Search

Highly porous Germanium surfaces with uniformly distributed columnar nanovoid structures are fabricated over a large area (wafer scale) by large fluence Sn+ irradiation through a thin silicon nitride layer. The latter represents a one-step highly reproducible approach with no material loss to strongly increase photon harvesting into a semiconductor active layer by exploiting the moth-eye antireflection effect. The ion implantation through the nitride cap layer allows fabricating porous nanostructures with high aspect ratio, which can be tailored by varying ion fluence. By comparing the reflectivity of nanoporous Ge films with a flat reference we demonstrate a strong and omnidirectional reduction in the optical reflectivity by a factor of 96% in the selected spectral regions around 960 nm and by a factor of 67.1% averaged over the broad spectral range from 350 to 1800 nm. Such highly anti-reflective nanostructured Ge films prepared over large-areas with a self-organized maskless approach have the potential to impact real world applications aiming at energy harvesting.

Metal oxide thin films show promising resistive switching properties, making them materials of reference for the development of memristive devices. TiO2 is probably one of the most studied materials and is being synthesized using various techniques, each of them having specific optimizable characteristics. In this paper, we report on an innovative approach by combining the sol–gel and the pulsed microplasma cluster source (PMCS) methods, exploiting the low temperature and low cost of the former process and precise control over nanocristallinity of the latter. We show that this approach overcomes the reported limitations that each technique shows in fabricating memristive devices when independently used. A side-by-side comparison of the TiO2 thin films produced by the PMCS, sol–gel, and PMCS/sol–gel hybrid methods (HM) demonstrates an improvement of the memristive properties and a reduction of the electrical shorts in the TiO2 based devices.

An international round-robin test (RRT) was performed to investigate a method to determine the interface location and the layer thickness of multilayer films by secondary ion mass spectrometry (SIMS) depth profiling as a preliminary study to develop a new work item proposal in ISO/TC-201. Two types of reference materials were used in this RRT. A SiGe alloy (Si52.4Ge47.6) reference film was used to determine the relative sensitivity factors of Si and Ge. A Si/Ge multilayer reference film was used to determine the relative sputtering rates of the Si and Ge layers. The layer thicknesses were measured from the interfaces determined by a 50 atomic percent definition. Seven laboratories from 5 countries participated in this international RRT. The RRT reference expanded uncertainties for Si and Ge layers in a Si/Ge multilayer with similar thicknesses as the reference film were 0.76 and 1.17 nm, respectively. However, those in a thinner Si/Ge multilayer film were slightly larger at 1.04 and 1.59 nm, respectively. Most of the thickness ratios in the 2 Si/Ge multilayer films were consistent with the RRT reference value within their expanded uncertainties.

Selective functionalization of the external/internal pore surface of porous silicon is of interest for the numerous potential applications of this material, in particular in pharmacology. With the aim of obtaining porous silicon platforms compatible with the aqueous environment and providing hydrophobic pores to load poorly water soluble molecules, we set-up a three step functionalization procedure consisting in two hydrosilylation reactions separated by the selective etching of the external surface. This procedure was applied both, to porous layers and porous microparticles. The characterization of the functionalized material by conventional techniques such as contact angle and FTIR showed a change of the properties of porous structures in line with the expected surface modifications. However, these techniques do not permit to clearly distinguish between internally and externally grafted functional groups. For this reason, an innovative procedure based on angle-resolved XPS was set-up and applied to differently functionalized pSi layers. By this technique, we obtained indications of prevalent grafting of hydrophilic moieties on the external surface and hydrophobic ones inside the pores.

Hardness is a key property in materials' characterization, particularly regarding micro and submicron functional coating technology. Hardness definition is commonly cited as the resistance of materials to plastic deformation. To assess the hardness value of a given material different methods are available and can be defined as macro-, micro- or nano-scale according to the forces applied and the displacements obtained. Nano-hardness measurements are essential for functional thin coatings characterization. In the present work we tested the possibility to evaluate coatings hardness by using a complementary method: the ion beam sputtering by Secondary Ion Mass Spectrometry (SIMS) equipment. This approach has two evident advantages: independence of measurements from the substrate and additional compositional and structural information provided by SIMS measurements. The applicability of this approach has been experimentally tested on several kind of hard coating films (as ZrN, TiN, TiAlCN et al.), including diamond-like (DLC) carbon materials. Some initial results pointed out the possibility to find a direct relationship between hardness and sputter rate, pointing out a good agreement with previous works. Nevertheless, detailed tests showed evident problems on DLC coatings. In this case the dependence between ion beam sputter rate and hardness is not evident and far to fit other experimental data. In our hypothesis this effect is related to the different nature of hardness on DLC materials.

