Depth Profile Of Density Research Articles

Density–depth profiles of an As‐implanted Si wafer studied by repeated planar sputter etching and total reflection x‐ray fluorescence analysis

AbstractThe implantation of a high dose of high‐energy ions into an Si wafer causes amorphization of the original monocrystalline structure within a near‐surface layer. The in‐depth distribution of both Si atoms of the wafer and As ions implanted at a dose of 1 × 1017 ions cm−2 and an energy of 100 keV is studied. A novel method combining a repeated planar and broad sputter etching with differential weighing, surface analysis by total reflection x‐ray fluorescence and Tolansky interferometry is used for this investigation. Different depth profiles are recorded on the nanometre scale for the concentration defined as the mole ratio of As and Si, for the mass density of the implanted layer and for the number density of As and Si. The results generally correspond with measurements of Rutherford backscattering spectrometry and only deviate when the assumptions made for the mass density do not fit well. An appropriate approach to this quantity involves the number density of implanted ions but, furthermore, considers a variation of the number density of Si atoms during implantation, especially for a high dose and high‐energy implantation. The variation can be taken into account by a factor γ, where γ > 1 indicates compression and γ < 1 indicates extension of the original crystalline structure. For the above mentioned implantation, γ is measured separately for each sublayer to obtain accurate depth profiles. Copyright © 2003 John Wiley & Sons, Ltd.

Phase-Sensitive Neutron Reflectometry

The recent development of exact methods for determining the phase in neutron specular reflectivity measurements makes it possible to extract the scattering length density (SLD) depth profile of a film directly, by first principles inversion. Ideally, other than the uncertainty introduced by statistical fluctuations in the measured reflectivities, any ambiguity in a SLD profile obtained by these phase-sensitive methods is a consequence only of the limited range of wavevector transfer Q over which the reflectivity data can, in practice, be collected. Where practically possible to employ such phase-sensitive techniques, which require the use of reference layers or variable surrounding media, not only is the uniqueness of the resultant SLD profile well defined, but also the ability to perform an inversion without any adjustable parameters eliminates the vagaries associated with fitting reflectivity data alone. A comprehensive summary of the experimental and theoretical methodologies of phase-sensitive neutron reflectometry is presented and illustrated by a current application.

Characterization of a Biomimetic Polymeric Lipid Bilayer by Phase Sensitive Neutron Reflectometry

Neutron reflectivity measurements were performed on a cell membrane mimic system, developed by Liu et al., consisting of a polyelectrolyte multilayer plus synthetic terpolymer plus a phospholipid layer (PE + TER + PC) assembly, at all of the intermediate steps in the assembly of the composite, to obtain neutron scattering length density depth profiles.1 The polyelectrolyte multilayer functions as a soft, water-containing “cushion” for the membrane mimic, formed by the synthetic terpolymer and the phospholipid layer. The assemblies were studied dry and in 92% humidity using a phase-sensitive neutron reflectometry technique. By use of two water conditions (D2O and H2O mixtures) on the PE + TER + PC assembly, the distribution of water in the layers was obtained. It was found that under 92% humidity conditions, the supported membrane mimic has 40% water content in the “cushion” polyelectrolyte multilayer and 10% water content in the terpolymer−phospholipid region. The overall thickness change, due to water uptake, was found to be 20 Å. Because fusing phospholipid vesicles onto the polyelectrolyte multilayer plus synthetic terpolymer assembly shows an overall 30 Å increase in thickness of the composite, it can be inferred that a phospholipid monolayer was formed.

Defect profiling in semiconductor layers by the electrochemical method

Special based equipment for selective electrochemical etching is presented which is appropriate for in situ observation of the defect structure. The operation of the setup is demonstrated on InGaAs/GaAs (001) heteroepitaxial systems in which the epitaxial layer thickness is above the critical layer thickness. Through incremental layer removal, the depth profile of the dislocation density is mapped. The measured defect density is inversely proportional to the layer thickness and corresponds to the theoretical model.

