The objective of this work is to propose new low-dimensional systems such as two-dimensional nets of quantum wires. The preparation method is based on electrochemical etching of multilayer quantum structures, controlled by measurements of the photoluminescence spectra. Modification of the surface structure is also traced with scanning probe microscopy. The process of a porous structure formation in superlattices leads to additional spatial confinement in the InGaAs layer plane through it, to formation of two-dimensional (2D) nets of quantum wires, as well as transformation of two-dimensional (2D) energy spectra of electrons into one-dimensional (1D) spectra. This procedure was realized on structures with a single InGaAs layer and on multilayer periodic structures. Electrophysical measurements of the electron gas characteristics were carried out on porous structures with 2D nets of quantum wires.

Process-induced defects are a serious issue for modern sub-micron Si LSIs. To characterize such defects, two different techniques are useful: electrically detected magnetic resonance (EDMR) and transmission electron microscope (TEM), which can detect small (point) and extended defects, respectively. We applied EDMR and TEM to the issue of defect-induced leakage currents in dynamic-random-access memory (DRAM) cells. For our DRAM samples (a 0.25-μm-rule series), although TEM showed no extended defects, EDMR successfully detected two types of point defects: V2+Ox (Si divacancy-oxygen complexes) and larger Si vacancies (at least larger than V6). We confirmed that these defects are the source of DRAM leakage currents. The observed defects were formed by ion implantation processes, but were more thermally stable than those in bulk Si crystals. The origins of this enhanced stability are attributed to the presence of oxygen atoms and a strong mechanical strain in LSIs. To clarify the origin of the complicated strain in LSI structures, we can directly measure the local-strain distribution in DRAM samples by means of convergent-beam electron diffraction (CBED) using TEM, which provides us with a valuable hint for understanding the formation mechanism of process-induced defects.

The increase of the gate leakage current of advanced CMOS technologies makes standard electrical characterization methods as C(V) measurement or charge pumping more complex and uncertain. In this paper, and based on C(V) characteristics, main elements that directly affect the electrical measurements of ultra thin MOS devices are clarified. Then, classical parameter extraction techniques are reviewed, pointing out their absolute limitations or giving potential keys of improvement.

The current status of SiC bulk growth is reviewed, while specific attention is given to the effect of defects in SiC substrates and epitaxial layers on device performance and yield. The progress in SiC wafer quality is reflected in the achievement of micropipe densities as low as 0.92 cm−2 for a 3-inch n-type 4H-SiC wafer, which provides the basis for a high yielding fabrication process of large area SiC power devices. Using a Murphy Probe Yield Analysis for the breakdown characteristics of 10 kV PiN diodes we have extracted an “effective” defect density for 4H-SiC material to be as low as 30 cm−2, providing valuable information to further isolate and address the specific material defects critical for device performance. We address the problematic degradation of the forward characteristics (Vf-drift) of bipolar SiC PiN diodes [CITE]. The underlying mechanism due to stacking fault formation in the epitaxial layers and possible effects of device processing are investigated. An improved device design is demonstrated, which effectively stabilizes this Vf-drift. We show the progression in the development of semi-insulating SiC grown by the sublimation technique from extrinsically doped material to high purity semi-insulating (HPSI) 4H-SiC bulk crystals of up to 100 mm diameter without resorting to the intentional introduction of elemental deep level dopants, such as vanadium. Uniform resistivities in 3-inch HPSI wafers greater than 3 × 1011 Ω-cm have been achieved. Secondary ion mass spectrometry, deep level transient spectroscopy and electron paramagnetic resonance data suggest that the semi-insulating behavior in HPSI material originates from deep levels associated with intrinsic point defects. MESFETs produced on HPSI wafers are free of backgating effects and have resulted in the best combination of power density and efficiency reported to date for SiC MESFETs of 5.2 W/mm and 63% power added efficiency (PAE) at 3.5 GHz.

Two approaches to defect-selective etching used for revealing and analysis of defects in GaN and SiC are described and critically evaluated. These are: (i) orthodox etching which results in formation of pits on the defect sites and (ii) electroless etching, which yields protruding etch features. The mechanisms of surface reactions that are responsible for the distinct differences in the morphology of defect-related etch features are discussed. The most frequently used etching systems for GaN and SiC and the methods of verification of their reliability in revealing different types of defects are described.

Non-destructive techniques are ideal for process control and process development. They do not involve any special wafer treatment and the wafers can be further processed in the manufacturing line. A relatively new technique that can be applied to materials characterizations is room temperature photoluminescence (RT-PL). Defects can be revealed because of the enhanced non-radiative recombination at the defect site. The efficiency of detecting defects in SiGe was evaluated. Defects could be detected in blanket film SiGe layers, and there was a good correlation with bright field microscopy inspection and laser scanning methods. The effect of ex-situ cleaning, pre-bake and Boron doping conditions have been investigated. The results follow the expected trends, more defects for non-standard processing. The defects detected by RT-PL were confirmed using transmission electron microscopy (TEM). Defects could also be detected in part-processed patterned wafers, giving a direct indication of the material quality in product wafers.

