Silicon Contamination Research Articles

VUV light induced surface interaction and accelerated diffusion of carbon, silicon, oxygen and other contaminants in LiF crystals

One hundred and fifty-seven nanometers irradiation of LiF crystals accelerates the penetration of C, O, N, Si and other contaminants within the crystal bulk. New bonding between the contaminant elements in both the irradiated and the non-irradiated areas of the crystal was identified. The chemical changes eventually form inhomogeneous nano/micro-islands, which further agglomerate to larger structures in the form of an interfacial layer. The concentration and diffusion (∼100 nm) of contaminants in the irradiated part of the crystal is higher than for the non-irradiated one. The accelerated diffusion is due to a force perpendicular to the crystal surface. This force is generated by the laser's electric field gradient in the direction of beam propagation due to the inhomogeneous absorption at the interfacial layer. A theoretical diffusion model in the presence of the laser field allows rationalizing and interpreting the experimental results.

Investigation of silicon contamination of Ta filaments used for thin film silicon deposition

After extensive utilisation of tantalum (Ta) catalyst filaments for hot wire chemical vapor deposition (HWCVD) of thin silicon films a strong degradation takes place. A high concentration of silicon was found not only on the surface but also in the bulk of the tantalum filament. Visual microscopic investigations, Secondary Ion Mass Spectrometry (SIMS), X-ray Diffraction (XRD) and Energy Dispersive X-ray Analysis (EDX) indicate appearance of various silicides and formation of thick silicon layer (> 50 μm) on the filament surface. The high-power backscattered scanning electron microscopy (SEM BSE) and optical microscopic analysis of the filament cross section reveal a complicated, non-uniform structure of filaments after use. By XRD a recrystallisation of tantalum kernel was detected. The EDX analysis indicates that silicides on the filament surface have the highest concentration of Si.

Influence of contamination on bonding to zirconia ceramic

The purpose of this study was to investigate the influences of contaminations and cleaning methods on bonding to dental zirconia ceramic. After saliva immersion and using silicone disclosing agent, airborne-particle abraded ceramic specimens were cleaned with isopropanol (AL), acetone (AC), 37% phosphoric acid (PA), additional airborne-particle abrasion (AA), or only with water rinsing (SS). Airborne-particle abraded specimens without contaminations (CL) were used as control group. For chemical analysis specimens of all groups were examined with X-ray photoelectron spectroscopy (XPS). Plexiglas tubes filled with composite resin were bonded to ceramic specimens using a phosphate-monomer containing composite luting resin. After 3-day water storage, tensile bond strengths (TBS) were tested. XPS analysis of group SS showed the presence of saliva and silicone (Si) contamination on the surface. The ratios of carbon/zirconium and oxygen/zirconium for groups PA and AA were comparable to those ratios obtained for group CL, indicating the removal of the organic saliva contamination. Airborne-particle abrasion and acetone completely removed Si contamination from ceramic surfaces. Isopropanol had little cleaning effect on the two contaminants. TBS (median +/- standard deviation) in MPa of the groups SS (11.6 +/- 3.1), AL (10.0 +/- 2.9), and AC (13.0 +/- 2.8) were statistically lower than those of groups PA (33.6 +/- 5.5), AA (40.1 +/- 3.6), and CL (47.0 +/- 8.1) (p < 0.001), while no differences were found in TBS between groups AA and CL (p > 0.5). Contamination significantly reduced bond strengths to zirconia ceramic. Airborne-particle abrasion was the most effective cleaning method.

Capillary Force Lithography: Fabrication of Functional Polymer Templates as Versatile Tools for Nanolithography

AbstractThe implementation of high‐resolution polymer templates fabricated by capillary force lithography (CFL) is explored both in nanoimprint lithography (NIL) and in the wet‐etching of metals. Several different thermoplastic and UV‐curable polymers and types of substrates are incorporated into the general CFL procedure to meet the diverging requirements of these two applications. The mechanical stability of UV‐curable templates for imprinting in polymers, as examined by atomic force microscopy (AFM), and their anti‐adhesive properties are excellent for application in NIL. The conditions for curing the UV‐curable polymer are optimized in order to obtain high‐stability polymer templates. Gold patterns on silicon with a lateral resolution of 150 nm are fabricated by subsequent lift‐off in acetone. Similar patterns with a lateral resolution of 100 nm are fabricated using templates of thermoplastic polymers on gold layers on silicon as an etch mask. The transfer of stamp residues during CFL with these polymer templates is proven by X‐ray photoelectron spectroscopy (XPS) and AFM friction analysis. For poly(methylmethacrylate) (PMMA), the presence of large amounts of silicon‐containing residues is found to compromise the processability of the resulting template in subsequent O2 reactive‐ion etching (RIE) treatment. The extent of silicon contamination is up to six times less for polystyrene (PS). At this level, the etch performance of the PS etch mask is not affected, as was the case for PMMA. Accurate downscaling of the lateral dimensions of the resulting metal patterns by several factors with respect to the dimensions of the PS etch mask is achieved by over‐etching of the gold. Overall, the results in this paper demonstrate the potential of CFL templates as tools for high‐resolution soft lithography.