Phase sensitive reflectometry and the unambiguous determination of scattering length density profiles

Exact methods for determining the complex neutron reflection amplitude for a thin film, which make use of multiple measurements of the specularly reflected intensities of composite systems, composed of the film adjacent to a reference layer and/or surrounding media, have been developed over the past several years. These techniques are valid even where the Born or distorted wave Born approximations break down. Thus, given both the modulus and phase of the specular reflection, a first-principles inversion can be performed which yields the scattering length density (SLD) depth profile of the film directly. Ideally, if the reflection amplitude is known for all wave vector transfers Q, the associated SLD profile is unique. Applying the aforementioned methods to a purely real SLD profile, which, effectively, is almost always that encountered in neutron reflection, at least two distinct reflectivity curves, corresponding to two different composite film systems, are required to determine the phase by direct algebraic computation, independently at each value of Q. Each of the composite systems consists of the common unknown part of the film plus a different reference layer segment and/or surrounding medium (e.g., the backing). Recently, investigations of certain classes of SLD profiles have been reported in the literature which examine whether a single X-ray reflectivity curve, given certain a priori knowledge about the system, i.e., about known parts of the film SLD and/or substrate, suffices to reconstruct the phase. Employing the exact formulation of phase sensitive reflectometry, we consider several illustrative and realistic cases in which a minimum of two reflectivity curves are required to distinguish the true SLD profile.

Characterization of chemical-vapor-deposited low-k thin films using x-ray porosimetry

Trimethylsilane-based carbon-doped silica films prepared with varying chemical-vapor-deposition process conditions were characterized using x-ray reflectivity and porosimetry to measure the film thickness, average film density, density depth profile, wall density, and porosity. Samples deposited under single or dual frequency conditions with either N2O or O2 as an oxidant were compared. The structural parameters were correlated with the chemical bond structure measured by Fourier transform infrared spectroscopy. The density profiles of the porous films were uniform with a slight densification at the film surface. The distribution of pores was also uniform through the film. Films prepared under a single frequency and/or N2O atmosphere had the lowest film density, wall density, and dielectric constant. The porosities of the films were similar and the pore sizes were less than 10 Å.

Interpretation of the depth‐dependent etch pit density in InGaAs/GaAs heterostructures

We study the defect structure of MBE-grown (100) InGaAs/GaAs mismatched heteroepitaxial layers by selective electrochemical (anodic) etching. By incremental layer removal, we map the depth profile of the etch pit density. The defect density is inversely proportional to the layer thickness and increases with In content. The results are explained by a quantitative theoretical model of threading dislocation annihilation. The importance of the layer residual strain in the annihilation process is shown.

Peculiarities of the long-range effects in GaAs-Based transistor structures upon combined irradiation with ions of various masses

Anomalous distinctions in the depth profiles of charge carrier density in the active regions of GaAs-based structures were observed for samples irradiated from the rear side (through substrate) with molecular hydrogen and argon ions either separately or in a certain combined sequence. The maximum effect of ion irradiation was observed in the structures characterized by increased density of defects in the buffer layer. The experimental results are explained by reconstruction of the impurity-defect complexes at the layer interfaces under the action of elastic waves arising both as a result of relaxation of the atomic displacement peaks in the ion stopping zone and due to the reverse piezoelectric effect.

Methods for determining representative density–depth profiles using nuclear density gauges

Most fills when compacted in layers will, to some degree, experience variations in density with depth. When measured using a nuclear density gauge in direct transmission mode, such variations in density through the layer will lead to unrepresentative density–depth profiles. This is because each measurement used to generate the profile is the mean density between the surface and the depth of measurement, and no correction for this is currently made. Uniform sands and silts, and pulverised fuel ash (PFA), often go through a process of over-stressing during compaction, leading to significantly lower densities in the upper part of a compacted layer. Under these circumstances the errors recorded in nuclear density gauge determinations of the density–depth profile may become significant. In extreme cases the usual guidance is to remove the upper part of the compacted layer prior to density testing. This paper describes numerical techniques to determine the mean density between successive measurement depths for the cases of both equal and unequal depth increments, thus giving a density–depth profile that is much closer to reality.

