Quantifying local and cooperative components in the ferroelectric distortion of BaTiO(3): learning from the off-center motion in the MnCl(6)(5-) complex formed in KCl:Mn(+).

The delicate balance between cooperative and local contributions in the ferroelectric distortions of BaTiO3 is explored by means of ab initio calculations. As a salient feature, it is found that a single Ti(4+) ion in BaTiO3 is not allowed to move off-center at ambient pressure, while this is no longer true if the lattice is expanded by only ∼5%, stressing the high sensitivity of the local contribution to chemical and hydrostatic pressures. In order to further understand the effect of local contributions on the phase transition mechanism of ferroelectrics, we have investigated the surprising C3v → C4v → Oh local transformations occurring in the 10-50 K temperature range for the MnCl6(5-) complex formed in KCl:Mn(+) that mimic the behavior of BaTiO3. From Boltzmann analysis of the vibronic levels derived from ab initio calculations and considering decoherence introduced by random strains, the present calculations reproduce the experimental phase sequence and transition temperatures. Furthermore, our calculations show that the off-center instability in KCl:Mn(+) would be suppressed by reducing by only 1% the lattice parameter, a situation that then becomes comparable to that found for BaTiO3 at ambient pressure. The present results thus stress the deep link between the structural phase transitions of ferroelectric materials and local phase transitions displayed by transition-metal impurities in insulators.

Crystal fields of porphyrins and phthalocyanines from polarization-dependent 2p-to-3d multiplets.

Polarization-dependent X-ray absorption spectroscopy is combined with density functional calculations and atomic multiplet calculations to determine the crystal field parameters 10Dq, Ds, and Dt of transition metal phthalocyanines and octaethylporphyrins (Mn, Fe, Co, Ni). The polarization dependence facilitates the assignment of the multiplets in terms of in-plane and out-of-plane orbitals and avoids ambiguities. Crystal field values from density functional calculations provide starting values close to the optimum fit of the data. The resulting systematics of the crystal field can be used for optimizing electron-hole separation in dye-sensitized solar cells.

Sharp lines due to Cr³âº and Mn²âº impurities in insulators: going beyond the usual Tanabe-Sugano approach.

This work is aimed at understanding the different behavior of optical sharp lines (corresponding to 10Dq-independent transitions) of Mn(2+) and Cr(3+) in normal and inverted perovskites that cannot be explained within the usual Tanabe-Sugano approach. In particular, we want to clarify why on passing from KMgF3:M to LiBaF3:M (M = Mn(2+), Cr(3+)) the energy, E((6)A1 → (4)A1), for Mn(2+) decreases by Δ = 1100 cm(-1), while Δ < 100 cm(-1) for the energy E((2)E →( 4)A2) corresponding to Cr(3+). The origin of this surprising difference in these model systems is clarified by writing the transition energies of MF6 complexes through the ten Coulomb and exchange integrals consistent with the cubic symmetry and not considered in the usual Tanabe-Sugano approach. It is shown that E((6)A1 → (4)A1) depends on exchange integrals K(3z(2) - r(2), xy) and K(x(2) - y(2), xy), while E((2)E → (4)A2) depends on K(xz, yz) where the two involved electrons display a π character. These exchange integrals have been calculated just considering a MF6 unit subject to the internal electric field due to the rest of the lattice ions. In addition to a reasonably reproduction of the main trends observed experimentally for the model systems, the present calculations prove that the exchange integrals are not related in a simple way to the covalency of involved orbitals. Particular attention is also paid to explain why the transitions, which are 10Dq-independent are less sensitive to the host lattice change than those which do depend on 10Dq. The present work shows that K(xz, yz) for Cr(3+) is particularly insensitive to the host lattice change and thus sheds light on the origin of the near independence of E((2)E → (4)A2) along the series of oxides doped with such an impurity .

Origin of small barriers in Jahn-Teller systems: quantifying the role of 3d-4s hybridization in the model system NaCl:Ni+.

