Band 97, Heft 6

The performance of thermal barrier coatings (TBCs) in jet engines is strongly influenced by thermal and mechanical properties of the bond coat. In this investigation two different types of Pt-modified diffusion coatings developed for application as oxidation protection or as a bond coat in TBC systems were studied with respect to their microstructure and mechanical properties. The microstructural analyses based on different microscopy techniques and X-ray diffraction measurements revealed distinct differences between the two coatings: the coating aluminized by a gas phase process consisted solely of a (NiPt)Al phase, while the coating fabricated by pack-cementation consisted mainly of the two phases (NiPt)Al and Al 2 Pt, leading to a higher strength in comparison with the single-phase coating. In contrast, the ductility determined with 3-point bend tests between room temperature and 700 °C was clearly better for the single-phase coating. Additionally, the substrate curvature of coated sheets was measured to gain some information about coating stresses.

Carbon-rich Si –O–C polymer-derived ceramics (PDCs) were investigated by various spectroscopic techniques, in order to characterize the evolution of their predominantly amorphous microstructure upon thermal treatment up to 1450 °C. Particular attention was addressed to modifications of the excess free carbon phase present in these materials. Surprisingly, the carbon clusters exhibited high stability above the pyrolysis temperature. Despite the high volume fraction of carbon, only a very limited carbothermal reduction process was detected. This study is divided into two parts: Part I deals with characterization tools that reveal a rather low lateral resolution and are hence termed here as integral spectroscopic techniques, i. e., solid-state NMR and Raman spectroscopy. In contrast, Part II illustrates the experimental results obtained from the very same materials characterized by spectroscopic and imaging techniques with high lateral resolution, i. e., electron energy-loss spectroscopy (EELS), high-resolution transmission electron microscopy (HRTEM), and energy-filtered TEM. In addition to materials characterization, emphasize of both papers is also to compare the information gained by either integral or local spectroscopy techniques and to highlight the strengths and weaknesses of either approach.

In this Part II of the paper on the evolution of carbon-rich Si–O–C polymer-derived ceramics (PDCs), emphasis is placed on the strengths and limitations of transmission electron microscopy (TEM) techniques with high lateral resolution, i. e., electron energy-loss spectroscopy (EELS), high-resolution TEM (HRTEM) and energy-filtered TEM (EFTEM). Here, the identical SiOC materials, as described in Part I. EELS studies confirmed the progression of structural rearrangements within the SiOC matrix temperatures exceeding 1200 °C. High-resolution TEM imaging showed that the SiOC matrices are indeed predominantly amorphous even upon high thermal treatment. Energy-filtered TEM analysis revealed, in contrast to the results obtained by Raman spectroscopy (Part I), that the excess free carbon phase undergoes a pronounced rearrangement within the amorphous microstructure. HRTEM characterization revealed the distribution of phases within the amorphous SiOC matrix; information that is not accessible by integral spectroscopic techniques. Discrepancies between the interpretation of experimental results obtained by local versus integral characterization tools are discussed in detail.

Three borazine-modified polycarbosilanes were synthesized by reaction of poly[1,2-ethynediyl-(methylsilylene)], [HSi(Me)C≡C] n , and borazine, B 3 N 3 H 6 using different stoichiometries. The polymeric precursors were transformed into inorganic Si–B–C–N materials by solid state thermolysis at 1400 °C each in 88% yield. High temperature thermogravimetric analysis in an argon atmosphere showed no substantial decomposition reactions below 1800 °C. Crystallization behavior up to 2000 °C in a nitrogen atmosphere was studied by post-thermolysis heat treatment of as-obtained ceramics at various temperatures and subsequent examination of samples at room temperature using powder X-ray diffraction. Ceramics annealed at 1800 °C were additionally characterized by transmission electron microscopy.

Thermodynamic modeling has been used to explain structural transformations induced by heat treatment of amorphous Si –C–N ceramics derived from polymers. Nanocrystalline silicon carbide and nanocrystalline silicon nitride identified in the ceramic microstructure have been regarded as metastable NASIC and NASIN phases in the Si–C–N system. The Gibbs energies G (NASIC) and G (NASIN) have been derived and used together with the previously modeled Gibbs energy of the amorphous am-SICN to compute metastable phase diagrams. Computational results allow explanation of the crystallization process of amorphous Si –C–N ceramics. According to this model, the temperature of invariant reaction between carbon and silicon nitride changes with the growth of nanocrystallites, which explains the dependence of the thermal stability on the ceramic microstructure.

