Search

In the present study, the micro-mechanical behavior of CoCrFeMnNi high-entropy alloy was investigated using an in-house micro-tensile setup at room temperature and 550 °C at different strain rates. Micro-mechanical properties are compared with those obtained using a commercial macro-tensile setup to check a potential sample size effect. Results show that mechanical properties such as yield strength, ultimate tensile strength and uniform elongation are independent of the sample size. However, the total elongation-to-failure of micro-samples is found to be lower than those of macro-counterparts. Apart from this, the material exhibits serrated plastic flow, which is strain rate dependent in terms of the onset strain and shape of serrations at 550 °C. Furthermore, transmission electron microscopy investigations were performed to correlate the occurrence of serrations to the observed distinct dislocation structures. Microstructural results provide direct evidence that dislocations are curved and hence effectively pinned and unpinned at the lowest applied strain rate, which might be responsible for the occurrence of serrated plastic flow.

In the present work, low cycle fatigue (LCF) behavior of an equiatomic CoCrFeMnNi high entropy alloy (HEA) is correlated to the microstructural evolution at 550 °C. The fully reversed strain-controlled fatigue tests were conducted in air under strain amplitudes ranging from 0.2% to 0.8%. The measured cyclic stress response showed three distinct stages which include initial cyclic hardening followed by a quasi-stable cyclic response until failure. The rate and amount of cyclic hardening increased with the increase in strain amplitude. In comparison to common austenitic stainless steels, CoCrFeMnNi HEA shows comparable strength and improved LCF lifetime at similar testing conditions. Electron-microscopy investigations after failure reveal no noticeable change in grain size, texture and annealing twins density. Initial cyclic hardening is attributed to the dislocations multiplication and dislocation-dislocation as well as dislocation-solute atom interactions. The quasi-stable cyclic response is associated with the equilibrium between dislocation multiplication and annihilation, which also leads to the formation of complex dislocation structures such as ill-defined walls and cells, particularly at higher strain amplitudes. Besides, the material exhibits serrated plastic-flow due to interactions between mobile dislocations and diffusing solute atoms (such as Cr, Mn and Ni). Lastly, segregation in the form of Cr- and NiMn-enriched phases were observed near grain boundaries, which appears to have a detrimental effect on the fatigue life.

Creep-fatigue (CF) interaction in a tempered martensitic Fe-9%Cr-based oxide dispersion strengthened (ODS) steel was studied at 650 °C by introducing hold-time of up to 30 min at peak tensile strain of 0.7%. The symmetrical loops under pure fatigue/continuous cycling (PF/CC) became asymmetrical due to stress relaxation during hold-time. Moreover, this also resulted in a reduction of cyclic life. For the investigated hold-time durations, the increase in tensile hold (TH) period had a negligible effect on peak stresses, but led to a further reduction in lifetime. Electron backscatter diffraction (EBSD) and transmission electron microscopy (TEM) were used to visualize microstructural evolution under both PF/CC and TH waveforms. In general, PF/CC resulted in: (1) rearrangement and/or annihilation of dislocations, (2) partial elimination of the original sub-grain structures, (3) grain growth, (4) M23C6 carbides coarsening and (5) Cr-W enriched Laves phase precipitation. Nevertheless, upon introducing tensile hold-time, no substantial additional microstructural changes were identified. To uncover reasons for specimens premature failure under TH waveforms, detailed investigations on their surfaces, cross-sections and fracture surfaces were carried out. These investigations led to two important conclusions. Firstly, due to comparatively longer high-temperature exposure, the extent of oxidation increased upon introducing TH which expedited damage progression. Secondly, TH induced intergranular damage in the form of creep cavities does not only provide additional crack initiation sites but also their growth/coalescence under tensile stresses act as a bridging link for accelerated crack propagation. These two findings are associated with a reduction of cyclic life due to introduction of hold-time. Hence, the effect of hold-times is primarily due to (1) oxidation-fatigue interaction and (2) creep-fatigue interaction.

In this work, a high-temperature low-cycle fatigue (LCF) behavior of a newly developed austenitic oxide dispersion strengthened (ODS) steel is investigated. The LCF tests were performed in air at 650 °C under three different strain amplitudes (±0.4, ±0.5, and ±0.7%) with a nominal strain rate of 10−3 s−1. The measured cyclic stress response showed four distinct stages which include short initial stable cyclic response followed by a prolonged hardening with subsequent short saturation and finally crack initiation and growth stage. The rate of hardening and the duration of stages are a function of applied strain amplitude. Microstructural investigations were carried out to shed light on the deformation mechanisms. After cycling, the overall microstructure appears stable without any modifications in grain shape and size. In addition, twinning and stacking fault fractions remain unchanged. However, cyclic hardening is an aftermath of dislocation multiplication whose rate is also a function of applied strain amplitude. Furthermore, oxide particles, as well as fine grains, inhibit strain localization by restricting three-dimensional dislocation structure formation that are associated with the development of extrusions and intrusions and are readily observed in conventional austenitic non-ODS steels.

