Computational and Experimental Study of Thermodynamics of the Reaction of Titania and Water at High Temperatures.

Gaseous titanium hydroxide and oxyhydroxide species were studied with quantum chemical methods. The results are used in conjunction with an experimental transpiration study of titanium dioxide (TiO2) in water vapor-containing environments at elevated temperatures to provide a thermodynamic description of the Ti(OH)4(g) and TiO(OH)2(g) species. The geometry and harmonic vibrational frequencies of these species were computed using the coupled-cluster singles and doubles method with a perturbative correction for connected triple substitutions [CCSD(T)]. For the OH bending and rotation, the B3LYP density functional theory was used to compute corrections to the harmonic approximations. These results were combined to determine the enthalpy of formation. Experimentally, the transpiration method was used with water contents from 0 to 76 mol % in oxygen or argon carrier gases for 20-250 h exposure times at 1473-1673 K. Results indicate that oxygen is not a key contributor to volatilization, and the primary reaction for volatilization in this temperature range is TiO2(s) + H2O(g) = TiO(OH)2(g). Data were analyzed with both the second and third law methods using the thermal functions derived from the theoretical calculations. The third law enthalpy of formation at 298.15 K for TiO(OH)2(g) at 298 K was -838.9 ± 6.5 kJ/mol, which compares favorably to the theoretical calculation of -838.7 ± 25 kJ/mol. We recommend the experimentally derived third law enthalpy of formation at 298.15 K for TiO(OH)2, the computed entropy of 320.67 J/mol·K, and the computed heat capacity [149.192 + (-0.02539)T + (8.28697 × 10-6)T2 + (-15614.05)/T + (-5.2182 × 10-11)/T2] J/mol-K, where T is the temperature in K.

Infrared spectroscopy of matrix-isolated neutral polycyclic aromatic nitrogen heterocycles: The acridine series.

The matrix-isolated, mid-infrared spectra of seven acridine-based polycyclic aromatic nitrogen heterocycles (PANHs) have been measured and compared to their non-nitrogen containing parent molecule. The acridine species investigated include acridine, benz[a]acridine, benz[c]acridine, dibenz[a,j]acridine, dibenz[c,h]acridine, dibenz[a,h]acridine and dibenz[a,c]acridine. The previously reported results for 1 and 2-azabenz[a]anthracenes are included for comparison. The experimentally determined band frequencies and intensities are compared with their B3LYP/6-31G(d) values. The overall agreement between experimental and theoretical values is good and in line with our previous investigations. Shifts, typically to the blue, are noted for the C-H out-of-plane (CHoop) motions upon insertion of a nitrogen atom. The formation of a bay region upon addition of additional benzene rings to the anthracene/acridine structure splits the solo hydrogen motions into a bay region solo and an external solo hydrogen, with the bay region solo hydrogen coupling to the quartet hydrogen motions and the external solo hydrogen coupling with the duo hydrogen motions resulting in an extreme decrease in intensity for the CHoop solo hydrogen band when the external hydrogen is replaced by a nitrogen atom. The C-C and C-H in-plane region of this acridine series exhibits the characteristic two fold increase in intensity, noted previously for PANHs. The strong ≈1400cm-1 band, which was identified in the previous PANH study, is noted in several molecular species as well as another strong PANH feature between 1480 and 1515cm-1 for several molecules. The presence of these strong bands appear to be primarily responsible for the two-fold increase in the C-H in-plane region's (1100-1600cm-1) intensity. The C-H stretching region can be characterized by contributions from the solo (bay or external), duo and quartet hydrogens, similar to what was observed in the dibenzopolyacene compounds.

Optical properties of (GaAs)n clusters (n = 2-16).

The electronic and geometrical structures of the lowest triplet states of (GaAs) n clusters ( n = 2-16) are studied using density functional theory with generalized gradient approximation (DFT-GGA). It is found that the triplet-state geometries are different from the corresponding singlet-state geometries; for n = 2-8, 10, and 11, the triplets and singlets have different topologies, while the (GaAs) 9, (GaAs) 12, (GaAs) 15, and (GaAs) 16 triplets possess a reduced symmetry, due to Jahn-Teller distortions. Except for GaAs, the singlet states are the ground states. Excitation energies and oscillator strengths are computed for excitations from the ground state to ten singlet states of all (GaAs) n clusters using time-dependent density functional theory. The adiabatic singlet-triplet gap is compared to the vertical gap, and the difference in the eigenvalues of the highest-occupied and lowest-unoccupied molecular orbitals (the HOMO-LUMO gap). While these three values show large oscillations for small n, they approach each other as the cluster size grows. Thus, the HOMO-LUMO gap computed using the DFT-GGA approach presents a rather reliable estimate of the adiabatic singlet-triplet gap.

