We describe a procedure for dispersion bulk Si into ultrasmall nanoparticles, much smaller than what is available today, with high throughput and excellent size definition and control. The quantum dots are ∼ 1 nm in diameter and contain 29 Si atoms[1-2]. We show most unexpected and totally surprising phenomenon. Unlike bulk Si, an optically inert indirect gap material, the particles are extremely active optically in the blue, exceeding the activity of fluorescein, such that single particles are readily detected and imaged, using two-photon near- infrared femto second excitation. This comes nearly a decade after Canham first surprised the scientific community by reporting red luminescence from anodized Si[3]. In addition to being ultrabright, the particles have other useful properties that are unlike those of bulk including being a source of stimulated emission and collimated beam emission[4-5], and harmonic generation[6]. Single electron charging and the confinment energy spacing, both are much larger than the thermal energy[7]. They can be formed into colloids, crystals, films, and collimated beams for applications in the electronics, optoelectronics and biomedical industries. The synthesis utilizes straightforward, low-cost, environmentally benign present-day commercial technologies and material: abundant Si, common chemicals and electricity. The procedure is readily amenable to large-volume production, easy recovery, and controlled delivery.

Using density functional with exchange correlation, configuration interaction and Monte Carlo theory, we constructed a structural propotype for 29 atoms (magic number for the Td symmetry and spherical shape), giving a size, an energy gap and an absorption spectrum that agree with experiment[8]. In the prototype, twenty four atoms form a network of reconstructed Si-Si species on the surface, with five atoms constituting a tetrahedral core.

The results suggest that the surface reconstructed network is the source of the new properties. In these particles, the elasticity drops, atoms become amenable to large displacement, and are subjected to unballanced atomic forces, conditions under which novel molecular structure or configuration, distinct from bulk, and which otherwise does not exist may form. This structure may constitute a transition between the bulk and atomistic phases.

The relationship between chemical states and optical properties for porous silicon (PS) under initial oxidation by heating at low temperatures (150, 200, 250 °C) in air was investigated using an infrared (IR) spectroscopy and photoluminescence (PL) measurements. The measurements in the IR region show that the IR absorption peaks for the Si-H stretching band (2050-2150 cm−1) change with appearance of the Si-O-Si-H stretching band (2100-2300 cm−1) when the thermal oxidation time is increased. On the other hand, the peak in the PL spectrum shows a blue shift from 820 nm to 710 nm with the oxidation time. The observed blue shift of the PL spectrum is due to the decrease in the initial PL peak intensity at 820 nm and the increase at 710 nm. Moreover, the peak intensities in the PL spectra at 820 nm and 710 nm have clear relationship to the amounts of the Si-H bonds and Si-O-Si-H bonds, respectively, as a function of the oxidation time.

These results indicate that luminescence center (LC) for as-prepared PS is ascribed to complexes including Si-H bonds. Also the LC for oxidized PS under the oxidation process at low temperatures is ascribed to complexes including Si-O-Si-H bonds covering the inner surface of PS.

Dangling bond defects in Si1−xGex alloy nanocrystals (nc-Si1−xGex) as small as 4 nm in diameter embedded in SiO2 thin films were studied by electron spin resonance (ESR), and the effects of the defects on photoluminescence (PL) properties were discussed. It was found that the ESR spectrum is a superposition of signals from Si and Ge dangling bonds at the interfaces between nc-Si1−xGex and SiO2 matrices (Si and Ge Pb centers). As Ge concentration increased, the intensity of the signal from the Ge Pb centers increased, while that from the Si Pb centers was almost independent of Ge concentration. The increase in the number of Ge Pb centers was accompanied by strong quenching of the PL. The observed correlation between the two measurements suggests that Ge Pb centers act as efficient non-radiative recombination centers for photogenerated carriers, resulting in the quenching of the PL.

Rutherford Backscattering Spectroscopy (RBS) and channeling studies with 1.4 MeV−4He ions as well as Heavy-Ion Elastic Recoil Detection Analysis (HI-ERDA) with 230 MeV 129Xe ions have been applied to characterize structural properties and the impurity content of thin Si films. The analytical potential of these ion-beam techniques is demonstrated for two types of samples: (1) μc-Si films prepared by dc magnetron sputtering in a pure Ar plasma and (2) homoepitaxial Si films deposited by low-temperature electron-cyclotron resonance PECVD at the transition from oriented to disordered growth. For μc-Si the atomic area density N.d obtained by RBS was compared with the optical thickness n.d (n=refractive index) derived from the interference structure of IR reflection spectra. It is shown that the ratio R=n.d/N.d of these quantities can serve as a figure of merit for the degree of crystalline order. An apparent similarity was found in the case of the epitaxially grown films between the Si disorder profiles evaluated from the RBS channeling spectra and the hydrogen and oxygen profiles determined by HI-ERDA. This suggests that hydrogen and oxygen are preferentially incorporated in the disordered parts of the films.