The phase equilibria in the Ni-rich part of the Ce–Ni binary system were re-determined using DTA, metallography and SEM/EDX. CeNi 5+ x presents a Ni solubility up to x = 0.53 at 1073 K. The melting point of CeNi 5 is 1608 ± 5 K and the CeNi 5+ x –Ni eutectic has been observed at 1477 ± 5 K and x (Ni) = 0.91. The enthalpy of formation of stoichiometric CeNi 5 has been measured by drop dissolution calorimetry in liquid aluminium and a value of Δ f H = – 29.1 ± 2.2 kJ/moles of atoms was derived. Based on this new experimental data and literature values, a complete thermodynamic modelling of the Ce–Ni diagram was performed.

A thermodynamic dataset was developed for the central part of the Ce–O system covering the range between Ce 2 O 3 and CeO 2 up to about 2000 K. All literature data for the thermodynamic functions, the phase diagram and crystallography were critically assessed. The modeling of the phases takes into account their crystal chemistry including order – disorder behavior. From the dataset, diagrams were calculated for the (1) heat capacities of CeO 2 and Ce 2 O 3 , (2) enthalpy increment of CeO 2 , (3) chemical potentials of oxygen in CeO 2– x , (4) partial enthalpies of oxygen in CeO 2– x , (5) the partial phase diagram of Ce–O and (6) chemical potentials of oxygen in the two-phase areas. These diagrams reproduce fairly well the experimental data from literature.

Annealing at low temperatures ( T < 670 K) of mixtures of iron nitride powders and iron powder is shown to induce phase transformations through nitrogen transfer between different powder particles. This nitrogen transfer occurs via loose mechanical contact points between different powder particles; nitrogen transport via the gas phase can be excluded. Unique scientific possibilities offered by this type of phase transformation studies are discussed.

Organic self-assembled monolayers with nitrogen-containing surface functionalities (amine or alkylammonium) were deposited on acid-etched titanium substrates. These were then coated with apatite deposited from 150% simulated body fluid at 37 °C and pH = 7.6 for 2, 4, and 6 days (after a 2-day “nucleation” step in contact with bioactive glass in 100 % simulated body fluid). Scanning electron microscopy showed that precipitates were numerous on amine- and alkylammonium-modified titanium surfaces after nucleation for 2 days. X-ray diffraction at grazing incidence detected crystalline hydroxyapatite on the modified and unmodified substrates after growth for 2 days. Tape peel tests followed by surface chemical analysis showed that the adhesion of the hydroxyapatite coating to the substrates was improved by amine and by alkylammonium modification of the etched titanium surfaces.

Based on density-functional calculations, interfacial reconstruction and structural transition at diamond(111)/M/copper (M = Ag, Cu, Ni, Co) interfaces are studied by analyzing the atomistic structures of the interfaces. Tuning the strength of chemical bonds across the interfacial plane (out-of-plane), we show that interfacial reconstruction happens as a result of competition between the out-of-plane chemical bonds and the in-plane bonds within the diamond substrate. A relatively weak out-of-plane bond may easily lead to reconstruction of interfacial structures. A structural transition from one-dangling-bond- to three-dangling-bond-terminated interface is also observed as the out-of-plane bond is sufficiently strong, which leads to a dramatic enhancement of the adhesion of the interface.

The grain boundary phases in SiC/Si 3 N 4 nano/micro-composite changes with time of high temperature treatment at 1750 °C in nitrogen atmosphere. Stable phases were obtained after 26 hours of annealing. Moreover, the average grain diameter distribution was significantly changed. The hardness and fracture toughness of the SiC/Si 3 N 4 nano/micro-composites decreased with the duration of the high-temperature treatment. Both findings can be explained by the different secondary crystalline phases formed during the high-temperature treatment. Phases Si 2 N 2 O, Y 2 SiO 5 , and Y 2 Si 2 O 7 , typical for the samples treated up to 4 h, were replaced by the YNSiO 2 phase in the samples treated longer than 26 h. Anisotropy of the measured mechanical properties in different sample cross-sections caused by the Si 3 N 4 grain alignment become negligible after long term high-temperature treatment.

SiC ceramics containing Lu 2 O 3 –AlN sintering additives were gas-pressure sintered, and creep testing in four-point bending geometry was performed at temperatures between 1200 °C and 1500 °C in air. Compared with the more conventional additive system Y 2 O 3 –AlN, the creep resistance of the Lu 2 O 3 -containing ceramics is higher. In particular, strain rates down to 6 ⨯ 10 –10 s –1 were measured after 60 h at both 1300 °C/300 MPa and 1400 °C/100 MPa; in no case was steady-state creep reached. Annealing the ceramics at 1950 °C for extended periods of time led to significant changes in the amount and distribution of intergranular phases. However, in contrast to earlier work using Y 2 O 3 –AlN additives, the influence of annealing on the creep resistance was negligible. Compliance testing as a function of temperature was conducted in order to monitor the softening behavior of materials with different additive compositions. In favorable cases, the drop in stiffness was only 12% up to a test temperature of 1450 °C.