The present study reports on the creep-fatigue (CF) behavior of a ferritic bimodal 12Cr-ODS steel. The tests were performed in air at 550 °C and 650 °C by introducing hold-time at peak tensile, peak compressive and both peak tensile as well as compressive strains of ±0.5%. The symmetrical loops under pure fatigue/continuous cycling (CC) became asymmetrical upon introducing hold-time at either peak tensile or peak compressive strain. On the contrary, hold-time on both sides had no influence regarding symmetry but exhibited minor variation in peak stresses. For the investigated hold-time durations, the increase in hold-time period had a negligible effect on peak stresses. Due to stress relaxation during hold-time, elastic strain gets partially converted into inelastic strain which is relatively lower at 650 °C than at 550 °C. This eventually led to lower mean stresses; and finally, a relatively lower reduction in lifetime at 650 °C than at 550 °C. In comparison to similar CC conditions, CF loading resulted in an expeditious and prominent microstructural evolution, which assists in accumulating additional inelastic strain. Nevertheless, microstructural evolution appeared independent of the nature of the applied hold-time waveform which became pronounced with increase in temperature. The dislocation arrangements varied from grain to grain and even within the grains as a consequence of different grains sizes, orientations, and oxide particle distributions. Furthermore, in comparison to CC, an extensive W-enriched Laves phase precipitation/coarsening in the form of continuous or semi-continuous grain boundary networks was realized under CF loading. The damage studies revealed a single surface initiated transgranular crack under CC and multiple surface initiated transgranular cracks under CF loading. The transgranular crack path under CC possessed an intergranular tendency under CF loading which is obviously associated with the lower cyclic life.

The present study reports on the high-temperature low-cycle fatigue (LCF) behavior of a ferritic bimodal 12Cr oxide dispersion strengthened steel. The fully reversed strain-controlled LCF tests were conducted at a constant total strain rate with different axial strain amplitude levels. The measured cyclic stress response showed four distinct stages which include instant initial cyclic softening followed by gradual cyclic hardening thereafter continuous linear cyclic softening and finally crack initiation and growth stage. The rate and amount of these stages depend on both strain amplitude and testing temperature. The cyclic stress-strain and strain-life relationships were obtained through the test results, and related LCF parameters were calculated. Microstructural investigations, using electron microscopy, were carried out to shed light on the deformation mechanisms. At 550 °C, no appreciable microstructural evolution occurred, which is consistent with the manifestation of Masing behavior. A propensity towards cross-slip and hence the formation of three-dimensional dislocation structures increased with increase in loading amplitude and testing temperature, which assisted in accumulating higher inelastic strain and led to slightly higher cyclic hardening at these conditions. Furthermore, W-enriched Laves phase particles precipitated at grain boundaries. The evolution of cell structures with strain amplitude at 650 °C resulted in a small deviation from Masing behavior. The damage studies revealed multiple surface initiated transgranular cracks under higher strain amplitudes and a single surface initiated transgranular crack under lower strain amplitudes. The different stages of the crack path, whose durations depend on the testing conditions, revealed distinct fatigue fracture features.

In this study, the tensile properties of a tempered martensitic 9Cr oxide dispersion strengthened (ODS) steel are investigated. The tensile tests were performed in the temperature range of 25 °C to 800 °C at a nominal strain rate of 10−3 s−1. At room temperature, the material exhibited high yield strength and ultimate tensile strength of 929 and 1052 MPa, respectively. A decrease in strength was observed with increase in temperature down up to 156 MPa of ultimate tensile strength at 800 °C. The total elongation-to-failure that was 7.6% at 25 °C increased sharply with increase in temperature and reached a maximum of 38.4% at 700 °C. In order to compare the influence of heat treatment the tensile tests were also performed on specimens when in a ferritic state. Interestingly, at elevated temperatures both states presented a similar trend of strength and elongation. Transmission electron microscopy after deformation revealed a modification of the deformation mechanism with the temperature. The dislocation activity that was homogeneously distributed at room temperature was localized close to grain boundaries at elevated temperatures. A strong particle–dislocation interaction was observed at all testing temperatures. Orowan mechanism is supposed to govern particle–dislocation interaction at moderate temperatures. At elevated temperatures, an attractive particle–dislocation interaction phenomenon called interfacial pinning was identified. The additional microstructural evolution that includes reduced dislocation density, the transformation of lath structure into coarse equiaxed grains and the M23C6 carbide coarsening, resulted in a loss of strength at elevated temperatures. Fracture surface investigation at room temperature revealed intragranular fracture with dimples. As the temperature increased, the fracture surface formed with a shear-lip zone at the outer periphery and dimples that were larger and deeper in comparison to the ones formed at lower temperatures. These observations are associated with the increased ductility. A change in fracture mechanism from intragranular to intergranular fracture was observed at 650 °C. This change became fully apparent at 800 °C where it is associated with a reduced ductility. The change in damage mechanism is due to the modification of the deformation mechanism. In comparison to other commercial, as well as experimental, ODS steels, the material offers an excellent compromise between strength and ductility.

