US20130059121A1

US20130059121A1 - Nanocrystalline silicon in sio2 composite and freestanding silicon nanoparticles

Info

Publication number: US20130059121A1
Application number: US13/667,138
Authority: US
Inventors: Jonathan Gordon Conn Veinot; Colin Michael Hessel
Original assignee: University of Alberta
Current assignee: University of Alberta
Priority date: 2005-05-27
Filing date: 2012-11-02
Publication date: 2013-03-07
Also published as: JP5424638B2; KR101331435B1; WO2006125313A1; EP1883949B1; KR20080042047A; CA2609537C; JP2008545826A; EP1883949A1; US20090117392A1; US8323731B2; CA2609537A1; EP1883949A4

Abstract

A nanocrystalline-Si/SiO₂composite containing silicon nanoparticles prepared by curing hydrogen silsesquioxane (HSQ) under reductive thermal conditions in the presence of a gas containing hydrogen is provided. Freestanding photoluminescent silicon nanoparticles are prepared by acid etching the nanocrystalline-Si/SiO₂composite.

Description

FIELD OF THE INVENTION

The present invention relates to methods of preparing nanocrystalline silicon embedded in SiO₂, in particular from a silsesquioxane precursor via reductive thermal curing. Further the present invention relates to the preparation of freestanding silicon nanoparticles by acid etching said nanocrystalline silicon embedded in SiO₂.

BACKGROUND OF THE INVENTION

Silicon nanostructures including porous silicon (p-Si),¹silicon rich oxides (SROs),²and freestanding Si nanoparticles have been the focus of intense research because of their unique chemical and optical characteristics. The electronic structure of bulk silicon provides an indirect bandgap of 1.12 eV with the lowest point of the conduction band and the highest point in the valence band occurring at different coordinates in reciprocal space. These restraints make the bandgap optical transition dipole-forbidden, limiting practical optoelectronic application of the bulk crystal due to the low photoluminescence intensity and slow carrier dynamics (i.e., long lived excited states). As the dimensions of a semiconductor particle decrease into the “nano” size regime, the bandgap energy increases and pseudo-continuous bands become discrete energy levels that are populated according to quantum mechanical selection rules. When the particle dimension nears the Bohr exciton radius (ca. 5 nm for silicon), quantum confinement effects emerge and photoluminescence (PL) shifts into the visible spectrum and becomes more intense. Some researchers suggest that the PL observed from photoexcited Si nanoparticles arises because the bandgap transition becomes weakly dipole allowed in this size regime.³Others claim that the photoemission originates from the passivation of surface traps present in bulk Si.⁴Regardless of the explanation, a characteristic photoemission maximum at approximately 1.7 eV is seen for Si-based nanostructures; including the “Si quantum wires” reported by Canham et al.,⁵nanocrystalline nc-Si/SiO₂composites,²and freestanding Si nanoparticles prepared via solution,^6,7,8precursor pyrolysis,^{9, 10}and physical techniques.^11,12,13The unique optical properties and electrochemical stabilityl¹⁴of nanoscale elemental Si offer significant potential for a variety of light emission applications. Furthermore, the biocompatibility of Si and SiO₂makes these materials potentially useful in sensing applications where toxic, electrochemically active compound semiconductor nanoparticles are impractical.
Measuring the direct effect of Si nanocrystal size on the PL spectrum of Si/SiO₂nanocomposites and freestanding Si nanocrystals^9,10,13appears complicated, with interface effects and particle interactions¹⁵playing key roles. Consequently, the size effects and the influence of the indirect bandgap of bulk Si on PL behavior of Si nanocrystals remain poorly understood. Methods for relating PL energy maximum to particle size are diverse and a variety of models have been proposed including: the effective mass approximation,¹⁶empirical tight binding band theory,^17,18empirical pseudopotential approximation,^{19, 20}and ab-initio local density approximation.^{21, 22}To facilitate better understanding of Si nanoparticle optical and chemical response, straightforward, cost-effective, scaleable methods for preparing materials of controlled size, crystal structure, and surface chemistry are necessary. Well-established physical techniques for preparing Si nanostructures such as porous silicon (p-Si) and silicon rich oxides (SRO) often employ highly corrosive reagents (e.g., hydrofluoric acid), costly procedures (e.g., ion implantation,^23,24vacuum evaporation,²⁵sputtering,²⁶and laser ablation²⁷) and provide only partial tailoring of film chemical composition. Furthermore, many of these methods are not easily scalable and are impractical for preparing macroscopic quantities (ca. >500 mg) of material.
SROs are a promising class of nanostructured materials made up of luminescent, crystalline Si nanoparticles embedded in environmentally inert SiO₂-like matrices. A common method for preparing SROs employs a multi-step process; the first stage involves deposition of thin “SiO” precursor films using physical methods such as vapor deposition, physical sputtering, or e-beam evaporation to deposit films onto flat substrates.²These recipe-based approaches control the Si:O ratio by maintaining a specific oxygen flow rate (i.e., partial pressure) during reactive deposition of “SiO” or by controlling the co-deposition rates of Si, “SiO”, and SiO₂to produce SiO_xfilms (0≦x≦2).²⁸Films are subsequently annealed at high temperature in a reducing atmosphere (typically 4% H₂, 96% inert gas) to promote formation of Si nanocrystals.²⁹Iterative variation of experimental parameters and post deposition micro-probe analyses have previously shown that films with composition ratios close to Si_1.0:O_1.5that are annealed at ca. 1100° C. produce the strongest PL. Still, fundamental questions remain regarding the mechanism for nanoparticle formation, the relationship between Si particle size and peak PL energy, among others.²Unfortunately, the exact composition, structure, and purity of “SiO” is subject of a longstanding controversy^30,31,32and is strongly dependent upon processing conditions.³⁰Further, the exact chemical structure of “SiO_x” remains largely ill-defined.³³These uncertainties potentially hinder rational study of chemical composition and its influence on the material properties of SRO nanoparticle composites. Other practical limitations of the abovementioned physical “SiO_x” deposition techniques are nontrivial control of chemical composition (e.g., straightforward introduction of dopants), and conformal coverage of textured and “non-flat” substrates (e.g., fiber optic cables) is somewhat limited by these line-of-sight techniques.
Although SROs can be used to provide insight into the photonic, electronic, and chemical interactions of nc-Si within solid matrices, it is essential that these same properties be studied and understood for freestanding Si nanoparticles. This is particularly true if research is to move beyond fundamental investigation and efficient applications are to be realized. Studying the chemical reactivity and tunability of freestanding Si nanoparticles, in concert with single particle spectroscopy could lead to a better understanding and optimization of particle photoemission. To this end effective methods for preparing freestanding Si nanoparticles must be established. While freestanding particles have been dislodged from p-Si surfaces,³⁴published data suggests individual Si nanoparticles remain trapped in larger (i.e., ≧1 μm) pieces of the p-Si structure.^11,12,13Similar liberation and bulk preparation of Si nanoparticles from SRO matrices is impractical given extremely small sample sizes.³⁵Laser induced precursor pyrolysis has recently been reported as an efficient method for preparing large quantities of Si nanoparticles from silane at rates of 20-200 mg/hour; this approach however relies on expensive laser equipment and custom designed reactors not available in most synthetic laboratories.^9,10Solution-based procedures,^{8,36,37,38,39,40,41,42,43}provide some post synthesis material processability but are often plagued by material purity, ill-defined particle surface chemistry, and limited ambient stability—all criteria crucial to the eventual application of these materials in optoelectronic devices.
Silsesquioxanes are commercially available, solution processable, discrete, structurally well-defined molecules composed of silicon-oxygen frameworks with empirical formulae (RSiO_1.5) where R may be a variety of chemical functionalities (e.g., H, alkyl, silyl, and aromatic). The chemistry of these compounds is well-established and a variety of cage structures are known.⁴⁴Hydrogen silsesquioxane (HSQ), a totally inorganic silsesquioxane (H₈Si₈O₁₂), is one of the most widely studied and has been investigated as a model silica surface,^45,46,47,48luminescent material,⁴⁹and a catalytic support.^50,51Examples of high purity silica have also been prepared from silsesquioxane precursors.⁵²It is generally accepted that upon oxidative thermal curing, the silsesquioxane cage structure of HSQ collapses to release SiH₄ ⁵³and a SiO₂-like network solid forms whose dielectric,⁵⁴mechanical, and processing characteristics depend on the curing conditions. Dielectric films produced by thermal curing of HSQ currently find application as spin-on, planarizing dielectric interlayers in the microchip industry.⁵⁴To date, no Si nanoparticle preparation employing silsesquioxanes has been reported.
There remains a need for a method for preparing large quantities of Si nanoparticles that, ideally, is uncomplicated, cost-effective and reproducible.