The study clearly shows that different metal cations were efficiently removed from water by employing coated silica particles with SO 3 H-terminated functionalized surfaces. Significant amounts of Fe, Mn, Ca, Mg, Na, and K could be removed from natural water samples with different concentrations varying from 0.075 mg/l for Mn up to 12 000 mg/l for Na by stirring the functionalized silica particles in the water samples for 1 h at room temperature. ICP–MS verify the removal of the elements and ToF-SIMS analyses clearly showed the presence of the elements on the particles after the treatment procedure. The removal of the metal cations is believed to be due to electrostatic attraction of the positively charged presumably hydrolysed metal cations to the negatively charged deprotonated SO 3 H-groups. However, from the current study it can not be excluded that chemisorption of the cations by the SO 3 H-group may be possible.

NiO–LSGM reactivity during sintering at high temperatures in air leads to the formation of LaNiO 3 -type phase which is found to be insulating for oxygen ions. XRD analysis of the NiO-LSGM samples after sintering at 1350 °C in air showed cubic (Pm-3m) LaNiO 3 -based phase with a lattice parameter of 3.856 Å in addition to the cubic structure of LSGM20-17. STEM/EDX analysis of the NiO– LSGM interface showed the presence of a Ni-containing perovskite phase (La, Sr) (Ga, Ni, Mg)O 3 – x in the diffusion layer of LSGM. The probable composition of the LaNiO 3 - type phase that formed during sintering is discussed.

Electrolytic anodization was used to form amorphous niobium oxide layers on niobium and V-doped niobium powder compacts which act as dielectric layers, e. g. in niobium-based solid electrolyte capacitors. The microstructure development within the layered structure niobium-niobium oxide was studied by scanning- and transmission electron microscopy in order to investigate the influence of processing parameters. It could be shown that the thickness as well as the quality of the oxide layers on niobium vary in a considerable range, depending on processing parameters. Examination of the influence of heat treatments on the structural properties of the oxide layers revealed remarkable changes at the niobium – niobium oxide interface upon heat impact exceeding 300 °C. In contrast, the interface remains stable with respect to the as-anodized sample when annealing was performed at lower temperature.

Vaporisation studies on Al–Ni alloys with x (Ni) = 0.43, 0.42, 0.35, and 0.32 were carried out by Knudsen effusion mass spectrometry. Partial pressure-temperature relations of Al(g) over the two phase regions {AlNi + Al 3 Ni 2 } and {Al 3 Ni 2 + Al 3 Ni} were determined in the temperature ranges 1104 – 1278 K and 1021 –1122 K, respectively. The corresponding equations are log( p (Al)/Pa) = –(20 039 ± 84)/ T (K) + (13.464 ± 0.070) for {AlNi + Al 3 Ni 2 } and log( p (Al)/Pa) = – (16 836 ± 235)/ T (K) + (11.345 ± 0.219) for {Al 3 Ni 2 + Al 3 Ni}. The vaporisation reactions, Al 3 Ni 2.025 (s) = 2.6436AlNi 0.766 (s) + 0.3564Al(g) and Al 3 Ni(s) = 0.572Al 3 Ni 1.747 (s) + 1.284Al(g) were evaluated and subsequently the enthalpy and Gibbs energy of formation of Al 3 Ni 1.747 (s) were derived.

Crystalline In(OH) 3 and UV-emitting In 2 O 3 thin films with a unique columnar morphology have been successfully prepared for the first time via a simple aqueous solution deposition method. In the present study, sulfonate-terminated self-assembled monolayers on single-crystal Si wafers were used as substrates for the film deposition. Uniform In(OH) 3 thin films formed on self-assembled monolayers in aqueous solutions of InCl 3 · 4H 2 O at 60°C. These films, containing columnar In(OH) 3 crystals, showed preferential growth orientation along the [100] direction. The current understanding of the film formation mechanism was discussed. After annealing in air at 400 °C, these films were converted to nanocrystalline In 2 O 3 films without modification of the morphology. Photoluminescence emission from In 2 O 3 films was reported at approximately 3.4 eV, which was ascribed to the near-band-edge emission.

Results from our recent simultaneous measurement of all three component activities by Knudsen effusion mass spectrometry, together with results available in the literature from vacancy concentration and calorimetric measurements, have been used in a thermodynamic modelling study. The results of this modelling for the binary B2-AlFe and B2-AlNi phases are presented in this communication. A four-sublattice cluster energy model in the point approximation has been used for the configurational contributions to the Gibbs energy. Excitational free energies are incorporated into the configurational energy and non-configurational contributions are considered to be decoupled from the configurational. The modelling has been found to lead to good agreement between calculated and experimental results for both the thermodynamic properties and the vacancy concentrations. It is clear that B2-AlFe is an antisite defect compound at both the lowest and highest temperatures with vacancy defects only predominating, and to only a small degree, at stoichiometry in an intermediate temperature range. In the case of B2-AlNi, however, vacancy defects predominate at stoichiometry over the whole temperature range so that it can be correctly described as a near-triple defect compound.