This article describes the tensile properties and fracture characteristics of a 12Cr oxide dispersion strengthened (ODS) ferritic steel with unique elongated bimodal grain size distribution. The tensile tests were carried out at four different temperatures, ranging from room temperature to 700 °C, at a nominal strain rate of 10 −3 s −1 . At room temperature the material exhibits a high tensile strength of 1294 MPa and high yield strength of 1200 MPa. At 700 °C, the material still exhibits relatively high tensile strength of 300 MPa. The total elongation-to-failure exceeds 18% over the whole temperature range and has a maximum value of 29% at 600 °C. This superior ductility is attributed to the material's bimodal grain size distribution. In comparison to other commercial, as well as experimental, ODS steels, the material shows an excellent compromise between strength and ductility. The fracture surface studies reveal a change in fracture behavior from a mixed mode fracture at room temperature to fully ductile fracture at 600 °C. At 700 °C, the fracture path changes from intragranular to intergranular fracture, which is associated with a reduced ductility.

The collaborative study is focused on the relationship between microstructure and yield stress for an ODS Fe-9%Cr-based transformable alloy and an ODS Fe-14%Cr-based ferritic alloy. The contributions to the total room temperature yield stress arising from various strengthening mechanisms are addressed on the basis of a comprehensive description of the microstructures uncovered by means of transmission electron microscopy (TEM), electron backscatter diffraction (EBSD), small-angle neutron scattering (SANS) and atom probe tomography (APT). While these methods provide a high degree of complementarity, a reasonable agreement was found in cases of overlap of information. The derived set of microstructure parameters along with reported strengthening equations was used to calculate the room temperature yield stress. The estimates were critically compared with the measured yield stress for an extended set of alloys including data reported for Fe-Cr model alloys and steels thus covering one order of magnitude or more in grain size, dislocation density, particle density and yield stress. The comparison shows that particle strengthening, dislocation forest strengthening, and Hall-Petch strengthening are the major contributions and that a mixed superposition rule reproduces the measured yield stress within experimental scatter for the whole extended set of alloys. The wide variation of microstructures additionally underpins the conclusions and goes beyond previous work, in which one or few ODS steels and narrow microstructure variations were typically covered.

In the present work, high temperature deformation behavior of oxide dispersion strengthened T91 was investigated and linked to the corresponding microstructure. First, tensile properties are presented and discussed in terms of yield strength, tensile stress and total elongation as a function of temperature. The results are compared to the matrix material and other ODS alloys. Second, transmission electron microscopy was applied to as received and deformed tensile test specimens. It is shown that the Y2O3 particle diameter increases slightly upon deformation at elevated temperatures. Additionally, distinctive coarsening of M23C6 carbides occurs at prior austenite grain boundaries. At temperatures above 500°C, dislocations are straight and pile up at grain boundaries due to thermally activated climbing. Oxide dispersion strengthened T91 provides high strength due to strong particle/dislocation interactions and good toughness properties.

Transmission electron microscopy (TEM) was applied to study the microstructure and composition of a novel austenitic oxide dispersion strengthened (ODS) steel (16 wt.% Cr, 14 wt.% Ni, 1 wt.% W). Special attention was paid to characterize the dispersed nano-oxide particles. It was found, that these embedded particles have a size distribution ranging from 1 to 70 nm. The particles with sizes up to 20 nm, have a density of ∼1023 m−3, are mainly Y2O3 or Y2Ti2O7. The Y2Ti2O7 particles with size up to 3 nm were found to be fully coherent with the γ-Fe matrix (without defects) with relationships 〈001〉 Y2Ti2O7 II 〈001〉 Fe, 〈111〉 Y2Ti2O7 II 〈111〉 Fe and (110) Y2Ti2O7 II (110) Fe. The larger Y2Ti2O7 were found semi-coherent. Due to the misfit of ∼6% between the Y2Ti2O7 crystal lattice and the matrix, which have similar FCC structures, misfit dislocations were observed. The Y2O3 particles, having a smaller misfit of ∼2%, but with BCC structure, reveal coherence with the matrix with relationship 〈001〉 Y2O3 II 〈001〉 Fe, 〈111〉 Y2O3 II 〈111〉 Fe, and {110} Y2O3 II {110} Fe. It was observed that different nanoparticles have a tendency to agglomerate which leads to the loss of their coherency. Independent of the sizes and composition, the nano-oxide particles (which are either fully- or semi-coherent) were able to pin dislocations under repulsive interaction which is known to improve mechanical properties.

