The formation of small clusters from model molecules was studied using density functional theory calculations and microkinetic modeling. Exothermic dimerization provides depletion of the gas phase by the dominant isotope when the system is cooled to a temperature between condensation of the monomer and the dimer. It was found that temperature correlates with the interatomic distances in the monomer and the SF6 dimer in accordance with Hooke’s law, which allowed us to estimate the equilibrium constant of the dimerization reaction. The separation coefficients α32 0.07, α33 9, and α34 13 for 32S, 33S, and 34S, respectively, which were obtained for SF6 (101.325 kPa at 300 K) cooled to the condensation temperature, significantly exceed the fractionation values for the kinetic reason α 1.03. Similar behavior is expected for He atoms and CF4, NF3, H2, and H2O molecules, which exhibit exothermic clustering, but not for O2, N2, and NO, which exhibit endothermic clustering.
The modification of gamma-Al2O3 after NO treatment and during the NO + H-2 reaction was studied using X-ray photoelectron spectroscopy (XPS) and density functional theory. The XPS N1s spectra contain features with binding energies BE = 399.0 and 403.0 eV; the first is preserved, and the second disappears when NO is removed from the gas phase. The features were identified from experimental and modeled O-KLL regions of possible modification products - NAln oxynitrides (n = 3; 4) of the gamma-Al2O3(110) surface and adsorbed states. Calculations show that NAln adsorbs NO, forming normal states with a distance to the surface d = 1.32-1.70 & Aring; and distant (d-NO) states (d = 1.91-2.06 & Aring;) in which the spin-polarized NO molecule is retained due to magnetic NAln and Coulomb interaction. An increase in the similarity of the O-KLL spectra in the sequence of NO/gamma-Al2O3(110), NO/NAln and d-NO/NAln at an adsorption heat of similar to 0.53, 1.33 and 1.26 eV, respectively, suggests that the modification includes the formation of NO/NAl4 corresponding to BE = 399.0 eV and d-NO/NAl4, which corresponds to BE = 403.0 eV and, being chemically unbound, is able to leave the surface in the absence of NO in the reaction medium.
The separation of isotopes of natural Krypton at the gas-liquid and liquid-solid phase interfaces was studied under nonequilibrium conditions using a cryogenic cell and mass spectrometry. The formation of condensate upon cooling Kr from the ambient temperature begins at an equilibrium temperature, which corresponds to the partial pressure of the dominant isotope 84Kr, and is accompanied by depletion of the gas phase 84Kr with a separation coefficient of ~0.92; but the isotopic composition returns to the original under conditions close to equilibrium. The formation of a solid phase near the freezing point is accompanied by depletion of the gas phase by heavy isotopes. The separation coefficients 86Kr and 84Kr are ~2 and ~12, respectively, when ~3.2% of the atoms pass into the solid phase. The solid phase with its fraction below 8.8%, 5.8% and 5.7% does not contain 80Kr, 82Kr and 83Kr with separation coefficients above ~90, ~110 and ~70, respectively, to compensate for the enrichment of the gas and liquid phases. Pressure-selective condensation can be used to separate components with close boiling points when distillation and temperature-selective condensation methods are ineffective, and freezing-out of heavy isotopes can be used to enrich elements with practically important isotopes.
In this work, the interaction of the surface of γ-Al2O3(110) with NO and H2 was studied using density functional theory calculations. Free γ-Al2O3(110) adsorbs NO and binds H atoms, but repels the H2 molecule. A triplet of low-coordinated OII-AlIII-OII atoms provides the catalytic activity of γ-Al2O3(110) along the path: (i) the adsorption of NO/AlIII is followed by the binding of H2 to form a hydroxylamine derivative NHOH through an intermediate NO/AlIII + 2 × H/OII complex; (ii) recombination of NHOH with the release of N2 through an intermediate NHOH/AlIII + NHOH/AlIV or adsorption of NO followed by the release of N2O through the intermediate NHOH/AlIII + NO/AlIV; the pathway ends with the regeneration of γ-Al2O3(110). The calculated adsorption heats ensure the diffusion of H atoms from the deposited Pt to the surface (110), initiating the formation of the NH2/AlIII + H/OII complex, which releases NH3 endothermically and is stable enough to inhibit stage (ii) of the above reaction pathway. An excess of O2 in the NO + H2 mixture excludes H/Pt and eliminates inhibition. The formation of oxynitrides is suppressed, but not excluded by more exothermic surface processes. The N-doped conductivity of bulk and surface oxynitrides Al32O47N and the dependence of the heat of adsorption of H atoms on the band gap width were revealed, which suggests a relationship between the band gap width and catalytic activity.
