanochemical reaction kinetics.

Quasi-static density functional theory calculations of the rate of mechanochemical decomposition of methyl thiolate species adsorbed on Cu(100) accurately reproduce the experimental normal-stress dependent rates measured in ultrahigh vacuum by an atomic force microscopy tip. This allows precise analytical models for mechanochemical reaction kinetics to be developed.

A new method for performing polarization modulation infrared reflection-adsorption spectroscopy of surfaces.

A new and relatively simple polarization modulation technique is presented and tested that enables the whole spectral range to be detected between 400 and 4000 cm(-1). This experiment is conventionally carried out using a photoelastic modulator that modulates incident plane polarized light through 90 degrees . This suffers from the drawback that it enables spectra to be collected only over a relatively narrow spectral range. As an alternative, a polarizer is placed in the beam and oriented at 45 degrees to the sample normal. This produces incident radiation fluxes with identical intensities for both s- and p-polarized light. A second polarizer is then modulated through 90 degrees and the surface spectrum is then extracted in the usual manner from the difference between these signals, normalized to their sum. The method is demonstrated for a self-assembled monolayer of 11-mercapto-undecanoicacid (11-MUA) on gold on mica, and it is shown that, while the resulting spectra are extremely sensitive to optical alignment, the method yields spectra that are in excellent agreement with published data.

Chemical self-assembly strategies for designing molecular electronic circuits.

Design principles are demonstrated for fabricating molecular electronic circuits using the inherently self-limiting growth of molecular wires between gold nanoparticles from the oligomerization of 1,4-phenylene diisocyanide.

Ethylene decomposition at undercoordinated sites on Cu(410).

We demonstrate the selective, low-temperature chemistry of ethylene on the strongly undercoordinated sites of Cu(410) by investigating its adsorption by high-resolution electron energy loss spectroscopy (HREELS), X-ray photoelectron spectroscopy (XPS), and temperature-programmed desorption (TPD). After dosing ethylene at approximately 110 K, apart from the expected pi-bonded species adsorbed on terraces, di-sigma-bonded ethylene and carbon are formed at the step edges. The latter product results from the complete dehydrogenation of ethylene and blocks sites for further dissociation and/or di-sigma-adsorption. However, these processes can be restored merely by heating the sample to 900 K, by causing the carbon to diffuse into the bulk. The presented results support the relevance of copper-based catalysts for the steam reforming process.

Development of a ReaxFF Force Field for Cu/S/C/H and Reactive MD Simulations of Methyl Thiolate Decomposition on Cu (100).

It has been shown that the rate of decomposition of methyl thiolate species on copper is accelerated by sliding on a methyl thiolate covered surface in ultrahigh vacuum at room temperature. The reaction produces small gas-phase hydrocarbons and deposits sulfur on the surface. Here, a new ReaxFF potential was developed to enable investigation of the molecular processes that induce this mechanochemical reaction by using density functional theory calculations to tune force field parameters for the model system. Various processes, including volumetric expansion/compression of CuS, CuS2, and Cu2S unit cells; bond dissociation of Cu-S and valence angle bending of Cu-S-C; the binding energies of SCH3, CH3, and S atoms on a Cu surface; and energy for the decomposition of methyl thiolate molecular species on copper, were used to identify the new ReaxFF parameters. Molecular dynamics simulations of the reactions of adsorbed methyl thiolate species at various temperatures were performed to demonstrate the validity of the new potential and to study the thermal reaction pathways. It was found that reaction is initiated by C-S bond scission, consistent with experiments, and that the resulting methyl species diffuse on the surface and combine to desorb ethane, also as found experimentally.

Enantioselective chemisorption of propylene oxide on a 2-butanol modified Pd(111) surface: the role of hydrogen-bonding interactions.

The enantioselective chemisorption of R- and S-propylene oxide has been measured either on clean Pd(111) that has been exposed to S-2-butanol at various temperatures to vary the proportion of 2-butanol and 2-butoxide species or by adsorbing S-2-butanol on oxygen-covered Pd(111) to form exclusively 2-butoxide. The results reveal that enantioselective chemisorption is only found when 2-butanol is present on the surface. This is ascribed to enantiospecific hydrogen-bonding interactions between 2-butanol and propylene oxide. Measurements of the variation in enantiospecificity with 2-butanol exposure suggest that propylene oxide can interact either with a single adsorbed 2-butanol molecule or, at higher coverages, with two adsorbed 2-butanol species to form enantioselective sites.

Modeling Mechanochemical Reaction Mechanisms.

The mechanochemical reaction between copper and dimethyl disulfide is studied under well-controlled conditions in ultrahigh vacuum (UHV). Reaction is initiated by fast S-S bond scission to form adsorbed methyl thiolate species, and the reaction kinetics are reproduced by two subsequent elementary mechanochemical reaction steps, namely a mechanochemical decomposition of methyl thiolate to deposit sulfur on the surface and evolve small, gas-phase hydrocarbons, and sliding-induced oxidation of the copper by sulfur that regenerates vacant reaction sites. The steady-state reaction kinetics are monitored in situ from the variation in the friction force as the reaction proceeds and modeled using the elementary-step reaction rate constants found for monolayer adsorbates. The analysis yields excellent agreement between the experiment and the kinetic model, as well as correctly predicting the total amount of subsurface sulfur in the film measured using Auger spectroscopy and the sulfur depth distribution measured by angle-resolved X-ray photoelectron spectroscopy.