Search

The isotopic comparative method (ICM) has been used to measure the concentrations and relative ion yields of Si+, Ge+ (Si-, Ge-) in SiGe alloys ranging almost from pure silicon to pure germanium by secondary-ion mass spectrometry under Ar+ bombardment at 8 keV, ~39° incidence, (17 keV, ~25°). This method requires only two specific ICM reference samples to be fabricated. The negative-secondary-ion yields YR−(Si) and YR−(Ge) are found to be nearly constant over the entire range of SiGe alloy. The measurement conditions are free of matrix effects, thereby allowing accurate and easy quantification of SiGe alloys. In contrast, the positive-secondary-ion yield YR+ (Si) decreases strongly over the full range of SiGe alloy, whereas YR+(Ge) varies only slightly for cGe

With the isotopic comparative method (ICM), we determined the ion yields of B+, O+, Si+, B−, O−, and Si− versus oxygen concentration in silicon under Ar+ bombardment. By multi-energy implantation, we obtained samples containing a near-uniform low concentration of 18O, which we then implanted with a high dose of 16O. The ICM allows us to determine the oxygen concentration profile as a function of substrate depth during analysis by secondary-ion mass spectrometry, following which the ion yield is measured as a function of oxygen concentration from 0 to 33 at.%. The results show that, whatever the charge of secondary ions, all relative ion yields are enhanced with increasing of oxygen concentration. Very strong matrix effects due to oxygen are found for B+ (>100), whereas moderate or slight matrix effects are found for Si+ and O− (>10) and for Si−, B−, and O+ (

Specific samples containing 18 O and 16 O are used to measure the variations of the relative ion yields of boron, oxygen and silicon as a function of oxygen concentration. 18 O and 16 O are used to implement an Isotopic Comparative Method (ICM) which allows to correct the matrix effects involved by the presence of a high concentration of oxygen in the sample: the near-flat profile of 18 O, measured in the 'dilute', linear regime (weak concentration) is used to calculate the real concentration of 16 O. The ion yields of B + , O + , Si + , O - and Si - are measured as a function of the oxygen concentration. For B + ion yield, the variation is important whereas they are weak for Si ± and O ± ion yields for the range [0-12 at.%]. This ICM applied to oxygen in silicon can be considered as an interesting complementary method of previous 16 O implantation method' and of 18 O single marker method'.

This work extends the Isotopic Comparative Method, which has been previously introduced with the aim of correcting matrix effects, to quantify high concentrations of boron in silicon up to 40 at.% by SIMS technique. It requires a specific sample containing two isotopes of the element to quantify: for this study silicon sample with isotope 10 B present in constant, sufficiently dilute concentration, quantifiable with the conventional Relative Sensitivity Factor method and the other isotope 11 B implanted at very high concentrations. The intensity of 10 B, whose concentration ratio with respect to silicon is known and constant, allows to measure the matrix effects induced by the presence of 11 B and to deduce the total boron concentration. The variations of the ion yields of boron Y B + and silicon Y Si + as a function of the boron concentration C B are investigated up to 40 at. %. It is found that under Ar + sputtering there are no matrix effects on Si + ions, but a doubling of Y B + . Under O 2 + , Y B + and Y Si + increases almost linearly and similarly up to 2.5 times their value in dilute regime. With saturated O 2 flooding under Ar + or O 2 + , Y B + and Y Si + increase similarly and weakly. These results allow defining an Extended Relative Sensibility Factor method for the quantification of concentrated boron profiles in silicon. It has been successfully implemented in a ~5 at.% boron layer obtained by Gas Immersion Laser Doping. ICM has been implemented for quantifying high dose boron implantation in silicon (C B up to 40 at.%).

The continued reduction in the scaling of semiconductor components has pushed the technique of SIMS to achieve unprecedented degrees of high depth resolution. For this purpose, various deconvolution methods have been developed during the last decade, among which is the iterative algorithm with hard constraints and Miller regularization, the heart of this contribution. In this paper, a comparative study in both cases of considering and neglecting the regularization term characterizing this algorithm is provided and analyzed. Both algorithms are tested on several theoretical structures and implemented for the deconvolution of a real structure. It was shown that the algorithm without the regularization term provides regular solutions close to the results given by the basic algorithm using this term, with a reduction in the number of iterations. Besides, gains in depth resolution and in maximum peak intensity without the regularization term are better than those achieved using it. These features may favor the first approach, and attribute it more considerations in SIMS. However, the proposed approach in cases of Gaussian and thick-layer profiles has the semiconvergence property. In this work, the behaviour of the deconvolved profiles as a function of the uncertainties in the depth resolution function (DRF) fit is also addressed.

