A novel explanation for the increased conductivity in annealed Al-doped ZnO: an insight into migration of aluminum and displacement of zinc.

A combined experimental and first-principles study is performed to study the origin of conductivity in ZnO:Al nanoparticles synthesized under controlled conditions via a reflux route using benzylamine as a solvent. The experimental characterization of the samples by Raman, nuclear magnetic resonance (NMR) and conductivity measurements indicates that upon annealing in nitrogen, the Al atoms at interstitial positions migrate to the substitutional positions, creating at the same time Zn interstitials. We provide evidence for the fact that the formed complex of AlZn and Zni corresponds to the origin of the Knight shifted peak (KS) we observe in 27Al NMR. As far as we know, the role of this complex has not been discussed in the literature to date. However, our first-principles calculations show that such a complex is indeed energetically favoured over the isolated Al interstitial positions. In our calculations we also address the charge state of the Al interstitials. Further, Zn interstitials can migrate from AlZn and possibly also form Zn clusters, leading to the observed increased conductivity.

Structural and electronic properties of defects at grain boundaries in CuInSe2.

We report on a first-principles study of the structural and electronic properties of a Σ3 (112) grain boundary model in CuInSe2. The study focuses on a coherent, stoichiometry preserving, cation-Se terminated grain boundary, addressing the properties of the grain boundary as such, as well as the effect of well known defects in CuInSe2. We show that in spite of its apparent simplicity, such a grain boundary exhibits a very rich phenomenology, providing an explanation for several of the experimentally observed properties of grain boundaries in CuInSe2 thin films. In particular, we show that the combined effect of Cu vacancies and cation antisites can result in the observed Cu depletion with no In enrichment at the grain boundaries. Furthermore, Cu vacancies are unlikely to produce a hole barrier at the grain boundaries, but Na may indeed have such an effect. We find that Na-on-Cu defects will tend to form abundantly at the grain boundaries, and can provide a mechanism for the carrier depletion and/or type inversion experimentally reported.

Extension of the basis set of linearized augmented plane wave (LAPW) method by using supplemented tight binding basis functions.

In order to increase the accuracy of the linearized augmented plane wave (LAPW) method, we present a new approach where the plane wave basis function is augmented by two different atomic radial components constructed at two different linearization energies corresponding to two different electron bands (or energy windows). We demonstrate that this case can be reduced to the standard treatment within the LAPW paradigm where the usual basis set is enriched by the basis functions of the tight binding type, which go to zero with zero derivative at the sphere boundary. We show that the task is closely related with the problem of extended core states which is currently solved by applying the LAPW method with local orbitals (LAPW+LO). In comparison with LAPW+LO, the number of supplemented basis functions in our approach is doubled, which opens up a new channel for the extension of the LAPW and LAPW+LO basis sets. The appearance of new supplemented basis functions absent in the LAPW+LO treatment is closely related with the existence of the u̇l-component in the canonical LAPW method. We discuss properties of additional tight binding basis functions and apply the extended basis set for computation of electron energy bands of lanthanum (face and body centered structures) and hexagonal close packed lattice of cadmium. We demonstrate that the new treatment gives lower total energies in comparison with both canonical LAPW and LAPW+LO, with the energy difference more pronounced for intermediate and poor LAPW basis sets.

The role of the VZn-NO-H complex in the p-type conductivity in ZnO.

Past research efforts aiming at obtaining stable p-type ZnO have been based on complexes involving nitrogen doping. A recent experiment by (J. G. Reynolds et al., Appl. Phys. Lett., 2013, 102, 152114) demonstrated a significant (∼10(18) cm(-3)) p-type behavior in N-doped ZnO films after appropriate annealing. The p-type conductivity was attributed to a VZn-NO-H shallow acceptor complex, formed by a Zn vacancy (VZn), N substituting O (NO), and H interstitial (Hi). We present here a first-principles hybrid functional study of this complex compared to the one without hydrogen. Our results confirm that the VZn-NO-H complex acts as an acceptor in ZnO. We find that H plays an important role, because it lowers the formation energy of the complex with respect to VZn-NO, a complex known to exhibit (unstable) p-type behavior. However, this additional H atom also occupies the hole level at the origin of the shallow behavior of VZn-NO, leaving only two states empty higher in the band gap and making the VZn-NO-H complex a deep acceptor. Therefore, we conclude that the cause of the observed p-type conductivity in experiment is not the presence of the VZn-NO-H complex, but probably the formation of the VZn-NO complex during the annealing process.

Native point defects in CuIn(1-x)Ga(x)Se2: hybrid density functional calculations predict the origin of p- and n-type conductivity.

