Recrystallization Of Silicon Research Articles

Purification of Metallurgical Silicon up to ‘‘Solar’’ Mark Silicon

The paper presents experimental results showing the possibility of obtaining the "solar" mark silicon by recrystallization of metallurgical silicon in low-melting metals on the example of tin and the cultivation of the obtained flakes of single crystal silicon by the Czochralski method. The experiments on the purification of low-melting metal (tin) after the completion of the cycle of the silicon flakes are conducted in order to re-use it. Qualitative and quantitative analysis of raw materials of silicon and tin and the composition after different processing stages were studied by XRF analysis on the instrument Elvax light. Structural features of the obtained silicon flakes were examined using scanning electron microscopy on the device РЕММА106И. The measurements of the conductivity type and electrical resistivity of the obtained monocrystalline ingot of silicon are carried out by a four-probe technique.

Hardening silicon-on-insulator nMOSFETs by multiple-step Si+ implantation

A novel technique is proposed to improve total irradiation dose (TID) hardness of buried oxides in a 0.13μm silicon-on-insulator (SOI) technology. Multiple-step Si ion implantation is implemented to avoid silicon film amorphization. Each implant step introduces silicon ion implantation of a lower dose into buried oxides which creates an amorphous/crystalline (a/c) interface inside the silicon layer. Rapid thermal annealing (RTA) removes implant-induced lattice damages by silicon recrystallization reflected in a/c interface moving towards the top silicon surface. The thermal process prevents top silicon layers from total amorphization arising in the technique of single high dose implantation method. X-ray Diffraction (XRD) spectrum confirms the existence of the a/c interface and determines the single implant dose. Experimental results on pseudo-MOS and H-gate partially-depleted SOI n-type MOSFETs show radiation tolerance up to 1.0Mrad(Si) though introduced metastable electron traps lead to I–V hysteresis and bias instabilities.

Amorphization and nanocrystallization of silicon under shock compression

High-power, short-duration, laser-driven, shock compression and recovery experiments on [001] silicon unveiled remarkable structural changes above a pressure threshold. Two distinct amorphous regions were identified: (a) a bulk amorphous layer close to the surface and (b) amorphous bands initially aligned with {111} slip planes. Further increase of the laser energy leads to the re-crystallization of amorphous silicon into nanocrystals with high concentration of nano-twins. This amorphization is produced by the combined effect of high magnitude hydrostatic and shear stresses under dynamic shock compression. Shock-induced defects play a very important role in the onset of amorphization. Calculations of the free energy changes with pressure and shear, using the Patel-Cohen methodology, are in agreement with the experimental results. Molecular dynamics simulation corroborates the amorphization, showing that it is initiated by the nucleation and propagation of partial dislocations. The nucleation of amorphization is analyzed qualitatively by classical nucleation theory.

Atomistic modeling of epitaxial growth of semiconductor materials

The epitaxial growth of semiconductors is a very important step for the processing of microelectronic devices. Solid Phase Epitaxial Regrowth (SPER) is involved in both the recrystallization of amorphous silicon and germanium areas. Selective epitaxial growth (SEG) is routinely used in manufacturing modern 3D devices like finFETs, raised source/drain transistors, gate-all-around transistors, etc. In this work, lattice kinetic Monte Carlo modeling is used as the main tool for the modeling and simulation of epitaxial growth in IV and III–V based materials. This is done by detecting particular local configurations at the interfaces, and assigning local growth rates depending on such configurations. This simple idea is used to identify recrystallization planes or to compute different formation energies for the trapping, detrapping and migration of atoms. Introduction of defects is also possible by allowing the formation of twins and small defective areas. We will show different cases where our model is used, namely: SPER and defect formation of Si, and we will explain how the models can be extended to simulate the epitaxial growth of III–V materials and to obtain the different facet formation and growth kinetics. All these cases are compared with relevant experimental results from the literature to provide an assessment of their current capabilities but also possible limitations.