The microstructure of InN layers grown by plasma-assisted molecular beam epitaxy on Si(111) substrates and an AlN buffer layer was investigated. InN layers with a thickness of ∼500 nm were deposited at substrate temperatures between 325 °C and 375 °C under otherwise identical conditions. The structural characterization was performed by scanning electron microscopy and different transmission electron microscopy techniques including selective-area electron diffraction, electron-energy loss spectroscopy and energy-dispersive X-ray spectroscopy. The microstructure of the InN layers changes considerably despite the comparably small interval of growth temperatures.

AlN layers with a thickness of 250 nm were grown by plasma-assisted gas source molecular-beam epitaxy on Si(111) at substrate temperatures between 600 °C and 900 °C. The surface morphology and microstructure of the AlN layers were analyzed by scanning and transmission electron microscopy. Different defect types are observed in the AlN layers and at the AlN/Si(111) interfaces as a function of the temperature: inclusions of pure Al in the Si-substrate, crystallites of the cubic AlN phase, dislocations, stacking faults and inversion domain boundaries. The formation and concentration of the defects depends strongly on the substrate temperature during the growth. X-ray diffraction rocking curves for the (0002) reflection yield minimum full width at half maximum values for the sample grown at the 900 °C under Al-rich conditions indicating optimum structural quality. However, the discussion of the entity of defects will show that a more differentiated view is required to assess the overall quality of the AlN layers.

In this study, we investigated the effect of in situ annealing on InAs quantum dots site-selectively grown on pre-structured GaAs substrates. A morphological transition is observed with original double dots merging into one single dot during annealing. This is accompanied by a reduction of quantum dots originally nucleating between defined sites. The photoluminescence intensity of annealed site-selective quantum dots is compared to annealed self-assembled dots with linewidths of single dot emission of about 170 and 81 μeV, respectively. UV-ozone cleaning is used to optimize the sample cleaning prior to quantum dot growth.

In an effort to understand why short wavelength (~1000 nm) GaAs-based photonic crystal slab nanocavities have much lower quality factors (Q) than predicted (and observed in Si), many samples were grown, fabricated into nanocavities, and studied by atomic force, transmission electron, and scanning electron microscopy as well as optical spectroscopy. The top surface of the AlGaAs sacrificial layer can be rough even when the top of the slab is smooth; growth conditions are reported that reduce the AlGaAs roughness by an order of magnitude, but this had little effect on Q. The removal of the sacrificial layer by hydrogen fluoride can leave behind a residue; potassium hydroxide completely removes the residue, resulting in higher Qs.

The In-concentration in InGaAs quantum dots located within a GaAs matrix was determined with the composition evaluation by lattice fringe analysis (CELFA) technique. However, the results obtained with this method cannot account for the three-dimensional shape of quantum dots and their embedding in GaAs. A correction procedure was developed that takes into consideration the shape of the quantum dots and the TEM sample thickness and quantum-dot size. After correction, In-concentration profiles show an increasing Incontent towards the top of the quantum dots which is consistent with the effect of In-segregation and earlier studies using other experimental techniques. Dedicated to Prof. Wolfgang Neumann on the occasion of his 65 th birthday

An ensemble of InAs quantum dots with ground state transition energies centered at 1.216 eV and density 1011dots/cm2 has been studied by time-resolved photoluminescence (PL). The wavelength of the 100-fs excitation pulse was tuned through the ground (excited) state transitions, resulting in resonant (optical phonon sideband) PL. The decay of the PL was time resolved with a streak camera in the interval 1.5–3 ns to avoid scattered laser light. The intensity of the PL was recorded with its polarization both parallel with and perpendicular to the excitation polarization (along one of the crystal’s cleave axes); the ratio is 2.22 at low temperatures and low excitation. A phenomenological rate equation analysis is made, separating the excitations into two classes, one polarized along the excitation polarization and the other unpolarized (either that way immediately after the excitation pulse or scattered from the first class). Excellent fits to the data lead to the conclusion that both classes decay radiatively with a lifetime of 1 ns, and a transfer from the polarized to the unpolarized species takes place with a distribution time of 12 ns at low temperatures and low excitation, dropping rapidly toward zero for temperatures above 30 K and for intense excitation levels. The polarization of a coherently excited ground state exciton should dephase with a rate equal to the sum of the radiative rate plus the inverse of this distribution time.