The principles of the convergent beam electron diffraction technique, which is a point-to-point method for local strain analysis of thin crystals in the transmission electron microscope, are briefly outlined. The availability in modern instruments of scanning attachments coupled with high-angle annular dark-field detectors (STEM/HAADF) has recently enabled the automatic acquisition of diffraction patterns in a large number of points, selected by digitally rastering the probe in a two dimensional region of the sample. As the components of the strain tensor can be calculated at each point, 2D strain mapping has thus become possible. An example of application of the technique to deep sub-micron shallow-trench isolation structures in silicon is reported.

Application of scanning techniques is strongly required for in-line/off-line diagnostics of semiconductors with spatial inhomogeneity of electronic properties due to uneven defect distribution. The defect mapping allows performing an initial screening and follow-up local analyses in large-scale materials and devices for electronic applications. This approach is illustrated on multi-crystalline silicon (mc-Si) wafers for solar cells. Four different mapping techniques were applied concurrently to full-size as-grown and processed mc-Si. They include the mapping of (1) minority carrier lifetime, (2) room-temperature photoluminescence, (3) dislocation density, and (4) temperature dependent EBIC and cathodoluminescence. Analyses revealed a clear correlation between the “defect” PL band intensity with the maximum at about 0.8 eV and concentration of the dislocations contaminated by impurities and defects. The 0.8 eV band has a distinctive location at wafer areas with reduced minority carrier lifetime, i.e. regions with enhanced recombination activity of defects. It is demonstrated that the concentration of the 0.8 eV defects has analytical relation to the concentration of non-radiative defects in dislocated regions. The model of these defects is discussed.

This paper discusses two techniques that have, at first sight, completely different applications: photon emission microscopy (PEM) and micro-Raman spectroscopy (μRS). We explain the principles of these techniques, their application domain, and we will show that they can in some cases offer complementary information, and be applied to common or similar problems.

Spectroscopic techniques are necessary for the characterization of optoelectronic materials. Among the experimental means used for such characterization, luminescence techniques play a relevant role because they supply information about the mechanisms generating light and are sensitive to structural aspects. Raman spectroscopy provides useful information about alloy instabilities and is sensitive to stress and temperature. Results concerning optoelectronic materials are presented with special emphasis on the application of these techniques to the degradation of laser diodes.

The luminescence of amorphous silicon layers either implanted with Er or co-implanted with Er and O has been studied by photoluminescence (PL) and cathodoluminescence (CL) in the scanning electron microscope. Annealing in nitrogen causes the formation of oxide species and Er-Si complexes or precipitates as well as spectral changes in the visible and infrared ranges. The main CL emission takes place in the visible range while PL spectra reveal intense visible and infrared emission. CL spectra show blue-violet, or green, emission bands whose relative intensities depend on the post-implantation annealing temperature. The PL spectra show a blue-violet band with a series of lines in the violet region related to phonon assisted transitions as well as different emission bands in the range 1200–1500 nm. The influence of the annealing-induced structural changes on the observed spectra is discussed.

Electron microscopy and cathodoluminescence (CL) microanalysis were used for a comparative study of porous layers fabricated by electrochemical etching of n-GaP and n-InP substrates in aqueous solutions of sulfuric and hydrochloric acids. Both the CL and morphology of porous layers were found to depend upon the anodic current density. At high current density (100 mA/cm2) anodization leads to the formation of so-called current-line oriented pores while at low current densities the pores grow along $\langle 111\rangle$ crystallographic directions. The porosity relief was found to give rise to spatial modulation of the CL intensity. The composition microanalysis proved the stoichiometry of porous GaP and InP skeletons, although we found considerable traces of oxygen in porous GaP layers. Self-induced voltage oscillations giving rise to a synchronous modulation of the diameter of pores and CL intensity were evidenced.

Self-assembled β-FeSi2 quantum (QD) dots were grown on n-type Si and investigated in this work. Secondary electron images show the shape and distribution of the quantum dots depends on the temperature and thickness of the Fe deposition. Electrical characteristics were measured in MIS devices prepared by covering the quantum dots by an oxide or by a photoresist layer. The cold deposited and subsequently annealed Fe layers were found to generate large concentration of deep level defects, compensating the upper few microns layer of the silicon wafer. Reactive deposition epitaxy (RDE) growth of QDs – where the iron is deposited on hot substrate—generated much lower concentration of defect, in some devices with characteristics comparable to the reference wafer. The significant scatter on the surface is attributed to inhomogeneous growth conditions and to residual surface contamination on the surface.