We have performed a first-principles study of the p- and n-type conductivity in CuIn(1-x)Ga(x)Se2 due to native point defects, based on the HSE06 hybrid functional. Band alignment shows that the band gap becomes larger with x due to the increasing conduction band minimum, rendering it hard to establish n-type conductivity in CuGaSe2. From the defect formation energies, we find that In/GaCu is a shallow donor, while V(Cu), V(In/Ga) and Cu(In/Ga) act as shallow acceptors. Using the total charge neutrality of ionized defects and intrinsic charge carriers to determine the Fermi level, we show that under In-rich growth conditions InCu causes strongly n-type conductivity in CuInSe2. Under increasingly In-poor growth conditions, the conductivity type in CuInSe2 alters to p-type and compensation of the acceptors by In(Cu) reduces, as also observed in photoluminescence experiments. In CuGaSe2, the native acceptors pin the Fermi level far away from the conduction band minimum, thus inhibiting n-type conductivity. On the other hand, CuGaSe2 shows strong p-type conductivity under a wide range of Ga-poor growth conditions. Maximal p-type conductivity in CuIn(1-x)Ga(x)Se2 is reached under In/Ga-poor growth conditions, in agreement with charge concentration measurements on samples with In/Ga-poor stoichiometry, and is primarily due to the dominant acceptor Cu(In/Ga).

The origin of p-type conductivity in ZnM2O4 (M = Co, Rh, Ir) spinels.

ZnM2O4 (M = Co, Rh, Ir) spinels are considered as a class of potential p-type transparent conducting oxides (TCOs). We report the formation energy of acceptor-like defects using first principles calculations with an advanced hybrid exchange-correlation functional (HSE06) within density functional theory (DFT). Due to the discrepancies between the theoretically obtained band gaps with this hybrid functional and the - scattered - experimental results, we also perform GW calculations to support the validity of the description of these spinels with the HSE06 functional. The considered defects are the cation vacancy and antisite defects, which are supposed to be the leading source of disorder in the spinel structures. We also discuss the band alignments in these spinels. The calculated formation energies indicate that the antisite defects ZnM (Zn replacing M, M = Co, Rh, Ir) and VZn act as shallow acceptors in ZnCo2O4, ZnRh2O4 and ZnIr2O4, which explains the experimentally observed p-type conductivity in those systems. Moreover, our systematic study indicates that the ZnIr antisite defect has the lowest formation energy in the group and it corroborates the highest p-type conductivity reported for ZnIr2O4 among the group of ZnM2O4 spinels. To gain further insight into factors affecting the p-type conductivity, we have also investigated the formation of localized small polarons by calculating the self-trapping energy of the holes.

Casting light on the darkening of colors in historical paintings.

The degradation of colors in historical paintings affects our cultural heritage in both museums and archeological sites. Despite intensive experimental studies, the origin of darkening of one of the most ancient pigments known to humankind, vermilion (α-HgS), remains unexplained. Here, by combining many-body theoretical spectroscopy and high-resolution microscopic x-ray diffraction, we clarify the composition of the damaged paint work and demonstrate possible physicochemical processes, induced by illumination and exposure to humidity and air, that cause photoactivation of the original pigment and the degradation of the secondary minerals. The results suggest a new path for the darkening process which was never considered by previous studies and prompt a critical examination of their findings.

Perovskite transparent conducting oxides: an ab initio study.

We present an ab initio study of the electronic structure and of the formation energies of various point defects in BaSnO3 and SrGeO3. We show that La and Y impurities substituting Ba or Sr are shallow donors with a preferred 1 + charge state. These defects have a low formation energy within all the suitable equilibrium growth conditions considered. Oxygen vacancies behave as shallow donors as well, preferring the 2 + charge state. Their formation energies, however, are higher in most growth conditions, indicating a limited contribution to conductivity. The calculated electron effective mass in BaSnO3, with a value of 0.21 m(e), and the very high mobility reported recently in La-doped BaSnO3 single-crystals, suggest that remarkably low scattering rates can be achieved in the latter. In the case of SrGeO3, our results point to carrier density and mobility values in the low range for typical polycrystalline TCOs, in line with experiment.

Origin of magnetism and quasiparticles properties in Cr-doped TiO2.

Combining the local spin density approximation (LSDA)+U and an analysis of superexchange interactions beyond density functional theory, we describe the magnetic ground state of Cr-doped TiO2, an intensively studied and debated dilute magnetic oxide. In parallel, we correct our LSDA+U (+ superexchange) ground state through GW corrections (GW@LSDA+U) that reproduce the position of the impurity states and the band gaps in satisfying agreement with experiments. Because of the different topological coordinations of Cr-Cr bonds in the ground states of rutile and anatase, superexchange interactions induce either ferromagnetic or antiferromagnetic couplings of Cr ions. In Cr-doped anatase, this interaction leads to a new mechanism which stabilizes a (nonrobust) ferromagnetic ground state, in keeping with experimental evidence, without the need to invoke F-center exchange. The interplay between structural defects and vacancies in contributing to the superexchange is also unveiled.

Quasiparticle band structure of rocksalt-CdO determined using maximally localized Wannier functions.