DEVELOPMENT OF CLEARING TECHNIQUE OF METALLURGICAL SILICON TO SOLAR GRADE SILICON

Experimental results demonstrating the possibility of obtaining solar grade silicon by recrystallization of metallurgical silicon in fusible metals, e.g. tin, and pulling of single crystal silicon ingot obtained from silicon scales by the Czochralski method have been presented. Experiments for the purification of a fusible metal (tin) after the end of a cycle of silicon scales obtaining for the purpose of its reuse have been carried out. We purified tin by vacuum decontamination of tin melt, its filtration and finally zone recrystallization. Qualitative and quantitative analysis of the initial materials (silicon and tin) and their structure after sequential stages of the technological process has been carried out by X−ray fluorescent analysis on an Elvax light device. The structural features of the silicon scales have been examined using scanning electron microscopy on a REMMA106I device. The conductivity type and the electrical resistivity of the obtained single crystal silicon ingot have been measured using the four−probe method on a PIUS−1UM−K device. We show that the composition of the pulled single crystal ingot is not worse than 99.999 wt.% Si, it has the n type of conductivity and its electrical resistivity is not less than 2.0 О hm• с m. These parameters meet the requirements to solar grade silicon and confirm the possibility of its obtaining from metallurgical silicon by recrystallization in fusible metals, e.g. tin.

Thermally induced recrystallization of textured hydrogenated nanocrystalline silicon

By an analysis of the local crystallinity based on model potential molecular dynamics simulations we investigated the effect of dissolved hydrogen on the thermally induced recrystallization of nanocrystalline silicon. By using the Kolmogorov-Johnson-Mehl-Avrami theory to analyze the atomistic data, we find that the recrystallization rate decreases exponentially with the hydrogen contamination. At low concentration, the kinetics is moderately affected by the H atoms that tend to migrate to the boundaries increasing their effective interface. At higher H content, we find an increasing number of Si${}_{m}$H${}_{n}$ hydrides that affect the crystalline order of the material and severely impede recrystallization. The analysis based on crystallinity is supported by the atomic scale study of the recrystallization mechanism, here identified as an inverted bond-switching process, and by the ability of hydrates to pin the amorphous-crystalline boundaries.

3D Modelling of Temperature Dependent Power Distribution During Microwave Annealing of Doped c-Si

Recrystallized silicon solar cells using microwave annealing will require adequate modelling of microwave power distribution to predict the uniformity of re-crystallization of silicon, specially at high temperatures. Since, recrystallization will have to be carried out of doped silicon, doping dependence of such processes at high temperatures is also very important. This work reports modelling results for temperature and doping dependence of microwave spatial power distribution in c-Si wafer during microwave annealing. Temperature dependence was introduced through bandgap, effective mass and lattice scattering terms in Drude model of free-carrier concentration. Dependence of doping was introduced through free-carrier dependent complex dielectric function. Absorption of power was then calculated from the complex part of the dielectric constant. The 3D Modelling was carried out using Exact Finite Difference Method. All results presented in this paper are for Industrial, Scientific and Medical (ISM) frequency of 2.45 GHz.

Synthesis of Hollow Nanocubes and Macroporous Monoliths of Silicalite-1 by Alkaline Treatment

A simple strategy involving desilication and recrystallization of silicalite-1 in tetrapropylammonium hydroxide (TPAOH) solution was successfully developed to prepare hollow zeolite nanocubes and three-dimensionally macroporous zeolite monoliths. Large voids were introduced to silicalite-1 crystals by controlled silicon leaching with OH– and thin intact shells were formed by the recrystallization of silicon with TPA+. The size of nanocubes could be easily controlled from ∼150 nm to ∼600 nm by simply adjusting the size of parent silicalite-1. Apart from template function to increase the yield of hollow silicalite-1, TPA+ adsorbed on the zeolite protects the parent crystal surface where the recrystallization occurred. The size of the mesopores and/or macropores in the hollow zeolite shell can be controlled by varying the amount of competitive Na+ adsorbent added to the TPAOH solution. Furthermore, three-dimensional macroporous zeolite monoliths can be formed when an electrolyte, such as NaCl, was added to t...

Boron removal by dissolution and recrystallization of silicon in a sodium–silicon solution

Low-purity Si powder was dissolved in Na solvent, and Si recrystallized by the evaporation of Na from the Na–Si solution at 1173K in a magnesium oxide crucible. Several impurity elements precipitated as compounds at the bottom of the crucible. Glow-discharge mass spectrometry showed a decrease in the concentration of most impurity elements, except Na and Mg, in the crystallized Si. The boron (B) concentration was reduced from 9.1 to 0.3mass ppm by recrystallization. In addition, solidification segregation was confirmed by the depth profile of the B concentration from the crystal surface.

Pulsed nanosecond annealing of magnesium-implanted silicon

Single-crystalline silicon is implanted by magnesium ions at room temperature and then subjected to pulsed ion-beam annealing. The surface morphology, crystallinity, and optical properties of the implanted silicon are studied before and after annealing. It is shown that ion implantation makes a near-surface layer of silicon about 0.1 m thick amorphous. Pulsed nanosecond ion-beam annealing results in silicon recrystallization and the formation of crystalline magnesium silicide precipitates. Optimal values of the implantation dose and pulse energy density for the formation of magnesium silicide precipitates in the near-surface layer of silicon are found.