SUMMARY OF THE INVENTION

Herein, the first application of hydrogen silsesquixane (HSQ) as a precursor to adherent, conformal, planarizing, photoluminescent nanocrystalline-Si/SiO₂(nc-Si/SiO₂) composites via straightforward reductive thermal annealing is reported. Further photoluminescent, freestanding silicon nanocrystals were readily liberated from the nc-Si/SiO₂composites upon etching.
Accordingly, the present invention relates to a method for preparing a nanocrystalline-Si/SiO₂composite comprising curing a precursor comprising hydrogen silsesquixane (HSQ) under reductive thermal conditions for a time and at a temperature sufficient to produce a nanocrystalline-Si/SiO₂composite.
In an embodiment of the present invention, the precursor comprising hydrogen silsesquixane (HSQ) is a film. Accordingly, the present invention also relates to a method for preparing a nanocrystalline-Si/SiO₂composite film comprising:
(a) forming a film comprising hydrogen silsesquioxane; and
(b) curing the film under reductive thermal conditions for a time and at a temperature sufficient to produce a nanocrystalline-Si/SiO₂composite film.
The invention also includes nanocrystalline-Si/SiO₂composites prepared using the method of the present invention as well as the use of such composites for the preparation of, for example, optoelectronic films and patterned films for incorporation into a variety of device structures, including, but not limited to, chemical sensors, optical amplifiers and waveguides.
The structural tunability of HSQ offers control over film composition and structure previously unattainable with status quo physical deposition methods.
The invention also includes a method for preparing photoluminescent silicon nanoparticles comprising acid etching the nanocrystalline-Si/SiO₂composites prepared using the method described above under conditions effective to produce photoluminescent silicon nanoparticles.
Other features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples while indicating preferred embodiments of the invention are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described in relation to the drawings in which:

FIG. 1 is a schematic showing the thermal processing of hydrogen silsesquioxine (HSQ) for preparing silicon nanoparticle/SiO₂-like (nc-Si/SiO₂) nanocomposites.

FIG. 2 shows photographs of A. optical grade silica substrate coated with a transparent nc-Si/SiO₂film, B. bulk sample of reductively annealed HSQ, C. mechanically ground B.

FIG. 3 shows orange/red photoluminescence from hydrofluoric acid liberated silicon nanocrystals upon exposure to a standard handheld ultraviolet light.

FIG. 4 is a graph showing the heating rate dependence of HSQ weight loss. Solid: nitrogen atmosphere. Dashed: 4% hydrogen:96% nitrogen atmosphere.