The structure and energetics of (GaAs)n, (GaAs)n(-), and (GaAs)n+ (n=2-15).

Electronic and geometrical structures of neutral, negatively, and positively charged (GaAs)n clusters are computed using density functional theory with generalized gradient approximation. All-electron computations are performed on (GaAs)2-(GaAs)9 while effective core potentials (ECPs) are used for (GaAs)9-(GaAs)15. Calibration calculations on GaAs and (GaAs)9 species support the use of the ECP for the larger clusters. The ground-state geometries of (GaAs)n(-) and/or (GaAs)n+ are different from the corresponding neutral ground-state geometry, except for n=7, 9, 12, 14, and 15, where the neutral and ions have similar structures. Beginning with n=6, all atoms are three coordinate, except for (GaAs)10+ and (GaAs)13+. For the larger species, there is a competition between fullerenes built from hexagons and rhombi and geometrical configurations where Ga-Ga and As-As bonds are formed, which results in the formation of pentagons. As expected, the static polarizability varies in the order of anion>neutral>cation, but the values are rather similar for all three charge states. The thermodynamic stability for the loss of GaAs is reported.

Dissociative and associative attachment of NO to iron clusters.

Electronic and geometrical structures of iron clusters with associative (FeNO, Fe2NO, Fe3NO, Fe4NO, Fe5NO, and Fe6NO) and dissociative (OFeN, OFe2N, OFe3N, OFe4N, OFe5N, and OFe6N) attachments of NO, as well as the corresponding singly negatively and positively charged ions, are computed using density functional theory with generalized gradient corrections. Both types of isomers are found to be stable and no spontaneous dissociation was observed during the geometry optimizations. The ground states correspond to dissociative attachment of NO for all iron clusters Fe(n), except for Fe and Fe+. All of the OFe(n)N clusters have ferrimagnetic ground states, except for OFe2N, OFe2N-, OFe4N, and OFe4N-, which prefer the ferromagnetic coupling. In the ferrimagnetic states, the excess spin density at one iron atom couples antiferromagnetically to the excess spin densities of all other iron atoms. Relative to the high-spin Fe(n) ground state, the lowest energy ferrimagnetic state quenches the total magnetic moments of iron clusters by 7, which is to be compared with a reduction in the magnetic moment of one in the lowest energy ferromagnetic states. Dissociation of NO on the iron clusters has a pronounced impact on the energetics of reactions; the Fe(n)NO+CO-->Fe(n)N+CO2 channels are exothermic while the OFe6N+CO--> Fe6N+CO2 channels are nearly thermoneutral.

Structure and properties of Mn(n), Mn(n)-, and Mn(n)+ clusters (n = 3-10).

Electronic and geometrical structures of Mn(3)-Mn(10) together with their singly negatively and positively charged ions are computed using density functional theory with generalized gradient approximation. The ground-state spin multiplicities in the neutral series are 16, 21, 4, 9, 6, 5, 2, and 5, for Mn(3)-Mn(10), respectively. Thus, there is a transition from a ferromagnetic ground state to a ferrimagnetic ground state at Mn(5). The energy difference between ferrimagnetic and ferromagnetic states in Mn(n) grows rapidly with increasing n and exceeds 2 eV in Mn(10). The corresponding change from ferro- to ferrimagnetic ground state occurs at Mn(6)(-) and Mn(3)(+) in the anionic and cationic series, respectively. Beginning with Mn(6), the ion spin multiplicities differ from that of the neutral by +/-1 (i.e., they obey the empirical "+/-1 rule"). We found that the energy required to remove an Mn atom is nearly independent of the charge state of an Mn(n) cluster and the number of atoms in the cluster, except for Mn(3). The results of our calculations are in reasonable agreement with experiment, except for the experimental data on the magnetic moments per atom, where, in general, we predict smaller values than the experiment.

All-electron DFT modeling of SWCNT growth on iron catalysts from carbon monoxide feedstock.