Despite its relevance, the microscopic origin of the energy barrier, B, between the compressed and elongated geometries of Jahn-Teller (JT) systems is not well understood yet because of a lack of quantitative data about its various contributions. Seeking to clear up this matter, we have carried out both periodic and cluster ab initio calculations on the model system NaCl:Ni(+). This system is particularly puzzling because, according to experimental data, its barrier is much smaller than that for other d(9) and d(7) ions in similar lattices. All calculations performed on the model system lead, in fact, to values |B| ≤ 160 cm(-1), which are certainly smaller than B = 500 cm(-1) derived for NaCl:M(2+) (M = Ag, Rh) or B = 1024 cm(-1) obtained for KCl:Ag(2+). As a salient feature, analysis of calculations carried out as a function of the Qθ (∼3z(2) - r(2)) coordinate unveils the microscopic origin of the barrier. It is quantitatively proven that the elongated geometry observed for NaCl:Ni(+) is due to the 3d-4s vibronic admixture, which is slightly larger than the anharmonicity in the eg JT mode that favors a compressed geometry. The existence of these two competing mechanisms explains the low value of B for the model system, contrary to cases where the complex formed by d(9) or d(7) ions is elastically decoupled from the host lattice. Although the magnitude of B for NaCl:Ni(+) is particularly small, the tunneling splitting, 3Γ, is estimated to be below 9 cm(-1), thus explaining why the coherence is easily destroyed by random strains and thus a static JT effect is observed experimentally. As a main conclusion, the barrier in JT systems cannot be understood neglecting the tiny changes of the electronic density involved in small distortions. The present calculations reasonably explain the experimental g tensor of NaCl:Ni(+), pointing out that the d-d transitions in NiCl6(5-) are much smaller than those for CuCl6(4-) and the optical electronegativity of Ni(+) is only around 1.

Cu2+ in layered compounds: origin of the compressed geometry in the model system K2ZnF4:Cu2+.

Many relevant properties (including superconductivity and colossal magnetoresistance) of layered materials containing Cu(2+), Ag(2+), or Mn(3+) ions are commonly related to the Jahn-Teller instability. Along this line, the properties of the CuF6(4-) complex in the K2ZnF4 layered perovskite have recently been analyzed using a parametrized Jahn-Teller model with an imposed strain [Reinen, D. Inorg. Chem.2012, 51, 4458]. Here, we present results of ab initio periodic supercell and cluster calculations on K2ZnF4:Cu(2+), showing unequivocally that the actual origin of the unusual compressed geometry of the CuF6(4-) complex along the crystal c axis in that tetragonal lattice is due to the presence of an electric field due to the crystal surrounding the impurity. Our calculations closely reproduce the experimental optical spectrum. The calculated values of the equilibrium equatorial and axial Cu(2+)-F(-) distances are, respectively, R(ax) = 193 pm and R(eq) = 204 pm, and so the calculated distortion R(ax) - R(eq) = 11 pm is three times smaller than the estimated through the parametrized Jahn-Teller model. As a salient feature, we find that if the CuF6(4-) complex would assume a perfect octahedral geometry (R(ax) = R(eq) = 203 pm) the antibonding a(1g)*(∼3z(2) - r(2)) orbital is placed above b(1g)*(∼x(2) - y(2)) with a transition energy E((2)A(1g) → (2)B(1g)) = 0.34 eV. This surprising fact stresses that about half the experimental value E((2)A(1g) → (2)B(1g)) = 0.70 eV is not due to the small shortening of the axial Cu(2+)-F(-) distance, but it comes from the electric field, E(R)(r), created by the rest of the lattice ions on the CuF6(4-) complex. This internal field, displaying tetragonal symmetry, is thus responsible for the compressed geometry in K2ZnF4:Cu(2+) and the lack of symmetry breaking behind the ligand relaxation. Moreover, we show that the electronic energy gain in this process comes from bonding orbitals and not from antibonding ones. The present results underline the key role played by ab initio calculations for unveiling all the complexity behind the properties of the model system K2ZnF4:Cu(2+), opening at the same time a window for improving our knowledge on d(9), d(7), or d(4) ions in other layered compounds.