Background: The close boiling points of carbon tetrafluoride (CF4) and nitrogen trifluoride (NF3) and the chemical inertness of CF4 make it difficult to remove it from NF3. Methods: A method for cleaning NF3 from CF4 at the gas-liquid interface is proposed. The purification is carried out by a sequence of cycles: (1) fractional condensation of NF3 at a cryostat temperature (T-cry), at which the equilibrium pressure of CF4 exceeds its partial pressure in the feed mixture; and (2) pumping out the gas phase enriched with CF4 and high-boiling impurities. The cycles are repeated at a next T-cry corresponding to the new CF4 content, and are completed by cryogenic distillation of the condensate with the removal of low-boiling impurities. Findings: The CF4 content of 5000 ppm in a 25 g NF3 sample was reduced to less than 100 ppm in three cycles at T-cry = 120 K; a CF4 separation coefficient above 40 was achieved. The method can be used for the industrial production of high-purity NF3, with the cryostat refrigerant being the only consumable material, as well as for the separation of other substances with close boiling points, such as isomers or isotopes. (C) 2021 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
In this study, density functional theory calculations and microkinetic modeling were used to estimate the catalytic activity of Re delta Co1-delta(0001) alloys (delta = 0; 0.25; 0.5; 0.75, 1), and Co(0001)-supported single- Re-1 and dual-atom Re-2 centers in the synthesis of ammonia. Nitrogen atoms (N-ad) form an ordered structure, and the bridge state is most active among others on the Re(0001) plane; while N-ad are concentrated around the Re-1 and Re-2 centers, splitting in electronic and thermodynamic properties. The specific catalytic activity of the centers Re-1, Re-2 and the Re(0001) plane changes in the order of 8.0.10(3), 32.0, and 1.0, respectively, due to weakly bound but highly active N-ad, which are stabilized by strongly bound and low-active N-ad. The calculations revealed the NRe3 state on the Re3Co1 alloy with an abnormally high binding energy (BE) of -7.05 eV, unequal Re-N distances and Re5d fillings, in contrast to the similar NRe3 state on Re(0001) with BE = -6.71 eV. A variable synergism of Re-Co alloys was established: the Re3Co1 or Re1Co3 alloy exhibits maximum catalytic activity if the rate-determining step is the formation or hydrogenation of N-ad under appropriate experimental conditions, respectively.
Возможность получения новых данных стандартными методами электронной спектроскопии и квантовой химии показана на примере интеркалата C2FBr0.15 и серебряной фольги. Особенности протяженных рентгеновских фотоэлектронных спектров интерпретированы электронными переходами в валентной зоне аналогичных элементарных ячеек. Анализ экспериментальных и расчетных спектров выявил два состояния внедренного Br2 — молекулярное и цепочечное. Взаимодействие Ag и NO2 при 300—520 K ограничено образованием окисленного состояния в приповерхностном слое толщиной ~6 Å, при этом в спектрах Ag3d и Ag MNN доминирует металлическое состояние серебра. Геометрические параметры, состояние атомов и характер связей между ними согласуются с полученными ранее результатами.
The possibility to obtain novel data by standard electron spectroscopy and quantum chemical techniques is exemplified by C2FBr0.15 intercalate and silver foil. The features of extended X-ray photoelectron spectra are interpreted by electronic transitions in the valence band of similar unit cells. The analysis of experimental and calculated spectra reveals two states of intercalated Br2: molecular and chain-like. The interaction of Ag with NO2 at 300–520 K is limited by the formation of an oxidized state in the near-surface layer with a thickness of ∼6 A, with the metallic state of silver dominating in the Ag3d И Ag MNN spectra. Geometric parameters, states of atoms, and the character of bonds between them are consistent with the previously obtained results.
