Growth mechanism of thin silicon oxide films on Si(100) studied by medium-energy ion scattering

Bridging the gap between surface physics and photonics

Use and performance criteria of photonic devices increase in various application areas such as information and communication, lighting, and photovoltaics. In many current and future photonic devices, surfaces of a semiconductor crystal are a weak part causing significant photo-electric losses and malfunctions in applications. These surface challenges, many of which arise from material defects at semiconductor surfaces, include signal attenuation in waveguides, light absorption in light emitting diodes, non-radiative recombination of carriers in solar cells, leakage (dark) current of photodiodes, and light reflection at solar cell interfaces for instance. To reduce harmful surface effects, the optical and electrical passivation of devices has been developed for several decades, especially with the methods of semiconductor technology. Because atomic scale control and knowledge of surface-related phenomena have become relevant to increase the performance of different devices, it might be useful to enhance the bridging of surface physics to photonics. Toward that target, we review some evolving research subjects with open questions and possible solutions, which hopefully provide example connecting points between photonic device passivation and surface physics. One question is related to the properties of the wet chemically cleaned semiconductor surfaces which are typically utilized in device manufacturing processes, but which appear to be different from crystalline surfaces studied in ultrahigh vacuum by physicists. In devices, a defective semiconductor surface often lies at an embedded interface formed by a thin metal or insulator film grown on the semiconductor crystal, which makes the measurements of its atomic and electronic structures difficult. To understand these interface properties, it is essential to combine quantum mechanical simulation methods. This review also covers metal-semiconductor interfaces which are included in most photonic devices to transmit electric carriers to the semiconductor structure. Low-resistive and passivated contacts with an ultrathin tunneling barrier are an emergent solution to control electrical losses in photonic devices.

Ultra-compact SOS-based bi-metallic TM-pass polarizer

Ultra-compact transverse magnetic (TM)-pass polarizer based on silicon on sapphire (SOS) platform is proposed and analysed. Low power consumption, high linearity and high speed of transmission are the major advantages of the SOS platform in different commercial applications especially in the mid infrared region.The suggested structure has bimetallic configuration of aluminium doped zinc oxide (AZO) and zirconium nitride (ZrN) to highly attenuate the quasi transverse electric (TE) mode. This is due to the coupling between the fundamental TE and the surface plasmon modes. However, the transverse magnetic mode can propagate with minimal losses. At 2.0 µm operating wavelength, the proposed TM-pass polarizer realizes 20.3 dB extinction ratio (ER) with 0.14 dB insertion loss (IL) at a device length of 3.0 µm. Therefore, the reported design has advantages of compact length, high efficiency and CMOS-compatibility.

Modification of a H-terminated silicon surface by organic sulfide molecules: the mechanism and origin of reactivity

For the reactions of disulfide molecules (RSSR), the steric effect rather than the electronic effect of the R group is the main origin of the different reactivity. In the reactions of sulfide molecules (RSXR′, X = S, P, Si, O, N, C), charges on the S atom and dissociation energies of the S–X bonds have a great impact on the reactivity of these reactions.

CMOS-compatible hybrid bi-metallic TE/TM-pass polarizers based on ITO and ZrN

Transverse-magnetic (TM) and transverse-electric (TE) pass polarizers based on a silicon-on-insulator platform are studied and analyzed. The proposed structures are CMOS-compatible based on indium tin oxide and zirconium nitride as alternative plasmonic materials. The bi-metallic combination of the plasmonic materials exhibit large coupling between one of the modes (TE or TM) in the silicon core and the surface plasmon mode, while the other mode can propagate with low losses. The numerical simulations for the TE-pass polarizer predict 32.7 dB extinction ratio (ER) and 0.13 dB insertion loss (IL) at a compact device length of 1.5 μm. Additionally, the TM-pass polarizer has 31.5 dB ER and 0.17 dB IL at a device length of 2 μm at an operating wavelength of 1.55 μm.

Development of the ReaxFF Reactive Force Field for Inherent Point Defects in the Si/Silica System

