X-ray Photoelectron Spectroscopy Imaging Research Articles

Understanding delamination in microelectronic devices using AES, XPS and chemometrics

The recent development of XPS instrumentation with near-micron spatial resolution has advanced the capability of elemental and chemical-state imaging accompanied by small-area analysis (down to 15 µm). In this paper, the combined use of X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) at enhanced spatial resolution is shown to have significantly improved the understanding of interfacial delamination and related problems encountered in the production of electronic devices in the field of microelectronics. An example of the application of surface analysis for ITO/Mo adhesion problems will be presented. The mathematical procedure using principal component analysis (PCA) in the reduction of noise in XPS images will also be described. The dramatic improvements in the image contrast and chemical component determination from multispectral image data sets will be presented. This study is intended to explore the contributions given by advanced surface analysis tools to solve real-world problems. Copyright © 2007 John Wiley & Sons, Ltd.

Vapor phase formation of a well-ordered aldehyde-terminated self-assembled monolayer on a SiO 2 surface and formation of silver film on the surface based on the silver mirror reaction

A well-ordered aldehyde (CHO)-terminated self-assembled monolayer (SAM) was formed from a vapor of triethoxysilylundecanal (TESUD) onto both quartz glass plates and silicon substrates covered with native oxide. This vapor phase treatment produced a TESUD-SAM on both substrates without any marked change in surface morphology, as evidenced by X-ray photoelectron spectroscopy (XPS) and atomic force microscopy (AFM). The thickness of this SAM was estimated by ellipsometry to be 1.2 ± 0.1 nm. Silver (Ag) metallization was also demonstrated on TESUD-SAM-covered quartz glass substrates based on the silver mirror reaction using an ammoniacal silver nitrate solution. XPS and AFM images confirmed that the deposited film was composed of densely packed Ag nanoparticles with an average diameter of 300 nm. The average thickness of the film was about 1–2 μm. Due to this particulate Ag film formation, the transparent insulating quartz glass plate became conductive with a mirror-like surface. The CHO terminal groups of the TESUD-SAM were found to serve effectively as a reducing agent to deposit Ag metal at the solid/liquid interface.

Photolithographically Patterned Surface Modification of Poly(dimethylsiloxane) via UV-Initiated Graft Polymerization of Acrylates

Patterned surface modification of poly(dimethylsiloxane) (PDMS) is achieved by combining ultraviolet-initiated graft polymerization (UV-GP) and photolithography. Poly(acrylic acid) (PAA) and poly(methacrylic acid) (PMAA) patterns were grafted onto PDMS with micrometer-scale feature edge resolution. The morphology and chemical composition of the grafted layers were assessed by optical and atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), and XPS imaging. AFM section analyses demonstrated the deposition of 33 +/- 1 and 62 +/- 8 nm thick patterned films of PAA and PMAA, respectively. Spatially resolved C 1s XPS provided images of carboxylic acid functionalities, verifying the patterned deposition of acrylate films on PDMS. These observations demonstrate the general usefulness of UV-GP and photolithography for micropatterning.

Anisotropic Hydrogel Thickness Gradient Films Derivatized to Yield Three-Dimensional Composite Materials

We report the preparation of in-plane density gradients of amino-terminated molecules and gold particles through derivatization of laterally varying thickness gradients of poly(acrylic acid) (PAA) or poly(acrylamide) (PAAm) films. PAA and PAAm gradients were formed by Zn(II)-catalyzed electropolymerization of acrylic acid (AA) or acrylamide (AAm) in the presence of an in-plane electrochemical potential gradient applied to Au or indium-tin-oxide (ITO) working electrodes. PAA thickness gradients were converted into density gradients of fluorocarbons or biocompatible groups by derivatizing with NH(2)CH(2)(CF(2))(6)CF(3) or an Arg-Gly-Asp (RGD)-containing peptide, respectively. X-ray photoelectron spectroscopy (XPS) and XPS imaging were used to characterize the modified PAA gradients. Transition regions as narrow as 104 mum were achieved for fluorocarbon gradients. PAAm gradients were treated with gold particles to form a density gradient of gold particles. Surface plasmon resonance imaging and scanning electron microscopy (SEM) as well as UV-visible absorption measurements were used to characterize the gold particle density gradients. It is likely that the gold particles were attached both on the surface and inside the PAAm film.