FIG. 5 shows the stages of HSQ thermal degradation in inert atmosphere.

FIG. 6 shows infrared spectra of A. neat hydrogen silsesquioxine B. NEAT reductively annealed HSQ. HF etched annealed HSQ showing characteristic Si—H stretching (2100 cm⁻¹) and decreased Si—O—Si bending (1096 cm⁻¹). C. 60 min., D. 120 min.

FIG. 7 shows normalized photoluminescence spectra of nc-Si/SiO₂from HSQ processed at different temperatures: A: 900° C., B: 1000° C., C: 1100° C.

FIG. 8 shows typical normalized photoluminescence spectrum of nc-Si/SiO₂A. bulk material, B. HF liberated hydride surface terminated nc-Si.

FIG. 9 shows A. A brightfield transmission electron micrograph of 2.27±0.59 nm nc-Si in SiO₂. (inset: Selected area electron diffraction of nc-Si/SiO₂thin film composite showing {111}, {200}, and {311}, reflections characteristics of diamond lattice Si.) B. Size distribution of Si nanocrystals, n=115.

FIG. 10 shows A. Transmission electron micrograph of 2.1±0.3 nm Si nanocrystals liberated via HF etching from nc-Si/SiO₂. B. Size distribution of liberated hydride surface terminated Si nanocrystals, d_avg.=2.11, 2σ=0.82, n=50 particles.

FIG. 11 shows an X-ray powder diffraction of mechanically ground, thermally processed HSQ. Noted reflections are consistent with diamond lattice silicon (* are reflections arising from the aluminum sample holder).

DETAILED DESCRIPTION OF THE INVENTION

A straightforward method for preparing adherent, conformal composites of nc-Si/SiO₂that luminesce in the visible and near-IR regions of the spectrum is described. The composites may be formed from HSQ films or bulk samples. Thermal gravimetric analysis indicates that sample heating rate and processing atmosphere influence the weight loss arising from SiH₄evolution. TEM, selected area electron diffraction, and X-ray powder diffraction confirm the presence of nc-Si likely produced from thermal decomposition of SiH₄evolved during the rapid thermal processing of HSQ. The well-defined molecular structure of HSQ offers excellent control and understanding of reaction conditions while also affording a straightforward method for producing macroscopic quantities of freestanding hydride terminated silicon nanocrystals. In addition, the solution processability, ease of handling, and chemical tunability of silsesquioxanes will facilitate the preparation of patterned optoelectronic films with tailored chemical response for incorporation into a variety of device structures including, chemical sensors, optical amplifiers, and waveguides.
Accordingly, the present invention relates to a method for preparing a nanocrystalline-Si/SiO₂composite comprising curing a precursor comprising hydrogen silsesquixane (HSQ) under reductive thermal conditions for a time and at a temperature sufficient to produce a nanocrystalline-Si/SiO₂composite.
The term “reductive thermal conditions” as used herein means in the presence of a gas comprising hydrogen, and suitably an inert gas such as helium, argon or nitrogen, or a mixture thereof. In an embodiment of the invention, the gas comprises about 2% to about 6% hydrogen and about 94% to about 98% nitrogen, suitably about 4% hydrogen and about 96% nitrogen.
The time and temperatures sufficient to produce a nanocrystalline-Si/SiO₂composite may be determined by a person skilled in the art by varying the curing time and temperature until optimal production of a nanocrystalline-Si/SiO₂composite is observed. The observation of nanocrystalline-Si/SiO₂may be made using techniques known in the art, such as, for example, transition electron microscopy (TEM), selected area electron diffraction (SAED), X-ray diffraction (XRD) and photoluminescence spectroscopy. The photoluminescence maximum changes as the curing temperature changes. In an embodiment of the invention the reductive curing is carried out at temperatures in the range of about 900° C. to about 1200° C., suitably at about 1100° C., for about 30 minutes to about 2 hours, suitably for about 1 hour.
The precursor comprising HSQ may be in the form of a HSQ solution or in the form of a solid. When a solution is utilized, it is desirable that that HSQ solution be applied onto a substrate to form a film comprising HSQ. When the precursor is in the form of a solid, bulk nanocrystalline-Si/SiO₂composites are prepared. The solid HSQ may be obtained, for example, by removing the solvent from a stock solution comprising HSQ.
The HSQ solution may be any suitable solution comprising HSQ and a carrier solvent. Such solutions are available, for example, from Dow Corning.⁵⁵In an embodiment of the invention the solution comprises about 5% to about 20%, suitably about 10% HSQ, in a suitable solvent, such as methyl isobutyl ketone.
It is an embodiment of the invention that nanocrystalline-Si/SiO₂composite is prepared as a film. Accordingly, the present invention relates to a method for preparing a nanocrystalline-Si/SiO₂composite film comprising:
(a) forming a film comprising hydrogen silsesquioxane (HSQ); and
(b) curing the film under reductive thermal conditions for a time and at a temperature sufficient to produce a nanocrystalline-Si/SiO₂composite film.
In an embodiment of the invention, the HSQ film may be formed on a substrate using any known method, for example by spin coating a solution comprising HSQ onto the substrate. Spin coating the HSQ solution provides a film of uniform thickness on the substrate and allows the thickness of the film to be controlled by varying the spin rate. In an embodiment of the invention, about 0.1 ml to about 1.0 ml, suitably about 0.5 ml, of the HSQ solution is deposited onto the substrate and the substrate is spun from about 0 to about 7500 rpm, suitably from about 0 to about 5800 rpm, in about 3 seconds to about 10 seconds, suitably in about 5 seconds, followed by an additional spinning time of about 5 to about 60 seconds, suitably about 30 seconds, at about 4000 rpm to about 7000 rpm, suitably at about 6000 rpm. The substrate may be any inert substrate suitable for use in the methods of the invention. In an embodiment of the invention, the substrate is optical grade silica. Suitably the film is formed on the substrate in an inert environment.
The invention also includes nanocrystalline-Si/SiO₂composites prepared using a method of the present invention as well as the use of such composites for the preparation of, for example, optoelectronic films and patterned films for incorporation into a variety of device structures, including, but not limited to, chemical sensors, optical amplifiers and waveguides.
The invention also includes a method for preparing photoluminescent silicon nanoparticles comprising acid etching the nanocrystalline-Si/SiO₂composites prepared using a method described above under conditions effective to produce photoluminescent silicon nanoparticles.
The acid etching of the nanocrystalline-Si/SiO₂composites may be carried out using any suitable acid, for example hydrofluoric acid, nitric acid or mixtures thereof. In an embodiment of the invention, the acid is hydrofluoric acid, suitably dilute HF.
Conditions effective to produce photoluminescent silicon nanoparticles include treating the nanocrystalline-Si/SiO₂composite films with acid for about 2 minutes to about 30 minutes, suitably about 15-16 minutes, at temperatures in the range of about 20° C.-30° C.
The photoluminescence from the nanoparticles after etching can be stabilized significantly by chemical oxidization using, for example, HNO₃. The particles produce stable colloidal dispersions in diols and triols.
Thus, the present invention also includes a process for stabilizing photoluminescence of the silicon nanoparticles of the present invention by treating the photoluminescent silicon nanoparticles with an oxidizer under conditions effective to achieve particle surface oxidation.
Modification of the nanoparticle surfaces with organic compounds can also further stabilize the photoluminescence and make them dispersable in a wide range of solvents. Surface coating, such as attachment of organic molecules to hydrogen terminated and hydroxyl terminated surfaces of these nanoparticles has been shown to significantly stabilize the photoluminescence of the nanoparticles against degradation.
Thus, the present invention also includes a process for stabilizing photoluminescence of the silicon nanoparticles of the present invention that involves treating the photoluminescent silicon nanoparticles under conditions effective to produce photoluminescent silicon nanoparticles having an Si—H terminated surface, and then treating the Si—H surface-terminated nanoparticles under conditions effect to achieve particle surface hydrosilylation.
The present invention also includes a process for stabilizing photoluminescence of the silicon nanoparticles of the present invention that involves treating the photoluminescent silicon nanoparticles under conditions effective to produce photoluminescent silicon nanoparticles having an Si—OH terminated surface, and then treating the Si—OH surface-terminated nanoparticles under conditions effect to achieve particle surface silanization.
Silicon nanoparticles of the present invention may be used, for example, in full-colour displays, optical sensors, and fluorescent tags for biological imaging.
The following non-limiting examples are illustrative of the present invention:

