Search

In this paper nMOS transistors with GdSiO gate dielectric are studied using electrical methods ( C – V , I – V and charge pumping) in order to assess the quality of the dielectric-semiconductor interface. Mobility is estimated using the split C – V technique and the influence of voltage stress on interface trap generation and charge build-up in the oxide is investigated. Generation of additional interface traps is observed during negative voltage stress only, which may be attributed to hole tunneling from the semiconductor to electron traps. Multi-frequency charge pumping measurements reveal the presence of border traps.

In mixed logic/memory circuits manufactured as a system on chip, two different thicknesses of dielectric layers are required. Simultaneous formation of both layers is possible if oxidation of a silicon layer is preceded by local nitrogen implantation, since the rate of oxidation depends on the nitrogen implantation dose and its profile. Experiments presented in this work show a possibility and an attempt of practical realization of controlling a very thin dielectric layer’s thickness by preceding the oxidation with an ultra-shallow nitrogen implantation from r.f. plasma. As opposed to the methods presented in the literature so far, where classical implanters or the ion immersion implantation plasma (IIIP) technique were used for ultra-shallow implantation, our process is performed in a typical PECVD planar plasma reactor. The r.f. plasma nitrogen implantation has been carried out from NH3 plasma and then immediately followed by the thermalor plasma-oxidation process. The electrophysical properties of the obtained layers and systems (ultrathin dielectric layer, silicon) were characterized by electrical methods. Results of ellipsometric, XPS and ULE-SIMS measurements are also presented and discussed. r 2008 Elsevier Ltd. All rights reserved.

The study examines the possibility of fabrication of pedestal oxynitride layers for high-k gate stacks by means of nitrogen implantation from r.f. plasma alone or followed immediately by the plasma oxidation process. The obtained layers were characterized by means of ellipsometry, X-ray photoelectron spectroscopy (XPS) and ultra low energy secondary ion mass spectrometry (ULE-SIMS). The results of electrical characterization of NMOS Al-gate test structures fabricated with the investigated layers used as a gate dielectric are also discussed. Presented results seem to be very promising and presented methods allow to form ultrathin pedestal oxynitride layers with good properties (e.g. breakdown behavior) and we believe that presented method—ultrashallow nitrogen plasma implantation with plasma oxidation may be seriously considered for future VLSI technologies.

Ultra-thin (5 and 6 nm) silicon oxynitride layers have been fabricated by the plasma-enhanced chemical vapour deposition (PECVD) process. Split experiments with annealing of the deposited dielectric layers were performed using the RTP reactor and a standard furnace, both at 900 °C. Possible changes in properties, structure and chemical composition of the obtained layers were investigated by means of spectroscopic ellipsometry, X-ray photoelectron spectroscopy (XPS), secondary ion mass spectrometry (SIMS) and electrical characterisation of manufactured test structures (metal–insulator–semiconductor (MIS) capacitors and MISFETs). The results achieved have shown that annealing at high temperature causes improvement of the properties of ultra-thin silicon oxynitride layers (e.g. lower interface traps density, lower leakage currents within the dielectric layer and lower charge-pumping currents of the MISFETs). The observed improvement in electro-physical properties can be attributed to the increase of the SiON phase. Moreover, comparison between the physical thickness and the equivalent oxide thickness (EOT) of the layers shows a decrease in physical thickness obtained by using the silicon oxynitride layer instead of the classical silicon dioxide. These findings are important for the consideration of chances of PECVD oxynitride layer application for CMOS technology.