Algorithm for automatic x-ray photoelectron spectroscopy data processing and x-ray photoelectron spectroscopy imaging

It is well known that x-ray photoelectron spectroscopy (XPS) quantification based only on the measured XPS-peak intensity can lead to large errors. The problem, which is caused by the strong depth dependence of the intensity contribution due to inelastic electron scattering, was solved by developing algorithms for analysis of the energy distribution of emitted electrons. These algorithms are now widely used but since their practical application requires operator interaction, they are not well suited for automatic data processing. A simplified algorithm that is sufficiently robust (i.e., insensitive to variations in analysis procedure) to be suitable for automation was suggested recently [J. Vac. Sci. Technol. A21, 1081 (2003)]. For each XPS peak, this algorithm determines the total amount of the corresponding atoms within the outermost approximately three inelastic electron mean free paths and it also gives an estimate of their depth distribution. Since the algorithm can be automated, it should also prove useful in XPS imaging. In the present article, we investigate the applicability of the algorithm by analysis of nanometer thin Au films deposited on a Ni substrate. The XPS signals from both the overlayer Au 4d and the substrate Ni 2p peaks are analyzed and it is found that the analysis gives amounts of atoms that are in good agreement with nominal thicknesses determined by a more elaborate analysis as well as by Rutherford backscattering spectroscopy. Analysis of the substrate Ni 2p peak gives results that are in agreement with the results from analysis of the overlayer Au 4d peaks and the determined depth distributions of both elements are also in agreement with the known distributions.

NanoESCA: imaging UPS and XPS with high energy resolution

A novel imaging electron spectrometer has been used for laterally resolved ultraviolet photoelectron spectroscopy (UPS) and X-ray photoelectron spectroscopy (XPS) in the soft X-ray range. The instrument is based on a high-resolution emission microscope optics using a cathode lens and an imaging dispersive analyser. The analyser is corrected for the leading aberration term ( α 2-term) by means of two hemispherical analysers in antisymmetric configuration with an appropriate transfer lens. Small-area spectra as well as energy-filtered images have been taken in the soft X-ray range for a meteorite sample and in the range of the d-band of a Cu polycrystal. An energy resolution of 106 meV along with a high spatial resolution allows a detailed analysis of trace elements in the meteorite sample as well as local UPS spectra of the surface of the Cu sample.

Visualisation of UF resin on MDF fibre by XPS imaging

Using X-ray photoelectron spectroscopy (XPS) will distinguish urea formaldehyde (UF) resin on the surface of medium density fibreboard (MDF) fibre. High spatial resolution XPS imaging was employed to 2-dimensionally map the presence and concentration of carbon and nitrogen on pressed MDF fibre. Using image analysis routines, composite XPS images of fibre and UF resin were produced for laboratory and commercially prepared MDF panels using the C1s and N1s peaks respectively. Results showed that MDF could be analysed directly, and the presence of UF resin features imaged on fibre was comparable to UF resin characterised by confocal microscopy. Due to low surface nitrogen contents, the XPS imaging instrument was necessarily operated at a low resolution, and with a relatively small field of view. This may make the XPS imaging method less attractive for visualising UF resin on wood fibre compared to other microscopy techniques.

Investigations of the corrosion protection of ultrathin a-C and a-C:N overcoats for magnetic storage devices

The thickness-dependent corrosion protection of carbon overcoats for magnetic hard disks can be analyzed by collecting X-ray absorption near edge structure (XANES) spectra at the Co L 3-edge. Co is the main constituent of the protected magnetic media underneath. The spectra of the Co absorption edge display a strong peak for pure metallic, non-oxidized Co. This peak splits up into several sub-structures for oxidized Co. Therefore, XANES spectra provide a straightforward method to determine the overcoat thickness, leading to closed coverage and corrosion protection of the underlying material. A similar approach was carried out by X-ray photoelectron spectroscopy (XPS). Standard a-C:N overcoats deposited by magnetron sputtering (MS) were compared to a-C overcoats deposited by a novel filtered high current pulsed arc source (HCA) using a highly resolved imaging XPS system providing the combined measurement of XANES and XPS spectra in the same field of view. We determined that the HCA deposited a-C overcoats provide a closed coverage down to a layer thickness of 1.4 nm, whereas standard MS a-C:N overcoats do not cover the magnetic film completely until a thickness of 2–3 nm is reached.