Electronic and geometrical structures of Fe4Cn(CO)m (n + m < or = 6) and their singly negatively and positively charged ions are computed using density functional theory with generalized gradient approximation. Isomers with CO bonded directly to the cluster iron atoms and bonded to a carbon atom chemisorbed on the cluster surface are optimized for the Fe4C2CO, Fe4C2(CO)2, Fe4C3CO, and Fe4C4CO series. Optimizations of a large number of differently shaped Fe4C4, Fe4C5, and Fe4C6 clusters are performed to find trends in preferable arrangement of carbon atoms, in particular, to determine the relative energetics of structures with single C atoms versus those with C2 dimers or C3 trimers. The computed total energies are used to estimate the energetics of the Boudouard disproportionation reactions Fe4Cn(CO)m + 00-->Fe4Cn+1(CO)m-1 + CO2.

Periodic table of 3d-metal dimers and their ions.

The ground states of the mixed 3d-metal dimers TiV, TiCr, TiMn, TiFe, TiCo, TiNi, TiCu, TiZn, VCr, VMn, VFe, VCo, VNi, VCu, VZn, CrMn, CrFe, CrCo, CrNi, CrCu, CrZn, MnFe, MnCo, MnNi, MnCu, MnZn, FeCo, FeNi, FeCu, FeZn, CoNi, CoCu, CoZn, NiCu, NiZn, and CuZn along with their singly negatively and positively charged ions are assigned based on the results of computations using density functional theory with generalized gradient approximation for the exchange-correlation functional. Except for TiCo and CrMn, our assignment agrees with experiment. Computed spectroscopic constants (r(e),omega(e),D(o)) are in fair agreement with experiment. The ground-state spin multiplicities of all the ions are found to differ from the spin multiplicities of the corresponding neutral parents by +/-1. Except for TiV, MnFe, and MnCu, the number of unpaired electrons, N, in a neutral ground-state dimer is either N(1)+N(2) or mid R:N(1)-N(2)mid R:, where N(1) and N(2) are the numbers of unpaired 3d electrons in the 3d(n)4s(1) occupation of the constituent atoms. Combining the present and previous results obtained at the same level of theory for homonuclear 3d-metal and ScX (X=Ti-Zn) dimers allows one to construct "periodic" tables of all 3d-metal dimers along with their singly charged ions.

Competition between linear and cyclic structures in monochromium carbide clusters CrCn- and CrCn (n = 2-8): a photoelectron spectroscopy and density functional study.

Photoelectron spectroscopy (PES) is combined with density functional theory (DFT) to study the monochromium carbide clusters CrCn- and CrCn (n = 2-8). Well-resolved PES spectra were obtained, yielding structural, electronic, and vibrational information about both the anionic and neutral clusters. Experimental evidence was observed for the coexistence of two isomers for CrC2-, CrC3-, CrC4-, and CrC6-. Sharp and well-resolved PES spectra were observed for CrCn- (n = 4,6,8), whereas broad spectra were observed for CrC5- and CrC7-. Extensive DFT calculations using the generalized gradient approximation were carried out for the ground and low-lying excited states of all the CrCn- and CrCn species, as well as coupled-cluster calculations for CrC2- and CrC2. Theoretical electron affinities and vertical detachment energies were calculated and compared with the experimental data to help the assignment of the ground states and obtain structural information. We found that CrC2- and CrC3- each possess a close-lying cyclic and linear structure, which were both populated experimentally. For the larger CrCn- clusters with n = 4, 6, 8, linear structures are the overwhelming favorite, giving rise to the sharp PES spectral features. CrC7- was found to have a cyclic structure. The broad PES spectra of CrC5- suggested a cyclic structure, whereas the DFT results predicted a linear one.

Spectroscopic and theoretical investigations of vibrational frequencies in binary unsaturated transition-metal carbonyl cations, neutrals, and anions.