Colour due to Cr3+ ions in oxides: a study of the model system MgO:Cr3+.

Seeking to understand why the cubic centre in MgO:Cr(3+) has the same 10Dq value as emerald, ab initio cluster and periodic supercell calculations have been performed. It is found that the equilibrium Cr(3+)-O(2-) distance, R, in MgO:Cr(3+) is equal to 2.03 Å and thus 0.06 Å higher than that measured for the emerald. Calculations carried out on the isolated CrO(6)(9-) complex at R = 2.03 Å give 10Dq = 14,510 cm(-1), which is 10% smaller than the experimental figure for MgO:Cr(3+). Nevertheless, when the internal electric field, ER(r), due to the rest of the lattice ions is also taken into account, the calculated 10Dq = 16,210 cm(-1) coincides with the experimental value. Accordingly, the colour shift for different oxides doped with Cr(3+) can be well understood on the basis of this extrinsic contribution to 10Dq usually ignored in a ligand field description. The calculated electrostatic potential, VR(r), related to ER(r), is found to be attractive when the electronic density is lying along <110> directions and |r| > 1 Å driven by the first shell of twelve Mg(2+) ions. The action of VR(r) upon the CrO(6)(9-) complex slightly decreases the energy of t2g(xy,xz,yz) orbitals with respect to that for eg(3z(2) - r(2),x(2) - y(2)) orbitals, thus enhancing the 10Dq value by 0.2 eV. However, the addition of VR(r) induces very small changes in the electronic density, a relevant fact that is related to the (2)E(t(2g)(3)) −> (4)A(2)(t(2g)(3)) emission energy being nearly independent of the host lattice along the series of Cr(3+)-doped oxides.

Electronic structure of Fe- vs. Ru-based dye molecules.

In order to explore whether Ru can be replaced by inexpensive Fe in dye molecules for solar cells, the differences in the electronic structure of Fe- and Ru-based dyes are investigated by X-ray absorption spectroscopy and first-principles calculations. Molecules with the metal in a sixfold, octahedral N cage, such as tris(bipyridines) and tris(phenanthrolines), exhibit a systematic downward shift of the N 1s-to-π* transition when Ru is replaced by Fe. This shift is explained by an extra transfer of negative charge from the metal to the N ligands in the case of Fe, which reduces the binding energy of the N 1s core level. The C 1s-to-π* transitions show the opposite trend, with an increase in the transition energy when replacing Ru by Fe. Molecules with the metal in a fourfold, planar N cage (porphyrins) exhibit a more complex behavior due to a subtle competition between the crystal field, axial ligands, and the 2+ vs. 3+ oxidation states.

Physical and chemical nature of the scaling relations between adsorption energies of atoms on metal surfaces.

Despite their importance in physics and chemistry, the origin and extent of the scaling relations between the energetics of adsorbed species on surfaces remain elusive. We demonstrate here that scalability is not exclusive to adsorbed atoms and their hydrogenated species but rather a general phenomenon between any set of adsorbates bound similarly to the surface. On the example of the near-surface alloys of Pt, we show that scalability is a result of identical variations of adsorption energies with respect to the valence configuration of both the surface components and the adsorbates.

Understanding Periodic Dislocations in 2D Supramolecular Crystals: The PFP/Ag(111) Interface.