A simple method for the determination of a Si p+/n junction depth is presented. The method is designed to delineate the specific junction due to its importance in the field of Si solar cells where cost effective and fast characterization techniques are necessary. It consists of the electrochemical transformation of the p+ Si to porous Si. The determination of the porous Si depth with the use of cross-sectional Scanning Electron Microscope (SEM) images provides a direct, fast and easy to implement measurement of the junction depth. In addition, through a simple 4-point probe electrical measurement of the sheet resistance, the average dopant concentration is determined, which allows the creation of an abrupt junction approximation of the p+/n junction. The method is shown to produce accurate results in two types of doping techniques, namely implantation and spin-on-doping and a range of junction depths between 200 nm and 1500 nm, as compared to the well-established secondary ion mass spectrometry (SIMS) technique.

In the last years a lot of effort has been directed in order to reduce silicon defects eventually formed during the ion implantation/anneal sequence used in the fabrication of CMOS devices. In this work we explored the effect of ion implant dose rate and temperature on the formation of silicon defects for high fluence 49BF2 implantations. The considered processes (implantation and annealing) conditions are those typically used to form the source/drain regions of p-channel transistors in the submicron technology node and will be detailed in the document. Characterization of implant damage and extended silicon defects left after anneal has been performed by TEM. Dopant distribution and dopant activation has been investigated by SIMS and SRP analysis. We have verified that implant dose rate and temperature modulate the thickness of the amorphous silicon observed after implant, as well as the concentrations of silicon defects left after anneal. Effect of high dose rate low temperature implantation on product device was also evaluated, showing a reduction of leakage current on p-channel transistors. Experimental set up, results and possible explanation will be reported and discussed in the paper.

/Ge interface, while phosphorous dose loss and diffusion to the bulk are strongly suppressed in comparison to Gesubstrate without nitrogen implantation. Possible physical mechanisms (involving phosphorous-vacancies or phosphorous-nitrogencomplexes, and end-of-range interstitials) that explain this result are presented and discussed.© The Author(s) 2015. Published by ECS. This is an open access article distributed under the terms of the Creative CommonsAttribution Non-Commercial No Derivatives 4.0 License (CC BY-NC-ND, http://creativecommons.org/licenses/by-nc-nd/4.0/),which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is not changed in anyway and is properly cited. For permission for commercial reuse, please email: oa@electrochem.org. [DOI: 10.1149/2.0061506ssl]All rights reserved.Manuscript submitted February 26, 2015; revised manuscript received April 9, 2015. Published April 21, 2015.

A Versailles Project on Advanced Materials and Standards round robin test (RRT) has been conducted to evaluate the linearity of the instrumental intensity scale and correction method using an approximation intermediate extended dead time model with parameters derived from two different isotope depth profiles. Nine organizations in five countries participated. An arsenic-implanted silicon wafer and a film of BN diffused into a Si wafer were supplied by the National Institute of Advanced Industrial Science and Technology along with instructions for the RRT. The instruments used to analyze 103(AsSi)− and 105(AsSi)− from arsenic-implanted samples were five quadrupole-type SIMS and four magnetic-sector type SIMS. The instruments used to analyze 10B+ and 11B+ from the BN-diffused samples were three quadrupole-type SIMS, four magnetic-sector type SIMS, and one time-of-flight type SIMS. We validated the usefulness of the approximation intermediate extended dead time model to correct saturated intensities for all SIMS in this RRT. The optimum extension parameter ρ tends to be affected by the ratio of the maximum reliable intensity to the maximum intensity in raw profiles. From the ratio, ρ may be predicted when the intensity reaches full saturation. On the other hand, ρ is also affected by lateral non-uniformity of intensity. In practice, because the maximum intensity does not reach full saturation and the intensity is not laterally uniform, ρ is likely to be smaller than its predicted value. Copyright © 2015 John Wiley & Sons, Ltd.

Nanostructured germanium substrates are produced by gold ion implantation; they show periodic ripples of nanometer size, decorated on the top and partially on one side with a forest of curled nanowires that end with gold-rich nanoparticles. For the first time, through a novel two-step soft lithography transfer process, the multi-scale nanopatterns are replicated, with features well below 100 nm, on biocompatible 2-norbornene ethylene cyclic olefin copolymer substrates. Given the suitable aspect ratio of the nanoripples and the peculiarity of their multiscale structure, the final substrates are available for cell–material interaction studies that can shed light on the role of the hierarchy of nanostructured materials in controlling the large-scale cellular behavior on biocompatible scaffolds. This work also presents an original combination of numerical analyses of scanning force microscopy images, which allows an accurate quantitative description of the outputs of the two-step transfer process.