Self-aligned barriers are widely investigated either in replacement of dielectric liners to decrease the total interconnect k value or as a treatment prior standard dielectric barrier deposition to improve reliability performances. In this paper, a technique based on the modification of the Cu surface is proposed. It consists first in removing native Cu oxide, then, enriching Cu surface with Si atoms followed by a nitridation step to complete the so called CuSiN self-aligned barrier. The dependency of Cu surface silicidation on Cu crystallographic orientation is described, evidencing two silicidation mechanisms: Si interstitial incorporation into Cu and Cu atoms substitution by Si atoms. He diluted in H"2 cleaning plasma prior to silicidation is demonstrated both to decrease Cu grain surface ability to silicidation compared H"2 plasma and to limit Si incorporation into Cu at grain boundaries. Compared to a standard SiCN barrier, CuSiN self-aligned barriers integrated as a treatment prior to SiCN evidenced at least four times longer lifetime under electromigration tests.

In this paper, the deconvolution of SIMS profiles analysed at very low primary energy (0.5 keV/O 2 + ) is addressed. The depth resolution function (DRF) of the SIMS analysis in presence of roughness is established and a deconvolution procedure is implemented without or in presence of roughness on samples containing delta-doped layers of boron in silicon. It is shown that the deconvolution procedure can lead to a great improvement of the full width at half maximum (FWHM) of the measured peaks in the case where no roughness in detected in the crater bottom. In the case where it is present, the conditions required to use a deconvolution procedure are discussed, and the deconvolution is implemented using precise and restrictive assumptions.

In this paper, the SIMS beam induced roughness (BIR) in monocrystalline Si in presence of initial surface or bulk defects of nanometric size is studied. We follow the development of the BIR by monitoring the increase of Si2+ and SiO2+ signals during SIMS sputtering. The topography of the crater bottoms is measured at different steps of the evolution of the roughness using an atomic force microscope (AFM). We show that in presence of nanometric sized defects on the surface or in the bulk, the BIR develops far more rapidly than usual. It appears as soon as the crater reaches the defects and, as reported on Si free from any treatment, the same morphology evidencing waves perpendicular to the sputtering beam develops rapidly. This study of the behaviour of the BIR in presence of voluntarily introduced defects allows us to better understand the basic physical phenomena involved in its apparition.

Self-aligned barriers have been widely investigated in the replacement of standard PECVD dielectric liners to decrease coupling capacitance. As an alternative to CVD or electroless approaches, a two step process based on the modification of the Cu surface is proposed. This technique consists first in enriching Cu surface with Si atoms followed by a nitridation step to complete the so called CuSiN self-aligned barrier. In this paper, the silicidation mechanism is described and the crucial role of the nitridation step is demonstrated in terms of barrier stability under electrical stress. CuSiN efficiency against Cu diffusion and oxidation is also evidenced. Compared to a standard SiCN barrier, CuSiN self-aligned barriers revealed a gain of at least 3 decades in time-dependent dielectric breakdown lifetime and up to 7% decrease in intra-level coupling capacitance.

The oxygen ion-beam-induced surface roughening observed during SIMS depth profiling of both Si(1 0 0) and 10° disoriented Si(1 0 0) surface without flooding has been studied by atomic force microscopy (AFM). Facets have been created at the crater bottom and parameters such as RMS (root mean square) roughness, wavelength of the waves and orientation of the facets have been measured. The ripple amplitude increases with the depth of erosion from a critical depth of approximately 2.5 μm for O2+ (10–4.5 keV) bombardment. In this paper, we investigate the surface orientation dependence of the roughening in Si(1 0 0) and 10° disoriented-Si(1 0 0). The same transition phase of secondary ion intensities during monitoring of matrix secondary ions and a similar evolution of surface topography and RMS roughness have been observed on both kinds of samples.