We report on the investigation of CdSe/ZnSe heterostructures by transmission electron microscopy (TEM) and photoluminescence spectroscopy (PL). CdSe layers with nominal thicknesses tCdSe between 0.5 and 8 monolayers (ML) were embedded in a ZnSe matrix and grown on a GaAs(001) substrate by molecular-beam epitaxy at 280 °C. The Cd-distribution was obtained from high-resolution TEM lattice fringe images using composition evaluation by lattice fringe analysis technique. The measured minimum, average and maximum Cd-concentrations and the overall CdSe contents in the layers increase with the nominal CdSe layer thickness and reach a constant value at tCdSe = 4 ML. The measured CdSe content in the regions with the maximal Cd-concentration continues to increase for tCdSe ≥ 5 ML. The increasing of measured Cd-concentrations/CdSe contents is correlated with a red shift of PL spectra. In the CdSe layers with tCdSe ≥ 5 ML, formation of defects is observed. The increasing density of defects and decreasing intensity of PL spectra with tCdSe suggest that the critical thickness for defect formation during CdSe growth on the ZnSe(001) is between 4 ML and 5 ML. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We present a detailed transmission electron microscopy study of the structure and distribution of self-organized Fe-rich nanocrystals in (Ga,Fe)N grown via metal-organic vapor phase epitaxy (MOVPE). In samples with a concentration of magnetic ions above the solubility limit— x ⩾ 0.4 % —of Fe in GaN at the growth conditions we can identify three different phases of the Fe-rich enclosures, namely: (a) α -Fe segregated in proximity of the surface, (b) e -Fe 3 N in the volume beneath, though well above the nominal interface between the undoped GaN buffer and the Fe-doped overlayer, and (c) nanocrystals at an intermediate location showing the presence of adjacent α -Fe and e -Fe 3 N. These observations suggest that nitrogen evaporation at the growing surface play an important role in the formation of Fe-rich phases. The occurrence of α -Fe and e -Fe 3 N in the same nanocrystal indicates that the two phases are actually correlated: phase separation may first lead to the precipitation of e -Fe 3 N, which inherits the hexagonal structure of GaN but generates misfit dislocations, and then a further depletion of nitrogen promotes the conversion of e -Fe 3 N into α -Fe.

We have studied the crystal structure of GaAs nanowhiskers grown by molecular beam epitaxy (MBE) on gold-activated GaAs(111)B substrates. The results of reflection high-energy electron diffraction and transmission electron microscopy showed that the MBE-grown GaAs nanowhiskers can form a crystal structure of sphalerite, wurtzite, or an intermediate phase close to 4H polytype, depending on the deposition conditions and the size of catalyst droplets. The results are interpreted within the framework of a thermodynamic model.

We investigate the dynamics of electrons injected into InAs/GaAs quantum dots by initializing and further observing the spin state of the electrons. For this purpose, we use spin polarized light-emitting diodes where the electron spin is set in a semimagnetic ZnMnSe layer. We find that the degree of optical polarization depends strongly on the ground state energy of the quantum dot. A dependence of polarization on dopant concentration in the spin aligner suggests an influence of residual electrons in the quantum dots.

We investigated InGaAs layers grown by molecular-beam epitaxy on GaAs (001) with transmission electron microscopy (TEM) and photoluminescence spectroscopy. InGaAs layers with In-concentrations of 16, 25 and 28 % and respective thicknesses of 20, 22 and 23 monolayers were deposited at 535 °C. Island formation is observed for the layer with the highest In-concentration. Inconcentration profiles were obtained from high-resolution TEM images by composition evaluation by lattice fringe analysis. The measured profiles can well be fitted applying the segregation model of Muraki et al. [Appl. Phys. Lett. 61 (1992) 557] and are in excellent quantitative agreement with the photoluminescence peak positions. From our data we conclude that island formation occurs when the amount of Indium in the In-floating layer reaches 1.1±0.2 monolayers indium.

We report on progress in growth and applications of submonolayer (SML) quantum dots (QDs) in high-speed vertical-cavity surface-emitting lasers (VCSELs). SML deposition enables controlled formation of high density QD arrays with good size and shape uniformity. Further increase in excitonic absorption and gain is possible with vertical stacking of SML QDs using ultrathin spacer layers. Vertically correlated, tilted or anticorrelated arrangements of the SML islands are realized and allow QD strain and wavefunction engineering. Respectively, both TE and TM polarizations of the luminescence can be achieved in the edge-emission using the same constituting materials. SML QDs provide ultrahigh modal gain, reduced temperature depletion and gain saturation effects when used in active media in laser diodes. Temperature robustness up to 100 °C for 0.98 μm range vertical-cavity surface-emitting lasers (VCSELs) is realized in the continuous wave regime. An open eye 20 Gb/s operation with bit error rates better than 10−12has been achieved in a temperature range 25–85 °Cwithout current adjustment. Relaxation oscillations up to ∼30 GHz have been realized indicating feasibility of 40 Gb/s signal transmission.

