Competitive segregation of gallium and indium at heterophase Cu-MnO interfaces studied with transmission electron microscopy

This paper concentrates on the possible segregation of indium and gallium and competitive segregation of gallium and indium at atomically flat parallel {111}-oriented Cu-MnO interfaces. The segregation of gallium at Cu-MnO interfaces after introduction of gallium in the copper matrix of internally oxidized Cu-1 at.% Mn could be hardly detected with energy-dispersive spectrometry in a field emission gun transmission electron microscope. After a heat treatment to dissolve indium in the copper matrix, gallium has a weak tendency to segregate, that is 2.5 at.% Ga per monolayer at the interface compared with 2 at.% in the copper matrix. The striking result is that this gallium segregation is observable because it does not occur at the metal side of the interface but in the first two monolayers at the oxide side. Using the same heat treatment as for introducing indium in the sample, but without indium present, gallium segregates strongly at the oxide side of the Cu-MnO interface with a concentration of about 14.3 at.% in each monolayer of the two. In contrast, the presence of gallium has no influence on the segregation of indium towards Cu-MnO interfaces, because the outermost monolayer at the metal side of the interface contains 17.6 at.% In, that is similar to previously found results. This leads to the intriguing conclusions, firstly, that, in contrast with antimony and indium, gallium segregates at the oxide side of the interface and, secondly, that the presence of indium strongly hampers gallium segregation. The results from analytical transmission electron microscopy on gallium segregation are supported by high-resolution transmission electron microscopy observations.     

A Subnanoscale Investigation of Sb Segregation at MnO/Ag Ceramic/Metal Interfaces

We have studied Sb segregation at MnO/Ag(Sb) ceramic/metal heterophase interfaces employing three-dimensional atom-probe (3DAP) microscopy. Specimens are prepared by the internal oxidation of Ag(Mn) alloys, leading to the formation of nanometer-size MnO precipitates within a Ag(Mn) matrix. Sb is introduced into the internally oxidized specimens with a vapor diffusion treatment. Appreciable Sb segregation is observed only after a subsequent segregation anneal is performed, and the measured interfacial excess of Sb at the MnO/Ag(Sb) interfaces, _Sb ^MnO/Ag, is determined directly. The temporal evolution of the MnO precipitates is followed for the different processing steps employed. It is shown that the concentration of silver within the MnO precipitates decreases from an initial value of 45–50 at.% Ag to less than 5 at.% Ag with increasing annealing time at the different processing temperatures. Thus the MnO precipitates form under paraequilibrium conditions and the precipitates inherit Ag from the matrix. With increasing aging time orthoequilibrium conditions prevail and the MnO precipitates reject the silver atoms they inherited from the matrix.     

Diffusion of indium and gallium in Cu(In,Ga)Se2 thin film solar cells

The diffusion of indium and gallium in polycrystalline thin film Cu(In,Ga)Se₂ layers has been investigated. Bilayer structures of CuInSe₂ on top of CuGaSe₂ and vice versa have been fabricated in both a Cu-rich and Cu-poor process (in relation to the ideal stoichiometry). In each process molybdenum coated soda-lime glass with and without a sodium barrier was used. These bilayers were analyzed with secondary ion mass spectrometry, X-ray diffraction, scanning electron microscope and transmission electron microscope equipped with energy dispersive X-ray spectroscopy. It was found that the grain boundary diffusion was not significantly higher than the diffusion inside the grains, also for Cu-rich layers. The diffusion is suggested to mainly proceed via vacant metal sites in the lattice structure. In sodium free films a higher diffusion into the bottom layers, compared to films with sodium, was seen in all cases. This observation was explained with a larger number of vacancies, that facilitates indium and gallium diffusion, in the sodium free films. The difference in diffusion between indium in CGS layers and gallium in CIS layers, in both Cu-rich and Cu-poor processes, was small for layers with sodium.     