The strong Mo–N bond restrains the catalytic activity of metallic Mo in ammonia synthesis. In this study, the semi-empirical calculations in conjunction with the density functional theory calculations, Brønsted–Evans–Polanyi relationship and microkinetic modeling were used to evaluate the rate of ammonia synthesis on model active sites of Mo-based alloys, nitrides, and clusters with a modified Mo–N bond. It was found that active sites of binary alloys MoδMe1−δ (0 ≤ δ ≤ 1; Me = Co, Pt, Ir, Rh) show the synergetic behavior. The sites of ternary Mo3Me3N (Me = Mo, Co, Pt, Ir) and Mo2N-type nitrides revealed higher activities than sites on Mo planes due to an extra Mo bond with the lattice N atom. The sites of octahedral clusters Mo3Me3N (Me = Mo, Co, Ir, Pt) exhibited higher catalytic activity than the sites of nitrides because their Me–N bonds are weaker than Mo–N. It was also found that tetragonal Mo2Me2 (Me = Co, Pt, Ir) and bi-tetragonal clusters Mo3Me2 (Me = Co, Ir, Pt) are the best cases because their sites provide the optimal combination of local structure and thermodynamics. Catalytic activities of the most active sites, relative to the Fe–C7 center, were found to change in the row 18.4 (threefold site Mo2Ir1 in cluster Mo3Ir2), 7.3 (Mo2 in cluster Mo2Ir2), 3.9 (Mo2Pt1 in cluster Mo3Ir3N), 3.8 (M3 on alloy Mo0.78Ir0.22), 2.0 [Mo3Pt1 on the plane (100) of Mo3Pt3N], 0.57 [Mo3 on the plane (111) of Mo2N], and 0.03 [Mo4 on the plane Mo(110)–(1 × 2)]. The design of tailor-made catalytic sites suggested in this paper can probably be applied to other catalytic systems.
Earlier, a theoretical model was suggested to discriminate catalytic sites M-n (each consisting of n adjacent atoms M on the metal surface) according to their undercoordination Sigma. It has been shown that the maximum activity of a site M-n (M = Pt, Rh, Ir, Fe, Ru, Re; n = 2, 3, 4) in the catalytic synthesis of ammonia requires the "resonant" Sigma whose major part is inaccessible at perfect planes because of steric restrictions. The current study applies this model to binary alloys and clusters to construct an advanced catalytic site by adjustment of real Sigma to the resonance. The catalytic activity of a site M-n has been estimated by the Bronsted-Evans-Polanyi relation with respect to the formation of NH species. It was found that on alloy surfaces, sites M-3 and M-4 demonstrate synergetic behavior. This suggests that the most active catalyst (Ru or Re) can be improved by its alloying with the least active one (Pt or Rh). In the case of the noble metals, the sites M-3 and M-4 at 4-, 5-, and 11- atomic clusters are similar to 10-10(3) times more active than such sites at perfect planes, whereas the sites of Ru and Re show the opposite behaviour. The model was verified by comparison of the calculated specific catalytic activities of metals, centers Fe-C-7 and Ru-B-5, and single crystals with the published data. The superior activity of a catalytic site is generally enabled by its optimal thermodynamics, which is affected deeply by the first coordination shell. A correlation between local structure, thermodynamics, and activity of a site is likely valid for other catalytic systems.
