Investigation of point defect clusters in silicon using parallel molecular dynamics

A parallel molecular dynamics algorithm is presented for computingconfigurations of relatively large defects in crystalline silicon, as modelledby the Stillinger–Weber (SW) three-body interatomic potential. Thealgorithm is based on a partitioning of physical space among the N processorswith atoms migrating freely between the partitions. Implementation on aneight-processor IBM SP2 computer shows the increased efficiency withsimulation size expected because of the increased computational load perprocessor relative to communication overhead. The parallel efficiency reached70% for 21 952 atoms. Calculations are presented for the thermodynamics offormation of interstitial and vacancy clusters containing up to seven pointdefects. The clusters were relaxed within a host lattice of about 3000 siliconatoms subjected to periodic boundary conditions. Free energies of formationfor temperatures 500 K T 1600 K were computed using thermodynamicintegration. Computed equilibrium distributions for these clusters show ashift to the larger species at lower temperatures, as expected. The SWpotential predicts greater driving forces for interstitial aggregation thanvacancy aggregation across the entire temperature range. Model calculationsfor a large vacancy cluster are also presented to demonstrate the utility ofthe algorithm for exploring very large defects in silicon.     

Vacancy-oxygen defects in silicon: the impact of isovalent doping

Silicon is the mainstream material for many nanoelectronic and photovoltaic applications. The understanding of oxygen related defects at a fundamental level is essential to further improve devices, as vacancy-oxygen defects can have a negative impact on the properties of silicon. In the present review we mainly focus on the influence of isovalent doping on the properties of A-centers in silicon. Wherever possible, we make comparisons with related materials such as silicon germanium alloys and germanium. Recent advanced density functional theory studies that provide further insights on the charge state of the A-centers and the impact of isovalent doping are also discussed in detail.     

Application of defect engineering to silicon Raman lasers and amplifiers

Silicon Raman lasers and amplifiers are the only silicon-based, monolithic device structures in which optical gain has been unambiguously observed. The main limitation on this gain is optical absorption due to free carriers, generated by two photon absorption. Here we explore a means to mitigate carrier effects via defect engineering. The optical and electrical properties of the defected silicon waveguides are modeled for both a uniform defect distribution and a remote, localized defect distribution. Simulation results indicate that a uniform defect distribution provides no improvement with any increase in net gain provided solely by surface modification. In contrast, for devices with remote defect volumes the reduction of carrier lifetime and limited optical absorption results in a significant improvement to net gain.     

Application of defect engineering to silicon Raman lasers and amplifiers

Silicon Raman lasers and amplifiers are the only silicon-based, monolithic device structures in which optical gain has been unambiguously observed. The main limitation on this gain is optical absorption due to free carriers, generated by two photon absorption. Here we explore a means to mitigate carrier effects via defect engineering. The optical and electrical properties of the defected silicon waveguides are modeled for both a uniform defect distribution and a remote, localized defect distribution. Simulation results indicate that a uniform defect distribution provides no improvement with any increase in net gain provided solely by surface modification. In contrast, for devices with remote defect volumes the reduction of carrier lifetime and limited optical absorption results in a significant improvement to net gain.     

Reduction of the bulk absorption coefficient in silicon optics for high-energy lasers through defect engineering

We engineered a factor-of-4 reduction in the bulk absorption coefficient over the 2.6-to-3.0-μm bandwidth in single-crystal Czochralski silicon optics for high-energy infrared lasers with high-temperature annealing treatments. Defect engineering adapted from the integrated circuit industry has been used to reduce the absorption coefficient across the 1.5-to-5-μm bandwidth for substrates up to 5 cm thick. A high-temperature oxygen-dispersion anneal dissolves precipitates and thermal donors that are present in the as-grown material. The process has been verified experimentally with Fourier transform infrared spectroscopy, infrared laser calorimetry, and Hall measurements. Reduction of the absorption coefficient results in less substrate heating and thermal distortion of the optical surface. The process is appropriate for other silicon infrared optics applications such as thermal-imaging systems, infrared windows, and spectrophotometers.     