Role of Silicon Contamination on Calcification of Hydrophilic Acrylic Intraocular Lenses

To verify the presence of the element silicon on hydrophilic acrylic intraocular lenses (IOLs) explanted because of calcification. Interventional case series with clinicopathological correlation. Twenty explanted IOLs with surface deposits (MemoryLens), and 20 with deposits mostly within their optic substance (SC60B-OUV and Aqua-Sense; 10 each) were used. After gross, microscopic, and histochemical analyses to confirm the presence of deposits, the lenses underwent scanning electron microscopy (SEM) with energy dispersive x-ray spectroscopy (EDS) for elemental composition, on the external surface of MemoryLens IOLs, and on the surface and internal substance of SC60B-OUV and Aqua-Sense IOLs. The weight percentage of the element silicon was obtained at the level of deposits, and at adjacent deposit-free areas in all lenses. Scanning electron microscopy (SEM) coupled with EDS confirmed that the composition of the deposits was calcium/phosphate in all cases. The element silicon was found in all 40 lenses, on all areas analyzed. The silicon weight percentage was higher at the level of the deposits. The presence of aluminum on five MemoryLens IOLs, and in most of the SC60B-OUV and Aqua-Sense lenses might be related to scattering from the aluminum mounting stubs used for surface analyses. Silicone compounds have been implicated in the calcification of another hydrophilic acrylic design (Hydroview). They may also have a role in the calcification of other hydrophilic acrylic IOLs. Further investigation on the relationship between the presence of the element silicon and the silicone compounds found on calcified hydrophilic acrylic lenses is necessary.

Cobalt Contamination in Silicon

The properties of cobalt as a contaminant in p-type silicon are studied by using cobaltimplanted wafers annealed by RTP or by RTP plus a low temperature furnace annealing. It is shown that after RTP most cobalt is under the form of CoB pairs. A quantification of cobalt contamination is provided based upon SPV measurements and optical pair dissociation. However, this quantification fails in furnace-annealed wafers because of the formation of a different level. It is shown that the CoB level is located near the band edges, whereas the level formed upon a low temperature furnace annealing is located near midgap. Besides, when the cobalt concentration is high enough a small fraction of cobalt is in a level different from the CoB pair even in RTP samples. This level can probably be identified with a previously observed midgap level. It is suggested that the same level is formed in RTP plus low temperature furnace annealed samples and in high concentration RTP annealed samples, and that this level may consist in some cobalt agglomerate.

Electronic properties of interstitial iron and iron-boron pairs determined by means of advanced lifetime spectroscopy

As dissolved iron is one of the most common lifetime-killing contaminants in silicon, its coexisting defect configurations, interstitial iron (Fei) and iron-boron pairs (FeB), are investigated on an intentionally iron-contaminated silicon sample by means of temperature-dependent lifetime spectroscopy (TDLS) and injection-dependent lifetime spectroscopy (IDLS). In good agreement with the literature, the study identifies the known Fei donor level at Et−EV=(0.394±0.005)eV and determines its symmetry factor as k=σn∕σp=51±5, which is an order of magnitude lower than expected from the literature. Using the well-confirmed k factor, the poorly confirmed electron-capture cross section is redetermined as σn=(3.6±0.4)×10−15cm2 at 300K. In addition, the observed exponential σ(T) dependence identifies the multiphonon emission mechanism as the dominant capture mechanism for electrons with an activation energy E∞=0.024eV. Concerning the defect related to the iron-boron pair, the study unambiguously identifies the deep FeB acceptor level as the dominant FeB recombination center. While the known average value EC−Et=(0.26±0.03)eV for the energy level is well confirmed, the symmetry factor is determined as k=0.45, which represents an intermediate value among the scattered results from the literature. Additional spectroscopic information from the defect transformation allows both capture cross sections of the FeB defect to be determined (σn=2.5×10−15cm2 and σp=5.5×10−15cm2). Being suggested as a fingerprint of iron in the literature, the crossover position of the Fei- and FeB-dominated IDLS curves (at 300K) is experimentally proved to be doping dependent which is accurately predicted on the basis of the extracted set of Fei and FeB defect parameters. An additional fingerprint of iron is found by the qualitative change of the TDLS curve upon illumination and its S-like shape under dark conditions, which represents a robust criterion for unambiguous identification of iron in silicon. As the most sensitive technique to detect and determine an iron contamination in silicon heavily relies on the Fei and FeB defect parameters, the spectroscopic results are of special practical importance.