A sample, composed of inverted boron deltas/SiO2/boron deltas/silicon on an insulator substrate (SOI), was analyzed using a Cameca IMS 5f and a Cameca IMS Wf, with 500 eV O2+, oxygen flooding, and an electron gun. To synthesize this “double delta doping” sample, two identical boron multi-deltas were grown on SOI wafers and molecularly bonded upside down, then one SOI substrate was removed. The quality of this sample was checked by TEM and AFM. From the boron deltas’ SIMS depth profiles, a comparative study of the two SIMS instruments was carried out by looking in detail at the depth resolution parameters. It was found that depth profiles acquired with both tools are very similar to those measured by TEM. Both tools separate B deltas 2 nm apart. However, the primary beam density is higher with the IMS Wf, allowing a two times faster analysis time than with the IMS 5f. This sample structure also allowed us to acquire in one measurement both the backside and the conventional front side depth profiles, therefore allowing the contribution from both the epitaxial growth and the contribution from the instrumental SIMS profiling conditions to be separated.

Atomic force microscopy (AFM) is used in tapping and contact mode in order to study the roughness created in the crater bottom during secondary ions mass spectrometry (SIMS) analysis in silicon, using O2+ primary ions without oxygen flooding. A characteristic topography is observed, where ripples develop with an increasing peak to peak height amplitude and a triangular section showing very flat facets. We use the conductive-AFM (C-AFM, also called tunneling AFM (TUNA) mode) to measure the current flowing through the oxide. We show that when a rather strong electric field is applied between the tip and the sample, one side of the ripples (facets hidden from the beam) exhibits a far more important leakage current than the other (facets facing the beam) at a given fixed voltage, suggesting that the corresponding areas are thinner or covered with a poorer insulator. This is a direct observation at a nanometer scale of the different nature of the materials covering both sides of the facets. The authors propose that over the studied surface (r.m.s. roughness≈30 nm), both facets are sufficiently oxidized to behave like an insulator.

To improve depth resolution in secondary ion mass spectrometry (SIMS) low energy and heavy primary ions can be used. We present results obtained with xenon, at low energies (down to 500 eV), on specially chosen samples (B delta-doping, SiGe–Si superlattices and Al/Ti multilayers) to evaluate the depth resolution enhancement in ultra-shallow implants and abrupt transitions. The depth resolution parameters extracted are compared with those found with O 2 + primary beam. Analyses of B delta showed an improvement for xenon at energies between 2 and 3 keV. Below 1 keV, sputter-induced roughness adversely affected the advantages of xenon. For SiGe–Si, O 2 + beam, without O 2 flooding remains the best choice to obtain a smaller decay length. On Al/Ti multilayers, depth profiles acquired with xenon are more suitable. This can be explained by a poor reactivity of Xe + with metallic target atoms.

Measuring depth resolution in secondary ion mass spectrometry is of crucial importance. In the case of boron in silicon, the depth resolution is frequently inspected by profiling molecular beam epitaxy (MBE) grown multi-delta-like structures. Recently, it has been shown that the quality of such samples is dependent on the growth parameters. In this paper, we show that samples grown by reduced pressure-chemical vapor deposition feature a lower dynamic range but a superior high intensity sharpness compared to MBE. Thanks to this, the depth resolution parameters reported here are excellent: the full-width at half-maximum is equal to 1.29 nm and the exponential decay length at the trailing edge to 0.64 nm for a 500 eV O2+ primary beam at θ=44° with some oxygen flooding. We are even able to discriminate between two B deltas only 1.71 nm apart.

Quantitative determination of dose and junction depth in the source/drain region of future CMOS devices (typically 50 nm for the contact and 25 nm for the extension) requires SIMS instruments capable of measuring dopant profiles with high depth resolution and low detection limits. In this work we investigate the analytical usefulness of monitoring MCs 2 + ions for semiconductor profiling of n-type dopants such as P, Ge or As with a primary (Cs + beam) impact energy of 1 keV on a SIMS Cameca IMS-5f instrument. The results obtained on ultrashallow implants in Si and SiO 2 indicate an improvement in the characterization by using the MCs 2 + technique; this technique has the advantage over the MCs + technique of providing higher useful yields for electronegative elements by up to 10 2 and can therefore be applied to perform more precise junction measurements in terms of concentration, dose, decay length and detection limit.