Figure 18 presents the C-O stretching vibrational frequencies of the first-row transition-metal monocarbonyl cations, neutrals, and anions in solid neon; similar diagrams have been reported for neutral MCO species in solid argon, but three of the early assignments have been changed by recent work and one new assignment added. The laser-ablation method produces mostly neutral atoms with a few percent cations and electrons for capture to make anions; in contrast, thermal evaporation gives only neutral species. Hence, the very recent neon matrix investigations in our laboratory provide carbonyl cations and anions for comparison to neutrals on a level playing field. Several trends are very interesting. First, for all metals, the C-O stretching frequencies follow the order cations > neutrals > anions with large diagnostic 100-200 cm-1 separations, which is consistent with the magnitude of the metal d to CO pi * donation. Second, for a given charge, there is a general increase in C-O stretching vibrational frequencies with increasing metal atomic number, which demonstrates the expected decrease in the metal to CO pi * donation with increasing metal ionization potential. Some of the structure in this plot arises from the extra stability of the filled and half-filled d shell and from the electron pairing that occurs at the middle of the TM row; the plot resembles the "double-humped" graph found for the variation in properties across a row of transition metals. For the anions, the variation with metal atom is the smallest since all of the metals can easily donate charge to the CO ligand. Third, for the early transition-metal Ti, V, and Cr families, the C-O stretching frequencies decrease when going down the family, but the reverse relationship is observed for the late transition-metal Fe, Co, and Ni families. In most of the present discussion, we have referred to neon matrix frequencies; however, the argon matrix frequencies are complementary, and useful information can be obtained from comparison of the two matrix hosts. In most cases, the neon-to-argon red shift for neutral carbonyls is from 11 to 26 cm-1, but a few (CrCO) lie outside of this range. In the case of FeCO and Fe(CO)2, it appears that neon and argon trap different low-lying electronic states. In general, the carbonyl neutrals and anions have similar shifts but carbonyl cations have larger matrix shifts. For example, the FeCO+ fundamental is at 2123.0 cm-1 in neon and 2081.5 cm-1 in argon, a 42.5 cm-1 shift, which is larger than those found for FeCO- (11.7 cm-1) and FeCO (11.7 cm-1). It is unusual for different low-lying electronic states to be trapped in different matrices, but CUO provides another example. The linear singlet state (1047.3, 872.2 cm-1) is trapped in solid neon, and a calculated 1.2 kcal/mol higher triplet state is trapped in solid argon (852.5, 804.3 cm-1) and stabilized by a specific interaction with argon. The bonding trends are well described by theoretical calculations of vibrational frequencies. Table 5 compares the scale factors (observed neon matrix/calculated) for the C-O stretching modes of the monocarbonyl cations, neutrals, and anions of the first-row transition metals observed in a neon matrix using the B3LYP and BP86 density functionals. Most of the calculated carbonyl harmonic stretching frequencies are within 1% of the experimental fundamentals at the BP86 level of theory, while calculations using the B3LYP functional give frequencies that are 3-4% higher as expected for these density functionals and calculations on saturated TM-carbonyls. For second- and third-row carbonyls using the BP86 density functional and the LANL effective core potential in conjunction with the DZ basis set, the agreement between theory and experiment is just as good. For example, the 16 M(CO)1-4 neutral and anion and 2 MCO+ cation (M = Ru, Os) carbonyl frequencies are fit within 1.5%. The 16 species (M = Rh, Ir) are fit within 1%, but the Rh(CO)1-4+ calculations are 2-3% too low and Ir(CO)1-4+ computations are 1-2% too low. In addition to predicting the vibrational frequencies, DFT can be used to calculate different isotopic frequencies, and isotopic frequency ratios can be computed as a measure of the normal vibrational mode in the molecule for an additional diagnostic. For diatomic CO, the 12CO/13CO ratio 1.0225 and C16O/C18O ratio 1.0244 characterize a pure C-O stretching mode. In a series of molecules such as RhCO+, RhCO, and RhCO-, where the metal-CO bonding varies, the Rh-C, C-O vibrational interaction is different and the unique isotopic ratios for the carbonyl vibration are characteristic of that particular molecule. Table 6 summarizes the isotopic ratios observed and calculated for the RhCO+,0,- species. Note that RhCO+ exhibits slightly more carbon-13 and less oxygen-18 involvement in the C-O vibration than CO itself and that this trend increases to RhCO and to RhCO- as the Rh-C bond becomes shorter and stronger. Note also how closely the calculated and observed ratios both follow this trend. In a molecule with two C-O stretching modes, for example, bent Ni(CO)2 exhibits a strong b2 mode at 1978.9 cm-1 and a weak a1 mode at 2089.7 cm-1 in solid neon, and these two modes involve different C and O participations. The symmetric mode shows substantially more C (1.0242) and less O (1.0217) participation than does the antisymmetric mode with C (1.0228) and O (1.0238) involvement, based on the given isotopic frequency ratios, which are nicely matched by DFT calculations (a1 1.0244, 1.0224 and b2 1.0232, 1.0241, respectively). These investigations of vibrational frequencies in unsaturated transition-metal carbonyl cations, neutrals, and anions clearly demonstrate the value of a close working relationship between experiment and theory to identify and characterize new molecular species.

