Search

In this work we perform a theoretical analysis of the thermoelectric performance of polycrystalline Si nanowires (NWs) by considering both electron and phonon transport. The simulations are calibrated with experimental data from monocrystalline and polycrystalline structures. We show that heavily doped polycrystalline NW structures with grain size below 100 nm might offer an alternative approach to achieve simultaneous thermal conductivity reduction and power factor improvements through improvements in the Seebeck coefficient. We find that deviations from the homogeneity of the channel and/or reduction in the diameter may provide strong reduction in the thermal conductivity. Interestingly, our calculations show that the Seebeck coefficient and consequently the power factor can be improved significantly once the polycrystalline geometry is properly optimized, while avoiding strong reduction in the electrical conductivity. In such a way, ZT values even higher than the ones reported for monocrystalline Si NWs can be achieved.

As far as ultra-dense crossbars are related to correspondingly dense wire arrays, the crossbar route to tera-scale integration depends on the availability of preparation techniques for wire arrays with density of 106 cm−1 or more. This linear density implies, for a planar arrangement, a pitch of 10 nm or less, which not only is at the limits of the current technical possibilities, but also can modify appreciably the band structure of silicon. A dramatic increase of density could only be achieved if it were possible to organize the nanowires in a three-dimensional fashion still exploiting the planar technology. In this work processes are described for the fabrication of out-of-plane, vertically arranged, polycrystalline silicon nanowires via a rigorously top-down batch process. These techniques are consistent with the production of wire arrays with linear density (projected on the surface) larger than that achievable with any other proposed top-down process. Used for the fabrication of the bottom wire arrays of crossbars, these processes should eventually allow a cross-point amount per unit area in excess of 1012 cm−2, thus providing candidate technologies for ultra tera scale integration. The technique developed for such out-of-plane crossbars can be used to implement new functions like coils, solenoids and transformers.

Silicon has been, and continues to be, the material support of integrated circuit (IC) technology-the enabling tool of one of the most impressive technological, industrial and social revolution of mankind. Silicon (both in monocrystalline and polycrystalline form as well as epitaxial silicon, silicon on insulator and silicon on nothing) has grown with IC technology for its ability to match the numerous and complex needs of microelectronics. In recent years, however, IC technology is experiencing problems of maturity that are expected to be soluble only by huge investment in fabrication plants. This state of affairs has motivated the search for potentially cheaper alternatives to the basic IC element-the metal-oxide-semiconductor (MOS) field-effect transistor (FET). Solutions involving the use of low dimensional structures (like silicon nanowires for the crossbar structure, or silicon quantum dots for single-electron transistors) instead of conventional MOS-FETs have been proposed. Remarkably enough, such a nano-engineered silicon has been proved to be of interest not only in electronics but also in photonics, energetics, and biomedicine.

While the analysis of the spectra resulting from energy- or angle-resolved X-ray photoelectron spectroscopy allows the in-depth atomic composition in the probed region to be determined even for complex samples, the determination of the bonding configuration is less trivial. In this paper it is shown that a description of the chemical shift in terms of partial charge and Madelung potential (as results from local modelling of the atom) can provide information even in complicate situations, like that characterizing the hydrogen-terminated (1 0 0) Si prepared by HF aq etching of the native oxide.

Silicon nanocavities can be terminated with hydrogen by wet chemical etching. Their infrared spectra can to a large extent be interpreted in terms of silicon monohydrides on H(7 × 7)Si(111), H(1 × 1)Si(111) and H(2 × 1)Si(100), and of silicon dihydrides on H(1 × 1)Si(100). The time evolution under isothermal conditions (600 °C) of the (1 0 0) faces admits a description in terms of transformation from H(1 × 1)Si(100) into (2 × 1)Si(100) with simultaneous H2 adsorption onto (2 × 1)Si(100) neat dimers. In so doing the inner H2 pressure decreases by about one order of magnitude from the initial value of 3 × 103 Torr. The unique properties of nanocavities allow their use as nanoreactors; this has led to the determination of the sticking coefficient for H2 adsorption in conditions of pressure, temperature and adventitious contamination otherwise not achievable. © 2010 Elsevier B.V. All rights reserved.