Antiplatelet thienopyridines (ticlopidine, clopidogrel) and their thienopyrimidinone congeners, induce prostacyclin-dependent thrombolysis in vivo. Here we tested whether thienopyridines (ticlopidine, clopidogrel, and its enantiomer without antiplatelet properties) and structurally related thienopyrimidinones release NO from coronary endothelium in the isolated guinea pig heart, perfused according to Langendorff technique. The involvement of endothelium-derived NO in coronary vasodilation induced by these agents was assessed by effect of L-N(G)-nitro-arginine methyl ester (L-NAME). In addition, effect of thienopyridines or thienopyrimidinones on nitrite accumulation in cultured endothelium was assayed. Tienopyridines (10-100 micromol L(-1)) and thienopyrimidinones (10-30 micromol L(-1)) produced concentration-dependent increase in coronary flow comparable to that induced by acetylcholine (0.1 micromol L(-1)) or bradykinin (3 nmol L(-1)) which was inhibited by L-NAME (by 50-70%) but not by indomethacin. Furthermore, thienopyridines and thienopyrimidinones caused NO release from cultured endothelial cells. In conclusion, both thienopyridines independently from their antiplatelet action and their thienopyrimidinone congeners that are devoid of antiplatelet action stimulate coronary endothelium to release NO. Endothelial action of these compounds merits further investigation.

The “ITRS Roadmap” suggests the necessity of working out the processing methods allowing formation of ultrathin dielectric layers with higher values of the dielectric permittivity than for silicon dioxide. The silicon oxynitride layers (SiO x N y ) seem to be the most natural compromise. But still none of high temperature methods used for its formation can be seriously considered as final solution for future ULSI-CMOS ICs production due to the inevitable formation of nitride monolayers just at the silicon–insulator interface. The main scope of this investigation is to check if this is true. The oxynitride layers were produced by PECVD method. The process has already been optimised in order to allow repeatable and reliable formation of ultrathin layers ( 2− and the nitride-content and that nitride and oxynitride is distributed nearly homogeneously.

The aim of this study was to check experimentally the feasibility of processing ultrathin Si3N4 layers by low-temperature deposition of quality allowing in the future their application as the gate dielectric. The layers were grown in PECVD planar electrodes, RF plasma system. The optimisation of the technology was performed in order to obtain good reproducibility and control of the deposition process for ultrathin layers (thickness The test structures containing MOS capacitors and MOSFETs of different sizes with PECVD Si3N4 layers as the gate dielectric were manufactured in order to perform electrical characterisation. Apart from the classical approach (C–V, I–V characteristics) the charge pumping technique has been also used to characterise the nitride–silicon interface. The obtained results have proved successful in the process optimisation effort and reasonably good (comparing to that in recent publications) properties of the formed system.

The results of numerical modeling of the base transit time and collector current of SiGe-base heterojunction bipolar transistors with a Gaussian base doping profile and two Ge profiles (linearly graded and box) are presented for the first time. The importance of including the dependence of minority carrier mobility on the drift field and the dependence of the effective density of states on the Ge concentration along the base is demonstrated through the analysis of base transit time and collector current. A function describing the decrease of the density of states product in strained SiGe layers with increasing Ge concentration is proposed.

Surface photovoltage (SPV) measurements may provide an effective method for determining electrical properties of silicon on insulator (SOI) surfaces. In the experimental part of this work the use of the non-contact SPV-based method of surface charge profiling (SCP) in the monitoring of surface charges is explored. It was demonstrated that application of this method is constrained by the interactions between charges on the Si surface and at the interface between the Si active layer and buried oxide (box). These interactions are subsequently modeled and related to the SCP measurements. It is demonstrated that at a given doping level, the thickness of the active layer and density of charge associated with the box are factors predetermining the effectiveness of this method in SOI surface characterization. Through modeling, the SOI substrate parameter space for which the SCP method can yield useful information is defined.

The [M(CN)xNOy]n- complexes (where M = Cr(I), Mn(I), Mn(II), Fe(I), Fe(II), Fe(III)) were studied as potential NO-donors using both pharmacological and theoretical semi-empirical methods. Only iron complexes appeared to be pharmacologically active. The quantum chemical calculations indicated that these complexes have the highest predisposition to undergo a nucleophilic attack followed by the NO+ release. The results allowed us to interpret the metabolism of the [M(CN)xNOy]n- complexes in terms of the NO(+)-donation.

Several SOI MOSFET models based on 1-D solution of Poisson's equation are compared in terms of their accuracy. A brief discussion of the simplifying assumptions leads to a conclusion that the methods used for surface potential evaluation strongly influence the accuracy of a model. An analytical approximation for the front surface potential is presented and then introduced into the fully analytical Lim-Fossum model [1]. The accuracy of the modified model is better than that of the original Lim-Fossum model [1].