The ability to organize materials in two or three dimensional structures forms the basis for approach worldwide to construct nanometer sized arrangements. Here we show the interaction of 200 MeV silver ions with a Si( 100) single crystal lattice which has been studied to look for defects with atomic resolution. Employing scanning tunnelling microscopy (STM), we demonstrate that the deposited energy is not stored as random defected arrangements at the irradiation site but as spatially extended structures at predetermined locations. These artificially reordered structures consist of random Si atoms, placed atomically sharp next to the single crystalline lattice.

Silicon nanocrystals have made a recent appearance in the literature because of their potential importance in microelectronic[1] and optoelectronic [2] devices. For example, replacing the traditional Si floating gate in nonvolatile memory field effect transistors with nanocrystals has been shown to produce more reliable and lower power memory devices than traditional transistors [1]. Controlled oxidation of Si nanocrystals is desirable in the processing of Si quantum dots for these devices. If controlled oxidation is attainable, then it could be used to control nanocrystal size, achieve high nanocrystal density, and achieve a high quality interface between Si dot and SiO2 gate oxide layer in nanocrystal based memory. In this work, we report the oxidation properties of various sized silicon nanocrystals, deposited by low-pressure chemical vapor deposition, in different oxidizing environments. In the literature, when a Si column or dot is oxidized below the viscoelastic temperature of SiO2 (950 oC), the oxidation will self-limit to a Si core size that is dependent on the oxidizing conditions and the initial particle size. This self-limiting phenomenon is said to occur because of the presence of a compressive stress in the oxide layer that limits the diffusion of the oxidizing agent through the SiO2 to the Si- SiO2 interface. This compressive stress is present at the interface because of the difference in the density of the oxide and the Si and is enhanced by the radius of curvature of the dot. This self- limiting oxidation phenomenon will be studied experimentally using microscopy techniques and the effect of the constrained structure will be characterized.

The development of novel devices for optical communication and integrated sensor applications is mainly focused on their possible integration into dedicated integrated circuits. The main problem concerning integrated optical systems in silicon technology has always been the formation of an highly efficient silicon-based light emitter which is a key feature to make a real step into the world of integrated optoelectronics.

One of the most promising approaches to form such a silicon based light emitter is ion beam synthesis. In this paper we will report our recent progress in extracting blue-violet (∼400 nm) electroluminescence (EL) from an ion beam synthesized Ge-rich silicon dioxide layer. The power efficiency of the EL was as high as 0.5 % which is one of the best values ever reported for Si - based light emission. The lifetime of the EL-device can reach several hours without special precautions of stabilizing the EL-active layer against ion or other contamination. Moreover, results are reported dedicated to the investigation of the excitation mechanism of this strong EL.

Employing electroreflectance (ER) spectroscopy, we have studied optical transitions in porous Si (PSi) with a thickness of 100±50 nm made from crystalline Si (c-Si) substrates with different resistivities, 4–10 ωcm (p-), 0.1–1 ωcm (p), and < 0.018 ωcm (p+). The ER features observed at 1.1–2.8 eV in p+ PSi are analyzed with a simple effective mass approximation (EMA) model for confined electron-hole (e-h) pairs in a spherical quantum dot as we have previously done for those observed at 1.2–3.1 eV in p- PSi. From the ER analysis with the EMA model, the effective crystal size is estimated without destruction, and the kinetic energy and the Coulomb attraction energy of the confined e-h pairs are also deduced.

Furthermore, the ER features corresponding to the optical transitions at E1(E0') critical point (CP) are observed at 2.8–3.3 eV in p+ PSi. With an increase in the crystal size, the transition energy of E1(E0') CP in p+ PSi is decreased, while that in p- and p PSi is unchanged from that of 3.4 eV found in c-Si including the p+c-Si. From the Raman results, strain- and disorder-induced spectral changes are found to be negligible, so that the high doping induced effect is the most acceptable mechanism for the red-shift in E1(E0') transitions of Si nanocrystals with the crystal size of 2–3 nm.