A lot of work has been carried out to prepare chemically homogeneous (1 0 0) silicon surfaces. The hydrogen-terminated (1 0 0) silicon surfaces are the most promising ones, especially in view of their remarkable environmental stability. The simplest way to produce hydrogen-terminated surfaces (attack in water solution of HF of a sacrificial, thermally grown, oxide) results in strongly heterogeneous rough surfaces (although with prevailing dihydride terminations). These surfaces can, however, be flattened and homogenized by treating them in H 2 at high temperature ( > 850 ° C). The morphological and chemical changes undergone by the surface during the treatment are studied X-ray photoelectron spectroscopy, atomic force microscopy, scanning tunnelling microscopy, infrared absorption spectroscopy in the attenuated total reflection mode, reflection high energy electron diffraction and thermal programmed desorption, and the mechanisms responsible for them are discussed.

Denoting with 'nanodevice' any device with size in one dimension at least in the nanometre length scale (NLS), the basic constituent of integrated circuits (ICs), the metal-oxide-semiconductor (MOS) field-effect transistor (FET), is by several years a nanodevice. In fact, the thickness of the SiO"2 gate dielectric is around 3nm for logics or 5nm for nonvolatile memories. However, since the factors limiting the IC integration are horizontal sizes, in electronics one speaks of nanodevice when its size in one horizontal dimension at least is in the NLS. The definition of features in the NLS is impossible via optical lithography, but can be done using electron- or ion-beam lithography. These techniques, however, are very expensive and still in their fancy, at least for what concerns their exploitation in the industry practice. Geometries in the NLS can however be produced with relative ease by the spacer patterning technique, i.e., transforming vertical features (like film thickness) in the vicinity of a step of a sacrificial layer into horizontal features. The ultimate length producible in this way is controlled by: the steepness of the step defining the sacrificial layer; the uniformity of the deposited or grown films; and the anisotropy of its etching. While useful for the preparation of a few devices with special needs, the above trick does not allow by itself the development of a nanotechnology, where each layer useful for defining the FET should be in the NLS and aligned on the underlying geometries with tolerances in the NLS. Setting up such a nanotechnology is a major problem which will involve the IC industry in the post-Roadmap era. Irrespective of the detailed structure of the basic constituents of nanoICs (molecules, supramolecular structures, clusters, etc.), any nanoIC can hardly be prepared without the ability to produce arrays of conductive strips with pitch in the NLS. This work is devoted to describe a scheme (essentially based on the existing microelectronic technology) for their production without the use of electron- or ion-beam lithography and used as host of molecular devices.

An experimental study of the derivatization of Si(100) surfaces with 1-octyne has been carried out via x-ray photoelectron spectroscopy at different take-off angles, and infrared absorption spectroscopy in attenuated total reflection mode and atomic force microscopy have been used to characterize the pristine surface. Our study has shown that it is possible to prepare nearly flat, terraced, homogeneously hydrogen-terminated Si(100) surfaces via exposure to H2 at high temperature (1100 °C) and that these surfaces may be functionalized with hydrocarbon moieties by simple immersion in the unsaturated hydrocarbon at ∼200 °C. Copyright © 2004 John Wiley & Sons, Ltd.

Hydrogen-terminated, atomically flat, 1×1 (un)reconstructed, (100)-oriented, silicon surfaces have been functionalized with 1-octyne and then studied via X-ray photoelectron spectroscopy. The C 1s signal shows that the addition of the hydrocarbon chain occurs through the formation of environmentally stable Si-C bonds, in agreement with hydrosilation as the major grafting route.

The structure and thermal stability in N-2 of hydrogen-terminated (100) silicon has been studied by X-ray photoemission spectroscopy, transmission electron microscopy, atomic force microscopy, thermal programmed desorption, and reflection high energy electron diffraction. Device-quality surfaces were prepared in an open-chamber reactor by exposing single crystalline, (100) oriented silicon to H-2 at high temperature (850 degreesC or 1100 degreesC) for durations on the order of 10(2) s. The observed stability with respect to N-2 at 850 degreesC is inconsistent with the reported desorption kinetics and may be accounted for in terms of either physico-chemical properties of the system (e.g., the presence of a buried layer of H-2 or of hydrogen-decorated vacancies whose out-diffusion restores the hydrogen terminations on the surface) or the reactor (persistence of hydrogen in the atmosphere even after switching it off). The nitridation by N-2 of hydrogen-terminated silicon is less efficient (per unit exposure) than that by N2O by 4 orders of magnitude.