The influence of ionizing radiation on the dielectric strength and wear-out of electrically pre-degraded MOS structures is investigated. It is shown that irradiation significantly increases the number of defect-related failures in pre-degraded oxides, while does not affect an “intrinsic” dielectric strength. The observed effect increases as the amount of damage due to pre-irradiation F-N stress increases.

The degradation phenomena and charge build-up during high-field Fowler-Nordheim stress in silicon dioxide layers of MOS structures, pre-degraded by ionizing radiation, has been studied from the point of view of dielectric strength wear-out and long-term reliability. External bias of both polarities was applied to the structures during irradiation to separate generated carriers, force the current flow, and to enhance or inhibit radiation-induced degradation. It has been found that breakdown and wear-out properties of the oxides remain unchanged after the irradiation regardless of irradiation conditions. On the other hand, while the rate of electron trap generation during the prolonged stress has not been found sensitive to irradiation, we have found that the initial voltage transients during constant-current stress may be strongly affected by the changes in the radiation-induced hole trap occupancy. The character of the initial transient may indicate both net positive or negative charge build-up, depending on the hole trap occupancy prior to the high-field degradation.

The paper presents results of an investigation of the electrical properties of cubic boron nitride (c-BN) layers deposited onto silicon substrates by the reactive pulse plasma method. Current-voltage characteristics of the layers in metal/insulator/semiconductor (MIS) structures were investigated in a wide voltage range for films of various thicknesses, obtained under different technological conditions. On the basis of I-V measurements, the breakdown properties of BN layers were also evaluated. In spite of many similarities of c-BN films to diamond-like carbon films, we have found a slightly different character of electrical properties. In most cases we observed switching phenomena and voltage-controlled negative resistance. The dielectric strength of investigated layers shows a wide range of critical electric fields in the range 0.5–6.0 MV cm −1 . An attempt to explain the observed effects on the basis of existing models, valid for similar materials, is also made.

The role of silicon dioxide layers in microelectronics and the importance of their integrity are undisputable. From passivating coatings and masking layers for diffusion to ultra-thin tunneling films — all the silicon technology could not exist without silicon dioxide. This review deals with some aspects of the integrity of thin silicon dioxide films for VLSI applications. The problems of dielectric strength and wear-out are considered from the point of view of their mechanisms, models, oxide processing dependence, testing, and measuring. A brief presentation of statistical approaches commonly applied to reliability topics is also included.

A modified version of fully analytical Lim-Fossum model of SOI MOS transistor has been presented. The modified model takes into account the dependence of the front surface potential on both gate voltages. The great advantage of the model is that it has improved accuracy and has fully analytical form. Finally, in order to estimate the accuracy of the model, it is compared to the numerical model.

Modifications of Pierret-Shields model which take into account either apparent or physical bandgap narrowing are presented. The influence of bandgap narrowing on the I–V characteristics of a MOSFET is theoretically examined through comparison of the modified models with the original Pierret-Shields model for various device parameters. It is proved that bandgap narrowing affects significantly performance of a MOSFET in the subthreshold and the near-threshold regions.

The paper presents the results of investigations of wear-out properties of γ-irradiated oxides in MOS structures. It has been found that although irradiation does not influence the breakdown characteristics of the oxide, the transient processes (observed via voltage behaviour) during high-field constant-current stressing can be significantly affected. Post-irradiation annealing, which removes radiation-induced changes, considerably improves long-term reliability when compared to the reference (non-irradiated structure), while it does not change transient behaviour during wear-out. The relation between voltage transients during constant-current high-field stressing, charge characteristics, and long-term reliability is also briefly discussed.

The influence of ionizing radiation on breakdown characteristics of MOS structures with thin gate oxides is investigated. It is found that, as the result of irradiation, a significant reduction in the number of medium-voltage defect-related breakdown events takes place. This improvement effect suggests that there is an interaction between defects responsible for early breakdowns and defects generated by radiation, which results in the hardening of oxide weak spots. The observed shortening of the time-to-breakdown for irradiated structures is explained in terms of radiation-induced positive charge and electron trap generation which, during subsequent TDDB stress, leads to enhancement of the internal oxide field. As the result of irradiation, a significant reduction of the field acceleration factor of oxide breakdown (determined for high, near critical electric fields) is also observed.