In this paper, the problem of the deconvolution of SIMS depth profiles is addressed. In particular, the hypotheses that are necessary for the deconvolution to be possible (in the actual state of the art) in the case of the SIMS signal are reviewed. Then, the principle of regularization, which is a mandatory step in the resolution of an ill-posed problem, is clarified. Two regularization methods used in the field of SIMS analysis are compared: Miller regularization and maximum entropy regularization. In a second part the study of a possible deconvolution, using a depth resolution function (DRF) that is not the DRF that has experimentally convolved the profile, is justified and theoretically addressed. Two cases arise: the DRF used in the deconvolution process is either thinner than the experimental DRF or it is thicker. It is shown that deconvolution using a DRF that is wider than the DRF that actually convolved the profile is possible, and must be taken into consideration. Some examples of simulated deconvolutions with a false DRF are given, and some tools are proposed that are theoretically able to detect a problem when the DRF used in the deconvolution process is wider than the real DRF. In the last section, an example of experimental deconvolution shows that the deconvolution process is able to provide reliable information. In particular, deconvolution of a SIMS analysis conducted at 5.5 keV O2+ and 42.4° incidence (in a Cameca IMS 3/4f) reveals some features of the sample that require at least 1 keV O2+ and 60° primary beam incidence to be detected experimentally. © 1998 John Wiley & Sons, Ltd.

The analysis by secondary ions mass spectrometry (SIMS) is a powerful tool for the study of the diffusion of boron in silicon. It has been extensively used in order to determine diffusion coefficients. Most of the time, the SIMS profiles used to calculate the diffusion coefficients are sufficiently large so that the SIMS analysis do not modify the real concentration distribution in a significant manner. But more and more, the first steps of the diffusion are to be considered in order to obtain very thin diffused structures. In that case, where very thin layers are analyzed by SIMS, it is no more possible to measure directly the real concentration distribution because the SIMS analysis modify it rather significantly. When the real width or the real shape of the analyzed layers is needed, the measured profile has to be corrected in some way. We show that in the case of Gaussian original profiles, the SIMS profiles can be efficiently corrected, and the exact second order moment of the profile determined from the measured profile until the original second order moment is as low as 20 A. This can be done by the study of the properties of the convolution of the profiles by the SIMS analysis response function. In the case of non-Gaussian profiles, the real shape and width of the profiles can be determined by a deconvolution procedure that we have previously described. (B. Gautier, J.C. Dupuy, G. Prudon, J.P. Vallard, C. Dubois, Proceedings of SIMS X, The International Conference on Secondary Ion Mass Spectroscopy and Related Techniques, Wiley, Chichester, 1996, pp. 443; B. Gautier, R. Prost, J.C. Dupuy, G. Prudon, Surface and Interface Analysis 24 (11) (1996) 733). This procedure is applied to the case of the SIMS measurement of δ-doped layers of boron in silicon before and after rapid thermal annealing (RTA).

In this paper, an iterative algorithm is used in order to deconvolve some real and simulated SIMS profiles of boron-doped layers in silicon. The real SIMS profiles are obtained by the analysis of delta layers of boron-doped silicon in a silicon matrix, analysed in a Cameca IMS3/4f instrument at oblique incidence. It is shown that two layers that are completely convolved by the SIMS experiment can be completely separated. In order to try to solve the problem of reliability of the deconvolution results, a confidence level is defined, which indicates the level below which the deconvolved profile should not be taken into account. Moreover, a parameter defined previously is used to define whether a satisfying solution of the deconvolution problem exists or not. This parameter is applied to the experimental profiles. The simulated profiles are chosen so that they correspond to real cases encountered by the SIMS analysts and clarify what can be expected from the method. Several characteristic features are examined, such as the separability of Gaussian functions or the dynamic range of the deconvolution of rising or falling functions. It is shown that the minimum distance between two delta layers that can be separated using 3.5 keV/O2+ primary beam SIMS analysis at oblique incidence is equal to 60 A, and that abrupt error functions and exponential functions can be retrieved very satisfactorily. © 1997 John Wiley & Sons, Ltd.