The influence of the growth conditions during molecular-beam epitaxy on the properties of InAs/GaAs quantum dot structures were systematically investigated by low temperature photoluminescence spectroscopy and transmission electron microscopy. The circular polarization degree (CPD) of the electroluminescence was compared for two quantum-dot spin-injection light-emitting diodes. The CPD depends on the position of the emission energy in the luminescence band. This correlation is similar for both samples despite the strongly different quantum dot morphologies. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We investigate electrical spin injection from a semi-magnetic n-type ZnMnSe spin aligner into III–V p–i–n diodes with InGaAs quantum dots (QDs) in the active layer. Quantitative transmission electron microscopy techniques are applied to characterize the different dot types used. Analysis of the circular polarization degree (CPD) of the device emission indicates the spin polarization of the injected electrons. Values of more than 70% are obtained for the wetting layer and high-energy QD states. However, the CPD shows a strong spectral dependence due to spin relaxation at the stage, before the electrons are finally captured in the dots. This is, e.g., evidenced by an initial increase of the polarization degree with rising temperature, attributed to motional narrowing effects. As a prerequisite for more detailed studies, we also demonstrate electrical spin injection into single InGaAs QDs, which should provide the basis for future single spin manipulation experiments. Finally, we suggest GaInNAs as optically active material for the realization of spin-polarized light-emitting diodes with long-wavelength emission. First results indicate CPD values of up to 80% for λ = 1130 nm, suggesting this approach to be very promising. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We report on the injection of electron spins into InGaAs quantum dots with an efficiency of up to 60%. This injection is observed in p-i-n light-emitting diode structures using the diluted magnetic semiconductor ZnMnSe as spin aligner (spin-LED). The degree of spin polarization is deduced from the circular polarization degree of the photons emitted when the injected electrons recombine in the quantum dots with unpolarized holes. We observe a strong energy dependence of the polarization degree with a strong increase starting from zero to a high value on the high energy side of the emission spectrum. To study the origin of this dependence, we compare results of two quantum-dot samples with emission peaks at 1.2 eV and 1.33 eV, respectively. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We report on electrical spin injection into InGaAs-based nanostructures through a semi-magnetic ZnMnSe layer in a magnetic field. We show an optical polarization degree of the electro-luminescence of up to 35% in InGaAs quantum wells and up to 14% in InGaAs quantum dots. To be able to attribute the polarization of the emitted photons to the spin alignment in the semi-magnetic layer, we have fabricated reference devices and performed all-optical measurements.

Using the well established MBE process, we introduced colloidal nanocrystals (NCs) into monocrystalline semiconductor layers. CdSe nanorods (NRs) and CdSe(ZnS) core-shell nanodots (NDs) with radii R between 2 and 4 nm on a ZnSe surface are capped by thin (d ∼ 2R ) and thick (d > 2R ) MBE-grown ZnSe layers of a thickness d . This new growth technique does not rely on strain like the Stranski-Krastanow growth of CdSe/ZnSe quantum dot layers, and thus offers additional degrees of freedom for choosing the NC composition, shape and size. In- and ex- situ characterisations of the samples are showing a ZnSe cap layer of high crystalline quality. After the growth of the cap layer the dots preserve their shape and emission spectrum. High-resolution transmission electron microscope images show the direct connection of the NCs crystal lattice to the crystal lattices of the ZnSe buffer and the epitaxially grown cap layer. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We present a growth technique which combines wet-chemical growth and molecular beam epitaxy (MBE) to create complex semiconductor nanostructures with nanocrystals as active optical material. The obtained results show that wet-chemically prepared semiconductor nanocrystals can be incorporated in an epitaxally grown crystalline cap layer. As an exemplary system we chose CdSe nanorods and CdSe(ZnS) core-shell nanocrystals in ZnSe and discuss the two limits of thin (d approximately 2R) and thick (d>2R) ZnSe cap layers of thickness d for CdSe nanorods and nanodots of radii R between 2 and 4 nm. In contrast to the strain-induced CdSe/ZnSe Stranski-Krastanow growth of a quantum dot layer in a semiconductor heterostructure, the technique proposed here does not rely on strain and thus results in additional degrees of freedom for choosing composition, concentration, shape, and size of the nanocrystals. Transmission electron microscopy and X-ray diffractometry show that the ZnSe cap layer is of high crystalline quality and provides all parameters for a consecutive growth of Bragg structures, waveguides, or diode structures for electrical injection.