Methods for determining representative density–depth profiles using nuclear density gauges

Most fills when compacted in layers will, to some degree, experience variations in density with depth. When measured using a nuclear density gauge in direct transmission mode, such variations in density through the layer will lead to unrepresentative density–depth profiles. This is because each measurement used to generate the profile is the mean density between the surface and the depth of measurement, and no correction for this is currently made. Uniform sands and silts, and pulverised fuel ash (PFA), often go through a process of over-stressing during compaction, leading to significantly lower densities in the upper part of a compacted layer. Under these circumstances the errors recorded in nuclear density gauge determinations of the density–depth profile may become significant. In extreme cases the usual guidance is to remove the upper part of the compacted layer prior to density testing. This paper describes numerical techniques to determine the mean density between successive measurement depths for the cases of both equal and unequal depth increments, thus giving a density–depth profile that is much closer to reality.

Structural characterization of a porous low-dielectric-constant thin film with a non-uniform depth profile

High-resolution x-ray reflectivity (XR) and small-angle neutron scattering (SANS) are applied to characterize both the nonuniform depth profile and pore structure of a low-dielectric-constant (low-k) thin film as prepared on a silicon substrate. The XR data show that the density depth profile has a multilayered structure with a dense, nonporous top layer and a less dense, porous bulk layer. A scattering invariant analysis of the SANS data is used to determine the average chord length of the pores, (14.8±2.0) nm, independent of the depth profile. Given the elemental composition of the film, the XR and SANS data are combined to calculate the mass density of the top layer, (1.13±0.05) g/cm3, the porosity of the less dense layer, (0.28±0.10), and the wall density, (0.92±0.15) g/cm3.

Structural Comparison of Hydrogen Silsesquioxane Based Porous Low-k Thin Films Prepared with Varying Process Conditions

The structures of hydrogen silsesquioxane based porous low dielectric constant (low-k) films (XLK) prepared with varying process conditions are characterized using a combination of high-energy ion scattering, X-ray reflectivity, and small-angle neutron scattering. We measure the film thickness, average mass density, density depth profile, wall density, porosity, average pore size, pore spacing, pore connectivity, and atomic composition. We compare samples with varying dielectric constants and degrees of cure or Si−H bonding fraction. The structural parameters are correlated with the chemical bond structure as measured by Fourier transform infrared spectroscopy. The density profiles of the porous films were uniform, with a slight densification observed at the film surface. Films with similar k values but varying degrees of cure have almost identical structural characteristics. Lower dielectric constant films have larger porosities and average pore sizes but lower wall densities. The process conditions used to alter the dielectric constant affect not only the porosity but also many other structural parameters such as the wall density.

Effect of rapid thermal annealing on oxide precipitation behavior in silicon crystal

Oxide precipitation behavior after rapid thermal annealing (RTA) in Ar, N 2 or O 2 ambient was investigated. Lightly boron-, nitrogen- or carbon-doped p-CZ silicon wafers with a diameter of 200 mm ( [ O i ]=11.5–13.9×10 17 atoms/ cm 3 (old ASTM)) were prepared as samples. RTA temperature and cooling rate were changed between 1280 and 1200 °C, and between 70 and 5 °C/s, respectively. From the result of in Ar ambient, it was found that (1) M-like depth profile of precipitate density was observed in the lightly boron-doped wafer, (2) width of precipitate denuded zone (DZ) was decreased in the nitrogen-doped wafer compared with the lightly boron-doped wafer and (3) DZ width in the carbon-doped wafer was comparable to lightly boron-doped wafer. From the calculated result of the depth profile of vacancy (V) and silicon interstitial (I) concentrations, C V and C I, it was clarified that the relationships of thermal equilibrium concentration and the diffusion constant of point defects are to be C V ∗>C I ∗ and D V< D I, respectively, to appear the M-like profile in lightly boron-doped wafer. From the result of the decrease of the DZ width, it was deduced that the diffusion constant of vacancy decreased in the nitrogen-doped wafer. From the result of in lightly boron-doped wafer in N 2 or O 2 ambient, it was found that (1) the precipitate density was higher compared with that in Ar ambient and (2) the precipitate was not observed in O 2 ambient. These results due to the injection of the vacancy or the interstitial silicon during RTA in N 2 or O 2, respectively.