Aging in the large CDF axial drift chamber

The Central Outer Tracker (COT) is a large axial drift chamber in the Collider Detector at Fermilab operating with a gas mixture that is 50/50 argon/ethane with an admixture of 1.7% isopropanol. In its first two years of operation the COT showed unexpected aging with the worst parts of the chamber experiencing a gain loss of ~50% for an accumulated charge of ~35 mC/cm. By monitoring the pulse height of hits on good tracks, it was possible to determine the gain as a function of time and location in the chamber. In addition, the currents of the high voltage supplies gave another monitor of chamber gain and its dependence on the charge deposition rate. The aging was worse on the exhaust end of the chamber consistent with polymer buildup as the gas flows through the chamber. The distribution in azimuth suggests that aging is enhanced at lower temperatures, but other factors such as gas flow patterns may be involved. Elemental and molecular analysis of the sense wires found a coating that is mostly carbon and hydrogen with a small amount of oxygen; no silicon or other contaminants were identified. High resolution electron microscope pictures of the wire surface show that the coating is smooth with small sub-micron nodules. In the course of working with the chamber gas system, we discovered a small amount of O 2 is enough to reverse the aging. Operating the chamber with ~100 ppm of O2 reversed almost two years of gain loss in less than 10 days while accumulating les2 mC/cm

Investigation of the tantalum catalyst during the hot wire chemical vapor deposition of thin silicon films

After extensive utilization of tantalum (Ta) catalyst filaments for hot wire (HW) CVD of silicon films we found a high amount of silicon not only at the surface but also in the bulk of the tantalum catalyst (up to 8%). The amount of silicon accumulation in the bulk depends on the concentration of silane in the process gas. SIMS investigation indicates a silicide/silicon formation on the surface and significant diffusion of silicon into the bulk of the catalyst. This silicon contamination changes the filament resistance and seems to be the reason for rather short lifetime of the filament, which is a serious concern for the HWCVD deposition technology. Annealing of the filament by direct current (DC) at high temperature in vacuum enables to drastically reduce the silicon content (silicides) in a surface region of about 2–3 μm. However, it cannot remove the thick silicon deposits (> 10 μm) from the filament surface. In order to decrease the silicon content in the tantalum catalyst and the silicon formation on the filament surface we used alternating current (AC) (13.5 MHz) to heat the filament. Due to the “skin” effect of the radio frequency (RF) current the silicon indiffusion and the formation of silicon deposits on the surface are suppressed.

Quantitative copper measurement in oxidized p-type silicon wafers using microwave photoconductivity decay

We propose a method to measure trace copper contamination in p-type silicon using the microwave photoconductivity decay (μ-PCD) technique. The method is based on the precipitation of interstitial copper, activated by high-intensity light, which results in enhanced minority carrier recombination activity. We show that there is a quantitative correlation between the enhanced recombination rate and the Cu concentration by comparing μ-PCD measurements with transient ion drift and total reflection x-ray fluorescence measurements. The results indicate that the method is capable of measuring Cu concentrations down to 1010cm−3. There are no limitations to wafer storage time if corona charge is used on the oxidized wafer surfaces as the charge prevents copper outdiffusion. We briefly discuss the role of oxide precipitates both in the copper precipitation and in the charge carrier recombination processes.