In-plane dislocation networks arise in both inorganic and organic films as a way of relieving the elastic strain that builds up at the substrate interface. In molecule/surface systems, supramolecular interactions are weak and more complex (compared to the atomic bonds in inorganic films), and their interplay with molecule-substrate interactions is very subtle, making it difficult to single out the driving force for a nanoscale dislocation pattern. On the basis of a combined experimental and theoretical work, we here show that periodic dislocations in a molecular PFP film are mainly driven by the optimization of molecule-substrate interactions. Compared to inorganic networks however, it implies a much lower energy imbalance, allowing a thermally induced transition from a low-energy strain dislocation pattern to a high-energy incommensurate moiré.

Cr3+-doped fluorides and oxides: role of internal fields and limitations of the Tanabe-Sugano approach.

This work is aimed at clarifying the changes on optical spectra of Cr(3+) impurities due to either a host lattice variation or a hydrostatic pressure, which can hardly be understood by means of the usual Tanabe-Sugano (TS) approach assuming that the Racah parameter, B, grows when covalency decreases. For achieving this goal, the optical properties of Cr(3+)-doped LiBaF(3) and KMgF(3) model systems have been explored by means of high level ab initio calculations on CrF(6)(3-) units subject to the electric field, E(R)(r), created by the rest of the lattice ions. These calculations, which reproduce available experimental data, indicate that the energy, E((2)E), of the (2)E(t(2g)(3)) → (4)A(2)(t(2g)(3)) emission transition is nearly independent of the host lattice. By contrast, the energy difference corresponding to (4)A(2)(t(2g)(3)) → (4)T(1)(t(2g)(2)e(g)(1)) and (4)A(2)(t(2g)(3)) → (4)T(2)(t(2g)(2)e(g)(1)) excitations, Δ((4)T(1); (4)T(2)), is shown to increase on passing from the normal to the inverted perovskite host lattice despite the increase in covalency, a fact which cannot be accounted for through the usual TS model. Similarly, when the Cr(3+)-F(-) distance, R, is reduced both Δ((4)T(1); (4)T(2)) and the covalency are found to increase. By analyzing the limitations of the usual model, we found surprising results that are shown to arise from the deformation of both 3d(Cr) and ligand orbitals in the antibonding e(g) orbital, which has a σ character and is more extended than the π t(2g) orbital. By contrast, because of the higher stiffness of the t(2g) orbital, the dependence of E((2)E) with R basically follows the corresponding variation of covalency in that level. Bearing in mind the similarities of the optical properties displayed by Cr(3+) impurities in oxides and fluorides, the present results can be useful for understanding experimental data on Cr(3+)-based gemstones where the local symmetry is lower than cubic.

Communication: strong excitonic and vibronic effects determine the optical properties of Li2O2.

The band structure and optical absorption spectrum of lithium peroxide (Li(2)O(2)) is calculated from first-principles using the G(0)W(0) approximation and the Bethe-Salpeter equation, respectively. A strongly localized (Frenkel type) exciton corresponding to the π(∗)→σ(∗) transition on the O(2)(-2) peroxide ion gives rise to a narrow absorption peak around 1.2 eV below the calculated bandgap of 4.8 eV. In the excited state, the internal O(2)(-2) bond is significantly weakened due to the population of the σ(∗) orbital. As a consequence, the bond is elongated by almost 0.5 Å leading to an extreme Stokes shift of 2.6 eV. The strong vibronic coupling entails significant broadening of the excitonic absorption peak in good agreement with diffuse reflectance data on Li(2)O(2) which shows a rather featureless spectrum with an absorption onset around 3.0 eV. These results should be important for understanding the origin of the high potential losses and low current densities, which are presently limiting the performance of Li-air batteries.

Renormalization of optical excitations in molecules near a metal surface.

The lowest electronic excitations of benzene and a set of donor-acceptor molecular complexes are calculated for the gas phase and on the Al(111) surface using the many-body Bethe-Salpeter equation. The energy of the charge-transfer excitations obtained for the gas phase complexes are found to be around 10% lower than the experimental values. When the molecules are placed outside the surface, the enhanced screening from the metal reduces the exciton binding energies by several eVs and the transition energies by up to 1 eV depending on the size of the transition-generated dipole. As a striking consequence we find that close to the metal surface the optical gap of benzene can exceed its quasiparticle gap. A classical image charge model for the screened Coulomb interaction can account for all these effects which, on the other hand, are completely missed by standard time-dependent density functional theory.