Atomic-Scale Structure and Chemistry of Segregation at Matrix/Precipitate Heterophase Interfaces

We study experimentally the local chemistry and atomic structure of heterophase interfaces on an atomic scale. In this work, the heterophase precipitate/matrix interfaces of small molybdenum nitride precipitates in an -iron matrix are investigated on a subnanometer scale by 1-dimensional atom-probe field-ion microscopy (1D-APFIM) and 3-dimensional atom-probe microscopy (3DAPM). Molybdenum nitride precipitates are generated by annealing Fe- 2 at.% Mo- X, where X = 0.4 at.% Sb or 0.5 at.% Sn, at 550°C, in an ammonia/hydrogen atmosphere. Internal nitridation at this temperature produces thin, coherent platelet-shaped molybdenum nitride precipitates. 1D-APFIM selected area analyses are efficient in determining the composition of the precipitates and it is found that a possible Sn or Sb segregation at coherent matrix/precipitate interfaces is either nonexistent or below the detection limit of 1D-APFIM. 3DAPM analyses, however, provide significantly better counting statistics and detect a small, but significant segregation of Sb at the matrix/precipitate interface with a Gibbsian interfacial excess of 0.30 ± 0.15 nm^–2. This is in distinct contrast to the segregation behavior of Sn or Sb at the interfaces of semicoherent coarse precipitates produced by internal nitridation at 600°C, for which much larger Gibbsian interfacial excesses of Sn or Sb, up to 7 ± 3 nm^–2, have been measured. In contrast, the thin platelets are either coherent or have significantly fewer misfit dislocations than is geometrically necessary for a full compensation of the lattice parameter misfit between precipitate and matrix. This demonstrates that Sn or Sb segregation with an appreciable Gibbsian interfacial excess is related to the presence of misfit dislocations at the interfaces of the coarse precipitates.     

Electronic structure and adhesion on metal-aluminum-oxide interfaces

This paper reports on the results of the systematic analysis of the atomic and electronic structure of the Me/α-Al₂O₃(0001) interfaces for two series of isoelectronic metals (Me = Cu, Ag, Au and Ni, Pd, Pt), depending on the termination of the oxide substrate and the configuration of oxide films. The calculations have been performed by the pseudopotential method in the plane-wave basis set. The adhesion energy of metal films has been calculated depending on the cleavage plane. It has been shown that the adhesion energy is maximum at the oxygen interface, which is caused by the ion component in chemical bonding at this interface. The aluminum and aluminum-enriched interfaces are characterized by the metallic type of bonding. The local densities of states and the charge distribution near the interface have been analyzed. It has been demonstrated that oxygen vacancies at the interface substantially weaken the adhesion due to the partial breaking of Me-O bonds.     

Improved efficiencies of hybrid organic light‐emitting diodes using efficient electron injection layer

Hybrid organic‐inorganic light‐emitting diodes were developed with pristine ZnO (2.0 wt%) and Cu‐doped ZnO (2.0 wt%) as electron injection layer and iridium(III)‐bis‐2‐(4‐fluorophenyl)‐1‐(naphthalen‐1‐yl)‐1H‐phenanthro[9,10‐d]imidazole (acetylacetonate) [Ir(fpnpi)₂ (acac)] as green emissive layer (521 nm). The pristine ZnO and Cu‐doped ZnO are deposited at indium tin oxide cathode and emissive layer interface. The electroluminescent performances increased by electron injection layer–Cu‐doped ZnO compared with ZnO‐based device because Cu‐doped ZnO injects electron efficiently result in balanced h⁺ ? e^? recombination in emissive layer than ZnO‐based device. The Cu‐doped ZnO (2.0 %) device shows luminance (L) of 10 982 cd/m² at 23.0 V (ZnO, 1450 cd/m² at 23.0 V).     

Oxide nanoparticles in an Al-alloyed oxide dispersion strengthened steel: crystallographic structure and interface with ferrite matrix

Abstract

Oxide nanoparticles are quintessential for ensuring the extraordinary properties of oxide dispersion strengthened (ODS) steels. In this study, the crystallographic structure of oxide nanoparticles, and their interface with the ferritic steel matrix in an Al-alloyed ODS steel, i.e. PM2000, were systematically investigated by high-resolution transmission electron microscopy. The majority of oxide nanoparticles were identified to be orthorhombic YAlO₃. During hot consolidation and extrusion, they develop a coherent interface and a near cuboid-on-cube orientation relationship with the ferrite matrix in the material. After annealing at 1200 °C for 1 h, however, the orientation relationship between the oxide nanoparticles and the matrix becomes arbitrary, and their interface mostly incoherent. Annealing at 1300 °C leads to considerable coarsening of oxide nanoparticles, and a new orientation relationship of pseudo-cube-on-cube between oxide nanoparticles and ferrite matrix develops. The reason for the developing interfaces and orientation relationships between oxide nanoparticles and ferrite matrix under different conditions is discussed.     