Supersaturated (Ti 0.5 – x W0.5Cr x )B 2 ( x = 0, 0.1, 0,2) solid solutions were produced by reaction sintering of transition metal diboride powders at a pressure of 60 MPa at 1800 °C. The formation kinetics of W 2 B 5 precipitates were investigated at lower temperatures by X-ray diffractometry (XRD) and secondary electron microscopy (SEM). The precipitation rate constants as a function of reciprocal temperature show non-monotonic behaviour in the range between 1450 and 1700 °C with a maximum at about 1650 °C. At low temperatures, Arrhenius behaviour is approximately valid, where the precipitation kinetics are dominated by the diffusion of the transition metals. In contrast, at temperatures higher than 1650 °C, the precipitation rates decrease rapidly, since supersaturation, and consequently nucleation, is strongly reduced. The rate constants of precipitation are identical within error limits for Cr concentrations of x = 0.1 and x = 0.2, however, they are significantly lower for x = 0. This means that the addition of small amounts of Cr accelerates the precipitate formation, while further addition has no further influence. Samples resulting from powders which were high-energy milled with WC/Co hard balls before reaction sintering show an increased density of precipitated crystallites but no increased rate constants of precipitation. This result can be explained under the assumption that for these samples a significant amount of additional pre-existing nuclei was present before precipitation starts, which were formed by heterogeneous nucleation at Co impurities, introduced by the milling process.

Low alloyed zirconium sheets (Zr702) have been cold-rolled up to 80% thickness reduction and submitted to various isothermal treatments. The aim was to identify the mechanisms which produce microstructure and texture changes during recrystallization and grain growth. XRD and texture analysis, FEG–SEM with EBSD as well as TEM were used to analyze the various specimens. Three types of substructures were observed in the initial deformed material. Accordingly, the nucleation starts in the most deformed areas and continues in the somewhat less deformed areas, which corresponds to a non oriented nucleation and results in a set of new grains, the size of which is very rapidly stabilized. In the last stage of recrystallization, the grains which have resisted the recrystallization disappear progressively by several mechanisms including in situ recrystallization. Therefore, the texture at the end of the recrystallization resembles the one of the deformed state. Normal grain growth leads to a moderate grain size increase due to the precipitates which slow down the grain boundary motion. This is also the stage where the texture changes due to the size advantage, after recrystallization, of grains in some specific orientations.

Lead Bismuth eutectic alloy (Pb: 55.5 wt.%, Bi: 44.5 wt.%) is a potential candidate coolant material for high-temperature reactors because of its low melting point (124 °C), high thermal conductivity, heat capacity, and better neutronic properties. Out-of-pile chemical compatibility studies of this coolant with graphite (coolant channel) have been carried out by isothermal annealing of the liquid alloy in a graphite crucible at 800, 900, 1000, and 1100 °C for times ranging from 100 h to 1000 h. Formation of a reaction layer is observed. The growth rate of the reaction layer follows a parabolic law. Reaction layer thicknesses of 61.3 μm and 121 μm are estimated from the growth rate vs. time relation after 1 year and 5 years respectively. The growth of the reaction layer is diffusion-controlled and the activation energy of the reaction is estimated to be 100 KJ/mol.

A Co – Cr –Mo laser cladding applied on railway wheels is characterised by a combination of EBSD and EDX. A complete pass of the cladding is investigated to achieve a better understanding of the microstructure evolution during laser cladding. A microstructure with columnar grains extending over the whole thickness of the cladding is observed. The grains have a <001>-fibre texture with the fibre axis parallel to the normal of the substrate surface, and a substructure consisting of cells/dendrites. During the cladding process, two different kinds of precipitates form in the cell walls, which can be identified as M 6 C and a non-equilibrium phase. Furthermore, stacking faults are observed to occur in the cladding grains and are discussed with respect to the laser cladding process.

The Na–H system thermodynamic properties were assessed using Gibbs free energy model parameters obtained from best fit optimizations to combined experimental and first-principles predicted data. The first-principles finite temperature thermodynamic property predictions, based upon density functional theory ground state minimizations and direct method lattice dynamics, were used to supplement the Na–H dataset wherever experimental information was unavailable or unattainable. The predictions proved to be important for extending the evaluation of the heat capacity of the stable NaH phase to cover the complete 0 – 2000 K temperature range. The predicted thermodynamic properties of the hypothetical NaH 3 end-member representing complete interstitial H substitution in solid body-centered cubic Na, provided a physical basis for modeling H dissolution in the Na lattice. The modeling also showed satisfactory agreement with experimental measurements of NaH enthalpies of formation, NaH decomposition pressures, and H solubility in liquid Na.