EXAMPLES

Reagents and Materials

HSQ was purchased from Dow Corning (tradename FOx-12®) as a 10 weight percent solution in methyl isobutyl ketone. This stock solution was used as received and stored in subdued light and inert atmosphere prior to use. Gold label methanol was purchased from Fisher scientific and freed of oxygen and trace water using a commercial Pure-Solv solvent system. Electronic grade hydrofluoric acid was purchased from J. T. Baker as a 49% aqueous solution and was used as received. Optical grade fused quartz was purchased from Esco Products. Prime-grade one-side-polished single-crystal Si substrates with no oxide back seal were purchased from Silicon Quest International.
Thin Film nc-Si/SiO₂Composite Fabrication.
HSQ films were deposited onto substrates (for example optical grade silica from Esco Products, Si(111) and Si(100)) using a Laurell Technologies WS-400B-6NPP/LITE programmable spincoater in a nitrogen filled glovebox (<0.1 ppm H₂O; 0.6 ppm O₂). Colourless, uniform films were formed by dropping 0.5 ml of stock solution onto the substrate followed by spinning at 0-5800 rpm in 5 seconds, followed by an additional 30 seconds at 5800 rpm. Varying the maximum spin rate allows for straightforward thickness control of and precursor films. HSQ films were transferred in inert atmosphere to a high temperature furnace and annealed for one hour in 4% H₂and 96% N₂. While in the furnace the samples remained on a cool stage until the desired temperature was reached. The stage was subsequently raised into the furnace “hot zone” where the sample was rapidly heated to peak temperature where it remained for 1 hour. Thicknesses of clear colourless precursor and coherent, light orange, annealed films on all substrates were determined using a Gaertner Multiangle Elispometer and are summarized in Table 1.
Bulk nc-Si/SiO₂Composite Preparation and nc-Si Isolation.
Solvent was removed from the HSQ stock solution to yield a white, crystalline solid. The solid was placed in a quartz crucible and transferred in inert atmosphere to a high temperature furnace and annealed for one hour in a 4% H₂and 96% N₂atmosphere. As in the preparation of composite thin films, samples remained on a cool stage until the desired temperature was reached at which time the stage was then raised into the furnace “hot zone” where the sample remained for 1 hour. After cooling to room temperature, the amber solid sample was removed and mechanically ground in a mortar and pestle to yield a fine brown powder which was subsequently etched in 1:1:1 49% HF:H₂O:ethanol for 2 hours followed by repeated washing/centrifugation cycles in methanol. Yield: 4.8%

Thermogravametric Analysis.

Thermogravametric analysis (TGA) was performed using a Perkin Elmer Pyris 1 TGA equipped with Pyris Thermal Analysis 7.0 software. Samples were placed in a Pt pan and heated in N₂or 4% H₂: 96% N₂atmospheres from room temperature to 1100° C. at 10, 20, 50, 100° C./min.