Thermodynamics of Phosphorus Removal from Silicon in Solvent Refining of Silicon

AbstractRefining of silicon for solar applications using metallurgical approaches has attracted a considerable attention in the recent years. The present study involves employing solvent refining as a purification technique in which silicon recrystallization takes place from an iron-silicon alloy melt. It is believed that iron will perform as the impurity “getter” and purified silicon crystals grow from the alloy melt, while the impurities are segregated to the liquid alloy. The focus of this article is on removal of phosphorus, as a critical impurity in solar silicon. In order to assess the feasibility and efficiency of phosphorus removal through solvent refining, the distribution of phosphorus between solid Si and Fe-Si melt at 1483–1583 K (1210–1310 °C) was measured. Interaction parameter between phosphorus and iron was calculated by varying the concentration of phosphorus at each temperature.

Structural evolution of implanted vicinal Si(111) during annealing via analysis of the dipole contribution

The changes in the amplitude and orientation of the dipole at the surface layer of phosphorous (P) implanted vicinal Si(111) show the evolution of its restructuring. The different electronegativities between Si and participated P atoms changed the dipolar configuration of Si surface. The dipolar configuration of implanted vicinal Si(111) would be variant for the different annealing conditions at which silicon recrystallization and P activation occur. Reflective second harmonic generation (RSHG) is a sensitive technique for studying the symmetrical dipole structure on the surface. We offer a model to distinguish dipolar configurations of the top surface from the one of the implanted bulk Si(111) and explain their relative phase variation in the results of RSHG experiment.

Susceptor assisted microwave annealing for recrystallization and dopant activation of arsenic-implanted silicon

The increasing need for quicker and more efficient processing techniques motivates the study of the use of a single frequency applicator microwave cavity, along with an alumina-coated SiC susceptor, as an alternative to current post-implantation processing. The extent of Si recrystallization and repair of the damage caused by arsenic implantation into Si is determined by cross-section transmission electron microscopy and Raman spectroscopy. Dopant activation is evaluated by sheet resistance measurements. Secondary ion mass spectroscopy is used to compare the extent of diffusion that results from such microwave annealing with that experienced when using conventional rapid thermal annealing (RTA). The results show that, compared to susceptor-assisted microwave annealing, RTA caused undesired dopant diffusion. The SiC-alumina susceptor plays a significant role in supplying heat to the Si substrate and also acts as an assistor that helps a high-Z dopant, like arsenic, to absorb the microwave energy, using a microwave loss mechanism, which is a combination of ionic and dipole losses.

Vacancy Generation by Laser Preirradiation for Junction Leakage Suppression

Vacancy generation by laser preirradiation on silicon substrate before implantation for advanced junction engineering was demonstrated. Amorphized p+/n junction diodes subjected to preimplant laser irradiation show a twofold reduction on the off-state leakage current and a two-time improvement on the on-state current compared to control devices without any preirradiation. The defect-removal mechanism is achieved by the recombination of excess vacancies trapped at the maximum laser melt depth as a result of the molten silicon recrystallization with the implantation-generated interstitials. The effectiveness of laser-generated vacancies in annihilating residual defects is evident in the suppression of junction leakage current.

Observation of explosive crystallization during excimer laser annealing using in situ time-resolved optical reflection and transmission measurements

XeF excimer laser-induced melting and recrystallization of amorphous silicon was investigated using in situ on-line time-resolved reflection and transmission measurements with a nanosecond time resolution. Explosive crystallization was observed for 900 Å thick amorphous silicon on SiO 2 deposited on non-alkali glass substrate upon 25 ns pulse duration of excimer laser by continuous-wave He-Ne probe laser. Three distinct regrowth regimes were investigated on the basis of various excimer laser fluences. Scanning electron microscopy, Raman spectroscopy and atomic force microscopy were used to evaluate the excimer laser-irradiated region of the sample. Grain size, surface roughness and melt duration as a function of various laser fluences are also determined.