We report on the fabrication and characterization of Cu(In,Ga)Se2 (CIGS)-based thin film bifacial solar cells using Al-doped ZnO (AZO) as cost-effective and non-toxic transparent back contact. We show that, by depositing both CIGS and AZO by Low Temperature Pulsed Electron Deposition at a maximum temperature of 250 °C, a good ohmic contact is formed between the two layers and good quality solar cells can be fabricated as a result. Photovoltaic efficiencies as high as 9.3% (front illumination), 5.1% (backside illumination) and 11.6% (bifacial illumination) have been obtained so far. These values are remarkably higher than those previously reported in the literature. We demonstrate that this improvement is ascribed to the low-temperature deposition process that avoids the formation of Ga2O3 at the CIGS/AZO interface and favours the formation of a low-resistivity contact in agreement with device simulations.

Recently, dynamic SIMS (D-SIMS) is being used to analyze ions simultaneously over a wide range of concentrations, from matrix level to extremely low (ng g−1). However, D-SIMS detectors, which are mostly used in pulse counting systems, have problems with detector saturation. Thus, the ion intensities of SIMS must be corrected in their high-intensity region. It has been reported that the approximation intermediate extended dead-time model (a-IED model) can be used to correct saturated intensity in the higher region better than can the conventional model. In this study, we evaluated the usefulness of the a-IED model to correct partially saturated and saturated intensities for magnetic-sector-type D-SIMS and quadrupole-type D-SIMS for the effects of dead time. Nine organizations from five countries participated in this project. Analyzed specimens were an arsenic-implanted silicon wafer and a diffused BN onto Si wafer. The instruments used to analyze arsenic-implanted samples were five quadrupole-type SIMS and four magnetic-sector-type SIMS. The instruments used to analyze the BN-diffused samples were three quadrupole-type SIMS, four magnetic-sector-type SIMS, and one time-of-flight-type SIMS. We validated the usefulness of the a-IED model to correct saturated intensities for all SIMS in this round-robin test. The optimum extension parameter ρ tends to be affected by the ratio of the maximum reliable intensity to the maximum intensity. Copyright © 2014 John Wiley & Sons, Ltd.

With the continuously increasing demand for traceable and often accredited measurements in the scientific community, the quality management of laboratory activity according to international ISO standards can be a valuable and proficient solution. In this paper, we describe a decade of work in this direction. The implementation of quality management according to the ISO/IEC 17025 ‘General requirements for the competence of testing and calibration laboratories’ in a surface science research-oriented analytical laboratory is presented. We analyze difficulties and drawbacks as well as benefits and permanent improvements that followed the accreditation, its maintenance, and extension to new methods. Copyright © 2014 John Wiley & Sons, Ltd.

In the present work diffusion of implanted nitrogen in germanium is studied as a function of implantation energy and post-implantation annealing temperature. Implantations have been performed at a constant dose and energies leading to the amorphization of the substrate. Nitrogen diffusion in Ge is anomalous (not obeying the 2nd Fick's law) towards the substrate surface on which an Al2O3 layer has been deposited before annealing for protection. There is evidence that N diffusion in Ge is interstitial-assisted and takes place in the presence of Ge interstitials gradients between the Al2O3/Ge and the former amorphous/crystalline (a/c) interface of the substrate during annealing. (© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

High-k dielectrics as SrxTi1-xOy (STO) are of great interest for the development of dynamic random access memory devices. The characterization of these nanolayers is important. Secondary ion mass spectrometry (SIMS) depth profiling through TiN and STO films is affected by strong artifacts in different ways. The erosion process causes surface topography modifications both in the TiN/STO/TiN layer system and in the silicon substrate. Atomic force microscopy analyses have been carried out on pristine TiN and STO film surfaces and at various crater depths. Very different roughness evolutions are identified for TiN or STO films, within the SIMS craters sputtered with 500 eV Cs+ while using eucentric stage rotation. Copyright © 2012 John Wiley & Sons, Ltd.