Material Characterization and Instrumentation.

Optical absorption spectra of annealed films were measured using a Cary 6000i dual beam UV-vis-NIR spectrometer. Photoluminescence (PL) spectra were evaluated at room temperature using the 325 nm line of a He—Cd laser excitation source and emission was detected with a fiber-optic digital charge coupled device (CCD) spectrometer whose spectral response was normalized using a standard black-body radiator. Fourier-transform infrared spectroscopy (FTIR) of thin film samples was performed using a Nicolet Magna 750 IR spectrophotometer. Transmission electron microscopy (TEM) and electron dispersive x-ray (EDX) analysis were preformed using a JEOL-2010 (La B₆filament) electron microscope with an accelerating voltage of 200 keV. Thin TEM samples were prepared by lifting off a piece of film with a razor blade and mounting it onto a copper grid with a 400 μm diameter hole. Samples were subsequently ion milled to perforation and images were obtained from the edge of the milled hole. TEM samples of the liberated, freestanding Si nanoparticles were dropcoated from a methanol suspension onto carbon coated copper grids. Bulk crystallinity of nc-Si/SiO₂composites was evaluated using an INEL XRG 3000 x-ray diffractometer equipped with a Cu Kα radiation source.

Example 1

Thin Film and Bulk nc-Si/SiO₂Composite Preparation

Hydrogen silsesquioxane was spincoated from a commercially available FOx-12® as a conformal, clear, colourless film that, upon reductive thermal processing in a 4% H₂: 96% N₂atmosphere, maintains its transparency, appears faint orange, is adherent (Scotch tape test) and resistant to abrasion (FIG. 1, FIG. 2). Reductive annealing of bulk quantities of white, crystalline HSQ using the aforementioned conditions for thin film thermal processing yields dark amber, glassy products that upon mechanical grinding yield orange/brown powders in near quantitative yield (FIG. 2B,C). TEM, SAED, XRD, and photoluminescence spectroscopy confirm thermal processing of thin film and bulk HSQ samples yield luminescent, diamond lattice, elemental silicon nanocrystallites encapsulated in a SiO₂-like matrix (vide infra). Embedded Si nanoparticles are readily liberated upon exposure of the nanocomposite to hydrofluoric acid which preferentially etches away the silicon oxide matrix leaving freestanding highly luminescent nc-Si.⁵⁶(FIG. 3).

Example 2

Thermal Gravimetric Analysis

Thermal processing of neat HSQ samples was evaluated using thermal gravimetric analysis (TGA). In a nitrogen atmosphere, thermal traces of HSQ obtained at a heating rate of 10° C./min show four distinct regions of weight loss (ca. 50-225° C.; 1.8%, 225-375° C.; 2.1%, ca. 375-425° C.; 0.8%, ca. 507° C.; 43%). With increased heating rate (i.e., 10, 20, 50, 100° C./min) a dramatic decrease (FIG. 4) in the observed weight loss at 507° C. (i.e., 10° C./min, 43% vs. 100° C./min, 5%) was noted. TGA analysis of HSQ in 4% H₂:96% N₂shows ca. 5% weight loss regardless of heating rate.
The accepted stages of HSQ thermal processing in inert atmosphere (i.e., N₂or Ar) are summarized in FIG. 5 and have previously been attributed to: i) trace solvent loss (<200° C.), ii) cage network redistribution with associated loss of SiH₄(ca. 250-350° C.), iii) Si—H thermal dissociation accompanied by loss of SiH₄and H₂(350-450° C.), and iv) collapse of the pore structure (>450° C.). Detailed literature studies of HSQ thermal properties report on thermal processing of partially crosslinked HSQ gels and thin films, accounting for small differences in the present DSC temperature ranges.⁵⁷The structures of low temperature (i.e., 250° C.-350° C.) thermally processed HSQ films have been studied spectroscopically, however the identity of any gas byproducts remains unknown. This is not the case for high temperature region of the TGA for which the loss of SiH₄and H²has been confirmed by mass spectrometry.⁵⁸Consistent with our nitrogen atmosphere TGA weight loss observations, Belot et al. noted a decrease in SiH₄evolution and weight loss at 507° C. with increased heating rates and proposed that it might result from the rapid thermal decomposition of SiH₄into silicon and hydrogen.⁵⁸Herein, it was confirmed that the observed trend in weight loss at ca. 450° C. with increased heating rate is the result of the thermal decomposition of SiH₄and the formation of Si. Upon rapid heating of HSQ (i.e., a 50° C./min), thermally liberated SiH₄is unable to escape the rapidly forming silicon oxide matrix prior to thermally decomposing. This process yields SiO₂encapsulated Si nanocrystals (FIG. 5E). When thermally processed in a 4% H₂: 96% N₂atmosphere, HSQ loses a maximum of 6.3 wt. % for a heating rate of 10° C./min. While not wishing to be limited by theory, one possible explanation of the exact role of hydrogen in this processing atmosphere may be H₂-induced modifications to the HSQ decomposition mechanism (FIG. 5E). It is conceivable that low concentrations of H₂in the thermal processing atmosphere limit, and may even prevent the dehydrogenation of HSQ thereby increasing the SiH₄available for thermal decomposition.