This work presents an experimental study based on X-ray photoemission spectroscopy (XPS) of hydrosilation of 1-alkyne as a tool for the functionalization of the (100) surface of silicon. In particular, the following processes are considered: (i) hydrogen termination of silicon via HF etching and subsequent exposure to H2 at high temperature, and (ii) grafting alkene chains to the resulting hydrogen-terminated surface via hydrosilation of 1-octyne.

The early oxidation stages of hydrogen-terminated single-crystalline Si(100) exposed to a diluted N2/N2O atmosphere at 850° C for different durations have been studied by XPS, following the evolution of the Si 2p signal. Evidence is given that the usual analysis, in terms of five pairs of peaks attributed to silicon in oxidation states from 0 to +4, does not account for the observed Si 2p signal. The spectrum is accurately reproduced only by assuming the existence of silicon in bonding configurations different from those usually assumed to occur at the Si/SiO2 interface. Copyright © 2002 John Wiley & Sons, Ltd.

A method is proposed for the valuation of net atomic charge in chemical compounds, based on XPS measurements of two energy levels for the same atomic species and on the evaluation of their difference. This approach, by considering two core levels, allows us to cancel the Madelung term and any other external potential from the electrostatic equation that rules the chemical shift of the binding energy. In this way, it directly relates the experimental shift with the net charge on the considered atomic species. The approximation that Madelung potential does not vary significantly from one core level to another one is used. The method has been applied to solid compounds containing silicon and to aluminosilicates having different counter-ions and different Si/Al ratios. The experimental data of net charges have been compared, where possible, with Pauling and Sanderson charges obtained from electronegativity data.

A renormalization group approach to self-similar multilayer aggregation processes is suggested on the basis of the following assumptions. 1. The aggregate is generated by a layer-by-layer process, ...

)4, though it is expected to relax via formation of peroxidic bridges), and the oxygen-bridge vacancy (which in the absence of relaxation can be described as a diradical center (≡Si•)2– the E′′ center). These radicals react with the residual atmosphere according to completely new pathways: the bombardment in a CO2 atmosphere results in the formation of ester-like and carboxylate groups, stable up to 500 °C at least, inserted in the SiO2 network at the oxygen-bridge vacancies. The bombardment in a C2H4 atmosphere results in more complex configurations: the oxygen-bridge vacancy reacts at room temperature with ethylene forming a Lewis adduct which, after heating at 500 °C, presumably reverts to a -CH2-CH2- bridge in between silicon atoms; the silicon-link vacancy likely reacts with C2H4 forming CH3CHO. These conclusions, based on experimental data (mainly infrared spectroscopy), are also supported by extended quantum mechanical calculations (density-functional methods and ab initio molecular dynamics).

Density functional calculations have been performed on molecular models of silica defects and their C 2 H 4 adducts. The results are consistent with a previous experimental infrared characterization that was interpreted in terms of C 2 H 4 addition to the SiO 2 skeleton at both the oxygen-bridge and silicon-link vacancy defects produced by argon bombardment. The geometries and thermodynamical stabilities of the proposed species and other possible adducts have been evaluated. The calculated C  C, C  O and C–H stretching frequencies are in good agreement with the experimental values and confirm the validity of the previous interpretation, helping to clarify some unclear details.

The desorption kinetics from heterogeneous surfaces contain information on energy heterogeneity. Mathematical methods for extracting the energy distribution function from desorption kinetics are developed. The overall desorption kinetics are described as weighted averages of the local desorption kinetics from each energetically homogeneous zone forming the surface. Whatever the local kinetics, the energy distribution is determined from the overall kinetics by using the condensation approximation. The method is applied to two situations: the time−logarithm law observed in a lot of systems of practical interest in catalysis, and the time−power law observed in many relaxation kinetics of macromolecules.