In this work we demonstrate the fabrication and characterization of high performance junction diodes using annealing temperatures within the temperature range of 300-350?C. The low temperature dopant activation was assisted by a 50nm platinum layer which transforms into platinum germanide during annealing. The fabricated diodes exhibited high forward currents, in excess of 400A/cm2 at ~|0.7|V for both p+/n and n+/p diodes, with forward to reverse ratio IF/IR greater than 104. Best results for the n+/p junctions were obtained at the lower annealing temperature of 300?C. These characteristics compare favorably with the results of either conventional or with Ni or Co assisted dopant activation annealing. The low-temperature annealing in combination with the high forward currents at low bias makes this method suitable for high performance/low operating power applications, utilizing thus high mobility germanium substrates.

Ultra-shallow (below 20 nm) disorder profiles have been characterized by spectroscopic ellipsometry (SE). The implanted depth region has been divided into sublayers with dielectric functions calculated by the effective medium approximation using single-crystalline and disordered components. The damage depth profile has been parameterized using a box model, an independent multilayer model, a graded multilayer model, an error function, and Gaussian profiles. Literature values and Tauc-Lorentz (TL) parametrization as well as multi-sample and single-sample approaches have been compared to describe the dielectric function of the disordered component. The distribution of the implanted ions and/or damage have been cross-checked using medium energy ion scattering (MEIS), transmission electron microscopy and Monte Carlo simulations. We found a good agreement in the damage profiles obtained by the different methods. There is an offset between the SE and MEIS damage profiles due to the fact that SE is very sensitive to the surface roughness, in contrast to MEIS. The correlation between this offset and the surface roughness has been investigated using atomic force microscopy.

The last decade has seen considerable interest in the use of germanium for high-speed low-power electronics. Despite the demonstration of high performance p-channel Ge transistors in planar and non-planar device technology, fabrication of n-channel Ge transistors faces a number of scientific and technological challenges which hinder the development of CMOS logic circuits based entirely on Ge. Major challenge towards the establishment of Ge deep submicron n-channel MOSFETs technology is the control of the fast vacancy-mediated n-type dopant diffusion in Ge, which prevents the formation of shallow (xj20cm-3) n+/p junctions [1,2]. Additional issues in meeting this target are constituted by: (i) the significant simultaneous out-diffusion trend of n-type dopants and (ii) the Ge thermal etching during post-implantation annealing, which causes significant Ge substrate loss at temperatures above 500oC, thus imposing the presence of a capping layer on Ge surface [3]. Up to now, a variety of approaches have been followed to suppress n-type dopant diffusion in Ge, the most representative being [1,2]: (a) co-implants with impurities such as carbon, fluorine or nitrogen (b) co-implants of two n-type dopants (e.g arsenic with antimony) (c) bulk interstitial injection by Ge self-implantation or GeOx clusters dissolution and (d) high temperature-short time non-conventional annealing schemes (Flash Lamp Annealing, Laser Annealing, Microwave Annealing). More recently, the ability of surface capping layers to modify the near surface Ge vacancy concentration has been demonstrated [4]. This revealed “capping layer engineering” as an additional supporting technique to control n-type dopant diffusion in Ge. During the last years, our work in the field has been devoted to the beneficial action of co-implanted nitrogen in suppressing low energy-implanted phosphorous in and out-diffusion in Ge [4-6]. The action of nitrogen is based on its individual anomalous diffusion [Figure(1)] and the simultaneous formation of complexes with phosphorous atoms [Figure(2)]. The aim of this work is to review the field comparing the above mentioned approaches, with particular emphasis on the action of nitrogen as phosphorous diffusion blocker. A series of experiments performed in low energy/high dose N2/P co–implanted (100) Ge as function of the post-implantation annealing scheme and temperature, the surface capping layer stoichiometry and the level of simultaneous interstitial injection in the substrate will be presented and discussed. The experiments aspire to reveal the microscopic origin of nitrogen anomalous diffusion in Ge, the role of nitrogen in suppressing the damage caused in Ge substrate by non-conventional annealing schemes such as FLA [Figures (3)-(4)], as well as processing issues, limitations and strategies towards the formation of shallow and highly activated n+/p junctions in Ge based on N/P co-implants. References: [1] E. Simoen, M. Schaekers, J. Liu, J. Luo, C. Zhao, K. Barla and N. Collaert, Phys. Status Solidi A 213, No. 11, p. 2799–2808, (2016) [2] C. Monmeyran, I.F. Crowe, R.M. Gwilliam, J. Michel, L.C. Kimerling, A.M. Agarwal, Internat. Mat. Rev., http://dx.doi.org/10.1080/09506608.2016.1261223, (2016) [3] N.Ioannou, D.Skarlatos, C.Tsamis, C.A.Krontiras, S.N.Georga, A.Christofi, D.S. McPhail, Appl. Phys. Lett., 93, art.101910, (2008) [4] D.Skarlatos, V.Ioannou-Sougleridis, M.Barozzi, G.Pepponi, D.Velessiotis, M.C. Skoulikidou, N. Z. Vouroutzis, K. Papagelis, P. Dimitrakis, C. Thomidis, B. Colombeau, ECS J. of Solid State Sci. and Technology., 6 (7), p.418, (2017) [5] D.Skarlatos, M.Bersani, M.Barozzi, D.Giubertoni, N.Z.Vouroutzis and V.Ioannou-Sougleridis, ECS J. of Solid State Sci. and Technology., 1 (6), P315, (2012) [6] C. Thomidis, M. Barozzi, M. Bersani, V. Ioannou-Sougleridis, N. Z. Vouroutsis, B. Colombeau, and D.Skarlatos, ECS Solid State Lett. 4, p.47, (2015). Figure 1