Example 3

FT-IR Spectroscopy

FT-IR spectroscopy of NEAT HSQ shows a characteristic absorption at 2251 cm⁻¹that is readily assigned to Si—H stretching (FIG. 6A). Absorptions are also noted in the range ca. 1300 to 800 cm⁻¹and have previously been assigned to internal vibrations of the Si—O—Si cage framework.⁵⁹Following reductive annealing (FIG. 6B) the absorption assigned to Si—H stretching disappears suggesting the HSQ molecules are crosslinked and the cage structure has collapsed. The replacement of broad HSQ Si—O—Si vibrations with a broad featureless absorption centered at ca. 1096 cm⁻¹, that was assigned to Si—O—Si bending in an SiO₂-like network, was also observed. FIG. 6C shows a typical FT-IR spectrum of partially etched nc-Si/SiO₂powder following etching in 1:1:1 49% HF:H₂O:ethanol for 120 minutes. Characteristic Si-ft stretching at 2100 cm⁻¹confirms particle hydride termination. In addition, a marked decrease in intensity of absorptions attributed to Si—O—Si vibrations at s 1400 cm⁻¹was noted with increased etching time. These vibrations are completely absent from the spectra of fully etched samples (FIG. 6D). Etching times to achieve complete removal of the SiO₂-like matrix depend on grinding efficiency of the composite material and the corresponding composite particle size.

Example 4

Absorption and Photoluminescence Spectroscopy

The UV-vis-NIR spectra of HSQ, a typical annealed HSQ film, and liberated Si nanoparticles provide limited information regarding the presence of nc-Si. The HSQ spectrum is featureless. In contrast, the spectra of a typical annealed thin film and suspended nc-Si/SiO₂powder exhibit a low energy absorption onset at ca. 350 nm. Silicon nanocrystals typically have an absorption onset in the UV or blue, depending on the particle size.²Following etching, the UV-vis-NIR absorption spectrum of freestanding Si nanocrystals shows a low energy absorption onset at 350 nm.
HSQ precursor films show no detectable photoluminescence. The influence of annealing temperature on the photoluminescent response of thermally processed HSQ films (thickness=140 nm) is significant (FIG. 7). Films processed at or below 700° C. in a 4% hydrogen atmosphere show no detectable visible photoluminescence at the limits of the instrumentation. When thin films of HSQ are annealed at 900° C., a weak PL emission emerges that is peaked at ˜700 nm. As the annealing temperature was increased up to 1100° C., a trend toward lower energy emission was observed. Annealing at a maximum temperature of 1100° C. results in an intense PL emission centered at 810 nm that is characteristic of silicon nanocrystal films.
While not wishing to be limited by theory, the noted dependence of PL emission maximum on thermal processing temperature may be understood in the context of the known HSQ thermal decomposition processes and present TGA analyses (vide supra). Clearly, any SiH₄evolved during slow heating escapes the oxide network and does not thermally decompose to produce nc-Si as seen in TGA heating profiles (i.e., 20, 50° C./min) and the PL characterization of thin film samples that are processed at or below 700° C. With increased processing temperature, the sample temperature increases more rapidly in the furnace hot zone and less SiH₄escapes prior to decomposing; this is manifested in less weight loss and the appearance of visible photoluminesence. While the exact origin of the PL maximum for samples annealed at 900° C. is not completely clear, present TGA analysis suggests that not all evolved SiH₄escapes when samples are heated at moderate rates (ca. 50° C./min). Previous studies of “SiO_x” films show the crystallization of Si nanocrystals occurs between 800 and 900° C.;²⁸the photoluminesence of HSQ films annealed at 900° C. seems consistent with these observations. HSQ samples processed below 900° C. yield Si particles whose dimension are below the detection limits of standard TEM and XRD analyses. In this regard, a reasonable explanation for the observed low intensity, high-energy photoluminesence maxima for these films is the photoexcitation/emission of very small Si clusters arising from the thermal decomposition of small quantities of matrix trapped SiH₄. The observed shift to lower energy is easily understood in the context of larger particles that form after annealing at higher temperatures.
Photoluminescence characterization of methanol suspensions of highly luminescent HF liberated Si nanoparticles (d=2.1±0.3 nm) with a peak PL emission at 640 nm is shown in FIG. 8B. The observed blue shift in PL maximum upon exposure to HF is consistent with previous reports for etched nc-Si/SiO₂nanocomposites.³⁵

Example 5

Transmission Electron Microscopy, EDX and SAED

FIG. 9A shows a representative bright field TEM image of a thin film annealed at 1100° C., showing irregular nc-Si particles with an average diameter 2.27 nm (2σ=0.59 nm; n=115) (FIG. 9C). Silicon nanoparticles are clearly uniformly distributed throughout the film (FIG. 9A). EDX analysis confirms the presence of only Si and O, Selected area electron diffraction (FIG. 9A, inset) shows the particles are crystalline and have the characteristic diamond lattice of silicon. TEM analysis of HF-liberated nc-Si dropcoated from methanol suspensions onto a carbon-coated grid shows discrete Si nanoparticles with diameters of d=2.11 nm (2σ=0.82 nm; n=50) (FIG. 10). This apparent decrease in particle size and broadening of the size distribution is likely the result of limited particle etching and size selective precipitation by which large particles are removed and only the smallest particles remain suspended in the solvent throughout the etching process. (Note: the apparent error is more than the difference in the sizes) Selected area electron diffraction confirms particles to be diamond lattice Si with the {111}, {200}, and (311) reflections are visible.

Example 6

X-Ray Powder Diffraction

The crystallinity of the SiO₂-embedded Si nanoparticles was evaluated by x-ray powder diffraction (XRD) using Cu K_αradiation source (FIG. 11). The positions and intensities of all broadened reflections agree with selected area electron diffraction noted for the abovementioned thin film analogues and are consistent with a diamond lattice Si.
While the present invention has been described with reference to what are presently considered to be the preferred examples, it is to be understood that the invention is not limited to the disclosed examples. To the contrary, the invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
All publications, patents and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety. Where a term in the present application is found to be defined differently in a document incorporated herein by reference, the definition provided herein is to serve as the definition for the term.