We redeveloped the ReaxFF force field parameters for Si/O/H interactions that enable molecular dynamics (MD) simulations of Si/SiO₂ interfaces and O diffusion in bulk Si at high temperatures, in particular with respect to point defect stability and migration. Our calculations show that the new force field framework (ReaxFF_present), which was guided by the extensive quantum mechanical-based training set, describes correctly the underlying mechanism of the O-migration in Si network, namely, the diffusion of O in bulk Si occurs by jumping between the neighboring bond-centered sites along a path in the (110) plane, and during the jumping, O goes through the asymmetric transition state at a saddle point. Additionally, the ReaxFF_present predicts the diffusion barrier of O-interstitial in the bulk Si of 64.8 kcal/mol, showing a good agreement with the experimental and density functional theory values in the literature. The new force field description was further applied to MD simulations addressing O diffusion in bulk Si at different target temperatures ranging between 800 and 2400 K. According to our results, O diffusion initiates at the temperatures over 1400 K, and the atom diffuses only between the bond-centered sites even at high temperatures. In addition, the diffusion coefficient of O in Si matrix as a function of temperature is in overall good agreement with experimental results. As a further step of the force field validation, we also prepared amorphous SiO₂ (a-SiO₂) with a mass density of 2.21 gr/cm³, which excellently agrees with the experimental value of 2.20 gr/cm³, to model a-SiO₂/Si system. After annealing the a-SiO₂/Si system at high temperatures until below the computed melting point of bulk Si, the results show that ReaxFF_present successfully reproduces the experimentally and theoretically defined diffusion mechanism in the system and succeeded in overcoming the diffusion problem observed with ReaxFF_SiOH(2010), which results in O diffusion in the Si substrate even at the low temperature such as 300 K.

Toward the Understanding of Selective Si Nano-Oxidation by Atomic Scale Simulations

The continuous miniaturization of nanodevices, such as transistors, solar cells, and optical fibers, requires the controlled synthesis of (ultra)thin gate oxides (<10 nm), including Si gate-oxide (SiO₂) with high quality at the atomic scale. Traditional thermal growth of SiO₂ on planar Si surfaces, however, does not allow one to obtain such ultrathin oxide due to either the high oxygen diffusivity at high temperature or the very low sticking ability of incident oxygen at low temperature. Two recent techniques, both operative at low (room) temperature, have been put forward to overcome these obstacles: (i) hyperthermal oxidation of planar Si surfaces and (ii) thermal or plasma-assisted oxidation of nonplanar Si surfaces, including Si nanowires (SiNWs). These nano-oxidation processes are, however, often difficult to study experimentally, due to the key intermediate processes taking place on the nanosecond time scale. In this Account, these Si nano-oxidation techniques are discussed from a computational point of view and compared to both hyperthermal and thermal oxidation experiments, as well as to well-known models of thermal oxidation, including the Deal-Grove, Cabrera-Mott, and Kao models and several alternative mechanisms. In our studies, we use reactive molecular dynamics (MD) and hybrid MD/Monte Carlo simulation techniques, applying the Reax force field. The incident energy of oxygen species is chosen in the range of 1-5 eV in hyperthermal oxidation of planar Si surfaces in order to prevent energy-induced damage. It turns out that hyperthermal growth allows for two growth modes, where the ultrathin oxide thickness depends on either (1) only the kinetic energy of the incident oxygen species at a growth temperature below T_trans = 600 K, or (2) both the incident energy and the growth temperature at a growth temperature above T_trans. These modes are specific to such ultrathin oxides, and are not observed in traditional thermal oxidation, nor theoretically considered by already existing models. In the case of thermal or plasma-assisted oxidation of small Si nanowires, on the other hand, the thickness of the ultrathin oxide is a function of the growth temperature and the nanowire diameter. Below T_trans, which varies with the nanowire diameter, partially oxidized SiNW are formed, whereas complete oxidation to a SiO₂ nanowire occurs only above T_trans. In both nano-oxidation processes at lower temperature (T < T_trans), final sandwich c-Si|SiO_x|a-SiO₂ structures are obtained due to a competition between overcoming the energy barrier to penetrate into Si subsurface layers and the compressive stress (∼2-3 GPa) at the Si crystal/oxide interface. The overall atomic-simulation results strongly indicate that the thickness of the intermediate SiO_x (x < 2) region is very limited (∼0.5 nm) and constant irrespective of oxidation parameters. Thus, control over the ultrathin SiO₂ thickness with good quality is indeed possible by accurately tuning the oxidant energy, oxidation temperature and surface curvature. In general, we discuss and put in perspective these two oxidation mechanisms for obtaining controllable ultrathin gate-oxide films, offering a new route toward the fabrication of nanodevices via selective nano-oxidation.

Diffusion of interstitial oxygen in silicon and germanium: a hybrid functional study

The minimum-energy paths for the diffusion of an interstitial O atom in silicon and germanium are studied through the nudged-elastic-band method and hybrid functional calculations. The reconsideration of the diffusion of O in silicon primarily serves the purpose of validating the procedure for studying the O diffusion in germanium. Our calculations show that the minimum energy path goes through an asymmetric transition state in both silicon and germanium. The stability of these transition states is found to be enhanced by the generation of unpaired electrons in the highest occupied single-particle states. Calculated energy barriers are 2.54 and 2.14 eV for Si and Ge, in very good agreement with corresponding experimental values of 2.53 and 2.08 eV, respectively.