Chromium Contamination in Silicon: Detection and Impact on Oxide Performances

Chromium Contamination in Silicon: Detection and Impact on Oxide Performances

Surface modification of boron fibres for improved strength in composite materials

When boron fibres are combined with an organic matrix, such as an epoxy resin, a high-performance composite structure is created. This study investigates the surface chemistry of plasma- and organosilane-treated boron fibres with the key aim to improving the adhesion properties between the boron fibre and the epoxy matrix. Optimisation of this interfacial region plays a critical role in influencing the mechanical behaviour of composite materials and has considerable industrial applications in the aerospace and manufacturing industries. The surface chemistry of a model boron surface and boron fibres was monitored using a combination of X-ray photoelectron spectroscopy (XPS) and time-of-flight secondary ion mass spectrometry (TOF-SIMS). Initial investigation of the as-received fibres showed the presence of silicone contamination on the fibre surface, which would affect adhesion. Removal of this contaminant through solvent cleaning and plasma oxidation provided an ideal surface for attachment of the organosilane adhesion promoter. A model for the interaction of the organosilane with a boron surface is proposed. The pull-out strength of boron fibres, with different surface treatments, embedded in the epoxy resin was measured using a custom designed adhesiometer. Compared with as-received boron fibres, a 6-fold improvement in the apparent interfacial shear strength was achieved for the organosilane treated fibres. Optical microscopy was used to determine the failure mechanisms between the fibre and epoxy resin. Typically, as the surface treatment improved adhesion, the locus of failure changed from the boron–epoxy interface to failure within the epoxy and ultimately fibre breakage.

Characterisation of Plastic Foams used as Backer Rod in the Construction Industry

Six polyethylene (PE) and two polyurethane (PU) commercially available foams typically used in the construction industry were investigated in terms of their mechanical, morphological and surface chemistry properties. Both PU foams were open cell foams, while all PE foams had closed cells; this led to a significantly lower hardness and compressive strength, but to a higher water absorption in the open cell foams. It was found that a higher density leads to a smaller cell size, which resulted in lower water absorption. Optical microscopic pictures were taken and they revealed a broad distribution of cell size and cell shape; skin cells were generally smaller than bulk cells. A new concept, the “theoretical solid foam surface”, is introduced with which an estimation of the contact area between foam and sealant (or any other solid/highly viscous material) is possible. XPS and IR experiments showed that all PE foams had an oxidised surface, and several PE foams also contained oxygen in their bulk phase. The PU foams contained silicon at a high concentration; some of the PE foams also had silicon incorporated though to a lesser extent. Nitrogen was found in some PE foams. The silicon and nitrogen contamination was assumed to arise from additives such as release agents. These impurities probably led to the high surface free energies found on PE foam. In particular, the skins with larger amounts of contamination had higher surface free energies. The measurement of the surface free energy of foams with the VCA method proved to be difficult because a large distribution of distorted contact angles was measured, caused by the roughnesses of the foam surfaces. Recommendations are made for the practical application of the research in the context of construction.

High-efficiency p-i-n a-Si:H solar cells with low boron cross-contamination prepared in a large-area single-chamber PECVD reactor

In this work, a new type of short water vapor treatment of the interface between the p- and i-layer is presented. This novel treatment is performed under vacuum below 1 mbar for 5 min and considerably reduces the i-layer boron contamination in amorphous silicon (a-Si:H) p-i-n solar cells prepared in single-chamber reactors. A significant advantage is that the substrate with the p-layer can remain loaded in the reactor during this oxidation treatment. The high effectiveness of this treatment in reducing the boron cross-contamination is directly supported by secondary ion mass spectroscopy measurements, by tracing the boron concentration depth profile across the p–i interface and by quantum efficiency measurements of the deposited cells. By applying this water vapor treatment, 0.3-μm-thick a-Si:H p-i-n solar cells of 1 cm 2 with high initial conversion efficiencies of 10.1% are deposited in a commercial large-area (35×45 cm 2) single-chamber PECVD KAI™ reactor and can clearly compete with cells deposited in multi-chamber systems. Light soaking of these cells for 1200 h at 50 °C leads to stabilized efficiencies of 8.2%. The relative typical efficiency degradation of 20% of such 0.3-μm-thick single-junction cells demonstrates that this treatment does not affect the stability in a negative manner.

Back Side Damage Gettering of Cu Using a Cavitating Jet

The effectiveness of the back side damage introduced by a cavitating jet to getter contamination in silicon has been demonstrated. Back side damage required for gettering has been introduced successfully into a silicon wafer, utilizing impacts caused by the cavitating jet. A wafer, which had previously been back side damaged in a selected area, was intentionally contaminated and submitted to a thermal treatment. The effectiveness of gettering was confirmed by the observation of etch pits on the surface, induced by the intentional contamination, other than in regions below which there was back side damage. No etch pits were observed in regions with back side damage.