Size-selecting effect of water on fluorescent silicon clusters

Silicon clusters were produced by gas aggregation in vacuum and co-deposited with water vapour onto a cold target where the water vapour froze. Melting of the ice yielded fluorescent silicon nanoparticles suspended in water which were investigated by photoluminescence spectroscopy (PL) and atomic force microscopy (AFM). The PL spectrum showed a prominent band at 420 nm and other, less intense bands at shorter wavelengths. No fluorescence was observed below 275 nm. The shortest wavelength observed was related to a silicon cluster diameter of 0.9 nm using a simple particle-in-a-box model. Drops of the suspension were also deposited on freshly cleaved HOPG and investigated by AFM. The images showed single and agglomerated clusters with heights of typically 0.6 up to 2 nm. The sizes displayed by our measurements are not correlated to the average sizes that result from gas aggregation, indicating a size-selecting effect of the water suspension. The cluster-cluster interaction in water is governed by repulsion due to thermal energy and attraction due to van der Waals forces. For very small clusters repulsion dominates; at 3 nm diameter the two forces are balanced. We identify this stable phase of small clusters as the origin of exceptionally stable fluorescence.     

Impurity engineering of Czochralski silicon

Microelectronic devices with high integration level and functional complexity are always requiring larger diameter and more perfect Czochralski (CZ) silicon wafers. Therefore, the defects, playing the key role in the quality control of silicon materials, have to be well controlled during crystal growth and device fabrication. Co-doping nitrogen (N), germanium (Ge) or carbon (C) into CZ silicon to control defect dynamics and to change defect evolution, so-called “impurity engineering”, has been developed in recent years, and has been widely applied in the fabrication of higher quality CZ silicon used for microelectronics nowadays. This article is to present an overview of the current status of impurity engineering in CZ silicon, based on the co-doping technologies of N, Ge and C. The fundamental properties of these three co-dopants and their interaction with point defects in CZ silicon are firstly introduced. The bulk of the article is focused on the effects of co-dopants on the formation of oxygen precipitates related to internal gettering (IG) of devices for metal contaminants, and voids associated with the gate oxide integrity (GOI) of devices in CZ silicon. Finally, the improvement of CZ silicon mechanical strength by co-doping technology is described.     

Trade-off strategies in engineering design

A formal method to allow designers to explicitly make trade-off decisions is presented. The methodology can be used when an engineer wishes to rate the design by the weakest aspect, or by cooperatively considering the overall performance, or a combination of thesestrategies. The design problem is formulated with preference rankings, similar to a utility theory or fuzzy sets approach. This approach separates the designtrade-off strategy from the performance expressions. The details of the mathematical formulation are presented and discussed, along with two design examples: one from the preliminary design domain, and one from the parameter design domain.     

Semi-empirical modelling of the di-interstitial defect in silicon

Previous infrared spectroscopy studies of the defect spectrum of neutron irradiated Czochralski grown silicon (Cz-Si) revealed a band at 533 cm^?1, which disappears from the spectra at ~170 °C and exhibits a similar thermal stability with the Si-P6 Electron paramagnetic resonance (EPR) spectrum correlated with the di-interstitial defect. The proposed structural model for this defect consists of two self-interstitial atoms located symmetrically around a lattice site Si atom. The calculations reveal that the previously suggested structure of the Si-P6 defect has a vibrational frequency at about 513 cm^?1, which is close to the experimental value of 533 cm^?1. The modeling results indicate that the 533 cm^?1 infrared band originates from the same structure as that of the Si-P6 EPR spectrum.     

Molecular dynamics simulations of impact of energetic silicon clusters onto crystalline silicon

Cluster deposition is a promising technique for the growth of nanometric structures and is therefore the objective of many theoretical and experimental investigations. In this work molecular dynamics calculations have been performed to study the deposition of silicon clusters onto a crystalline silicon substrate. The evolution of the structure of the deposited cluster has been studied as a function dependence of the initial direction of incidence of the cluster, of the orientation of the exposed surface, of the cluster shape and of the interatomic forces. The comparison of deposition onto (100) and (110) shows the presence of channelling effects which lead to the implant of the cluster atoms and limit their spreading on the substrate surface. Compact clusters, obtained from the minimization of the quantum mechanical energy, show an enhanced resilience against fragmentation in comparison with clusters with a crystalline lattice. A similar resistance is obtained when using an interatomic potential accounting for a coordination larger than the one of crystalline silicon. At variance with the known behavior of metallic clusters, the effects of the incidence direction are hardly perceptible.     