The use of secondary ions mass spectrometry (SIMS) generally leads to very high sensitivity and depth resolution for silicon-based components. Low concentration elements such as dopants, are easy to quantify in silicon, whereas high concentration elements are subject to “matrix effects”, i.e. the intensity does not vary linearly with the concentration. Matrix effects are presently studied in SiGe alloys using two primary beams O 2 + and Cs + and a CAMECA IMS4F instrument. Rapid thermal chemical vapor deposition grown Si/Si 1− x Ge x /Si multilayer structures with x varying from 0 to 23.5% are preliminarily characterized by Rutherford backscattering spectrometry (RBS) in thickness and atomic composition. The linear variation of the SIMS intensity ratio I Ge + / I Si + with the RBS concentration ratio x /(1− x ) is confirmed for both O 2 + and Cs + beams. However, the linearity does not imply the absence of matrix effects. A variation of the ionization yields with Ge concentration and a similar behavior for τ Ge+ and τ Si+ actually lead to an exact compensation of the intensity ratio variation. At any energy, O 2 + experiments systematically over-estimate the real Ge concentration, whereas matrix effects under Cs + beam are weak. For polyatomic MCs + ions, strong matrix effects are observed in Si 1− x Ge x alloys. A quantitative analysis of major constituents and a dopant (boron) is used to determine the depth profiles of a thin base of a SiGe HBT heterostructure. Two different procedures are used depending on the primary beam. Consistent results are obtained for the concentrations of the major elements. The fully autonomous SIMS analysis with the Cs + beam is still penalized by the lack of sensitivity for boron in the secondary positive mode. © 1997 Elsevier Science S.A. All rights reserved.

Electron powder ribbon (EPR) plates elaborated from 99.9%–99.9999% pure silicon powder show crystallographic structure favourable to the achievement of nanocrystallites which are at the origin of luminescence. Fourier transform IR spectrometry and secondary ion mass spectrometry observations correlated with the comparative luminescence of porous layers produced on single-crystal and EPR polycrystalline silicon are carried out. The luminescence spectra are controlled by a combination of the electrochemical parameters and the starting material resistivity. An electroluminescent cell is elaborated and microcrystalline structures are obtained for sensor applications using the adsorption process (e.g. moisture measurements).

Microelectronic needs accurate analysis of very thin layers as delta doped structures. By secondary ion mass spectrometry (SIMS) with O2+ ion beam, we analyse mono-delta of boron in silicon located at 30 nm under the surface and then a multi-delta of boron in silicon located just under the surface ( ≈ 10 nm). For the mono-delta, we study modifications of the SIMS profiles within the projected range of primary ion beam Rp. From the dependence of FWHM within Rp, we deduce the real thickness of pseudo-delta thin layers by assuming a convolution process for SIMS measurements. This assumption is well confirmed by the analysis of a boron tri-delta doped layers.

We studied arsenic redistribution with annealing time in the SiW system during the formation of WSi2 by rapid thermal processing at 850 °C. The silicon was arsenic doped and the tungsten was deposited by sputtering. A significant amount of arsenic diffuses back and segregates at the WSi2Si interface, the remainder is incorporated in the disilicide. This behavior is quite different from that of boron, which diffuses simultaneously to the silicon through the disilicide layer.

Low-energy (3–100 eV) electrons are shown to be efficient promoters of the superficial nitridation of thin SiO2 films grown on boron-doped Si(100) using low pure NH3 pressures (⩽2 × 10−4 mbar) at ambient temperature. The initial concentration (≈1015 at. cm−3) of ionized acceptors in silicon near the SiO2/Si interface is first lowered by the nitridation process and then restored by hydrogen annealing at 450°C. The ultrahigh vacuum experimental set-up permits us to uncouple two important effects from a thermally-assisted (700–950°C) r.f. NH3 plasma nitridation process: the role played by low-energy (3–5 eV) electrons and the temperature. It is found that the existence of charge particles near the sample surface must be taken into account in the plasma reaction process and that the main effect of thermal activation is to enhance the reaction and the diffusion of nitrogen species through the SiO2 layer up to the SiO2/Si interface.

Low energy (3–100 eV) electrons are shown to be efficient promoters of the superficial nitridation of thin SiO 2 films grown on boron-doped Si(100) using low pure NH 3 pressures ( -1 mbar) at ambient temperature. The initial active boron concentration in silicon (≌10 15 cm -3 ) is first lowered near the SiO 2 /Si interface by the nitridation process and then restored by a hydrogen annealing at 450°C. The ultra-high vacuum experimental set-up permits us to uncouple two important effects from a thermally-assisted RF NH 3 plasma nitridation process : the role played by low energy (3–5 eV) electrons and the temperature. It is found that the presence of electrons near the sample surface must be taken into account in the plasma reaction process and that the main effect of thermal activation is to enhance the reaction and the diffusion of nitrogen species through the SiO 2 /Si interface.