A Very Low Dark Current Temperature-Resistant, Wide Dynamic Range, Complementary Metal Oxide Semiconductor Image Sensor

A very low dark current (VLDC) temperature-resistant approach which best suits a wide dynamic range (WDR) complementary metal oxide semiconductor (CMOS) image sensor with a lateral over-flow integration capacitor (LOFIC) has been developed. By implementing a low electric field photodiode without a trade-off of full well-capacity, reduced plasma damage, re-crystallization, and termination of silicon–silicon dioxide interface states in the front end of line and back end of line (FEOL and BEOL) in a 0.18 µm, two polycrystalline silicon, three metal (2P3M) process, the dark current is reduced to 11 e-/s/pixel (0.35 e-/s/µm2: pixel area normalized) at 60 °C, which is the lowest value ever reported. For further robustness at low and high temperatures, 1/3-in., 5.6-µm pitch, 800×600 pixel sensor chips with low noise readout circuits designed for a signal and noise hold circuit and a programmable gain amplifier (PGA) have also been deposited with an inorganic cap layer on a micro-lens and covered with a metal hermetically sealed package assembly. Image sensing performance results in 2.4 e-rms temporal noise and 100 dB dynamic range (DR) with 237 ke- full well-capacity. The operating temperature range is extended from -40 to 85 °C while retaining good image quality.

Thermal annealing of porous silicon to develop a quasi monocrystalline structure

Low porosity (~20–30%) porous silicon produced by electrochemical anodization was annealed in the temperature range 1000–1150 °C under pure hydrogen atmosphere and under hydrogen mixed with nitrogen in different proportions for a duration of 15–60 min. Porous silicon was transformed to quasi monocrystalline porous silicon (QMPS) in the temperature range 1050–1100 °C under pure hydrogen atmosphere. The crystallinity was confirmed by grazing incidence X-ray diffraction. Field emission scanning electron microscopic (FESEM) studies revealed that the surface layer is pore free with a few voids embedded inside the body. Atomic force microscopic (AFM) studies confirmed relatively smooth and uniform surfaces under the same annealing conditions. Our experimental results concluded that the recrystallization of porous silicon at 1100 °C and in presence of pure hydrogen exhibits lower reflection loss compared to bulk crystalline silicon. Also the electrical resistivity of QMPS makes it suitable for optoelectronic devices and solar cells.

Atomistic Simulation Techniques in Front-End Processing

ABSTRACTAtomistic process models are beginning to play an important role as direct simulation approaches for front-end processes and materials, and also as a pathway to improve continuum modeling. Detailed insight into the underlying physics usingab-initiomethods and classical molecular dynamics simulations will be needed for understanding the kinetics of reduced thermal budget processes and the role of impurities. However, the limited sizes and time scales accessible for detailed atomistic techniques usually lead to the difficult task of relating the information obtained from simulations to experimental data. The solution consists of the use of a hierarchical simulation scheme: more fundamental techniques are employed to extract parameters and models that are then feed into less detailed simulators which allow direct comparison with experiments. This scheme will be illustrated with the atomistic modeling of the ion-beam induced amorphization and recrystallization of silicon. The model is based on the bond defect or IV pair, which is used as the building block of the amorphous phase. It is shown that the recombination of this defect depends on the surrounding bond defects, which accounts for the cooperative nature of the amorphization and recrystallization processes. The implementation of this model in a kinetic Monte Carlo code allows extracting data directly comparable with experiments.

Polysilicon thin film transistors using sputtered HfO2 gate dielectric and SiGe source/drain

We have fabricated polysilicon thin film transistors (TFTs) using sputtered HfO2 gate dielectric and SiGe as source/drain. The polysilicon film is formed by low temperature furnace recrystallization of amorphous silicon. The effective dielectric constant of the HfO2 in the TFTs is 17.7, and the transconductance of the TFTs is about 4.5 times that of those using SiO2 as gate dielectric. The use of SiGe as source/drain requires a lower thermal budget than the conventional technique of impurity implantation in the polysilicon film.

Enlargement of Crystal Grains in Thin Silicon Films by Continuous-Wave Laser Irradiation

The cw laser recrystallization of amorphous silicon (a-Si) thin films deposited on SiO2 films with Si(100) substrates was investigated by controlling irradiation conditions. The effects of the laser spot shape and laser-irradiated region overlap were found important resulting in two dimensional crystal growth different from conventional lateral crystallization. A laser beam spot was designed to realize a gradual temperature slope in the laser-irradiated region. The crystal grain growth of the Si thin films was effectively enhanced. Raster scanning was performed to make the crystal growth direction regular and effectively enhanced the crystallization of the recrystallized Si thin films. The laser-irradiated region overlap also effectively enhanced the crystallization of the recrystallized Si thin films. Consequently, (100) well-oriented polycrystalline Si thin films with an average grain size of 1.60 µm, which is about 5 times larger than that obtained by conventional excimer laser recrystallization, were obtained.