电迁移对Ni/Sn3.0Ag0.5Cu/Au/Pd/Ni-P倒装焊点界面反应的影响

本文研究了150 ℃, 1.0× 10⁴ A/cm²条件下电迁移对Ni/Sn3.0Ag0.5Cu/Au/Pd/Ni-P倒装焊点界面反应的影响. 回流后在solder/Ni和solder/Ni-P的界面上均形成(Cu,Ni)₆Sn₅类型金属间化合物. 时效过程中两端界面化合物都随时间延长而增厚, 且化合物类型都由(Cu,Ni)₆Sn₅转变为(Ni,Cu)₃Sn₄. 电迁移过程中电子的流动方向对Ni-P层的消耗起着决定性作用. 当电子从基板端流向芯片端时, 电迁移促进了Ni-P层的消耗, 600 h后阴极端Ni-P层全部转变为Ni₂SnP层. 阴极界面处由于Ni₂SnP层的存在, 使界面Cu-Sn-Ni三元金属间化合物发生电迁移脱落溶解, 而且由于Ni₂SnP层与Cu焊盘的结合力较差, 在Ni₂SnP/Cu界面处会形成裂纹. 当电子从芯片端流向基板端时, 阳极端Ni-P层并没有发生明显的消耗. 电流拥挤效应导致了阴极芯片端Ni层和Cu焊盘均发生了局部快速溶解, 溶解到钎料中的Cu和Ni原子沿电子运动的方向往阳极运动并在钎料中形成了大量的化合物颗粒. 电迁移过程中(Au,Pd,Ni)Sn₄的聚集具有方向性, 即(Au,Pd,Ni)Sn₄因电流作用而在阳极界面处聚集.     

Thermal stability of Cu–Nb nanolamellar composites fabricated via accumulative roll bonding

In situ annealing within a neutron beam line and ex situ annealing followed by transmission electron microscopy were used to study the thermal stability of the texture, microstructure, and bi-metal interface in bulk nanolamellar Cu/Nb composites (h?=?18?nm individual layer thickness) fabricated via accumulative roll bonding, a severe plastic deformation technique. Compared to the bulk single-phase constituent materials, the nanocomposite is two orders of magnitude higher in hardness and significantly more thermally stable, e.g., no observed recrystallization in Cu at temperatures as high as 85% of the melting temperature. The nanoscale h?=?18?nm individual layer thickness is maintained up to 500°C, the lamellar structure thickens but is maintained up to 700°C, and recrystallization is suppressed even up to 900°C. With increasing temperature, the texture sharpens, and among the interfaces found in the starting material, the {112}_Cu?||?{112}_Nb interface with a Kurdjumov-Sachs orientation relationship shows the greatest thermal stability. Our results suggest that thickening of the individual layers under heat treatment coincides with thermally driven removal of energetically unfavorable bi-metal interfaces. Thus, we uncover a temperature regime that maintains the lamellar structure but alters the interface distribution such that a single, low energy, thermally stable interface prevails.     

The chemical composition of a metal/ceramic interface on an atomic scale: The Cu/MgO {111} interface

The chemical composition profile across a Cu/MgO {111}-type heterophase interface, produced by the internal oxidation of a Cu(Mg) single-phase alloy at 1173 K, is measured via atom-probe field-ion microscopy with a spatial resolution of 0.121 nm; this resolution is equal to the interplanar spacing of the {222} MgO planes. In particular, we demonstrate directly that the bonding across a Cu/MgO {111}-type heterophase interface, along a <111> direction common to both the Cu matrix and an MgO precipitate, has the sequence Cu|O|Mg... and not Cu|Mg|O...; this result is achieved without any deconvolution of the experimental data. Before determining this chemical sequence, it was established, via high-resolution electron microscopy, that the morphology of an MgO precipitate in a Cu matrix is an octahedron faceted on {111} planes with a cube-on-cube relationship between a precipitate and the matrix; that is, {111}_Cu//{222}_MgO and <110>_Cu // <110>_MgO.     