Tungsten films were deposited on As- and B-implanted Si substrates by the Si reduction of WF6 at 300°C. The composition of the W/Si structures was investigated by secondary ion mass spectrometry (SIMS). Sputtered W/Si standard structures were used for calibration and to establish the procedure of the SIMS analyses. The depth concentration profiles of Si, W, As, B, O and F atoms in the W/Si structures were determined. These results corroborated the data obtained by Rutherford backscattering spectroscopy and nuclear reaction analyses. The WSi interfaces grown by the Si reduction of WF6 were found to be very rough. The SIMS profiles supported the growth mechanism of the W films based on the pinhole model proposed in the literature.

The very low lattice mismatch between the ErSi1.7(0001) and Si(111) allows a convenient epitaxial growth of the former on silicon substrates. However, the morphology of the films formed is found to depend greatly on the epitaxy techniques used. We have thus undertaken a study of the different possible techniques (single metal deposition, alternate evaporation and co-evaporation) so as to determine the procedure which gives silicide thin films of good quality. Best results are obtained by co-evaporation: the films are free of pinholes and present a flat surface. The formation of the silicide films was followed in-situ by LEED and XPS. The ErSi1.7 silicide is characterized by a sharp ( 3 × 3 )R30° LEED pattern which always appears after annealing at temperatures ⩾ 650°C whatever deposition technique is used. The Si KLL spectra, on the other hand, are found very appropriate to detect the presence of holes. The texture of the films was studied ex-situ by SEM. Glancing incidence X-ray diffraction and SIMS have also been used to get insights into the crystalline parameters and the in-depth composition of the silicide layers. The lattice of the epitaxial film is laterally extended to adapt to the Si(111) lattice and compressed axially. No evidence of ordered Si vacancies in the silicide volume is detected. The Er : Si ratio is the same throughout the film. The formation of ErSi1.7 is discussed in relation to the nucleation controlled reaction and the atomic structure of the compound.

International audience; We report on a detailed analysis of the superconducting properties of boron-doped silicon films grown along the 001 direction by gas immersion laser doping. This technique is proved to be a powerful technique to dope silicon in the alloying range 2-10 at.% where superconductivity occurs. The superconducting transitions are sharp and well defined both in resistivity and magnetic susceptibility. The variation of Tc on the boron concentration is in contradiction with a classical exponential dependence on superconducting parameters. Electrical measurements were performed in magnetic field on the sample with cB = 8 at.% (400 laser shots) which has the highest Tc (0.6 K). No hysteresis was found for the transitions in magnetic field, which is characteristic of a type-II superconductor. The corresponding upper critical field was on the order of 1000 G at low temperatures, much smaller than the value previously reported. The temperature dependence of Hc2 is very well reproduced by the linearized Gorkov equations neglecting spin effects in the very dirty limit. These measurements in magnetic field allow an estimation of the electronic mean-free path, the coherence length, and the London penetration depth within a simple two-band free electron model.

The direct reaction between tungsten and silicon (or chromium and silicon) to obtain tungsten disilicide from a thin metallic layer deposited on a silicon wafer was investigated. The metallic layer (W, or WCr) was sputtered on As- or B-implanted silicon. The samples were annealed under neutral atmosphere or forming gas in an annealing furnace (AF) or by rapid thermal process (RTP), and then analysed by secondary ion mass spectrometry (SIMS) and X-ray diffraction. For a fixed annealing time (AF: 30 min; RTP: 20 s), the temperature above which the tungsten disilicide was detected is 800 °C for RTP and 700 °C for AF, irrespective of the dopant type; this temperature was considerably lower when an intermediate chromium layer was deposited between tungsten and silicon. In all cases, the growth kinetics, i.e. disilicide thickness proportional to the square root of the time, were consistent with the mechanism of silicon atom diffusion.