Density functional (DFT) calculations for different size cluster models of the hydrogen-terminated HFaq-etched (100) Si surface have been performed to verify that the quantities of interest (namely, atomic net charges and interatomic distances) in assigning the lines observed by X-ray photoelectron spectroscopy (XPS) vary weakly with cluster size. Net charge analysis based on Voronoi Deformation Density (VDD) method and accurate DFT geometry optimization calculations involving the smallest clusters as local models of various surface silicon atoms are used to assign chemical species to the features observed in the XPS spectra through evaluation of the chemical shifts, which are controlled by both the net charge and the Madelung potential truncated to nearest neighbours of the considered atoms.

Dimensional confinement in silicon nanowires (NWs) is well-known for enhancing phonon scattering, thus leading to a pronounced reduction of thermal conductivity κ with respect to bulk material. The effect of confinement on phonon scattering in nanolayers (NLs), however, has not been fully understood. In this work, thermal conductivity on polycrystalline silicon NLs with roughened surfaces and thicknesses ranging from 30 to 100 nm has been experimentally investigated. For measurement purposes, the nanostructures were fabricated with a dedicated surface nano-machining process, thus producing vertical silicon nanostructures suspended on Al/Si electrodes on a silicon substrate, using SiO2 as a sacrificial layer. By designing such structures in a four-terminal configuration, their κ could be determined by the current-voltage method. Boron doped silicon NLs were examined, at resistivity ranging between 2 and 10 m $$\Upomega$$ cm. We found an increase of phonon scattering from the confinement, since κ decreased steadily with the thickness from values typical of thick films (around 30 W m−1 K−1) down to

Energy filtering has been widely considered as a suitable tool to increase the thermoelectric performances of several classes of materials. In its essence, energy filtering provides a way to increase the Seebeck coefficient by introducing a strongly energy-dependent scattering mechanism. Under certain conditions, however, potential barriers may lead to carrier localization, that may also affect the thermoelectric properties of a material. A model is proposed, actually showing that randomly distributed potential barriers (as those found, e.g., in polycrystalline films) may lead to the simultaneous occurrence of energy filtering and carrier localization. Localization is shown to cause a decrease of the actual carrier density that, along with the quantum tunneling of carriers, may result in an unexpected increase of the power factor with the doping level. The model is corroborated toward experimental data gathered by several authors on degenerate polycrystalline silicon and lead telluride.

Heavily boron-doped polycrystalline silicon has been reported to be characterized by somewhat unexpectedly high power factor. High Seebeck coefficients are however unexpected in materials with high carrier densities. A semiquantitative model was proposed, showing that the potential barrier structure at grain boundaries, along with the nanometric grain size, leads to an unusual mechanism of carrier filtering, named adiabatic energy filtering. Actually, the presence of potential barriers associated with segregated boron disables charge transport by holes in the band deep tail. This leads to a decrease of the actual carrier density, as in the case of standard energy filtering. However, the nanometric grain size along with the inefficiency of the hole-hole relaxation mechanism in degenerate semiconductors actually prevents carriers from relaxing, causing an increase of the average (macroscopic) drift mobility. Thus, in spite of the decrease of drifting hole density the electrical conductivity is found to increase. In this communication a refinement of the model is presented, that will be discussed and corroborated with an extended body of experimental data gathered by several authors on degenerate polycrystalline silicon films. © 2012 American Institute of Physics.

Nanostructuring has opened new ways to increase the thermoelectric performance of a host of materials, mainly by decreasing their thermal conductivity κ while preserving the Seebeck coefficient S and electrical conductivity σ. The thermoelectric properties of degenerated polycrystalline silicon films with nanocavities (NCs) have been studied as a function of annealing temperature upon isochronous annealings in argon carried out every 50°C in the range 500 – 1000°C which were used to modify the shape of the NCs. We found that presence of the NCs had no negative effect on the electronic properties of the system. The measured values of S and σ were close to those previously reported for the blank polycrystalline silicon films with the same doping level. The thermal conductivity was also found to be close to the value measured on the blank sample, about half of the reported value in polycrystals. This led to a power factor of 15.2 mWm-1K-2 and a figure of merit of 0.18 at 300 K.