TABLE 1

	Maximum Spin	HSQ Thickness	nc-Si/SiO₂
Substrate	Rate (RPM)	(Å)^a	Thickness (Å)^a

Optical grade	5800	1393	923
SiO₂
Si(111)	5800	1620	1553
Si(100)	5800	1617	1524

^aFilm thicknesses were determined using η_Si= 3.85, η_SiO2= 1.46, and η_film= 1.41.

FULL CITATIONS FOR DOCUMENTS REFERRED TO IN THE SPECIFICATION

¹Buriak, J. M. Chem. Rev. 2002, 102, 1271.
²Meldrum, A. Resent Res. Devel. Nuclear Phys. 2004, 1, 93.
³Takagahara, T.; Takeda, K. Phys. Rev. B 1992, 46, 15578.
⁴Klimov, V. I.; Schwarz, C. J.; McBranch, D. W.; White, C. W. Appl. Phys. Lett. 1998, 73, 2603.
⁵Canham, L. T. Appl. Phys. Lett. 1990, 57, 1046.
⁶Baldwin, R. K.; Pettigrew, K. A.; Garno, J. C.; Power, P. P.; Liu, G.-Y.; Kauzlarich, S. M. J. Am. Chem. Soc. 2002, 124, 1150.
⁷Pettigrew, K.; Liu, Q.; Power, P. P.; Kauzlarich, S. M. Chem. Mater. 2003, 15, 4005.
⁸Rowsell, B. D.; Veinot, J. G. C. Nanotechnology 2005, 16, 732.
⁹Li, X. G.; He, Y. Q.; Swihart, M. T. Langmuir 2004, 20, 4720.
¹⁰Li, X. G.; He, Y. Q.; Talukdar, S. S.; Swihart, M. T. Langmuir 2003, 19, 8490.
¹¹Nayfeh, M. H.; Barry, N.; Therrien, J.; Akcakir, O.; Gratton, E.; Belomoin, G. Appl. Phys. Lett. 2001, 78, 1131.
¹²Nayfeh, M. H.; Akcakir, O.; Belomoin, G.; Barry, N.; Therrien, J.; Gratton, E. Appl. Phys. Lett. 2000, 77, 4086.
¹³Belomoin, G.; Therrien, J.; Smith, A.; Rao, S.; Twesten, R.; Chaleb, S.; Nayfeh, M. H.; Wagner, L.; Mitas, L. Appl. Phys. Lett. 2002, 80, 841.
¹⁴Ding, Z.; Quinn, B. M.; Haram, S. K.; Pell, L. E.; Korgel, B. A.; Bard, A. J. Science 2002, 296, 1293.
¹⁵M. Glover and A. Meldrum, Opt. Mater. 2005, 27, 977.
¹⁶Breitenecker, M.; Sexl, R.; Thirring, W. Zeit. Physik 1964, 182, 123.
¹⁷Hill, N. A.; Whaley, K. B. J. Electron. Mater. 1996, 25, 269.
¹⁸Hill, N. A.; Whaley, K. B. Phys. Rev. Lett. 1995, 75, 1130.
¹⁹Wang, L. W.; Zunger, A. J. Phys. Chem. 1994, 98, 2158.
²⁰Delerue, C.; Allan, G.; Lannoo, M. Phys. Rev. B 1993, 48, 11024.
²¹Delerue, C.; Allan, G.; Lannoo, M. J. Lumines. 1998, 80, 65.
²²Ogut, S.; Chelikowsky, J. R.; Louie, S. G. Phys. Rev. Lett. 1997, 79, 1770.
²³Shimizu-lwayama, T.; Ohshima, M.; Niimi, T.; Nakao, S.; Saitoh, K.; Fujita, T.; Itoh, N. J. Phys.: Condens. Matter 1993, 5, L375.
²⁴Meldrum, A.; Haglund, R. F.; Boatner, L. A.; White, C. W. Adv. Mater. 2001, 13, 1431.
²⁵Kehler, U.; Hofineister, H.; Opt. Mater. 2001, 17, 83.
²⁶Gorbilleau, F.; Portier, X.; Ternon, C.; Voivenel, P.; Madelon, R.; Riszk, R. Appl. Phys. Lett. 2001, 78, 3058.
²⁷Orii, T.; Hirasawa, M.; Seto, T. Appl. Phys. Lett. 2003, 83, 3395.
²⁸A. Meldrum, A. Hryciw, A. N. MacDonald, C. Blois, K. Marsh, J. Wang, and Q. Li, J. Vac. Sci. Tech. (in press).
²⁹Neufeld, E.; Wang, S.; Aptez, R.; Buchal, Ch.; Carius, R.; White, C. W.; Thomas, D. K. Thin Sol. Films 1997, 294, 238.
³⁰Holhl, A.; Weider, T.; van Aken, P. A.; Weirich, T. E.; Denninger, G.; Vidal, M.; Oswald, S.; Deneke, C.; Mayer, J.; Fuess, H. J. Non-Cryst. Sol. 2003, 320, 255. and references therein.
³¹Schulmeister, K.; Mader, W. J. Non-Cryst. Sol. 2003, 320, 143.
³²Commercially available “SiO” is an orange/brown/black solid, however SiO is reported to be a white, crystalline (cub.) solid with m.p. >1702° C. See: Handbook of Chemistry and Physics 73^rd Edition, D. R. Linde Editor, CRC Press, Boca Raton, Fla. 1997.
³³“SiO_x” is viewed as a kinetically inert, amorphous material that converts to the thermodynamically favored nanocrystalline Si and SiO₂upon thermal processing. See ref. 30.
³⁴Valenta, J.; Janda, P.; Dohnalova, K.; Niznansky, D.; Vacha, F.; Linnros, J. Opt. Mat. 2005, 27, 1046.
³⁵Recently bulk thermolysis of “SiO_x” was reported however no yield was provided. See: Liu, S. M.; Sato, S.; Kimura, K. Langmuir 2005, 21, 6324.
³⁶Heath, J. R.; Shiang, J. J.; Alivisatos, A. P. J. Chem. Phys. 1994, 101, 1607.
³⁷Baldwin, R. K.; Pettigrew, K. A.; Garno, J. C.; Power, P. P.; Liu, G. Y.; Kauzlarich, S. M. J. Am. Chem. Soc. 2002, 124, 1150.
³⁸Zou, J.; Baldwin, R. K.; Pettigrew, K. A.; Kauzlarich, S. M. Nanolett. 2004, 7, 1181.
³⁹Pettigrew, K. A.; Liu, Q.; Power, P. P.; Kauzlarich, S. M. Chem. Mat. 2003, 15, 4005.
⁴⁰Liu, Q.; Kauzlarich, S. M. Mat. Sci. and Eng. B 2002, 96, 72.
⁴¹Mayeri D, Phillips B L, Augustine M P, Kauzlarich S M Chem. Mater, 2001, 13, 765.
⁴²Baldwin, R. K.; Pettigrew, K. A.; Ratai, E.; Augustine, M. P.; Kauzlarich, S. M. Chem. Comm. 2002, 17, 1822.
⁴³Wilcoxon, J. P.; Provencio, P. P.; Samara, G. A. Phys. Rev. B 1999, 60, 2704.
⁴⁴Brook, M. B. Silicon in Organic, Organometallic, and Polymer Chemistry, John Wiley and Sons, Inc.: New York, N.Y., 2000.
⁴⁵Feher, F. J.; Budzichowski, T. A.; Blanski, R. L.; Weller, K. J.; Ziller, J. W. Organometallics 1991, 10, 2526.
⁴⁶Feher, F.; Newman, D. A. J. Am. Chem. Soc. 1990, 112, 1931.
⁴⁷Feher, F.; Newman, D. A.; Walzer, J. F. J. Am. Chem. Soc. 1989, 111, 1741.
⁴⁸Feher, F. J.; Budzichowski, T. A.; Rahimian, K.; Ziller, J. W. J. Am. Chem. Soc. 1992, 114, 3859.
Azinovic, D.; Cai, J.; Eggs, C.; Konig, H.; Marsmann, H. C.; Veprek, S. J. Lumines. 2002, 97, 40.
⁵⁰Tour, J. M.; Pendalwar, S. L.; Cooper, J. P. Chem. Mat. 1990, 2, 647.
⁵¹Brook, M. A.; Ketelson, H. A. M.; Pelton, R. H.; Heng, Y.-M. Chem. Mat. 1996, 8, 2195.
⁵²Arkles, B.; Berry, D. H.; Figge, L. K.; Composto, R. J.; Chiou, T.; Colazzo, H.; Wallace, W. E. Sol-Gel Sci. Tech. 1997, 8, 465.
⁵³Yang, C.,-C.; Chen, W.-C. J. Mat. Chem. 2002, 12, 1138.
⁵⁴Hummel, J.; Endo, K.; Lee, W. W.; Mills, M.; Wang, S. Q. Low-Dielectric Constant Materials V, The Materials Research Society, Warrendale, Pa., 1999, vol. 565.
⁵⁵http://www.dowcorning.com/content/etronics/etronicsspin/etronics_spin_imo v. asp.
⁵⁶Williams, K. R. Journal of Microelectromechanical Systems, 1996, 5, No. 4.
⁵⁷Detailed literature studies of HSQ thermal properties report on thermal processing of partially crosslinked HSQ gels and thin films, accounting for small differences in the present DSC temperature ranges. See: Yang, C., —C.; Chen, W.,-C. J. Mat. Chem. 2002, 12, 1138.
⁵⁸Belot, V.; Corriu, R.; Leclecq, D.; Mutin, P. H.; Vious, A. Chem. Mat. 1991, 3, 127.
⁵⁹Albrecht, M. G.; Blanchette, C. J. Electrochem. Soc. 1998, 145, 4019.