Depth profiles of Fe-Ti multilayer structure composed of 29 alternate Fe and Ti layers (68 and 84 A thick) were obtained by secondary ion mass spectrometry under 5.5 and 3 keV O2+ bombardment. The effect of oxygen flooding on the in-depth resolution and matrix effects was investigated. Without oxygen flooding during ion bombardment, induced roughness has been developed. With oxygen flooding at 10−5 Torr, the layers are easily visible, the depth resolution is constant with the depth and the Ti signal is representative of the composition profile in the multilayer sandwich (no matrix effects on Ti intensities).

Silicon nitride films have been deposited by the process of mercury-sensitized photochemical vapor deposition (photo-CVD), using a gaseous mixture of SiH4 and NH3 under 253.7 nm ultraviolet light irradiation. The main advantages of this technique are the low deposition temperature (100–400°C) and no radiation damage. Deposition rate of 8 nm/min and refractive index of about 2 were achieved. Infrared spectrum, SIMS spectrum and electrical properties of the films, including dielectric constant, flat-band voltage, breakdown field and mobile ion density were investigated.

Silicon dioxide films were nitrided at atmospheric pressure in ammonia gas at 1100°C. The atomic transport mechanism of oxygen during the nitridation was studied by using isotopic oxides (18O labeled oxides). Physico-chemical characterization of the films was carried-out by nuclear reaction analysis (NRA), secondary ion mass spectrometry (SIMS) and ellipsometry. The nitrogen incorporation in the oxide film is accompanied by a decrease of the oxygen amount. It is shown that this decrease of the oxygen amount is due to an autodiffusion phenomenon. A good fit between the calculated and experimental curves is obtained for a self-diffusion coefficient D ∗ of 3 × 10-15 cm2 s-1.

Molecular beam epitaxy layers doped by low-energy implantation of arsenic —either during growth or during interruptions of growth (δ-doped layers)— have been characterized with respect to incorporation probability and electrical activity of the implanted arsenic, electron mobility and steepness of the doping profiles. Results show that (1) even at low implantation energy, incorporation probability is equal to unity and all the arsenic atoms are electrically active, (2) bulk mobility is obtained in layers doped during growthj and an interesting mobility enhancement is observed in δ-doped layers and (3) steepness of the doping profiles obtained is beyond the depth resolution of secondary ion mass spectrometry. Simulations of doping profiles indicate that doping gradients of 20 A per decade should be expected.

In this paper, we investigated the problem of the deconvolution of SIMS depth profiles analyzed at low primary energy (1keV/O+ 2 ), in particular in presence of roughness. This phenomenon leads to a variant depth resolution function (VDRF). To describe the DRF variations with depth, we have established a matrix slightly different from the classical Teoplitz matrix. Then by using the matrix formalism we give a generalization to the case of VDRF of an algorithm implemented initially in Fourier space The variant deconvolution procedure is first tested on simulated profile by using this algorithm and Van-Cittert algorithm with Miller regularization; we show a great improvement of the depth resolution and we observed that our modified algorithm is far better. The algorithm is then implemented on samples containing delta-doped layers of boron in silicon. It was shown that if the adequate estimation of the VDRF is available, we can obtain the best result.

A hybrid CoWP/SiCN Cu passivation was integrated in a three-metal-level interconnect stack at 65 nm technology node using a porous ULK material (K=2.5). 5 and 20 nm thick Pd-free CoWP electroless barriers were evaluated using a standard trench first hard mask architecture (TFHM) integration scheme, with PVD, ALD or punch-through Ta-based metallization processes. This study evidenced strong interaction between CoWP and etching chemistries, inducing feature size modification. Results evidenced the successful integration of an ultra-thin electroless barrier with slight process tuning, whereas thicker one still requires specific etch process development or integration scheme modification

International audience; We have measured the depth resolution function of the SIMS analysis of boron in silicon for different experimental conditions and fitted this function with an analytical expression initially proposed by Dowsett et ul. We use this analytical depth resolution function for the implementation of an iterative deconvolution algorithm, taking into account several properties of the signal, such as positivity and regularity. This algorithm is described precisely. The algorithm is tested on several theoretical structures and then implemented for the deconvolution of real structures. Of borondoped silicon layers in silicon. In particular, a sample constituted by six consecutive delta layers and a 75 A thick layer are deconvolved. It is shown that the asymmetry of the profiles is complftely removed and that the full width at half-maximum of the deconvolved delta layers can be reduced down to 41 A. It is also shown that a layer whose real thickness is smaller than the measured width of the resolution function can be easily distinguished from a delta layer, and its thickness estimated.