Shallow defect states in hydrogenated amorphous silicon oxide

The theoretical cluster-Bethe-lattice method is used in this study to investigate the shallow defect states in hydrogenated amorphous silicon oxide. The electronic density of states (DOS) for the SiO₂ Bethe lattice of various Si–O–Si angles, non-bridging oxygen Si–O^√, peroxyl radical Si–O–O^√, threefold coordinated O₃ and Si–H bonds are calculated. The variation of the Si–O–Si bond angle causes the bandgap fluctuation and induces tail states near the conduction band minimum. The Si–O^√ and Si–O–O^√ bonds introduce shallow defect states in the energy gap near the top of the valence band. The Si–H bond induces a defect state, in the energy gap near the conduction band minimum, in a-SiO_x with high oxygen concentration, but not low oxygen concentration. The O₃ bond itself does not induce defect state in the energy gap. The O₃⁺D⁻ complex, formed by the O₃ and threefold coordinated silicon, induces shallow state in the energy gap near the conduction band minimum. This defect state can explain the energy shift of photoluminescence of a-SiO_x:H under annealing.     

Application of genetic algorithms to hydrogenated silicon clusters

We discuss the application of biologically inspired genetic algorithms to determine the ground state structures of a number of Si-H clusters. The total energy of a given configuration of a cluster has been obtained by using a non-orthogonal tight-binding model and the energy minimization has been carried out by using genetic algorithms and their recent variant differential evolution. Our results for ground state structures and cohesive energies for Si-H clusters are in good agreement with the earlier work conducted using the simulated annealing technique. We find that the results obtained by genetic algorithms turn out to be comparable and often better than the results obtained by the simulated annealing technique.     

Infrared signals correlated with self-interstitial clusters in neutron-irradiated silicon

Using infrared spectroscopy we have investigated the defect spectrum of neutron-irradiated Czochralski–silicon (Cz–Si). The study was focused on three weak signals, mainly on a band at 533 cm^?1, as well as on two other bands at 582 and 592 cm^?1. The band at 533 cm^?1 disappears from the spectra at ~170 °C exhibiting similar thermal stability with the Si-P₆ electron paramagnetic resonance spectrum, previously correlated with a di-interstitial defect. The suggested model for the latter defect, comprising two self-interstitials placed symmetrically a lattice site Si atom, is very similar with that of the allene molecule. This allowed the calculation of the vibrational frequency of the suggested di-interstitial structure giving a value close to the 533 cm^?1, in further support of the above assignment. The band at 582 cm^?1 is stable up to 550 °C. The possible correlation of its origin to large self-interstitial clusters is examined. Also, the origin of the 592 cm^?1 band, which is stable up to 200 °C is discussed, with indications tentatively pointing to a CV pair.     

Defect engineering of Czochralski single-crystal silicon

Modern microelectronic device manufacture requires single-crystal silicon substrates of unprecedented uniformity and purity. As the device feature lengths shrink into the realm of the nanoscale, it is becoming unlikely that the traditional technique of empirical process design and optimization in both crystal growth and wafer processing will suffice for meeting the dynamically evolving specifications. These circumstances are creating more demand for a detailed understanding of the physical mechanisms that dictate the evolution of crystalline silicon microstructure and associated electronic properties. This article describes modeling efforts based on the dynamics of native point defects in silicon during crystal growth, which are aimed at developing comprehensive and robust tools for predicting microdefect distribution as a function of operating conditions. These tools are not developed independently of experimental characterization but rather are designed to take advantage of the very detailed information database available for silicon generated by decades of industrial attention. The bulk of the article is focused on two specific microdefect structures observed in Czochralski crystalline silicon, the oxidation-induced stacking fault ring (OSF-ring) and octahedral voids; the latter is a current limitation on the quality of commercial CZ silicon crystals and the subject of intense research.     