The maximum photoresponse of a normal silicon photodetector, that uses a p-n junction as the active zone, is obtained when the incident radiation wavelength is around 750nm. This response diminishes significantly when the incident radiation is near or in the UV region. Meanwhile, nanocrystalline silicon (nc-Si) films with high transparency above 650nm and high absorbance in the UV can be prepared. By quantum confinement effects, a fraction of this absorbed UV energy is re-emitted as visible photons that can be used by the junction. We study the enhancement of the UV-photoresponse of two silicon detector prototypes with a silicon p-n junction active zone and with a photoluminescent nc-Si overlayer. One prototype is made with a porous silicon/n-type silicon/p-type silicon/p++-silicon/metal configuration and the other with an Eu-doped Si-SiO2 overlayer instead of the porous silicon one. The comparison between both prototypes and the control is presented and discussed stressing on the enhancement effect introduced by the photoluminescent overlayers, stability and reproducibility.

This report presents an investigation on carrier transport in LED structures based on oxide passivated nanocrystalline silicon (OPNSi), formed by oxidation of porous silicon. This material, like its precursor, can luminesce quite efficiently while demonstrating several advantages in stability (i.e. chemical, thermal, electrical and electroluminescence). OPNSi can be best described as a porous glass structure with defects that facilitate transport, and remaining embedded nanocrystals of silicon that support light emission. Although this study does not provide a direct measurement of the density of states in OPNSi, the following transport study suggests a high density of states having a broad energy distribution that readily exchange charge with the silicon electrodes. Experimental data also suggests the existence of deeper trap centers that do not facilitate transport, yet influence transport behavior significantly. The device operation is explained by bipolar injection from an electron-injection cathode and a hole-injection anode into the semi-insulating OPNSi layer. The device is modeled as a “field effect diode”, where untraditional concepts are applied in the interpretation of experimental observations. Extensive electrical characterization of OPNSi LEDs has lead to the development of a comprehensive transport model that is self-consistent with all experimental observations.

Thin porous silicon (PS) based metal/PS/p-c-Si structures were prepared on moderate-and high- resistivity substrates. Measurements of current-voltage (I-V) dependencies and impedance at various temperatures were used for the investigation of carrier transport in these structures. The exponential forward bias I-Vdependencies for both types of structures spread over several orders of magnitude with a low value of quality factor (close to 2) and have activation temperature dependencies with an activation energy equal to half the c-Si band gap. The reverse current has a square root dependence on the reverse bias voltage and the activation energy is equal to half the c-Si band gap. Therefore, it was concluded that the reverse and forward currents in thin PS-based device structures were determined by the generation and recombination of carriers in the depletion region of the c-Si substrate. It was shown that a large area spreading current exists in structures made on highly resistive substrates, which appears to be due to a highly conductive inverse (n-type) layer formed in the c-Si substrate at the PS/c-Si heterojunction.

Annealing of amorphous Si/SiO2 multilayer produces nanocrystals embedded between oxide interfaces with the size of the nanocrystals depending on the Si layer thickness. It is found that the crystallization temperature is strongly enhanced by the presence of the oxide interface and follows an exponential law. A model is presented, which derives the exponential dependence taking into account the interface energies, the thickness of the layers, the melting point of the system, and the bulk amorphous crystallization temperature. The critical crystallization radius and the critical thickness of the Si layer are discussed.

We review the orthodox theory of single charge tunneling in a semiconductor quantum dot and we extend it to treat both single electron and hole charging effects. We analyze recent tunneling spectroscopy experiments. We show that for sufficiently large bias voltages V, both electrons and holes can tunnel into the quantum dot, leading to specific features in the I(V) curve. We present detailed simulations of the I(V) curves based a tight binding method for the electronic structure. A very good agreement is obtained with available experiments on InAs nanocrystals, allowing a complete interpretation of the spectra. Finally, we make some predictions concerning Si nanocrystals.

Two-dimensional periodic arrays of inverted pyramid holes with nanometer scale have been patterned on the surface of a (100) silicon wafer and studied for possible application in nanoscale silicon based devices. The surface patterning employed a simple microelectronic processing scheme in which the standing wave intensity pattern from two interfering 458nm laser beams was used to expose holes in a photoresist layer. Subsequent dry etching through an underlying oxide mask layer, followed by a KOH etching step yielded a highly periodic, large area array of inverted pyramids. The pyramid geometry is formed during the anisotropic KOH etch, which stops at the (111) pyramid walls. Therefore, the tips of all inverted pyramids are formed by the intersection of (111) silicon crystal planes and have identical geometry. This study focuses on the use of these features as templates for the controlled crystallization of amorphous silicon layers and also as electric field concentrating “funnels” in MOS-type structures. We will discuss a proposed device in which silicon nanocrystals will be incorporated into the concentrated electric field region at the tip of each inverted pyramid. With this structure, the charging of identical addressable nanocrystals may be possible, leading to the development of practical nanoscale silicon devices.