We report the study of the thermoelectric properties of degenerate, boron-doped polycrystalline silicon on insulator structures. The occurrence of a regime where both the Seebeck coefficient and the conductivity increase is confirmed. This results in a power factor P of 13 mW K-2 m-1. We propose that such high values of P may be determined by adiabatic energy filtering occurring at grain boundaries decorated by segregated boron.

A process is developed for the fabrication of vertically arranged poly-silicon nanowires via a rigorously top-down batch process. The technique allows the production of wire arrays with larger linear density (projected on the surface) than those achievable with any of the other proposed top-down processes. © 2010 Elsevier B.V. All rights reserved.

The development of artificial devices (nanobots), working as blood white cells but addressed to the recognition and eventually the destruction of endogenous pathological states, is an ambitious goal. Swarm intelligence can be a key element to successfully tackle the challenges posed by this goal. Here we describe an applicative scenario, based on swarm of nanobots, by sketching the environment in which the nanobots operate, the constraints related to their physical implementation, and the tasks they have to tackle. In this scenario, we propose to use collisions between nanorobots as a way of communication inside the swarm. © 2010 Springer-Verlag Berlin Heidelberg.

An array of nanocavities in silicon offers an inner surface defined by nearly flat, highly ordered, planes which can be passivated by hydrogen injection from aqueous solution of HF. The infrared spectrum shows that the inner surface is mainly constituted by monohydrides in different morphologies. The corresponding signals are very narrow (full width at half maximum at room temperature below 5 cm-1, depending on the line), this being evidence for high local order. Such an inner surface can be used as a benchmark for the assignment of infrared signals observed on outer surfaces, where the sources of errors-and the consequent uncertainty in the spectral attribution-may be large. The comparison between the spectra from a well-characterized external Si(1 0 0) surface and from an internal one (in both cases hydrogen-terminated) has allowed a better qualification of the external surface. Copyright © 2010 John Wiley & Sons, Ltd.

The architecture of a surveillance system, able to make up for the poor ability of the immune system in recognizing endogeneous diseases (like cancer, arteriosclerotic vascular disease, Alzheimer disease, etc.), is designed. Based on the use of swarms of nanobots communicating with a resident central unit, this architecture can actually be implemented, though at a rudimentary level, with current technologies based on the functionalization of integrated circuits

The preparation of an array of nanocavities (NCs) in silicon via thermal treatments of high-fluence helium-implanted silicon is a well-established process. The NCs have offered a powerful tool for the preparation of well-defined and ordered internal silicon surface, enabling the experimental investigation of its free energy. Here, instead, we interpret the NCs as an ideal nanolaboratory for the study of the adsorption and desorption of H 2 on silicon. We will present the determination of the internal pressure through the equilibrium abundances of monohydride dimers and dihydrides obtained with infrared spectroscopy. Copyright © 2010 John Wiley & Sons, Ltd.

Density functional calculations for both periodic slabs and different size cluster models of the hydrogen-terminated (100) surface of silicon are used to study a new configuration, formed by a silylene center interacting with vicinal silicon dihydrides through nonconventional hydrogen bonds. A comparison between slab-model and cluster-model approaches to modeling surface silylene defect formation processes is presented. The cluster models are used to analyze the structure and bonding of the silylene with a Lewis acid and base, showing the Zwitterionic nature of the defect. The silylene is also demonstrated to behave as a strong Brønsted acid. The stabilization of the silylene defect via interaction with species unavoidably present in the HF(aq)-etching solution is investigated. Finally, the negative chemical shift observed by X-ray photoelectron spectroscopy in the HF(aq)-etched (100) Si surface is attributed to the occurrence of silylene defect.

A theoretical analysis of the hydrogen-terminated (100) surface of silicon leads to the identification of a new configuration, formed by a silylene center interacting with vicinal silicon dihydrides. This structure may be viewed as a metastable configuration of 2 x 1 (100) (SiH)(2). Silylene can however be stabilized via interaction with water. The paper proposes that some of the anomalous features observed at the hydrogen-terminated or oxidized (100) Si can be attributed to silylene centers datively stabilized by oxo groups or to structures resulting from their decomposition.