Quantitative high-resolution transmission electron microscopy was applied to determine the indium and nitrogen distributions in a GaInNAs/GaAs heterostructure. Two almost identical samples grown by gas-source molecular beam epitaxy on GaAs (001) substrates were investigated. The samples contained InGaAs and GaInNAs wells with the same thickness and In concentration. According to high-resolution X-ray diffractometry the average In concentration is 27% in both samples, with an N concentration of 1.1% in the GaInNAs quantum well. The evaluation of the photoluminescence peak position of the InGaAs sample and additional photoreflectance measurements yielded an In concentration of 30%. In-concentration profiles were obtained with the composition evaluation by the lattice fringe analysis (CELFA) technique for the InGaAs well using the chemical sensitivity of the (002) reflection. The averaged In concentration in the InGaAs quantum well was determined to be (30±1)% in good agreement with the other measurement techniques. The N-concentration profiles in the GaInNAs wells were measured by the comparison of the CELFA results in the samples with and without nitrogen. The N distribution is characterised by two maxima close to the GaInNAs/GaAs interfaces. The measured N concentration in the central part of the quantum well is (2.5±1.0)%.

(3 1 1)A GaAs/AlAs corrugated superlattices (CSLs) and satellite (3 1 1)B and (1 0 0) SLs were studied using Raman spectroscopy, high-resolution transmittance electron microscopy (HRTEM) and photoluminescence (PL). The thickness of GaAs layers was varied from 1 monolayer (ML) to 10 ML , the thickness of AlAs barriers was 10 ML in (3 1 1) direction. The strongest modification of the Raman spectra is found for the case of partial ( nm ) GaAs filling of the AlAs surface. The calculated and experimental Raman spectra demonstrated a good agreement for both complete (⩾1 nm ) and partial ( nm ) GaAs filling of the AlAs surface. According to Raman and HRTEM data, in the case of partial filling of (3 1 1)A AlAs surface, GaAs forms quantum well wires of finite length (quantum dots). A drastic difference of PL from grown side-by-side (3 1 1)A and (3 1 1)B SLs was observed. A strong room temperature PL in the green–yellow spectral region was observed in GaAs/AlAs (3 1 1)A CSLs containing GaAs type-II quantum dots.

We investigate cubic Zn1−xMnxSe layers (0 ≤ x ≤ 0.43) grown on GaAs(001) by elemental source molecular-beam epitaxy. To determine the Mn concentration, energy-dispersive X-ray analysis, Rutherford backscattering and particle-induced X-ray emission are applied. In the next step we utilize X-ray diffraction to measure both the strained lattice parameter of the films along and perpendicular to the growth direction. The unstrained values are then derived from elasticity theory using elastic constants for cubic Zn1−xMnxSe calculated from published data for hexagonal material. Finally, an extrapolation to x = 1 yields the lattice parameter of cubic MnSe (a0 = 5.90 A). (© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We investigated the influence of the temperature of the ZnSe cap layer growth on the Cd distribution in CdSe/ZnSe heterostructures by transmission electron microscopy, high-resolution X-ray diffraction and photoluminescence spectroscopy. ZnSe buffer layers were grown by molecular-beam epitaxy on GaAs (0 0 1). For the CdSe layers and ZnSe cap layers, migration-enhanced epitaxy was used. Increasing the temperature for the cap layer growth reduces the Cd concentration and the total CdSe amount in the (Cd,Zn)Se layers leading to a blue shift of the photoluminescence spectra. This is owing to a stronger Cd desorption during the cap layer growth for a higher substrate temperature.

We report the transformation of Shockley partial dislocations (PDs) into Frank PDs in lattice-matched ZnS 0.04 Se 0.96 /GaAs(001) as investigated by transmission electron microscopy. The ZnS 0.04 Se 0.96 layers, with a nominal thickness of 70 nm, were grown on GaAs(001) by metal-organic chemical vapour deposition at 350°C. We mainly find stacking-fault pairs on the (111) and planes that are bound by Shockley PDs with Burgers vector . Different reactions are observed between PDs taking place in situ in the electron microscope, leading to the transformation of Shockley PDs into Frank PDs with and stacking faults on the or planes.

The influence of the growth conditions during capping of CdSe/ZnSe quantum structures grown on GaAs(001) by molecular-beam epitaxy (MBE) were systematically investigated by high-resolution x-ray diffraction, transmission electron microscopy, and temperature dependent, partly time-resolved photoluminescence spectroscopy. The results clearly indicate formation of quantum wells with potential fluctuations if conventional MBE is used for capping the CdSe by ZnSe. In contrast, quantum dot formation occurs using migration enhanced epitaxy for this growth step. In the latter case, quantum dots can be obtained without formation of stacking faults.