Claims

1. A nanocrystalline-Si/SiO₂composite containing silicon nanoparticles prepared by curing hydrogen silsesquioxane (HSQ) under reductive thermal conditions in the presence of a gas containing hydrogen.

2. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 1 in the form of a film.

3. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 1 in which said reductive thermal conditions comprise exposing said hydrogen silsesquioxane to a temperature in the range of from about 900° C. to about 1200° C. for about 30 minutes to about two hours.

4. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 3 wherein said reductive thermal conditions include the presence of an inert gas.

5. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 4 wherein said inert gas is nitrogen.

6. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 4 wherein the gas comprises about 2% to about 6% hydrogen and about 94% to about 98% nitrogen.

7. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 2 in which said film is on a substrate.

8. The nanocrystalline-Si/SiO₂composite containing silicon nanoparticles as claimed in claim 7 in which said substrate is optical grade silica.

9. Freestanding photoluminescent silicon nanoparticles prepared by acid etching the nanocrystalline-Si/SiO₂composite as claimed in claim 1.

10. Freestanding photoluminescent silicon nanoparticles as claimed in claim 9 wherein the silicon nanoparticles are exposed to an said acid for about 2 minutes to about 30 minutes and at a temperature in the range of from about 20° C. to about 30° C.

11. Freestanding photoluminescent silicon nanoparticles as claimed in claim 9 in which the acid is hydrofluoric acid, nitric acid, or mixtures thereof.