The control of the chemical state of the inner surfaces of nanocavities (NCs) produced by the annealing of helium-implanted silicon has influence on lifetime control, gettering and wafer bonding. In this work it is demonstrated that the etching in HFaq of (1 0 0) silicon containing a buried array of NCs produces a giant injection of hydrogen with the consequent passivation of the inner surfaces, mainly via the formation of silicon monohydride at (1 1 1) faces and monohydride dimers at 2 × 1 reconstructed (1 0 0) faces. These terminations are very stable and survive heat treatments at 700 °C. © 2009 IOP Publishing Ltd.

Adding via batch processing assigned molecules to sublithographic features. Needless to say, strategy XB_ requires the ability to attach organic moieties to the silicon surface, thus forming a hybrid interface with assigned electrical properties. It is also noted that since the crosspoint regions are recessed, not only is the use of scanning probe techniques disallowed by hypothesis, but also impossible. The needs of Increasing the reactivity of the silicon surface to the desired organic derivatization Attaining the maximum coverage degree Allowing the molecules to self-assemble in an ordered structure, determined uniquely by intermolecular lateral forces and steric hindrance are generally assumed to require a prior homogeneous termination of the surface. The nature of terminal species (halogen, hydrogen, hydroxyl, or others) is fixed by the grafting route.

Net-charge analysis, involving siladamantane moieties as local models of various surface silicon atoms, is used in combination with infrared spectroscopy to assign chemical species to the features observed in the x-ray photoelectron spectra from hydrogen-terminated (1 0 0) Si prepared by HFaq etching of the native oxide. (Some figures in this article are in colour only in the electronic version)

The geometric description of the world is at the root of Western thought: to the first (perhaps mythic) philosopher, Thales, we owe a theorem of plane geometry. The role of geometry in the construction of physical theories is equally important: on the cosmological scale, gravitation is explained in terms of space curvature; on the submicroscopic scale, quantum mechanics may be explained in terms of fluctuation of the metric tensor [303]. On the laboratory length scale, the most important geometric property of any thermodynamic system is most likely its shape. The general mathematical description of surfaces is a notoriously complicated affair – especially if one tries to describe real surfaces.

Understanding the ultimate scale below which a device no longer behaves as extrapolated from its behavior on a larger scale is not trivial. A special case, of topical importance, is understanding which are the physical limits below which electronic computation is impossible. Most ICs are indeed produced for handling data. They are usually classified into two families, logics and memories, according to the way they operate in combinatorial or sequential way. From the architectural point of view, the logic gates are arranged and linked in a way controlled by the specific logic functions they implement; on the contrary, memories are organized in a periodic arrangement of their constituting cells. In view of their greater architectural simplicity, the limits of computation are better understood considering memories. Though restrictive, not only is the forthcoming analysis addressed to the largest class of electron devices, but also to the one that is expected to meet such limits sooner because of the fastest growth rate in the increase of complexity. Following the analysis of [38], the ultimate limits of computation will be determined in the hypotheses that computation is a physical process occurring in a medium immersed in a thermal reservoir at an assigned temperature and thus obeying the underlying physical laws. The computational figure of merit will be calculated considering the need for transforming the microscopic computation outcome into a macroscopic event.

Nanotechnology may be defined as the set of techniques1 and methods for the exploitation of the properties of matter on the nanoscale. Nanotechnology may also be considered as the practical realization of the dreams of Feynman and Drexler: In a legendary talk titled There’s plenty of room at the bottom: An invitation to enter a new world of physics, delivered on 29 December 1959 at the Winter Meeting in the West of the American Physical Society [1],2 Feynman proposed the use of atoms as information storage elements, thus allowing a planar bit density on the scale of 1015 cm−2 – the petascale integration (PSI). Subsequently, in a series of publications started in 1982 and culminating in a book on Nanosystems – Molecular Machines, Manufacturing and Computation [3], Drexler suggested that the optimal scale for the exploitation of microscopic objects at the macroscopic scale is the molecular or supermolecular one.3 Drexler’s conclusion, although reducing the maximum density of a surface arrangement from the petascale to the terascale, does, however, clarify that the exploitation of the potentials offered by molecular or supramolecular arrangements requires the development of a technology for assembling, addressing and probing objects on the nanometer length scale – just nanotechnology.