Germanium (Ge) surfaces have been irradiated with 26 keV gold (Au) ions at a constant fluence and at incidence angles varying from 0° to 85°. The evolution of the emerging nanostructures is studied by atomic force microscopy (AFM), scanning electron microscopy, x-ray photoelectron spectroscopy (XPS), and cross-sectional transmission electron microscopy. The obtained results are compared with findings reported in the literature. Periodic rippled patterns with the wave vector parallel to the projection of the ion beam direction onto the Ge surface develop between 30° and 45°. From 75° the morphology changes from parallel-mode ripples to parallel-mode terraces, and by further increasing the incidence angle the terraces coarsen and show a progressive break-up of the front facing the ion beam. No perpendicular-mode ripples or terraces have been observed. The analysis of the AFM height profiles and slope distributions shows in the 45°-85° range an angular dependence of the temporal scale for the onset of nonlinear processes. For incidence angles below 45°, the surface develops a sponge-like structure, which persists at higher incidence angles on the top and partially on the face of the facets facing the ion beam. The XPS and the energy-dispersive x-ray spectroscopy evidence the presence of Au nano-aggregates of different sizes for the different incidence angles. This study points out the peculiar behavior of Ge surfaces irradiated with medium-energy Au ions and warns about the differences to be faced when trying to build a universal framework for the description of semiconductor pattern evolution under ion-beam irradiation.

Nanocrystalline silicon embedded in dielectric matrices is currently studied for Si-photonics, memory devices and solar cells. A common method for the preparation of silicon nanocrystals embedded in oxides is the phase separation of silicon rich oxide (SRO) in SiO 2 and Si via thermal annealing. Phase separation, nucleation and crystallization of SRO are known to depend on the thickness of the SRO layer. Here we investigate the structural changes in a sample consisting of alternated nanometer-thick SRO and SiO 2 layers - a so-called superlattice (SL)-during thermal annealing. Under a thermal treatment the material undergoes a series of modifications due to sintering, phase separation, crystallization and layer mixing. In this work we investigate these transformations in an SL grown by plasma enhanced chemical vapor deposition (PECVD) with several analytical techniques: XPS, variable angle ellipsometric spectroscopy (VASE), SIMS, and X-ray reflectivity (XRR). Both SIMS and XRR measurements clearly reveal the periodicity of the samples. XPS analysis reveals that phase separation of SRO in silicon and SiO 2 occurs in the annealing temperature range 600-925 °C. The process is accompanied by reduction in overall thickness of the samples (ongoing also at higher temperatures) as evidenced from the ellipsometric spectra. A maximum form birefringence is achieved at 925 °C and stays nearly constant until 1100 °C. Eventually, the form birefringence decreases at the highest annealing temperature of 1150 °C, which according to SIMS measurements is caused by a partial oxidation of silicon in the outermost SRO layer.

Nanocrystalline semiconductors embedded in dielectric matrices are currently under investigation for use in Si-photonics and in memory devices. The aim of a joint research activity in the FP6-ANNA*) project (http://www.i3-anna.org) is to develop and improve metrologies for the measurement of nanocrystal properties.

Silicon nanocrystals (Si-ncs) embedded in a dielectric matrix (e.g. Si 3 N 4 , SiO 2 ) attract significant interest for their use in silicon-based devices such as memories, light amplifiers, and light emitting diodes. Si-ncs in an SiO 2 matrix are generally obtained through thermal annealing of silicon-rich oxide (SRO) at elevated temperatures (900-1200 °C). In this study, we have investigated the composition of annealed SRO layers close to the interfaces with Si 3 N 4 , or c-Si with XPS, SIMS, and atomic force microscopy (AFM). The analyses indicate an approximately 2-nm thick Si-depleted region close to both the Si 3 N 4 and c-Si interface formed after annealing at 1050 °C. The formation of this Si-depleted layer is of importance for both electrical transport and charge storage in annealed SRO.