Silicon Contamination Adsorbed on Pure Titanium Plate during Soaking Test in Hanks' Balanced Saline Solution

Silicon contamination adsorbed on pure Titanium plate during soaking test in Hanks’ balanced saline solution (HBSS) and distilled water was investigated. Amounts of silicon eluted from various materials of vessels were determined by inductively coupled plasma atomic emission spectrometry (ICP-AES). The amounts of silicon deposited on the surface of pure titanium plates after soaking were evaluated by XPS. The amounts of eluted silicon in the solutions were dependent on the materials of vessels and pH value of solutions. The surface concentration of silicon on titanium plates reflected the amounts of eluted silicon in solutions. It is possible to reduce the silicon contamination by choosing suitable vessels as containers for solutions. However, in spite of using a vessel with no silicon elution, surface concentration of silicon on titanium plates was detected because of the silicon impurity of solution itself.

Silicone elastomers for electronic applications. II. Effects of noncrosslinked materials

AbstractWe performed applied functional tests to examine how the noncrosslinked fraction of a cured silicone elastomer affected the performance of electrical devices and the solderability and adhesion on gold‐plated printed circuit boards. To obtain information about the spreading tendencies of commercial silicone elastomers, and to complement the function tests, we conducted so‐called creep tests to examine the migration and spreading of the noncrosslinked fraction. The electrical tests, performed with both reference oligomers and cured commercial poly(dimethyl siloxane)s, showed that failure occurred even at very low silicone contamination levels. The solderability and adhesion, however, were not affected by the presence of poly(dimethyl siloxane)s, especially not the reference solutions, which consisted mainly of volatile siloxanes. The creep tests, performed on commercial silicone elastomers, indicated that the noncrosslinked, nonvolatile part had to be taken into account in a discussion of silicone contamination. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 90: 3780–3789, 2003

Forced Vapour-Phase Decomposition (FVPD) in Combination with e.g. TXRF - a Method to Determine Contamination in Silicon

Forced Vapour-Phase Decomposition (FVPD) in Combination with e.g. TXRF - a Method to Determine Contamination in Silicon

Poly-crystalline silicon thin films prepared by plasma-assisted hot-wire chemical vapor deposition

A combination of hot-wire chemical vapor deposition (HWCVD) and RF plasma, referred to as plasma-assisted HWCVD (P-HWCVD) was used to prepare poly-crystalline silicon (poly-Si) thin films. The effects of the plasma on the film properties were studied by varying the RF power ( P w) from 0 to 40 W. The results indicate that, compared with that of HWCVD samples, the film crystalline fraction ( X c) is enhanced at low P w assistance, whereas it decreases at higher P w. The uniformity of the film thickness is considerably improved by introducing plasma. It is also found that the porosity of the film, indirectly detected from infrared spectra, is much reduced. Auger analysis of the tantalum filament used in the P-HWCVD process shows much lower silicon contamination than that in HWCVD.

Production-scale processing of a new intermetallic NiAl–Ta–Cr alloy for high-temperature application: Part I. Production of master alloy remelt ingots and investment casting of combustor liner model panels

Several master alloy remelt ingots of about 100 kg each with nominal composition 45 at.% Ni, 45 at.% Al, 7.5 at.% Cr and 2.5 at.% Ta were produced by vacuum induction melting. These ingots were used for a series of investment castings with variation of casting parameters in order to develop and optimise the investment casting process for this new intermetallic NiAl-based alloy. The microstructures of the remelt ingots and of various investment castings were characterised by light optical and scanning electron microscopy (SEM). The compositions of the constituent phases were determined by energy dispersive and wavelength dispersive analysis. The 0.2% yield stress, secondary creep rates and oxidation rates were determined and compared with those of materials produced on laboratory scale. Except of a decreased toughness of the production scale material the properties were otherwise comparable to those of the laboratory scale material. Though properties of the cast alloys are found not to be sensitively affected by impurities such as oxygen, carbon, iron, hafnium and zirconium, silicon was found to promote segregation during casting which is believed to result in embrittlement of the production scale material. Silicon contaminations could be avoided by usage of a ceramic-free cold-wall induction melting aggregate, but due to the coarse resulting microstructures this did not improve toughness of the production scale material.