Distribution of oxygen vacancies and gadolinium dopants in ZrO2–CeO2 multi-layer films grown on α-Al2O3

Gdolinia doped ZrO₂ and CeO₂ multi-layer films were deposited on α-Al₂O₃ (0001) using oxygen-plasma-assisted molecular-beam epitaxy. Oxygen vacancies and Gd dopant distributions were investigated in these multi-layer films using X-ray diffraction (XRD), conventional and high-resolution transmission electron microscopy (HRTEM), annular dark-field imaging in scanning transmission electron microscopy (STEM), X-ray energy dispersive spectroscopy (EDS) elemental mapping and X-ray photoelectron spectroscopy (XPS) depth profiling. EDS and XPS depth profiling reveal that the Gd concentration in the ZrO₂ layer is lower than that in the CeO₂ layer. As a result, a higher oxygen vacancy concentration exists in the CeO₂ layers compared to that in the ZrO₂ layers. In addition, Gd is found to segregate only at the interfaces formed during the deposition of CeO₂ layers on ZrO₂ layers. On the other hand, the interfaces formed during the deposition of ZrO₂ layers on CeO₂ layers did not show any Gd segregation. The Gd segregation behavior at every other interface is believed to be associated with the low solubility of Gd in ZrO₂.     

Magnetism of ultrathin Fe(110) films

Application of conversion electron Mössbauer spectroscopy (CEMS) to structural and magnetic analysis of ultrathin films and their interfaces is reviewed. Fe(110) films were prepared on W(110) under UHV conditions and analyzed in situ. CEMS provides detailed information on the mode of growth and film structure and on magnetic hyperfine fields, B _hf. Local structure of B _hf across the film is discussed in relation to modifications of magnetic order caused by the finite (including monolayer) film thickness and by the electronic structure of the interface.     

CEMS characterisation of Fe/high-κ oxide interfaces

The interfaces between Fe and different high-κ oxides are investigated by means of conversion electron Mössbauer spectroscopy (CEMS). Information on the magnetic ordering at the interface is obtained from the magnetic hyperfine splitting of the Mössbauer spectra. The reactivity of the Fe atoms at the interface (intermixing) is also estimated by CEMS. X-ray diffraction (XRD) and X-ray reflectivity (XRR) provide additional information on the intermixing and different phases present at the interface. CEM-spectra show the presence of both ferromagnetic and paramagnetic phases. CEMS and XRD results show that the Fe/HfO₂ and Fe/Al₂O₃ interfaces are the least reactive. The degree of intermixing between Fe and the high-κ oxide is determined by the oxide surface roughness.     

The electrical properties of metal-oxide-semiconductor devices fabricated on the chemically etched n-InP substrate

The electrical properties of the Cu/n-InP and Al/n-InP Schottky barrier diodes (SBDs) with and without the interfacial oxide layer have been investigated by using current-voltage (I-V) measurements. The oxide layer on chemically cleaned indium phosphide (InP) surface has been obtained by exposure to water vapor at 1 ml/min at 200 °C before metal evaporation. The chemical composition of surface oxides grown on the InP is investigated using X-ray photoelectron spectroscopy (XPS). Phosphorus is present as In(PO₃)₃, InPO₄, P₂O₅ and P₄O₁₀. The values of 0.437 ± 0.007 and 0.438 ± 0.003 eV for the barrier height of the reference Cu/n-InP and Al/n-InP SBDs were obtained, respectively. Furthermore, the values of 0.700 ± 0.030 and 0.517 ± 0.023 eV for the barrier height of the oxidized Cu/n-InP and Al/n-InP SBD were obtained, respectively. The transport properties of the metal-semiconductor contacts have been observed to be significantly affected by the presence of the interfacial oxide layer. Devices built on the oxidized surfaces show improved characteristics compared with those built on chemically cleaned surfaces. The chemical reactivity of the metal with oxide and n-InP is important to the formation of the Schottky barriers. The reactive metal Al gave a low barrier height due to the reduction of oxide and reaction with InP. The transmission coefficients for the oxidized Cu/n-InP and Al/n-InP are equal to 2.23 × 10⁻⁵ and 4.60 × 10⁻², respectively.     

Effects of cooling rate on the microstructure and solute partitioning in near eutectoid Ti–Cu alloys

The effect of cooling rate on eutectoid decomposition in a near eutectoid (Ti-5.5 at.% Cu) alloy has been investigated in a systematic manner by coupling scanning electron microscopy, transmission electron microscopy and atom probe tomography studies. Thus, the competition between nucleation and growth of proeutectoid α plates from pre-existing β grain boundaries, and eutectoid decomposition (α?+?Ti₂Cu) via a pearlitic mechanism has been studied as a function of cooling rate, using a Jominy-end quenched sample that was cooled from the high-temperature single β phase. When the alloy was subjected to very fast cooling (160?K/s), proeutectoid α plates, supersaturated in Cu, are formed along with a highly refined lamellar eutectoid product between these α plates. In contrast, intermediate (9?K/s) and slow (2?K/s) cooling results in considerably coarser proeutectoid α plates as well as lamellar eutectoid products. With the decrease in the cooling rate, there was a substantial increase in the volume fraction of the lamellar eutectoid product and the composition of all decomposition products approached their equilibrium values. Also, the slowest cooled sample (2?K/s) exhibited substantially rougher and irregular interfaces between the proeutectoid α and the lamellar eutectoid product, which seems to promote the cooperative growth of lamellar α?+?Ti₂Cu. Irrespective of the cooling rate, nucleation of the lamellar eutectoid (α?+?Ti₂Cu) product appears to only occur at the interface between the proeutectoid α plates and the β matrix.     

