Different mechanisms have been proposed for Fischer-Tropsch synthesis, the conversion of CO and H 2 to long-chain alkanes. Density functional theory calculations indicate that CO activation has a barrier of 220 kJ/mol on Co(0001), and hence the concentration of surface C or CH 2 species is likely too low to explain the high chain growth probability. Hydrogenation lowers the C-O dissociation barrier to 90 kJ/mol for HCO and to 68 kJ/mol for H 2 CO; however, CO hydrogenation has a high energy barrier of 146 kJ/mol and is +117 kJ/mol endothermic. We propose an alternative propagation cycle starting with CO insertion into surface RCH groups. The barrier for this step is 80 kJ/mol. RCHCO is subsequently hydrogenated to RCH 2 CHO, which undergoes C-O dissociation with a barrier of 50 kJ/mol. The hydrogenation barriers are 120 and 48 kJ/mol along the dominant reaction path. The calculated CO turnover frequency for the proposed CO insertion mechanism is 1 to 2 orders of magnitude faster the hydrogen-assisted CO activation mechanism and 4 orders of magnitude faster than direct CO activation on a model Co(0001) surface.

No abstract

No abstract

X-ray photoelectron spectroscopy (XPS) is a powerful and popular surface characterization technique, and the measured shifts in the core electron binding energies are sensitive to the chemical structure and local environment of the surface species. C 1s binding energies were calculated with density functional theory (DFT) for 17 structures including eight well-characterized structures on a Co(0001) surface and nine on a Pt(111) surface, while B 1s binding energies were calculated for six well-characterized structures and compared with experimental values. DFT calculations describe the 2.8 eV variation in the C 1s binding energies on Co surfaces, the 4.2 eV variation in the C 1s binding energies on Pt surfaces, and the 5.5 eV variation in the B 1s binding energies in the test sets with average deviations of 85, 73, and 53 meV, respectively. The shift in the C 1s and the B 1s binding energies can be correlated with the calculated charges, though only within homologous series. To illustrate how binding energy calculations can help elucidate catalyst structures, the nature of the resilient carbon species deposited during Fischer–Tropsch synthesis (FTS) over Co/γ-Al2O3 catalysts was studied. The catalysts were investigated using XPS after reaction, and the measured C 1s binding energies were compared with DFT calculations for various stable structures. The XPS peak at 283.0 eV is attributed to a surface carbide, while the peak at 284.6 eV is proposed to correspond to remaining waxes or polyaromatic carbon species. Boron promotion has been reported to enhance the stability of Co FTS catalysts. Again, the combination of XPS with DFT B 1s binding energy calculations helped identify the nature and location of the boron promoter on the Co/γ-Al2O3 catalyst.

No abstract

Supported Co catalysts exhibit favorable activity and selectivity for Fischer−Tropsch synthesis (FTS) but deactivate slowly. To explore deactivation by carbon deposition, the stability of various forms of deposited carbon was evaluated using density functional theory (DFT). A surface carbide and graphene islands were calculated to be thermodynamically stable. Two forms of deposited carbon are also distinguished experimentally after 200 h of FTS. On the basis of this mechanistic insight, boron was proposed as a promoter to enhance the stability of Co catalysts. DFT calculations indicate that boron and carbon display similar binding preferences, and boron could selectively block the deposition of resilient carbon deposits. To evaluate the theoretical predictions, supported 20 wt % Co catalysts were promoted with 0.5 wt % boron and tested under realistic FTS conditions. Boron promotion was found to reduce the deactivation rate 6-fold, without affecting selectivity and activity.

Dispersed nickel sulfate ͑NiSO 4 ͒ microclusters on Si substrates were fragmented by pulsed excimer laser irradiation to serve as catalysts for carbon nanotube/nanofiber ͑CNT/CNF͒ growth. At proper fluences, NiSO 4 clusters were pulverized into nanoparticles. The sizes of clusters/nanoparticles were found to be dependent on laser fluence and laser pulse number. By increasing the laser fluence from 100 to 300 mJ/ cm 2 , the size of disintegrated particles decreased drastically from several micrometers to several nanometers. It was found that laser-induced disintegration of as-dispersed NiSO 4 clusters was mainly due to physical fragmentation by transient thermal expansion/ contraction. Thermal melting of nanoparticles in a multipulse regime was also suggested. Hot-filament chemical vapor deposition ͑HFCVD͒ was used for growth of CNTs from the pulsed-laser treated catalysts. For samples irradiated at 100 and 200 mJ/ cm 2 , CNFs were dominant products. These CNFs grew radially out of big NiSO 4 clusters, forming dendritic CNF bunches. For samples irradiated at 300 mJ/ cm 2 , dense multiwalled carbon nanotubes ͑MWCNFs͒ with uniform diameters were obtained. It is suggested that elemental Ni was formed through thermal decomposition of NiSO 4 clusters/nanoparticles during HFCVD. The size and the shape of the Ni aggregation, which were determined by the initial size of NiSO 4 clusters/nanoparticles, might affect the preference in the synthesis of CNTs or CNFs.

No abstract