CdO in the rocksalt structure is an indirect band gap semiconductor. Thus, in order to determine its band gap one needs to calculate the complete band structure. However, in practice, the exact evaluation of the quasiparticle band structure for the large number of k-points which constitute the different symmetry lines in the Brillouin zone can be an extremely demanding task compared to the standard density functional theory (DFT) calculation. In this paper we report the full quasiparticle band structure of CdO using a plane-wave pseudopotential approach. In order to reduce the computational effort and time, we make use of maximally localized Wannier functions (MLWFs). The MLWFs offer a highly accurate method for interpolation of the DFT or GW band structure from a coarse k-point mesh in the irreducible Brillouin zone, resulting in a much reduced computational effort. The present paper discusses the technical details of the scheme along with the results obtained for the quasiparticle band gap and the electron effective mass.

Electronic structure of transparent oxides with the Tran-Blaha modified Becke-Johnson potential.

We present electronic band structures of transparent oxides calculated using the Tran-Blaha modified Becke-Johnson (TB-mBJ) potential. We studied the basic n-type conducting binary oxides In(2)O(3), ZnO, CdO and SnO(2) along with the p-type conducting ternary oxides delafossite CuXO(2) (X=Al, Ga, In) and spinel ZnX(2)O(4) (X=Co, Rh, Ir). The results are presented for calculated band gaps and effective electron masses. We discuss the improvements in the band gap determination using TB-mBJ compared to the standard generalized gradient approximation (GGA) in density functional theory (DFT) and also compare the electronic band structure with available results from the quasiparticle GW method. It is shown that the calculated band gaps compare well with the experimental and GW results, although the electron effective mass is generally overestimated.

The quasiparticle band structure of zincblende and rocksalt ZnO.

We present the quasiparticle band structure of ZnO in its zincblende (ZB) and rocksalt (RS) phases at the Γ point, calculated within the GW approximation. The effect of the p-d hybridization on the quasiparticle corrections to the band gap is discussed. We compare three systems, ZB-ZnO which shows strong p-d hybridization and has a direct band gap, RS-ZnO which is also hybridized but includes inversion symmetry and therefore has an indirect band gap, and ZB-ZnS which shows a weaker hybridization due to a change of the chemical species from oxygen to sulfur. The quasiparticle corrections are calculated with different numbers of valence electrons in the Zn pseudopotential. We find that the Zn(20+) pseudopotential is essential for the adequate treatment of the exchange interaction in the self-energy. The calculated GW band gaps are 2.47 eV and 4.27 eV respectively, for the ZB and RS phases. The ZB-ZnO band gap is underestimated compared to the experimental value of 3.27 by ∼ 0.8 eV. The RS-ZnO band gap compares well with the experimental value of 4.5 eV. The underestimation for ZB-ZnO is correlated with the strong p-d hybridization. The GW band gap for ZnS is 3.57 eV, compared to the experimental value of 3.8 eV.

Temperature-dependent Debye-Waller factors for semiconductors with the wurtzite-type structure.

We computed Debye-Waller factors in the temperature range from 0.1 to 1000 K for AlN, GaN, InN, ZnO and CdO with the wurtzite-type structure. The Debye-Waller factors were derived from phonon densities of states obtained from Hellmann-Feynman forces computed within the density-functional-theory formalism. The temperature dependences of the Debye-Waller factors were fitted and fit parameters are given.

Computation and parametrization of the temperature dependence of Debye-Waller factors for group IV, III-V and II-VI semiconductors.

We calculated the temperature dependence of the Debye-Waller factors for a variety of group IV, III-V and II-VI semiconductors from 0.1 to 1000 K. The approach used to fit the temperature dependence is described and resulting fit parameters are tabulated for each material. The Debye-Waller factors are deduced from generalized phonon densities of states which were derived from first principles using the WIEN2k and the ABINIT codes.

Determination of the mean inner potential in III-V semiconductors, Si and Ge by density functional theory and electron holography.

The mean inner potentials of various III-V semiconductors, Si and Ge have been calculated by density functional theory methods. For that purpose, the Coulomb potential of slabs consisting of a crystal and vacuum region has been computed and averaged inside the crystal region. The computed values are in agreement with experimental values obtained by electron holography for Si and GaAs. For the other semiconductors, the deviations are smaller than 0.8 V. The results from density functional theory are approximately 10% smaller than the values derived from atomic scattering factors computed by Hartree Fock calculations.

Valence electronic charge density of distorted C60- monomers in polymerized KC60 and RbC60.

We investigate the valence electronic charge density of the C(60) (-) monomers in (C(60) (-))(n) polymer chains in K- and RbC(60) by means of a nonorthogonal tight-binding formalism using experimental data on the positions of the carbon atoms. Various configurations of the C(60) cages are considered. Starting from the ideal icosahedral C(60) structure and moving to the realistic, experimentally determined spatial configuration of the C(60) cages in K- and RbC(60), we observe a systematic increase of the electric quadrupole moments on the C(60) (-) monomers. We also confirm the validity of factorizing the charge density of a C(60) (-) monomer into an angular and a radial part.