Structure and composition of the segregated Cu in V2O5/Cu system

We have investigated segregation of copper at the surface of V₂O₅ films deposited onto Cu substrate by employing surface analysis techniques. X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) confirmed that the Cu is segregated at the surface and its chemical state is Cu₂O. According to secondary ion mass spectroscopy (SIMS) and glow discharge spectroscopy (GDS), the Cu concentration inside the deposited V₂O₅ layer is low. Ultraviolet photoelectron spectroscopy (UPS) and scanning tunneling spectroscopy (STS) revealed the segregation alters the surface local density of states. Surface analysis of deposited samples in ultra high vacuum (UHV) condition verified that the segregation occurs during the deposition. We have extended kinetic tight binding Ising model (KTBIM) to explain the surface segregation during the deposition. Simulation data approve the possibility of surface segregation during room temperature deposition. These results point out that on pure Cu substrate, oxidation occurs during the segregation and low surface energy of Cu₂O is the original cause of the segregation.     

The effects of indium segregation on the valence band structure and optical gain of GaInAs/GaAs quantum wells

The effects of indium segregation on the valence band structures and the optical gain in GaInAs/GaAs quantum wells are theoretically investigated using 4×4 Luttinger–Kohn Hamiltonian matrix. The method for the band structure calculation is based on the finite difference method, then the optical gain is calculated using the density matrix approach. For segregation coefficient R less than 0.7, indium segregation has little influence on optical gain, but for segregation coefficient R more than 0.7, it has a significant influence on optical gain, the gain spectra can be blue-shifted with the increase of segregation coefficient R, and the peak gains are decreased as segregation coefficient R increases, which is mainly due to the reduction of the carrier population inversion.     

Room temperature growth of InxGa1−xN thin films by mixed source modified activated reactive evaporation

Polycrystalline In_xGa_1−xN thin films were prepared by mixed source modified activated reactive evaporation (MARE) technique. The films were deposited at room temperature on glass substrates without any buffer layer. All the films crystallize in the hexagonal wurtzite structure. The indium concentration calculated from XRD peak shift using Vegard's law was found to be varying from 2% to 92%. The band gap varies from 1.72 eV to 3.2 eV for different indium compositions. The indium rich films have higher refractive indices as compared to the gallium rich films. The near infra-red absorption decreases with gallium incorporation into InN lattice which is mainly due to decrease in the free carrier concentration in the alloy system. This fact is further supported from Hall effect measurements. MARE turns out to be a promising technique to grow In_xGa_1−xN films over the entire composition range at room temperature.     

Specific contact resistance of IGZO thin film transistors with metallic and transparent conductive oxides electrodes and XPS study of the contact/semiconductor interfaces

In this work, the specific contact resistance (ρ_c) between amorphous indium-gallium-zinc-oxide (IGZO) semiconductor and different contact electrodes was obtained from thin film transistors (TFTs). Ti/Au (10/100 nm), aluminum doped zinc oxide (AZO, 100 nm) and indium tin oxide (ITO, 100 nm) were used as source/drain electrodes to fabricate IGZO TFTs. Chemical states of the contacts/semiconductor interfaces were examined by depth profile X-ray photoelectron spectroscopy (XPS) analysis to explain the origin of the differences on specific contact resistance. The lowest ρ_c achieved using Ti/Au was related to the formation of a TiO_x interlayer due to oxygen atoms diffusing out from the semiconductor under layer, increasing the carrier concentration of IGZO at the interface and lowering the ρ_c. On the contrary, no interfacial reactions were observed between IGZO and AZO or ITO source/drain. However, IGZO resistivity increased with ITO contacts likely due to oxygen vacancies filling during ITO deposition. This fact seems to be the origin of the high contact resistance between IGZO and ITO, compared to IGZO-AZO and IGZO-Ti/Au interfaces.