Germanium quantum dots embedded in silicon have been used in the past to improve the opto-electronic properties of Si based materials. The idea is to overcome the limitation of the indirect band gap of Si by a strong localization of the carriers in quantum dots. However, the Ge quantum dots provide a strong carrier confinement only for the holes, the electrons are only weakly confined in the Si. In this study we embedded the Ge quantum dots in strained Si quantum wells grown on relaxed SiGe buffer layers. The strained Si quantum wells provide a confinement of the electrons in the vicinity of the Ge dots.

The structures were deposited on planar as well as on patterned substrates by molecular beam epitaxy. The structural and optical properties of the samples were analyzed using high resolution cross sectional transmission electron microscopy (TEM) as well as low temperature photoluminescence. The size of the mesa structures have been used as experimental parameter. Relaxed buffer layers grown on line shaped mesa structures show a strongly reduced dislocation density. Consequently the deep luminescence attributed to dislocations in the buffer layers is strongly reduced and pronounced photoluminescence of the quantum structures grown on top of the buffer layers can be observed.

The crystallization of SiGe-heterostructure in which a thin Ge layer put between Si layers has been investigated. When the Ge layer is inside the Si layer, the crystallization phenomena are similar to those of SiGe alloy layer; Ge completely diffuses in Si layer during the crystallization. When the Ge layer is at the interface between the Si layer and the quartz substrate, significant enhancement of crystallization has been found. In this structure, decrease in surface free-energy increases the nucleation rate at the interface and causes anomalous localization of Ge at the interface. When the Ge layer is on the surface of the Si layer, the crystallization property is quite different from the other structures. The underlying Si layer has large and aligned grains when the furnace annealing is assisted by the laser-annealing.

A part of this work was carried out under the ASET program supported by NEDO, Japan.

In this work, the imaginary part of the dielectric function of porous silicon is studied by means of both the tight-binding and the effective medium approaches, in the latter exact result is obtained for the case of 50% porosity. Within the tight-binding approximation, the dielectric function is calculated by using the interconnected and chessboard-like supercell models for the Si skeleton. These microscopic models give quantitatively similar results, which are by a factor of three larger than those from the effective medium theory.

Microwave chemical vapor deposition (MCVD) is utilized to deposit nanocrystalline silicon (nc-si) thin films onto a variety of substrates for application to thin film transistors (TFT's) and solar cells. It is especially important to gain reproducible control of the processing. Thus, an in-situ mass spectrometer (MS) records the plasma conditions with variation of process conditions such as gas selection, pressures, partial pressures, and substrate temperature. These data are correlated with electrical and optical properties of the films. Raman spectra show a FWHM of 11/cm with position at 522/cm as desired for crystalline Si. Typical film thickness is 100nm with grain size of 20-30 nm, using standard deposition, and 50-80 nm when the substrate is intensely optically illuminated during deposition, called photon assist (PA). Hydrogen dilution serves to increase the crystallinity of the films. The ratio of photo-to dark conductivity exceeds 10+5 with dark conductivity as low as 1.5 × 10-10 S/cm. Thin film transistors have been fabricated with Ion/Ioff of 10+7. Hetrojunction solar cells were fabricated using amorphous Si/ nc-Si/ crystlline Si giving a conversion efficiency of above 10.5%, without an antireflection coating. The use of MS in device design will be emphasized.

Shock synthesis of nanocrystalline Si, Ge and CdTe was accomplished using high- velocity thermal spray. Si or Ge powders were injected into a high energy flame, created by a thermal spray gun, where the particles melt and accelerate to impact on a substrate. The shock wave generated by the sudden impact of the droplets propagated through the underlying deposits, which induces a phase transition to a high pressure form. The decompression of the high-pressure phase results in the formation of several metastable phases, as evidenced by transmission electron microscopy and x-ray diffraction studies. The peak pressure is estimated to be ≈23GPa with a pulse duration of 1-5 ns. Transmission electron microscopy revealed that the metastable phases of Si with a size range of 2 to 5 nm were dispersed within Si-I. In Ge, a metastable phase, ST-12, was observed. This is a decompression product of Ge-II which possesses the β-Sn type of structure. In the case of CdTe, a fine dispersion of hexagonal CdTe particles, embedded in cubic-CdTe with an average size of 2 nm was obtained.