The prospects of the complementary use of X-ray photoelectron spectroscopy (XPS) and density functional theory (DFT) have been demonstrated by the examples of highly oriented pyrolytic graphite, half-fluorinated graphite C2F, and half-fluorinated graphite C2F intercalated with Br C2FBr0.15. It has been shown that the photoelectron energy losses in XPS spectra conform well to valence band electron transitions resulted from the DFT calculations for relevant unit cells. This conformity justified the other results of joined XPS and DFT studies, which have revealed two arrangements of the Br2 embedded into the C2F framework. The first arrangement corresponds to separate Br pairs in which the Br state is similar to a free Br2 molecule, whereas the second one is an ultra-dense Br chain in which the Br state is between free Br2−1 and Br10 species. The specific energy losses in the XPS Br3d spectrum of C2FBr0.15 indicate a comparable content of both Br arrangements in a sample. Besides, a distinct structure in the difference F1s XPS spectrum is assigned to the expected strengthening of the C-F bond in a C2F matrix under the Br2 intercalation. The state and orientation of intercalated Br2 are juxtaposed with experimental studies by Near Edge and Extended X-ray Absorption Fine Structure spectroscopy and by Raman spectroscopy. A successful confluence of XPS and DFT can be useful in the field of material science, providing the local geometry, the state and bonding between atoms in a sample, and thereby revealing the wear performance of the material, regardless of its application.
Electronic configuration of chemically bound atoms at the surface, including adsorbed species, or in the bulk of a solid contains a set of natural traps for energy absorption provided by valence band transitions or plasmon oscillations. The core level excitation of any origin is generally coupled with those traps, forming a multichannel route for nonradiative energy dissipation. Using an example of Pt and graphite-based materials, the study shows experimental tracing over these channels by means of elastic electron scattering and X-ray photoelectron spectroscopy. As a complement to the experimental data, calculations of the density of states provide information on chemical behavior and local geometry of the atoms in a sample.
The regularities of the electron energy dissipation found in the subsurface atomic layer are valid in the bulk of a solid, too. On the example of model graphite-based materials it is shown that energy losses in X-ray photoelectron spectra agree with the calculated valence electron excitation spectra in analogous unit cells. The control of conjugate electron transitions opens the way to gain new data on the geometry, character, and order of bonding between atoms in the sample by the conventional electron spectroscopy and quantum chemistry methods.
Unique catalytic potential of metal surfaces has encouraged a great number of basic and applied studies. The manuscript highlights the general regularities in a field on the grounds of strong interrelation between catalytic, kinetic and thermodynamic behaviour of the reaction system. The trials of the catalytic NH3 synthesis and the oscillatory NO+H2 reaction have revealed that the thermodynamics of the local structure determines the properties and multiplicity of the reaction intermediates enabling the peculiar macroscopic kinetics and specific catalytic activity. Structure and activity of catalytic sites are correlated within a realistic model, where total undercoordination of adjacent surface atoms and enthalpy of local reaction is taken as a descriptor for structure and activity, respectively. The model has specified the resonant catalytic centers for NH3 synthesis on metal surfaces in close agreement with experimental data. The basal planes of noble metals are less active than Fe- and Ru-based catalysts, whereas an extraordinary activity of small Pt, Ir and Rh clusters can be expected. A strong advantage of imperfections compared to perfect areas in the surface wave nucleation is evaluated. Isothermal rate oscillations in open heterogeneous catalytic reaction systems are expected under the multiplicity of reaction intermediates fairly different in activity, providing the steady state and reaction rout multiplicity. Switching between active and inactive kinetic brunches gives rise to the explosive coverage changeover that can be visualized as a traveling wave. A single pattern of oscillations in the NO+H2 reaction includes the key role of intermediate NHad species providing the catalytic removal of strongly bound nitrogen. The driving forces, the feedback, and chemical interactions within the traveling waves are clearly understood.
Adsorption sites Mn consisted of n adjacent atoms M, each bound to the adsorbed species, are considered within a realistic model. The sum of bonds Σ lost by atoms in a site in comparison with the bulk atoms was used for evaluation of the local surface imperfection, while the reaction enthalpy at that site was used as a measure of activity. The comparative study of Mn sites (n=1–5) at basal planes of Pt, Rh, Ir, Fe, Re and Ru with respect to heat of N2 dissociative adsorption QN and heat of Nad+Had→NHad reaction QNH was performed using semi-empirical calculations. Linear QN(Σ) increase and QNH(Σ) decrease allowed to specify the resonant Σ for each surface in catalytic ammonia synthesis at equilibrium Nad coverage. Optimal Σ are realizable for Ru2, Re2 and Ir4 only, whereas other centers meet steric inhibition or unreal crystal structure. Relative activity of the most active sites in proportion 5.0×10−5: 4.5×10−3: 1: 2.5: 3.0: 1080: 2270 for a sequence of Pt4, Rh4, Fe4(fcc), Ir4, Fe2–5(bcc), Ru2, Re2, respectively, is in agreement with relevant experimental data. Similar approach can be applied to other adsorption or catalytic processes exhibiting structure sensitivity.