FZ silicon samples were multi-implanted with He ions at energies ranging from 0.8 MeV to 1.9 MeV. The dose was 5×1016 He cm−2 for all high energies but 0.8 MeV (3×1016 He cm−2). After implantation, the wafers were submitted to different annealing processes in an Argon atmosphere (A samples: 900 °C and 700 °C for 2 h; B sample 550 °C for 14 h) to optimize both density and size of cavities in order to reach a structure potentially light emitting like that of porous silicon. The samples were studied by cross section transmission electron microscopy (XTEM) and photoluminescence spectroscopy. XTEM results showed that a very thick region containing cavities has been obtained by a multi step implantation and that an annealing at low temperature is sufficient to recover the implantation damage. PL spectra arising from different depths have been obtained. PL results confirm that silicon in the region containing the highest density of cavities remains crystalline with a sufficiently low defect density for optoelectronic applications.

Lateral conductivity effects have been investigated in self-organised InAs quantum dot (QD) structures grown in a GaAs matrix with different cap layers. Current-voltage (I-V), capacitance-voltage (C-V), DLTS, capacitance and conductance frequency dependence, fast defect transient (FDT), and electron beam induced conductivity (EBIC) measurements were applied. The conductivity in the QD plane decays within a distance of 10 microns. The capacitance transients are dominated by the local QD-plane transversal conductivity and by the free carrier transport in the cap layer. The nonequilibrium free carrier created by electron beam excitation develop a potential barrier at macroscopic distances from the electrical contacts.

The results of investigations of undoped GaN heteroepitaxial layers with Atomic Force Microscopy and Kelvin Force Microscopy are presented. The measurements of surface potential were performed in the dark and under illumination below and above the energy gap of GaN. In the dark no fluctuations of potential were observed. Moderate illumination revealed fluctuations of (1) short range (~ 100 nm and ~ 0.1 V), related to threading dislocations with a screw component and (2) long range, independent of morphology (~ 5 µm and ~ 0.3 V). Intense illumination removed long range fluctuations, while dislocations remained surrounded by areas of lower potential.

The industrial use of instruments based on Atomic Force Microscopy that started in the mid 1990's has seen a wide spectrum of applications for product and process controls in semiconductor industry where the continuous effort of smaller and smaller size defect reduction is a driving force of the technological development. The paper will examine a subsection of this world, the electronic grade silicon wafer manufacturing and product development. In this area, the technique is applied to the measurement of surface defect geometry down to a size of a few angstroms and a surface micro-roughness at various areas of inspection from 1E-10 cm2 to 1E-4 cm2. First, a brief description of the measurement principle is given in order to discuss aspects related to accuracy and stability of the technique when micro-roughness values are considered. Then, a few examples involving the inspection of crystal related defects detectable on the silicon wafer surface as well as different textures following various surface-finishing processes will be discussed. Finally, examples of the use of Atomic Force Microscopy on Silicon On Insulator and strained silicon materials will give an insight about the relevance of this technique in the field of New Materials development in the electronics industry.

Two kinds of widegap-semiconductor nanowires, Tb-doped AlN nanowire arrays on Si and free-standing BN nanowires, were fabricated by different catalyst-free methods. Well-aligned Tb-doped AlN nanowire arrays were grown on the Si (111) substrate by magnetron sputtering method. Free-standing BN nanowires were grown by heat-treatment of B-N-O precursor and graphite powders. The crystal structure of nanowires was characterized by using X-ray diffraction and transmission electron microscopy. Cathodoluminescence (CL) observation was performed with a field emission scanning electron microscope operating with an electron beam lower than 5 kV. CL spectra mapping as well as monochromatic CL imaging clearly revealed not only the variation of the luminescence spectra of different nanowires but also that along the single wire. Our results demonstrates the advantage of CL for the characterization of nanowires.

We report on the transformation behaviour of metastable defects labelled EM1 (Ec-0.29 eV), EM2 (Ec-0.41 eV) and EM3 (Ec-0.55 eV) which are induced in n-type silicon by hydrogen implantation. Hydrogen implantation was performed at 88 K with an energy of 90 keV to a dose of 2 × 1010 cm−2. After fabrication of Schottky diodes on the implanted surfaces, deep level transient spectroscopy measurements were made to monitor metastable behaviour of defects. All three metastable defects are regenerated with reverse-bias cooling and removed with zero-bias cooling. 10-min isochronal annealing reveals that EM1 is regenerated around 270 K and is removed around 220 K. The EM2 (EM3) regeneration temperature is around 270 K (270 K) and its removal temperature 220 K (260 K). Isothermal annealing treatments for EM1 show that its transformation follows first order kinetics for both regeneration and removal. An activation energy is 0.94 eV and a frequency factor 6 × 1014 s−1 for regeneration of EM1, and 0.73 eV and 3 × 1013 s−1 for its removal.