New frontiers in X-ray photoelectron spectroscopy

X-ray photoelectron spectroscopy (XPS) has seen widespread use in applied and basic surface science studies since 1970. Its capabilities appear to be undergoing a series of improvements, due both to technological changes and to the culmination of efforts by the XPS community to solve a series of scientific questions that underlie the technique. This paper describes some of the advances made in the past 5 yr and uses two particular studies to illustrate the improvements: the use of mathematical deconvolution to study the initial and extended oxidation of nickel metal, and the use of XPS imaging to identify electrochemical processes during the pitting corrosion of a nickel-based alloy. Finally, the use of modern synchrotrons in further improving the spectroscopic capabilities of XPS will be described.

Oxidative Graft Polymerization of Aniline on Modified Si(100) Surface

Chemical modification of the argon plasma-pretreated Si(100) surface by UV-induced surface graft polymerization with either glycidyl methacrylate (GMA) or glycidyl acrylate (GA) was carried out. The GA graft polymerized Si substrate was further subjected to coupling reaction with aniline (An) and finally oxidative graft polymerization of An. The composition and microstructure of the graft-polymerized Si(100) surfaces were studied by X-ray photoelectron spectroscopy (XPS) and imaging XPS, respectively. The graft concentrations of the GMA polymer, GA polymer, and An polymer increased with increasing concentration of the respective monomer used for graft polymerization. The graft polymerization efficiency of GA on the Ar plasma-pretreated Si(100) was much higher than that of GMA. Ethanol, when used as a solvent, catalyzed the coupling reaction of the epoxide groups of GA with An and should be of more than 40 vol % in concentration to achieve the optimum effect. The protonation−deprotonation characteristics, interconvertible intrinsic redox states, and metal reduction behavior of the polyaniline (PANI) chains, obtained from subsequent oxidative graft polymerization of An on the modified Si(100) surface, were grossly similar to those of the PANI homopolymer. The resistance of the modified Si(100) surface from consecutive graft polymerization with GA and An was on the order of 107 Ω/sq.

Profiling of the SiO2 - SiC Interface Using X-ray Photoelectron Spectroscopy

ABSTRACTThe implementation of SiC based sensors and electronics for operation in chemically harsh, high temperature environments depends on understanding the SiO2/SiC interface in field effect devices. We have developed a technique to fabricate wedge polished samples (angle ∼ 1×10−4 rad) that provides access to the SiO2/SiC interface via a surface sensitive probe such as xray photoelectron spectroscopy (XPS). Lateral scanning along the wedge is equivalent to depth profiling. Spatially resolved XPS images of the O 1s and Si 2p core levels were obtained of the interfacial region. Samples consist of device-quality thermally grown oxides on 4H-SiC single crystal substrates. The C 1s spectrum suggests the presence of a graphitic layer on the nominally bare SiC surface following thermal oxidation.

The influence of the yttrium content on the structure and properties of Ti1−x−y−zAlxCryYzN PVD hard coatings

Abstract Ti 1− x − y − z Al x Cr y Y z N and Ti 1− x − y Al x Cr y N coatings have been grown in a multiple target PVD coating unit using the combined steered arc evaporation/unbalanced magnetron deposition technique. Ti 1− x − y − z Al x Cr y Y z N film properties and yttrium distributions have been found to change significantly when three targets with equal yttrium content (Ti 0.495 Al 0.495 Y 0.01 ) or one yttrium-containing target (Ti 0.48 Al 0.48 Y 0.04 ) with two neighbouring pure Ti 0.5 Al 0.5 targets to produce a non-homogeneous distribution were used. The structural characteristics and residual stresses present in the coatings were determined by X-ray diffraction using both Bragg–Brentano and glancing angle parallel beam (sin 2 ψ method) geometries. In all cases the state of stress was found to be compressive, with values at 1° incidence angles of −6.5 GPa using a single 4 at.% yttrium containing target and −3.2 GPa using three TiAlY targets each containing 1 at.% Y. The residual stress value at 1° angle of incidence for the Ti 1− x − y Al x Cr y N film was similar to that of the homogeneously grown yttrium containing film, i.e. −3.8 GPa. The surface chemistry of the as-deposited and thermally treated films has been further analysed using high-resolution monochromatic X-ray photoelectron spectroscopy (XPS). Large area (700 μm×300 μm) and small spot (55 μm) XPS analysis was used to examine the role of low concentrations of the elements Y and Cr during the oxidation process. XPS imaging using O 1s, Ti 2p, Cr 2p and Fe 2p photoelectron peaks has been used to study the distribution of oxides formed on the surface after heat treatment. High resolution analysis of the Y 3d core level identified two chemical species for the as-deposited TiAlCrYN films at positions Y 3d 156.0 and 158.1 eV respectively, whilst only single species were identified in the oxidised film at Y 3d 158 eV.