Spectrochemical series and the dependence of Racah and 10 Dq parameters on the metal-ligand distance: microscopic origin.

The origin of the spectrochemical series and the different dependence of crystal-field splitting (10Dq) and Racah parameters on the metal-ligand distance, R, is explored through ab initio calculations on Cr(3+)-doped K2NaScF6, Cs2NaYCl6, Cs2NaYBr6, and Cs2NaYI6 lattices. For this purpose both periodic and cluster calculations have been performed. An analysis of ab initio results proves that 10Dq values mostly come from the small admixture of deep nLs ligand orbitals present in the antibonding eg(∼ x(2)-y(2),3z(2)-r(2)) level and not from the dominant covalency with valence nLp ligand orbitals, which is actually responsible for the reduction of Racah parameters. This study thus reveals the microscopic origin of the stronger dependence upon R of 10Dq when compared to that observed for Racah parameters, thus explaining why electronic transitions which are 10Dq-independent give rise to sharp optical bands. As a salient feature, while the covalency with nLp levels increases significantly on passing from CrF6(3-) to CrI6(3-), the nLs admixture in eg is found to be practically unmodified. This fact helps to understand the progressive decrease of 10Dq through the series of CrF6(3-), CrCl6(3-), CrBr6(3-), and CrI6(3-) complexes embedded in the corresponding host lattices when compared at the corresponding equilibrium distance at zero pressure. The growing importance of the nLs admixture is well-depicted using deformation density diagrams on passing from the ground state (4)A2(t2g(3)) to the (4)T2(t2g(2)eg) excited state depicted at several R values.

Copper-phthalocyanine based metal-organic interfaces: the effect of fluorination, the substrate, and its symmetry.

Metal-organic interfaces based on copper-phthalocyanine monolayers are studied in dependence of the metal substrate (Au versus Cu), of its symmetry [hexagonal (111) surfaces versus fourfold (100) surfaces], as well as of the donor or acceptor semiconducting character associated with the nonfluorinated or perfluorinated molecules, respectively. Comparison of the properties of these systematically varied metal-organic interfaces provides new insight into the effect of each of the previously mentioned parameters on the molecule-substrate interactions.

Communication: Systematic shifts of the lowest unoccupied molecular orbital peak in x-ray absorption for a series of 3d metal porphyrins.

Porphyrins are widely used as dye molecules in solar cells. Knowing the energies of their frontier orbitals is crucial for optimizing the energy level structure of solar cells. We use near edge x-ray absorption fine structure (NEXAFS) spectroscopy to obtain the energy of the lowest unoccupied molecular orbital (LUMO) with respect to the N(1s) core level of the molecule. A systematic energy shift of the N(1s) to LUMO transition is found along a series of 3d metal octaethylporphyrins and explained by density functional theory. It is mainly due to a shift of the N(1s) level rather than a shift of the LUMO or a change in the electron-hole interaction of the core exciton.

Optical to ultraviolet spectra of sandwiches of benzene and transition metal atoms: Time dependent density functional theory and many-body calculations.

The optical spectra of sandwich clusters formed by transition metal atoms (titanium, vanadium, and chromium) intercalated between parallel benzene molecules have been studied by time-dependent density functional theory (TDDFT) and many-body perturbation theory. Sandwiches with different number of layers, including infinite chains, are considered. The lowest excitation energy peaks in the spectra are characteristic of the robust bonding in these complexes. The excitation energies vary in a systematic way with the metal atoms and with the cluster size, and so these materials could be used to tune the optical properties according to specific functionality targets. The differences in the spectra could be used to identify relative abundances of isomers with different spins in experimental studies. As a salient feature, this theoretical spectroscopic analysis predicts the metallization of the infinite (TiBz)(infinity) chain, which is not the case of (CrBz)(infinity).