High concentration dopant distributions in silicon like those required to form ultra shallow junctions can affect SIMS analyses introducing matrix effects on secondary ions, otherwise not observed in more dilute regimes. In this work the effect of high arsenic concentration on negative secondary ion yield has been investigated when sputtering with a 0.5 keV Cs+ primary beam in a magnetic sector instrument. Two effects have been distinguished: the first one is a variation of matrix signals observed on as implanted samples in correspondence with the projected range, probably due to the damage induced by the high dose implant; the second is a step observed for several species and related to the arsenic distribution. The latter seems due to the effect of the active dopant on secondary ion energy distributions. A wide and well-centred energy slit can limit this effect allowing a more accurate quantification of the activated high As concentrations.

SIMS and medium energy ion scattering (MEIS) have been applied to the characterization of ultra shallow distribution of arsenic in silicon obtained by ion implantation at 1 and 3 keV and successive annealing at low temperature (lower than 700 °C). In case of heavy elements in light matrices, the MEIS ultimate depth resolution and its ability to obtain quantitative results from first principles result in a good reference for SIMS depth profiling. The comparison of the results obtained by the two techniques allows to discriminate among different SIMS quantification processes in order to individuate the best in terms of accuracy in the initial transient width and at the SiO 2 –silicon interface: the simple normalization of 28 Si 75 As − curve to the average of 28 Si 2 − results in the best agreement between SIMS and MEIS in the surface region. Moreover SIMS profile of the 3 keV as implanted sample resulted 1.9 nm shallower than correspondent MEIS profiles whereas samples annealed at either 650 °C or 700 °C for 10 s show a good alignment of the As segregation peak at the SiO 2 /Si interface. The sample annealed at 550 °C for 200 s shows a reduced shift between SIMS and MEIS measured As peak: a possible effect of the residual amorphous layer on the sputtering rate is pointed out as responsible of these different shifts.

Surface modifications and etching mechanisms of several nanoporous low-k dielectrics (spin-on and PECVD) using fluorocarbon plasmas have been investigated by ToF-SIMS and AFM and compared with those of SiO2. The results show a fluoropolymer film growing on the materials using highly polymerizing discharges (C4F8). The fluoropolymers are converted into fluoroether-like compounds upon etching low-porosity dielectrics in ion-rich C4F8/90% Ar plasmas. These layers mitigate the influx of plasma species and inhibit the etching rate. No surface roughness develops for these conditions, keeping the plasma/material interaction regime stationary. On the other hand, the surface coverage by fluoroethers is reduced when the porosity exceeds a given threshold. Consequently, direct plasma/dielectric interactions including ion bombardment take place, causing an increase of the etching rate, surface roughening and severe modifications of the pristine dielectric.

The characteristics of silicon-nitride films deposited with plasma-enhanced chemical vapour deposition (PECVD) can vary not only in composition, but also in terms of physical properties. An important parameter in the fabrication of micro-electro-mechanical systems (MEMS) devices is the control and tailoring of the film stress. In fact, silicon-nitride film stress can vary from tensile (up to 800 MPa) to compressive (-500 MPa), depending on the PECVD deposition conditions (e.g. plasma pressure, plasma frequency and power). In this work, we have investigated the role of frequency mixing on the resulting film structure and stress. The layers were prepared with a commercial PECVD deposition chamber, keeping all deposition parameters constant except the ratio between low (380 kHz) and high (13.56 MHz) frequency. Film stress values were compared with compositional and chemical data from X-ray photoemission spectroscopy (XPS) and secondary ion mass spectrometry (SIMS) measurements. These two techniques are complementary. The comparison between quantified SIMS depth profiles and the XPS HF etch-back method shows a good agreement in nitrogen profile behaviour. In addition, XPS analyses have provided a suitable chemical characterisation of the different growth processes. We demonstrate that this analytical approach is really effective for the physico-chemical characterisation of PECVD grown silicon nitride.