Fe-Al multilayers have been mixed with Xe + ions at high temperatures. The composition depth profiles have been analyzed by secondary ion mass spectroscopy (SIMS). It is shown that SIMS reveals the Al-rich compound formation inside the initial Al layers, and that this point cannot be questioned by exaltation effect on Fe + or Al + ion intensities. Phase formation has been proved by X-ray diffraction at grazing incidence

Boron redistribution during WSi2 formation from a boron implanted Si (100), W sputtering deposition, and RTP, is studied within the annealing time at 850°C by SIMS profiling. It is shown that boron diffuses in the same time as silicon through the disilicide layer and accumulates at the W/WSi2 interface.

We describe in this paper the use of MEM in ultra high vacuum to visualize surface potential, and to obtain projected shadow images and the work function, applied to a silicon (100) surface. We show that the work function does not depend on the n or p doping of the sample and, simultaneously, we show that it is possible to visualize surface potential of approximately 10 mV, the junction bias being ensured by the electron beam.

Surface nitridation of very thin SiO2 films (≈13 nm) is achieved at low ammonia pressures by thermal activation (900°C < T

The corrosion of some gold base alloys was studied in artificial saliva through the use of electrochemical methods, primarily potentiokinetic curves. The effects of various alloying elements were investigated.

Silicon (100) cleaned in an ultra-high vacuum chamber is characterized by Auger spectroscopy and Low Energy Electron Diffraction. Starting with a clean sample (diffraction pattern 2×1 for Si(100) and Auger spectra typical of 4-Si bonds), the oxidation is carried out in situ under low oxygen pressure. The thickness and nature of the oxide layer are studied by Auger spectroscopy in the 0-30 A range and MOS device is realized by metal evaporation in situ without intermediate exposure of the sample to the air. Carbon which is the main surface contaminating agent is not introduced by the residual atmosphere during the oxidation process and is located near the Si−SiO 2 interface Le silicium (100), nettoye dans une enceinte ultravide, est caracterise par spectroscopie Auger et diffraction d'electrons de tres faible energie. A partir d'un echantillon propre (diagramme de diffraction Si(100) 2×1, spectre Auger ne contenant que les transitions relatives au silicium en liaison 4-Si), l'oxydation est conduite in situ sous basse pression d'oxygene. L'examen des spectres Auger renseigne sur la nature et sur l'epaisseur des couches d'oxyde, dans la gamme de quelques A a 30 A. Une evaporation metallique permet de realiser des structures MOS sans remise a l'air de l'echantillon. Le carbone, qui apparait frequemment comme principal contaminant de surface, n'est pas introduit par contamination a partir de l'atmosphere residuelle lors de l'oxydation thermique. Sa localisation se situe au voisinage de l'interface Si−SiO 2

New macrocyclic and macroacyclic Schiff bases have been obtained by condensation of 2,6-diformyl-4Z-phenols (Z = Cl, CH 3 , (CH 3 ) 3 C) with 4-N-dodecyldiethylenetriamine or the polyamines, l,5-diamino-3-azapentane, 1,5-diamino-3-thiapentane and octadecylamine. Their interaction with lanthanide(III) (Ln(III) = La, Ce, Sm, Eu, Tb, Dy, Yb) and the uranyl(VI) ions have been studied. The complexes, formulable as Ln ( H 2 L )( X ) 3 · nH 2 O or UO 2 L (H 2 L — unionized macrocyclic or macroacyclic ligands; X = NO 3 − , Cl − , n = 1, 0), have been synthesized by template procedure or by reaction of the appropriate metal salt with the preformed ligands. They have been characterized by elemental analysis, mass spectrometric measurements, infrared and nmr spectroscopy (decoupling and n.O.e. nuclear Overhauser effect, experiments). A study of the phase transfers of trivalent rare earths from nitrate and/or chloride media through an organic membrane to a receiving aqueous solution by the macrocyclic compounds has been undertaken and the results obtained for the extraction, transport and stripping of dysprosium(III), europium(III) and lanthanum(III) ions are reported. It was verified that these ligands are efficient in the extraction of lanthanide(III) ions without the addition of a synergic agent. A high selectivity has been observed for dysprosium(III) or europium(III) with respect to lanthanum(III) from a mixture of La, Eu and Dy.