Cr3+ in layered perovskites: do the electron paramagnetic resonance parameters only depend on the impurity-ligand distances?

The actual value of axial, R(ax), and equatorial, R(eq), impurity-ligand distances for Cr(3+) embedded in tetragonal K(2)MgX(4) (X = F, Cl) lattices has been explored by means of density functional theory (DFT) calculations on clusters involving up to 69 ions using two different functionals. For K(2)MgF(4):Cr(3+) R(eq) and R(ax) are found to be coincident within only 0.5 pm. When the g tensor of K(2)MgF(4):Cr(3+) is derived considering only the CrF(6)(3-) unit in vacuo at the calculated equilibrium geometry the g(⊥)-g(||) quantity fails to reproduce the experimental value by one order of magnitude. In contrast, when the active electrons localized in the CrX(6)(3-) complex (X = F, Cl) are allowed to feel the anisotropic electric field coming from the rest of the lattice ions the splitting in the first excited state, (4)T(2), increases by one order of magnitude. The present results thus show that the g tensor anisotropy and the zero-field splitting constant, D, observed for K(2)MgX(4):Cr(3+) (X = F, Cl) are not mainly due to a local deformation of the CrX(6)(3-) octahedron but to the action of the internal electric field, often ignored when seeking the microscopic origin of electronic properties due to impurities in insulating lattices. Accordingly, serious doubts on the validity of the superposition model are cast by the present work.

Pseudo-Jahn-Teller origin of the low barrier hydrogen bond in N(2)H(7) (+).

The microscopic origin and quantum effects of the low barrier hydrogen bond (LBHB) in the proton-bound ammonia dimer cation N(2)H(7) (+) were studied by means of ab initio and density-functional theory (DFT) methods. These results were analyzed in the framework of vibronic theory and compared to those obtained for the Zundel cation H(5)O(2) (+). All geometry optimizations carried out using wavefunction-based methods [Hartree-Fock, second and fourth order Moller-Plesset theory (MP2 and MP4), and quadratic configuration interaction with singles and doubles excitations (QCISD)] lead to an asymmetrical H(3)N-H(+)cdots, three dots, centeredNH(3) conformation (C(3v) symmetry) with a small energy barrier (1.26 kcalmol in MP4 and QCISD calculations) between both equivalent minima. The value of this barrier is underestimated in DFT calculations particularly at the local density approximation level where geometry optimization leads to a symmetric H(3)Ncdots, three dots, centeredH(+)cdots, three dots, centeredNH(3) structure (D(3d) point group). The instability of the symmetric D(3d) structure is shown to originate from the pseudo-Jahn-Teller mixing of the electronic (1)A(1g) ground state with five low lying excited states of A(2u) symmetry through the asymmetric alpha(2u) vibrational mode. A molecular orbital study of the pseudo-Jahn-Teller coupling has allowed us to discuss the origin of the proton displacement and the LBHB formation in terms of the polarization of the NH(3) molecules and the transfer of electronic charge between the proton and the NH(3) units (rebonding). The parallel study of the H(5)O(2) (+) cation, which presents a symmetric single-well structure, allows us to analyze why these similar molecules behave differently with respect to proton transfer. From the vibronic analysis, a unified view of the Rudle-Pimentel three-center four-electron and charge transfer models of LBHBs is given. Finally, the large difference in the N-N distance in the D(3d) and C(3v) configurations of N(2)H(7) (+) indicates a large anharmonic coupling between alpha(2u)-alpha(1g) modes along the proton-transfer dynamics. This issue was explored by solving numerically the vibrational Schrodinger equation corresponding to the bidimensional E[Q(alpha(2u)),Q(alpha(1g))] energy surface calculated at the MP46-311++G(**) level of theory.