Electron microscopy of defect clusters produced by radiation damage

The methods of characterization of different types of radiation-induced defect clusters bytem have been reviewed. Point defects produced in irradiated materials agglomerate in two or three dimensional clusters to reduce the strain energy associated with them. Two-dimensional clusters assume the configuration of vacancy or interstitial type dislocation loops which can be resolved if the size of the loops is large compared to the extinction distance associated with the imaging reflection. The small loops give rise to a black dot contrast under the kinematical and a black-white contrast under the dynamical imaging conditions. The method of characterization of dislocation loops which include the determination of the nature of the loop, the Burgers vector and the loop plane normal is discussed taking examples from the work done on the ion irradiated Ni₄Mo samples. A summary of available experimental results on the characterization of dislocation loops in different metals and alloys having fcc, bcc and hcp structures is presented. The contrast from stacking fault tetrahedra which form in some fcc metals and alloys after a certain degree of annealing is also discussed. The optimum conditions for imaging three-dimensional clusters or voids are derived on the basis of the contrast theory proposed for such defects. Special reference is made to the usefulness of “through focus analysis” in the imaging of very small cavities (with diameters as small as about 10 Å). It is shown that the formation of disordered zones resulting from displacement cascades in the ordered matrix can be utilized in determining the shape and the volume of cascades in the virgin state. The importance of different contributing factors like the strain contrast and the structure factor contrast in producing the overall contrast from the disordered zones is discussed. Detailed observations on the shape of the disordered zones are shown to be important to establish the occurrence of the replacement collision sequence and the formation of sub-cascades.     

Action strategies for strengthening industrial clusters in southern Taiwan

For the past decade, the Taiwanese government has applied the policy of “North heavy, South light” to put more emphasis on development in northern Taiwan instead of the south. This has resulted in uneven development between the northern and southern regions, especially when introducing, developing, and supporting high technology and resource allocation in education. Industry clusters have become the centerpiece of economic development policy in many parts of the world. In this study, it is assumed that there are two categories of less-advantaged regions in Taiwan: (1) older industrialized regions dominated by labor-intensive/capital-intensive industries, and (2) industrial regions that have merged with potential high-technology small firms but still lack infrastructure. This study discusses the following points: (a) How do industrial clusters work in action? (b) What barriers do less-advantaged industrial clusters face? (c) What action strategies promote less-advantaged industrial clusters?     

第I类硅笼中M@Si_(20)原子团的电子态研究

采用量子化学SCF-Xα-SW方法计算了第I类Si基笼状化合物中M@Si20(M=□,Na,Zr,Cs,Ba,Ce)原子团的态密度,及各体系中Si原子的结合势。结果表明在原子团M@Si20中笼内金属原子M与笼上硅原子间的结合强度随金属原子不同而变化,结合方式依赖于金属原子M的电子结构。笼内原子的d轨道主要分布在完整晶体的导带底,此外另有少部分与价带结合。作为新型硅笼材料设计,探讨了构成笼内含Ce、Zr的硅笼材料的可能性。     

Complexing of platinum in float zone and Czochralski silicon - electron paramagnetic resonance study of 6-platinum clusters and oxygen related 1-platinum midgap defect

Abstract

Silicon supersaturated with platinum has preferred metastable trigonal or nearly trigonal arrangements of six platinum atoms after annealing at 540°C. The structure is strongly affected by oxygen or oxygen related intrinsic defects. Co-doping with iron prevents the formation of platinum clusters, although in the temperature range of the clustering most of the interstitial iron is expected to have been precipitated. An oxygen related 1-Pt defect was detected by electron paramagnetic resonance (EPR). Its identification with the dominating recombination centre in platinum doped silicon is suggested by photo-EPR experiments and the preliminary hypothesis of an excited spin triplet state.     

Regular relief on a silicon surface as a structural defect getter

An investigation was made to determine how a regular relief on the silicon surface influences gettering in silicon-silicon-dioxide structures. The regular relief was created by a photolithographic technique before oxidation and comprised an orthogonal network of overlapping bands. The gettering was determined from the isothermal relaxation of the capacitance of a silicon-silicon-dioxide structure after switching from strong inversion to even stronger inversion. It is shown that a regular relief at the silicon-silicon-dioxide interface is an effective getter at a depth of several hundred micron. Pis’ma Zh. Tekh. Fiz. 25, 75–80 (January 12, 1999)     

Ground state structures and properties of small hydrogenated silicon clusters

We present results for ground state structures and properties of small hydrogenated silicon clusters using the Car-Parrinello molecular dynamics with simulated annealing. We discuss the nature of bonding of hydrogen in these clusters. We find that hydrogen can form a bridge like Si-H-Si bond connecting two silicon atoms. We find that in the case of a compact and closed silicon cluster hydrogen bonds to the silicon cluster from outside. To understand the structural evolutions and properties of silicon cluster due to hydrogenation, we have studied the cohesive energy and first excited electronic level gap of clusters as a function of hydrogenation. We find that first excited electronic level gap of Sin and SinH fluctuates as function of size and this may provide a first principle basis for the short-range potential fluctuations in hydrogenated amorphous silicon. The stability of hydrogenated silicon clusters is also discussed.