We investigate the effect of the pre-amorphisation damage on the structural properties, and dopant diffusion behaviour of indium and carbon co-implanted layers in silicon. Ion implantation of indium and carbon in silicon was used to produce co-implanted specimens. Rutherford Backscattering Spectroscopy and Secondary Ion Mass Spectroscopy have been performed on as-implanted and annealed samples to assess in detail the structural properties of the doped layers and the diffusion behaviour. The results have been compared with data obtained for similar implants performed into crystalline silicon to achieve a deeper understanding of the mechanisms driving the diffusion of the indium in silicon in presence of co-implanted species. In particular a reduction of the indium diffusion and a saturation level for the indium substitutional retained dose were observed.

Due to the sensitivity, the good depth resolution and the great interest in ultra shallow profile, secondary ion mass spectrometry (SIMS) is one of the prime techniques used in the semiconductor industry. Low impact energy beams are required to profile shallow distributions. Since Cs þ beam sputtering can cause morphological artifacts as well as O2 þ beam does, a detailed study is required to understand development and limiting analytical conditions. In this work we analyzed the effect of low energy Cs þ primary beam incident at 688 and 788 on different silicon samples. By using atomic force microscopy (AFM) and scanning electron microscopy (SEM) we underline their reliability and correlate the morphological effects to the SIMS analytical parameters and samples characteristics. # 2004 Elsevier B.V. All rights reserved.

In the present work the effect induced by a thin silicon oxide and its interface with silicon on the quantification of ultra-shallow arsenic distributions measured by secondary ion mass spectrometry (SIMS) has been studied. Three As implants on a 11 nm SiO2/Si stack have been measured: the projected ranges of the arsenic implants were chosen to result in the implants being in oxide, at the SiO2/Si interface and just beyond the interface, respectively. The measurements have been performed by different analytical methodologies using Cs+ primary ions at varying impact energies and collecting the molecular species SiAs−. Different methods of normalization have been studied and the effect of the oxide on RSF variation and dose measurement has been discussed.

In this article, we investigate the thermal growth of SiO2 films on 6H-SiC preamorphized by Ar+ ion implantation. The experimental conditions to grow oxide layers on SiC with high oxidation rate and low temperature have been determined. Rutherford back scattering channeling, secondary ion mass spectrometry, and x-ray photoelectron spectroscopy analyses have been used to investigate the oxide characteristics and composition. During wet oxidation at 800 degreesC, a linear growth rate of about 0.7 nm/min has been found for layers as thick as 240 nm. The presence of a SiO2Cx transition region between a stoichiometric SiO2 layer and the SiC substrate has been evidenced. The thickness of this transition layer increases with the oxidation time up to reach 118 nm for 390 min.

The higher degree of instrumental automation becomes increasingly important in order to reduce the operator attendance for SIMS analyses. Concurrently a high repeatability is demanded from these instruments to guarantee process reproducibility or material uniformity. This work is focused on the depth profiling repeatability obtained from the new magnetic sector SIMS instrument CAMECA SC-Ultra 300 for arsenic, boron and phosphorous implants in silicon. The short-term repeatability has been evaluated by performing a continuous sequence of automatic measurements: RSF repeatability better than 1% has been obtained over a single day. Additionally the long-term repeatability has been evaluated by repeating several weeks later the same measurements with the instrumental presets stored by software, i.e. without changes in the analytical parameters. The RSF repeatability values are about 3% for As with cesium source and 5% for B with oxygen primary beam.

To realise gate dielectrics in the present ULSI technology ultra thin oxynitrides are used. Nitrogen quantitative depth profiles are mandatory to characterise this material. Depth resolution is the key analytical parameter to obtain useful ultra thin oxides characterisation. To improve this resolution very low primary ion impact energy is required. In the present paper, we compare quantitative depth profiles carried out by dynamic-SIMS and ToF-SIMS, respectively. Dynamic-SIMS analyses have been performed using a Cameca 4-f and the new Cameca Sc-Ultra 300 instrument. Different impact energies and incidence angles were used in combination with MCs+ ion monitoring. The D-SIMS profiles at keV and sub keV primary beam impact energies, are discussed and compared with ToF-SIMS data obtained using a IONTOF IV instrument.

Preliminary total cross section data for positron scattering on argon and krypton between 250 and 5000 eV are presented. A Ramsauer-type apparatus, previously used for electron-scattering measurements has been modified. To check for the convergence between electron and positron total cross sections, a semiempirical analysis has been applied. The analysis for argon indicates an asymptotic convergence for electron and positron projectiles – the high energy parameters of the both fits are equal within the uncertainty limits. No such a clear conclusion is possible with the present data up to 5000 eV for krypton.