Sn Spheres Embedded in a SiO2 Matrix: Synthesis and Potential Application As Self-Destructing Materials

We introduce a simple process for the fabrication of SiO2 films embedded with β-Sn-rich nano/microspheres. Sn spheres with maximum and minimum sizes of 10 μm (near the SiO2 surface) and 5 nm (at the Si/SiO2 interface) were grown within a 0.7-5.7 μm-thick SiO2 layer by evaporating SnO powders onto an Si (100) substrate for 1-600 min at 600-900 °C and 0.001-5.0 Torr. A possible growth mechanism of these materials is discussed. The current-voltage characteristics of the as-fabricated samples were investigated to identify potential applications. During these tests, small flashes of light and the presence of damaged areas were observed at the oxide surfaces of the samples using an optical camera and a field emission scanning electron microscope, respectively. The electrical breakdown and shutdown of the devices observed in the current-voltage curves were attributed to the destruction of the SiO2 surface. In addition, the current-time responses show that the size of the damaged regions can be controlled by the voltage and duration of the applied stress, and are independent of the size and shape of the electrodes. The present materials thus possess great potential for applications in self-destructing devices.

Expanding the Repertoire of Molecular Linkages to Silicon: Si-S, Si-Se, and Si-Te Bonds

Silicon is the foundation of the electronics industry and is now the basis for a myriad of new hybrid electronics applications, including sensing, silicon nanoparticle-based imaging and light emission, photonics, and applications in solar fuels, among others. From interfacing of biological materials to molecular electronics, the nature of the chemical bond plays important roles in electrical transport and can have profound effects on the electronics of the underlying silicon itself, affecting its work function, among other things. This work describes the chemistry to produce ≡Si-E bonds (E = S, Se, and Te) through very fast microwave heating (10-15 s) and direct thermal heating (hot plate, 2 min) through the reaction of hydrogen-terminated silicon surfaces with dialkyl or diaryl dichalcogenides. The chemistry produces surface-bound ≡Si-SR, ≡Si-SeR, and ≡Si-TeR groups. Although the interfacing of molecules through ≡Si-SR and ≡Si-SeR bonds is known, to the best of our knowledge, the heavier chalcogenide variant, ≡Si-TeR, has not been described previously. The identity of the surface groups was determined by Fourier transform infrared (FTIR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and depth profiling with time-of-flight-secondary ionization mass spectrometry (ToF-SIMS). Possible mechanisms are outlined, and the most likely, based upon parallels with well-established molecular literature, involve surface silyl radicals or dangling bonds that react with either the alkyl or aryl dichalcogenide directly, REER, or its homolysis product, the alkyl or aryl chalcogenyl radical, RE· (where E = S, Se, and Te).

Fabrication of a plasmonic modulator incorporating an overlaid grating coupler

The fabrication of a novel plasmonic reflection modulator is presented and described. The modulator includes plasmon excitation using a diffraction grating coupler and is based on a metal-insulator-semiconductor structure on silicon. Fabrication includes a thin thermal oxide, a plasmonic metal surface defined by optical lithography, a metal grating coupler defined by overlaid e-beam lithography, a passivation layer with metalized vias, and electrical contacts. Physical characterization of intermediate structures is provided along with modulation measurements at λ0 ∼ 1550 nm which verify the concept.

Synthesis of homogeneous and high-quality GaN films on Cu(111) substrates by pulsed laser deposition

Homogeneous and high-quality GaN films with a RMS thickness inhomogeneity of less than 2.8% were grown on an AlN buffer layer using pulsed laser deposition and optimized laser rastering program.

Direct measurement of the growth mode of graphene on SiC(0001) and SiC(0001¯)

We have determined the growth mode of graphene on SiC(0001) and SiC(0001¯) using ultrathin, isotopically labeled Si(13)C "marker layers" grown epitaxially on the Si(12)C surfaces. Few-layer graphene overlayers were formed via thermal decomposition at elevated temperature. For both surface terminations (Si face and C face), we find that the (13)C is located mainly in the outermost graphene layers, indicating that, during decomposition, new graphene layers form underneath existing ones.

Synthesis and oxidation of luminescent silicon nanocrystals from silicon tetrachloride by very high frequency nonthermal plasma

Silicon nanocrystals have recently attracted significant attention for applications in electronics, optoelectronics, and biological imaging due to their size-dependent optical and electronic properties. Here a method for synthesizing luminescent silicon nanocrystals from silicon tetrachloride with a nonthermal plasma is described. Silicon nanocrystals with mean diameters of 3-15 nm are synthesized and have a narrow size distribution with the standard deviation being less than 20% of the mean size. Control over crystallinity is achieved for plasma pressures of 1-12 Torr and hydrogen gas concentrations of 5-70% through adjustment of the plasma power. The size of nanocrystals, and resulting optical properties, is mainly dependent on the gas residence time in the plasma region. Additionally the surface of the nanocrystals is covered by both hydrogen and chlorine. Oxidation of the nanocrystals, which is found to follow the Cabrera-Mott mechanism under ambient conditions, is significantly faster than hydrogen terminated silicon due to partial termination of the nanocrystal surface by chlorine.