The photoluminescence properties of type II GaAs/AlAs superlattices grown on the (311) surface are determined by their polarity. Previous HRTEM investigations demonstrated a corrugation (with height of 1 nm and period of 3.2 nm) of both GaAs/AlAs and AlAs/GaAs interfaces in samples grown on the (311)A surface. In the present study, a lateral periodicity of 3.2 nm is also revealed in HRTEM images of a superlattice grown on the (311)B surface and in their Fourier transforms. However, this periodicity is poorly pronounced, which is due to fuzzy corrugation and the presence of a long-wavelength (>10 nm) disorder. Photoluminescence spectra of the GaAs/AlAs superlattice on the (311)A surface are strongly polarized relative to the direction of interface corrugation, in contrast to the (311)B superlattice, in which the corrugation is weakly pronounced. It was found that the strong mixing between the Θ and X minima of the conduction band, occurring only in sublattices with strongly corrugated interfaces, allows generation of bright red luminescence at 650 nm up to room temperature. The distinctions revealed between the superlattices grown on the (311)A and (311)B surfaces confirm that it is precisely the interface corrugation, and not crystallographic orientation, that governs the optical properties of (311) superlattices.

While providing a general overview over the current status of self-organization of quantum islands in the II–VI semiconductor system, with the main focus on CdSe embedded in ZnSe, this paper shall give an introduction to the possibilities opened by a modification of the standard growth technique. In molecular beam epitaxy, we have substituted the generally used Cd-elemental source with a CdS-compound source. The sulfur is usually not included in the growing layer. However, its presence can be surfactant-like while the elevated Cd-temperature of the dissociated CdS leads to changed thermodynamic conditions on the growth front. Using migration enhanced epitaxy, nearly perfect quantum wells with respect to lateral homogeneity can be obtained by suppressing the inherent Cd segregation and clustering. These processes are generally responsible for the formation of small islands (SI) (lateral diameter 3–5 nm) even when not attempting to grow island like structures. The suppression of these SI was a first step to gain control over the island formation. Larger islands with central Cd concentrations above 40% are more of interest for device applications, since a population at room temperature is necessary. In particular, high-density systems are required. Using the modified growth mode, well correlated, stacked island systems were obtained. Their outstanding structural and optical properties will be discussed in detail. The absence of a closed wetting layer in the CdSe/ZnSe system and the appearance of island like structures, even at submonolayer nominal deposition, further corroborate the assumption that island formation does not readily occur in a standard Stranski–Krastanow growth mode, which is assumed for InAs/GaAs.

In this study we report the dependence of the defect configuration observed in CdSe/ZnSe quantum dot structures on the ZnSe cap layer thickness by transmission electron microscopy (TEM) and reflection high-energy electron diffraction (RHEED). The samples were grown by molecular-beam epitaxy at 350°C. The nominal thickness of the CdSe layers was approximately 3 monolayers and the cap layer thickness was varied in the range from 3 to 60 nm. In all samples, RHEED showed a transition from the two-dimensional (2D) to the three-dimensional (3D) growth mode during the CdSe deposition. By TEM we found pairs of stacking faults (SFs) lying on one of the two pairs of lattice planes given by {(111)-(1 11)} and {(111)-(111)}. The SFs are bound by Shockley partial dislocations. Both of the SFs forming one pair originate from the same stair-rod dislocation with Burgers vector b = ⅙{110} lying at the CdSe-ZnSe interface inside a Cd-rich region (island). Measuring the atomic displacements in the vicinity of the S...

GaAs/AlAs superlattices grown simultaneously on GaAs substrates with the (311)A and (311)B orientations have been studied by photoluminescence and high-resolution transmission electron microscopy with a Fourier analysis of images. A periodic interface corrugation is observed for (311)B superlattices. A comparison of the structure of (311)A and (311)B superlattices indicates that the corrugation occurs in both cases and its period along the \([01\overline 1 ]\)direction is equal to 3.2 nm. The corrugation is less pronounced in (311)B superlattices, wherein it exhibits an additional modulation (long-wavelength disorder) with the characteristic lateral size exceeding 10 nm. The vertical correlation of regions rich in GaAs and AlAs, which is well observed in (311)A superlattices, is weak in (311)B superlattices due to the occurrence of long-wavelength disorder. The optical properties of (311)B superlattices are similar to those of (100) ones and differ radically from those of (311)A superlattices. As distinct from (311)B, strong photoluminescence polarization anisotropy is observed for (311)A superlattices. It is shown that it is the interface corrugation rather than the crystallographic (311) surface orientation that determines the optical properties of (311)A corrugated superlattices with thin GaAs and AlAs layers.

Vertically correlated islands of a quality comparable to the model system InAs/GaAs have been fabricated using modified molecular beam epitaxy. While island widths range from 20-50 nm, island heights are very homogeneous, resulting in narrow photoluminescence (PL) emission line width of 20-40 meV even for stacks of 50 layers. Island centers contain Cd concentrations up to more than 80%. Distinct PL double peaks can be assigned to correlated island stacks with large islands and the uncorrelated region, containing much smaller, laterally interacting islands.