Chemical Modification of Silicon (100) Surface via UV-Induced Graft Polymerization

Modification of argon plasma-pretreated single-crystal silicon wafer surface via UV-induced graft polymerization with various functional monomers, such as acrylamide (AAm), N,N‘-diamethylaminoethyl methacrylate (DMAEMA), and 2,2,2-trifluoroethyl acrylate (TFEA), was achieved. The modified Si(100) surfaces were characterized by X-ray photoelectron spectroscopy (XPS), imaging XPS, atomic force microscopy (AFM), and water contact angle measurements. The graft polymerization was affected by plasma pretreatment time and UV irradiation time. XPS results suggest that mild and brief plasma treatment is sufficient to cause surface oxidation and to generate sufficient peroxides and hydroxyl peroxides for the subsequent UV-induced graft polymerization in the presence of a vinyl monomer. Prolonged plasma treatment and the accompanying overoxidation of the silicon surface have an adverse effect on the graft polymerization. For all the cases investigated, the XPS results revealed that the grafted polymers form a thin l...

Investigation of imaging X-ray photoelectron spectroscopy for surface analysis of atmospheric particulates

X-Ray photoelectron spectroscopy (XPS) is used to investigate surface species on atmospheric particulates as part of an investigation into the sources and health effects of such particulates. Oxygen and carbon species dominate the surface with trace quantities of Na+, NO3–, Cl– and SO42–. Multiple species for carbon and oxygen were evident, with major species identified as oxidised carbon, graphitic/aliphatic carbon and ruthenium oxide or a carbide. The potential of imaging XPS to show localised variation of surface species across atmospheric particulates is demonstrated.

Developmental Neurobiology Implications from Fabrication and Analysis of Hippocampal Neuronal Networks on Patterned Silane-Modified Surfaces

We have determined the parameters necessary to fabricate reproducible neuronal patterns which we are using to begin studying fundamental issues in developmental neurobiology. The addition of a beam homogenizer, as well as a new surface preparation, has enabled the routine production of reproducible, high-resolution (2−20 μm) organosilane patterns. The effects of surface preparation and beam dosage were monitored using X-ray photoelectron spectroscopy (XPS) and proof of patterning is provided by high-resolution imaging XPS. We also report the guidance of neuronal adhesion and neurite outgrowth and the creation of reproducibly defined circuits of embryonic (E18−19) rat hippocampal neurons using these patterned surfaces in vitro. We have achieved a >50% rate of pattern formation, and at times the rate approaches 90%. We are using these patterns to address the issue of how geometric pattern cues might be used to affect cell-to-cell communication and we report the preliminary results on the synaptic development of the hippocampal neurons using dual patch-clamp electrophysiology. We monitored neurite outgrowth and the emergence of both spontaneous and evoked synaptic activity for both patterned and unpatterned (control) hippocampal cultures. The results indicate the intriguing possibility that geometry itself may be a modulating or trophic factor for cell development.

Cell-based sensor microelectrode array characterized by imaging x-ray photoelectron spectroscopy, scanning electron microscopy, impedance measurements, and extracellular recordings

We are developing a cell-based biosensor consisting of a planar microelectrode array that allows detection of extracellular potentials and their modulation in the presence of toxins or other active agents. To improve cell–electrode coupling, the microelectrodes were electroplated with platinum black. We report on the use of imaging x-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), impedance measurements, and extracellular recordings to assess the effectiveness of this procedure. SEM provided highly detailed images of the shape and structure of well-formed deposits of thickness on the order of 1 μm or more. Because of its inherent high surface sensitivity, imaging XPS could reveal the presence of platinum deposits that were too thin to be detected by SEM. For typical, well-plated microelectrodes, impedance measurements showed reductions in the electrical resistance at 100 Hz from roughly 60 MΩ or more 1 MΩ. The overall electronic coupling of biopotentials to the microelectrodes was demonstrated by recordings obtained from beating rat myocytes and from rat spinal cord cells.