Hydrophobic films by atmospheric plasma curing of spun-on liquid precursors

Hydrophobic coatings have been produced on glass and acrylic samples by using a low-temperature atmospheric pressure plasma to polymerize liquid fluoroalkylsilane precursors. The fluoroalkylsilane precursor was dissolved in isooctane and spun onto the substrate at 550 rpm. The sample was then exposed to the reactive species generated from a nitrogen plasma. The plasma was operated with 2.3 vol % N2 in helium at 7.4 W/cm2 at a radio frequency of 27.12 MHz. The total and polar component of the coating's surface energy was found to equal 11.0 and 1.2 dyn/cm, respectively. Average water contact angles of 110 degrees and 106 degrees were measured on the coated glass and acrylic surfaces, respectively. X-ray photoelectron spectroscopy revealed that, after treatment, the fluoroalkyl ligands remained intact on the Si atoms, with a F/C atomic ratio of 2.23.

Surface Chemistry of Aerosolized Nanoparticles:Thermal Oxidation of Silicon

The kinetics of reaction between silicon nanoparticles and molecular oxygen were studied by tandem differential mobility analysis. Aerosolized silicon nanoparticles were extracted from a low-pressure silane plasma into an atmospheric pressure aerosol flow tube reactor. Particles were initially passed through a differential mobility analyzer that was set to transmit only those particles having mobility diameters of approximately 10 nm. The monodisperse particle streams were mixed with oxygen/nitrogen mixtures of different oxygen volume fractions and allowed to react over a broad temperature range (600-1100 degrees C) for approximately one second. Particles were size-classified after reaction with a second differential mobility analyzer. The particle mobility diameters increased upon oxidation by up to 1.3 nm, depending on the oxygen volume fraction and the reaction temperature. Oxidation is described by a kinetic model that considers both oxygen diffusion and surface reaction, with diffusion becoming important after formation of a 0.5 nm thick oxide monolayer.

Atomic Silicon in Siloxanic Networks: The Nature of the Oxo-Oxygen−Silicon Bond

The existence of atomic silicon cryptates in siloxanic networks has been studied theoretically via density functional calculations. By modeling with model molecules the candidate sites to host atomic silicon, we found that metastable adducts can be formed only in regions where the siloxanic network is not subjected to steric constraints; stationary states are instead unstable in highly reticulated siloxanic networks. The nature of the oxo-oxygen-silicon bond at the SiO2 surface is analyzed in detail. It is concluded that silicon is kept at the surface in atomic-like configuration by (i) sigma charge donation from oxo-oxygen atoms into the empty silicon psigma orbital; (ii) pi charge back-donation from singly occupied silicon 3ppi orbitals into empty sigma* model molecule orbitals. Surprisingly, these results attribute to atomic silicon the character of bifunctional Lewis acid.

Reaction of the oxygen molecule at the Si(100)-SiO2 interface during silicon oxidation

Using constrained ab initio molecular dynamics, we investigate the reaction of the O2 molecule at the Si(100)-SiO2 interface during Si oxidation. The reaction proceeds sequentially through the incorporation of the O2 molecule in a Si-Si bond and the dissociation of the resulting network O2 species. The oxidation reaction occurs nearly spontaneously and is exothermic, irrespective of the O2 spin state or of the amount of excess negative charge available at the interface. The reaction evolves through the generation of network coordination defects associated with charge transfers. Our investigation suggests that the Si oxidation process is fully governed by diffusion.

Oxygen diffusion through the disordered oxide network during silicon oxidation

An atomic-scale description is provided for the long-range oxygen migration through the disordered SiO2 oxide during silicon oxidation. First-principles calculations, classical molecular dynamics, and Monte Carlo simulations are used in sequence to span the relevant length and time scales. The O2 molecule is firmly identified as the transported oxygen species and is found to percolate through interstices without exchanging oxygen atoms with the network. The interstitial network for O2 diffusion is statistically described in terms of its potential energy landscape and connectivity. The associated activation energy is found in agreement with experimental values.

O2 diffusion in SiO2: triplet versus singlet

We address the diffusion of the oxygen molecule in SiO2, using first-principles spin-polarized total-energy calculations. We find that the potential energy surfaces for the singlet and triplet states are very different in certain regions, and that the O2 molecule preserves its spin-triplet ground state not only at its most stable interstitial position inside the solid but also throughout its diffusion pathway. Therefore, the singlet state is not a good approximation to describe the behavior of O2 inside SiO2, and spin-polarization effects are fundamental to understand the properties of this system.