X-Ray Reflectivity and FTIR Measurements of N 2 Plasma Effects on the Density Profile of Hydrogen Silsesquioxane Thin Films

The density depth profile and chemical bond structure of hydrogen silsesquioxane (HSQ) thin films treated with an plasma with varying power and exposure time were measured using specular X-ray reflectivity (SXR) and Fourier transform infrared (FTIR) spectroscopy. The SXR data indicated that the density profile of an untreated HSQ film is not uniform. At least four layers with different electron densities were required to fit the SXR data. For HSQ films treated with either increasing plasma power or plasma exposure time, the film roughness increased and a densified layer was observed at the film/air interface. The thickness of the densified layer increased with both plasma power and plasma exposure time. In the FTIR spectra of the plasma-treated films, the intensities of the Si-O peaks due to the network structure and the Si-OH peak due to water absorption increased and the intensities of the Si-H peaks decreased. The FTIR data show that the plasma converts the HSQ structure into a -like structure and are consistent with the densification observed in the SXR measurements. In general, the HSQ film is more sensitive to increasing plasma power than to increasing plasma exposure time. © 2001 The Electrochemical Society. All rights reserved.

In situ characterization of thin film growth: Boron nitride on silicon

Real-time ellipsometry (RTE) in combination with particle flux measurement is applied to ion beam assisted deposition of boron nitride (BN) films. RTE is used as a tool for process diagnostic to improve the deposition stability. A novel technique for the determination of absolute density depth profiles from dynamic growth rate data and film forming particle flux is employed. From real-time cantilever curvature measurement and simultaneously recorded film thickness data instantaneous stress depth profiles are derived with a depth resolution in the nm range. The synergistic effects on the information obtained from RTE, particle flux, and cantilever bending data are demonstrated. The density of turbostratic BN (tBN) is found to increase slightly with film thickness while the compressive stress decreases, indicating an increasing quality and/or size of crystallites in the course of film growth. Refractive index and density depth profiles in cubic BN (cBN) films correspond perfectly to structural information obtained from dark field transmission electron microscope graphs. The established tBN/cBN two-layer model is found to be a crude approximation that has to be replaced by a three-layer model including nucleation, grain growth, and coalescence of cBN. The instantaneous compressive stress in a homogeneous tBN film is found to decrease, while the density increases during growth. The instantaneous compressive stress depth profiles in cBN films are more complex and not easy to understand but reliable information on the structural evolution during growth can be extracted.

Effect of Rapid Thermal Annealing on Oxygen Precipitation Behavior in Silicon Wafers

The effect of rapid thermal annealing (RTA) on oxygen precipitation behavior in Czochralski silicon wafers was investigated with an emphasis on the RTA ambient, temperature and cooling rate. It was found that (i) anomalous oxygen precipitation (AOP) was observed in the case of RTA temperature at ≧1240°C with cooling rates of ≧25°C/s in Ar and in the case of RTA temperature at ≧1200°C with cooling rates of ≧5°C/s in N2, while AOP was not observed in O2, (ii) an M-like depth profile of precipitate density appeared for the cooling rates of ≧50°C/s in Ar, and ≧25°C/s in N2, and (iii) the width of the precipitate denuded zone was larger than the width of outdiffused oxygen in the case of RTA in Ar. The relationships of thermal equilibrium concentrations CJ* and diffusion constants DJ were estimated to be CI*<CV* and DI>DV (I: interstitial, V: vacancy) from the experimental results of RTA in Ar and the calculated results using the Voronkov model.