Standard equipment for electron spectroscopy empowers fingerprinting the adsorbed layer at atomic-molecular level. The resource results from the novel route for primary electron energy dissipation through a set of shake-off and shake-up electron transitions, each coupled with the threshold core level excitation. Experimental evidence for the route is based on regular fine structures in extended elastic electron spectra from the Pt(100) single crystal fitting the valence state structures of adsorbed species and substrate atoms as well as plasmon excitations; in total over 20 particular satellites have been detected in conventional experimental conditions. Theoretical justification of a route consists in the mechanism being a combination of well-known electron transitions. The phenomenon is considered as a fundamental regularity of electronsolid interaction disclosing the outer shell structure of near-surface atoms alike an Auger effect brings to light the inner shell structure.
The conjugate electron excitation (CEE) above the core-level thresholds of substrate atoms is a combination of the known electronic shake-off and shake-up electron transitions resulting in the location of excited electrons can at both vacuum level and vacant states near the Fermi level. The CEE satellites in elastic electron spectra provide the direct experimental data on the valence state structure of surface atoms and components of the adsorbed layer. The CEE control can be implemented as an additional option of the standard equipment for electron spectroscopy.
Elastic electron scattering by the adsorbate covered Pt(100) single crystal surface and density of states (DOS) calculations highlight two peculiar channels for the primary electron energy consumption via the conventional threshold core level excitation coupled with particular electron transitions. The first channel affects the substrate atoms and implies Pt DOS shake-off and shake-up transitions and multiple plasmon excitations; the second one includes shake-off processes in the adsorbed layer and enables the valence state structure of the adsorbed species. The mechanism of electron transitions assumes that one-dimensional DOS at the vacuum level, in addition to vacant DOS at the Fermi level, is an active spot for allocation of excited electrons. The observed phenomena are supposed to be the general regularity of electron-solid interaction and a useful tool for fingerprinting the adsorbed layer at molecular level.
The conventional mechanism of oscillatory phenomena in the NO + H2 reaction on noble metals is supplemented with the temporal reaction route, providing fast removal of atomic nitrogen from saturated Nad layer via NHad species as a catalyst. The total sum of lost valences at a given adsorption site is taken as a degree of local surface imperfection. A semi-empirical evaluation of this approach regards an enormous advantage of Nad atoms at a grain boundary against perfect terraces in the activity of NHad formation as a driving force of both spatiotemporal and rate oscillations. Besides, NHad species are expected to obey an “easy-come-easy-go” principle enabling their ready formation as well as high diffusivity and reactivity. Mathematical modelling at fixed realistic step constants reveals three kinetic region-attractors related to a steady state, regular oscillations, and total reaction suppression by adsorbed oxygen atoms.
The study revealed additional channels of inelastic electron scattering, which accompany the threshold excitation of the substrate Pt4d level — ionization of the valent states of adsorbed particles chemically bonded to the excited atom, and excitation of the surface plasmon vibrations. The conjugate excitation of this type shows up as a series of typical satellites in the spectra of disappearance potentials, which reflects the structure of valent states of adsorbed particles. Analysis of the satellite structure revealed the intermediate formation of NH x,ads particles in the reaction NOgas + Hads on the surface of Pt(100) single crystal and, taking into account the earlier data, made it possible to formulate a general mechanism of selfoscillations in the NO + H2 reaction on platinum metals. Mathematical modeling of reaction kinetics on the Pt(100) surface within the suggested mechanism demonstrated the presence of regular self-oscillations of the reaction rate at invariable values of the step constants.