X‐ray photoelectron imaging of germanosilicate optical fibre preforms

AbstractThe germanosilicate glass core (GeO2–SiO2) of an optical fibre preform has been analysed and imaged using small‐spot (∼450 μm) x‐ray photoelectron spectroscopy (XPS). The germanium oxide was most readily detected in the glass preform sample using the Al Kα x‐ray‐induced Ge L3MM Auger feature. The observed Ge L3MM lineshape and estimated oxidation state shift of 8.8 eV with respect to pure germanium indicate that the germanium is predominantly in the form of GeO2 (+4 oxidation state) within the germanosilicate core. Argon ion etching was found to sputter preferentially the oxygen from the glass, and the observed changes in Ge L3MM Auger lineshapes suggest that suboxide defects (GeO) are produced. Quantitative XPS images and linescans were used to map the spatial distribution of germanium within the preform and were in good agreement with measured refractive index profiles.

X-ray photoelectron spectroscopy characterisation of third body layers formed during automotive friction braking

X-ray photo electron spectroscopy (XPS) has been used to determine the nature of the “TRANSFER FILM” (TF) or “THIRD BODY LAYER” (TBL) formed at the friction interface of some simple friction couples using materials commonly employed in automotive braking systems. Imaging X-ray photo electron spectroscopy (IXPS) has been used with good effect to illustrate both the interaction and relative distribution of the various chemical species present on the grey cast iron surface after rubbing against a number of different asbestos free friction materials. Evidence is presented which highlights the relationship between transfer film chemistry, friction material composition and work done during braking. The role of lubricant additions to the friction material, in particular their effectiveness at a range of asperity temperatures, is examined in detail. The influence of chemical changes at the interface is also discussed. The results obtained in this study allows conclusions to be drawn about the dependence of transfer film chemistry on the asperity temperature and the composition of the friction material. This information is used to explain some of the differences in friction performance which may occur during automotive braking.

The influence of structural damage on the electronic properties of amorphous germanium : J. A. Olley. Solid St. Commun.13, 1441 (1973)

The thickness-dependent corrosion protection of carbon overcoats for magnetic hard disks can be analyzed by collecting X-ray absorption near edge structure (XANES) spectra at the Co L3-edge. Co is the main constituent of the protected magnetic media underneath. The spectra of the Co absorption edge display a strong peak for pure metallic, non-oxidized Co. This peak splits up into several sub-structures for oxidized Co. Therefore, XANES spectra provide a straightforward method to determine the overcoat thickness, leading to closed coverage and corrosion protection of the underlying material. A similar approach was carried out by X-ray photoelectron spectroscopy (XPS). Standard a-C:N overcoats deposited by magnetron sputtering (MS) were compared to a-C overcoats deposited by a novel filtered high current pulsed arc source (HCA) using a highly resolved imaging XPS system providing the combined measurement of XANES and XPS spectra in the same field of view. We determined that the HCA deposited a-C overcoats provide a closed coverage down to a layer thickness of 1.4 nm, whereas standard MS a-C:N overcoats do not cover the magnetic film completely until a thickness of 2–3 nm is reached.

A thin-film high sheet resistivity material : K. Heid. Solid St. Technol. (Sept. 1973). p. 56

The thickness-dependent corrosion protection of carbon overcoats for magnetic hard disks can be analyzed by collecting X-ray absorption near edge structure (XANES) spectra at the Co L3-edge. Co is the main constituent of the protected magnetic media underneath. The spectra of the Co absorption edge display a strong peak for pure metallic, non-oxidized Co. This peak splits up into several sub-structures for oxidized Co. Therefore, XANES spectra provide a straightforward method to determine the overcoat thickness, leading to closed coverage and corrosion protection of the underlying material. A similar approach was carried out by X-ray photoelectron spectroscopy (XPS). Standard a-C:N overcoats deposited by magnetron sputtering (MS) were compared to a-C overcoats deposited by a novel filtered high current pulsed arc source (HCA) using a highly resolved imaging XPS system providing the combined measurement of XANES and XPS spectra in the same field of view. We determined that the HCA deposited a-C overcoats provide a closed coverage down to a layer thickness of 1.4 nm, whereas standard MS a-C:N overcoats do not cover the magnetic film completely until a thickness of 2–3 nm is reached.