Investigation of N2 Plasma Effects on the Density Profile of Hydrogen Silsesquioxane Thin Films

ABSTRACTThe density depth profile and chemical bond structure of hydrogen silsesquioxane (HSQ) thin films treated with an N2 plasma with varying power and exposure time were measured using specular x-ray reflectivity (SXR) and Fourier transform infrared (FTIR) spectroscopy. The SXR data indicate that the density profile of an untreated HSQ film is not uniform and four layers with different electron densities were required to fit the SXR data. For HSQ films treated with either increasing plasma power or plasma exposure time, the film roughness increased and a densified layer was observed at the film/air interface. The thickness of the densified layer increased with both plasma power and plasma exposure time. As a result, up to seven distinct layers were used to model the experimental SXR data from plasma treated films. The FTIR data show that the plasma transforms the Si-H bonds in the HSQ film into Si-O bonds leaving more oxygen atoms around a Si atom. These data are also consistent with the densification observed in the SXR measurements. In general, the HSQ film is more sensitive to increasing plasma power than to increasing plasma exposure time.

Modeling and analysis of sound transmission in the Strait of Korea

The experiment called the Acoustic Characterization Test III was performed in the oceanographically complex Strait of Korea. It was designed to provide accurate measurements of sound transmission and array signal gain under known environmental conditions. Bottom sampling and sub-bottom surveys coupled with archival geophysical information provided the basis for the geoacoustic depth profiles of sound speed, density and attenuation. The bottom was a sand-silt sediment for which shear wave propagation at experimental frequencies was determined to be unimportant. Using the compressional wave profiles, the measured bathymetry and the water sound speed as input parameters, good agreement was obtained between measured and calculated narrowband transmission loss when an attenuation profile with a frequency dependence to the 1.8 power in the near sediment-water interface layer was used. This power law was determined by using an effective attenuation coefficient and a least squares comparison of calculated and measured sound transmission of five narrowband tones between 47 and 604 Hz. The resulting geoacoustic model was used to compare measured and calculated broadband sound transmission and signal spread and excellent agreement was found. These results are consistent with measurements in other sand-silt areas where site specific frequency dependent characteristics have been observed.

Correlations Between Structural Characteristics and Process Conditions of HSQ Based Porous Low-k Thin Films

ABSTRACTA novel methodology using a combination of ion scattering, x-ray reflectivity (SXR), and small angle neutron scattering was used to characterize the structure and properties of a hydrogen silsesquioxane (HSQ) based porous low-k dielectric films after varying process conditions. The dielectric constant and the remaining Si-H fraction (degree of cure) of the samples were varied from 1.5 to 2.2 and from 30 % to 52 %, respectively, by controlling the mass ratio of the solvent and the HSQ resin in the initial solutions and the wet ammonia treatment time. We determined the density depth profile, average mass density, wall density, porosity, average pore size, average wall thickness, pore connectivity and atomic composition. The chemical bond structures were also measured using Fourier transform infrared (FTIR) spectroscopy. The density profile of each porous low-k film was uniform and only two layers were required to fit the experimental SXR data. Higher dielectric constant films show significantly higher wall densities and lower porosities and pore sizes. The measured increases in the wall density with lower Si-H fractions are consistent with the FTIR results.

Depth profile of tritium in plasma exposed CX-2002U

Depth profiles of the Ag grains density in autoradiographs, which represent tritium concentration in CX-2002U samples exposed to high flux D/T particles under various conditions, were examined and the apparent diffusion coefficients were estimated from the profiles. Plasma discharge generating D/T atomized particles with low energies increases tritium inventory in the samples by introducing high tritium concentration on the surface exposed and following diffusion process into the deep region with apparent diffusion coefficients (1.7×10 −16 m 2/s at 293 K and 2.3×10 −15 m 2/s at 573 K), which are much larger than the diffusion coefficients in the bulk reported. Oxygen RF-plasma exposure might be effective to remove tritium retained even at a fairly deep region in carbon fiber composite (CFC) components.