Thin oxides deposited by low-pressure chemical vapour deposition (LPCVD) at high temperature are being widely investigated in VLSI technology to be used as possible substitutes for the usual thermal oxides. High-temperature oxide deposition involves reactions between N 2 O and dichlorosilane or silane. Considering possible active dielectric applications, the effect of further nitridation and annealing of deposited films has to be considered. Evaluation of the nitrogen content and the distribution profile of these dielectrics then becomes a relevant analytical issue. Moreover, it is also crucial to monitor the amount of possible contaminant species, such as chlorine, introduced by the precursor gases during the deposition. In this work we present the development of an analytical methodology for quantitative depth profiling of chlorine in thin oxides using SIMS and XPS. In addition, the evolution of chlorine distribution in an appropriate set of high-temperature oxide films submitted to nitridation and further thermal processes will be shown.

In the last years a lot of effort has been directed in order to reduce ion implantation damage, which can be detrimental for silicon device performances. Implantation's dose rate and temperature were found to be two important factors to modulate residual damage left in silicon after anneal. In this work high dose rate, low temperature, high dose arsenic and boron implantations are compared to the corresponding low dose rate, room temperature processes in terms of silicon lattice defectiveness and dopant distribution, before and after anneal is performed. The considered implant processes are the one typically used to form a source/drain region in a CMOS process flow in the submicron technology node. A spike anneal process was applied to activate the dopant. Low temperature, high dose rate implantations have found to be effective in reducing silicon extended defects with a negligible effect on the profile of the activated dopant. Experimental set up, results and possible explanation will be reported and discussed in the paper.

An apparatus for very low-energy positron scattering measurements in gases is presented. The apparatus uses electrostatic optics, two stages of positron moderation, magnetic focusing in the scattering chamber and a pseudorandom time-of-flight correlation technique. ” 2000 Elsevier Science B.V. All rights reserved.

Optimization of oblique incidence ultra low energy O2+ SIMS analysis of ultra shallow boron distributions has been investigated varying the atmosphere in the analysis chamber (ultra high vacuum or oxygen flooding) and evaluating the effect of a rotating stage allowing a 20 rpm rotation during the analysis. The impact of the different analytical approaches to the ripple formation on the crater bottom has been investigated on a boron delta doped silicon sample by AFM analysis. The combined use of a 0.5 keV O2+ beam at 68° of incidence with oxygen flooding and stage rotation of 20 rpm gave a decay length of 2.0 nm/decade at 60 nm depth without any appreciable detection of variation of sputtering rate.

In most cases sufficient sample quantity is available to cut the sample to fit the available window size of SIMS sample holders. However, in cases where only a small amount of material is available analysis of small ( 1 mm ×1 mm) samples is required. Some typical examples can be SiC specimens or decapsulated Si devices. The edge of a typical sample holder presents a large step, where the electric field lines are distorted and the primary/secondary beams alignment and measurement precision is dramatically affected. This edge effect becomes critical if the sample dimensions are comparable to the thickness of the sample window. For SIMS analyses, it is very important to have a flat surface in the analyzed and surrounding areas, therefore we have fabricated a special sample holder for very small specimens. A new mask has been designed and fabricated in which it is simple to analyze small sample sized and the precision of the sample holder has been tested showing good data repeatability.

We report the effect of neodymium (Nd) doping of TiO2 thin films produced on quartz, soda lime and silicon substrates by RF co-sputtering of two targets of TiO2 and metallic Nd. The Nd/Ti ratio was varied in the range of 0.9–100 at.%. A study based on a joint use of Auger electron spectroscopy and X-ray fluorescence analysis allowed us to define the chemical composition of all the deposited films. A piece of direct evidence of the structural changes (phase transformation from anatase to rutile for Nd/Ti atomic ratio up to 43 at.%) and amorphization for an Nd/Ti ratio equal to 100 at.%, a lattice distortion due to Nd incorporation and also the rare-earth ion location site was obtained by X-ray diffraction studies, in correlation with the optical properties analysis. NIR photoluminescence properties of the doped films in the wavelength range of 800–1200 nm were investigated by direct (green, blue-green and blue lines at 514.5, 496 and 476 nm respectively) and indirect excitations of Nd (using the UV line at 355 nm), in the latter case excitation through the titania host. In both cases, an intense NIR photoluminescence was observed at 907 and 1093 nm pertaining to the transitions between the excited 4F3/2 to 4I9/2 and 4I11/2 levels of the Nd3+ ions for the lowest ion concentration (Nd/Ti ratio equal to 0.9 at.%) and a correlation between the phase structure of the matrix and Nd concentration and the emission properties was found.