Closed-shell polycyclic aromatic hydrocarbon cations: a new category of interstellar polycyclic aromatic hydrocarbons.

Density functional theory has been employed to calculate the harmonic frequencies and intensities of a range of polycyclic aromatic hydrocarbon (PAH) cations that explore both size and electronic structure effects on the infrared spectroscopic properties of these species. The sample extends the size range of PAH species considered to more than 50 carbon atoms and includes several representatives from each of two heretofore unexplored categories of PAH cations: (1) fully benzenoid PAH cations whose carbon skeleton is composed of an odd number of carbon atoms (C(odd) PAHs); and (2) protonated PAH cations (HPAH+). Unlike the radical electronic structures of the PAH cations that have been the subject of previous theoretical and experimental work, the species in these two classes have a 'closed'-shell electronic configuration. The calculated spectra of circumcoronene, C54H18, in both neutral and (radical) cationic form are also reported and compared with those of the other species. Overall, the C(odd) PAHs spectra are dominated by strong CC stretching modes near 1600 cm(-1) and display spectra that are remarkably insensitive to molecular size. The HPAH+ species evince a more complex spectrum consistent with the added contributions of aliphatic modes and their generally lower symmetry. Finally, for both classes of closed-shell cations, the intensity of the aromatic CH stretching modes is found to increase with molecular size far out of proportion with the number of CH groups, approaching a value more typical of neutral PAHs for the largest species studied.

Carbon chain abundance in the diffuse interstellar medium.

Thanks to the mid-IR sensitivities of the ISO and IRTS orbiting spectrometers it is now possible to search the diffuse interstellar medium for heretofore inaccessible molecular emission. In view of the recent strong case for the presence of C(7-) (Kirkwood et al. 1998, Tulej et al. 1998),and the fact that carbon chains possess prominent infrared active modes in a very clean portion of the interstellar spectrum, we have analyzed the IRTS spectrum of the diffuse interstellar medium for the infrared signatures of these species. Theoretical and experimental infrared band frequencies and absolute intensities of many different carbon chain species are presented. These include cyanopolyynes, neutral and anionic linear carbon molecules, and neutral and ionized, even-numbered, hydrogenated carbon chains. We show that--as a family--these species have abundances in the diffuse ISM on the order of 10(-10) with respect to hydrogen, values consistent with their abundances in dense molecular clouds. Assuming an average length of 10 C atoms per C-chain implies that roughly a millionth of the cosmically available carbon is in the form of carbon chains and that carbon chains can account for a few percent of the visible to near-IR diffuse interstellar band (DIB) total equivalent width (not DIB number).

Infrared spectra of substituted polycylic aromatic hydrocarbons.

Calculations are carried out using density functional theory (DFT) to determine the harmonic frequencies and intensities of 1-methylanthracene, 9-methylanthracene, 9-cyanoanthracene, 2-aminoanthracene, acridine, and their positive ions. The theoretical data are compared with matrix-isolation spectra for these species also reported in this work. The theoretical and experimental frequencies and relative intensities for the neutral species are in generally good agreement, whereas the positive ion spectra are only in qualitative agreement. Relative to anthracene, we find that substitution of a methyl or CN for a hydrogen does not significantly affect the spectrum other than to add the characteristic methyl C-H and C triple bond N stretches near 2900 and 2200 cm-1, respectively. However, addition of NH2 dramatically affects the spectrum of the neutral. Not only are the NH2 modes themselves strong, but this electron-withdrawing group induces sufficient partial charge on the ring to give the neutral molecule spectra characteristics of the anthracene cation. The sum of the absolute intensities is about four times larger for 2-aminoanthracene than those for 9-cyanoanthracene. Substituting nitrogen in the ring at the nine position (acridine) does not greatly alter the spectrum compared with anthracene.

Quantum mechanical calculations to chemical accuracy.

Full configuraton-interaction (FCI) calculations have given an unambiguous standard by which the accuracy of theoretical approaches of incorporating electron correlation into molecular structure calculations can be judged. In addition, improvements in vectorization of programs, computer technology, and algorithms now permit a systematic study of the convergence of the atomic orbital (or so-called one-particle) basis set. These advances are discussed and some examples of the solution of chemical problems by quantum mechanical calculations are given to illustrate the accuracy of current techniques.