Breakdown tests of diamond-like carbon-silicon (DLC-Si) heterojunctions were performed. The DLC films were obtained by the r.f. chemical vapour deposition method under typical conditions and in the microelectronic standard technological laboratory. The differences between the breakdown diagrams of metal-insulator-semiconductor structures produced in the two ways were shown.

Since the long-channel behaviour of MOS devices can be retained in spite of a severe reduction of the MOS transistor channel length, there are still many applications for long-channel MOS transistor models. This paper is a review of such models and its aim is to elucidate the influence of two effects: the dependence of the surface potential on the gate voltage in the strong inversion and mobility modulation along the channel due to the effective transverse electric field variation. The paper contains also a brief discussion of quantum-mechanical effects and mobility modelling.

Several well-known MOSFET models are compared in terms of their accuracy. A brief discussion of the simplifying assumptions used by those models leads to the conclusion that the accuracy of a model is strongly influenced by the methods used for surface potential evaluation. In this paper an analytical approximation for the surface potential vs. gate-voltage dependence is strong inversion is presented. This approximation is further applied in modelling long-channel MOSFET I–V characteristics in strong inversion. The new description achieves greater accuracy than other models based on the same drain-current formula.

The paper is concerned with multistage interactive-auction systems for collective decision making. The systems make it possible to solve some typical decision problems; it is assumed that participants of the auction are directly - sometimes financially - interested in the final result of decision made. Two main tasks of collective decision making are presented. The first one is connected with the choice among discrete alternatives; the second - with the choice of the quantity and financing of the public divisible good. Taking into account that in general, procedures used to solve these problems can be considered as realizations of noncooperative N-person games with non-zero sum of payments, different types of equilibrium for such games are discussed. So called free-rider problem (manipulation) is also analysed. In conclusions some extensions of the approach used are suggested. It is pointed out that in further studies a special attention should be paid to such topics as mechanisms designed to prevent free-rider behaviour in more complex procedures for the collective decision making and stopping rules for the procedures.

New fromulae for determination of the thicknesses of gate oxide and active layer in SOI structures are presented. The extraction procedure is based on CV measurements of SOI MOSFETs or gated diodes with the parasitic capacitance of the buried oxide taken into account.

A new threshold definition is proposed for symmetrical undoped double gate MOS (DGMOS). Threshold voltage is calculated using the potential model described in [1] with only two fitting parameters, the values of which do not depend on device geometry. Comparison with the results of numerical simulations and other models of VT is presented and good accuracy of the new model is demonstrated.

Possible production of an SiO 2 aerogel coated with diamond-like carbon (DLC) by the r.f. decomposition of methane is presented. The r.f. decomposition of methane seems to be the most convenient method for the production of superhard carbon coating on silica aerogels. Aerogels are new materials obtained by removing solvent from gels under supercritical conditions. They are usually transparent, highly porous with very low density. Other unique properties such as a low heat transfer coefficient, low sound velocity, refractive index close to that of air, high purity, and easy doping by different inorganic and organic substances make their applications very promising in electronics, ceramics, catalysts and other fields. However, silica aerogels are problematic because they are brittle. The silica gel was obtained by hydrolysis and condensation of tetraethoxysilan. A two-step method was used. One set of gel samples was heated in the autoclave filled with alcohol and decompressed isothermally after reaching the critical point of the solvent. Another set was subjected to supercritical extraction of alcohol by carbon dioxide followed by isothermal decompression at a temperature higher than the critical temperature of CO 2 (304.2 K). In the processes described transparent aerogels of density 104–115 kg m -3 were obtained. In the paper the influence of DLC coating on the mechanical strength of the DLC-SiO 2 aerogel system is presented. The DLC layer was deposited onto two surfaces of an aerogel plate by r.f. plasma chemical vapour deposition. The DLC films are very hard and resistant to acids and bases. DLC, which is semitransparent for mass transfer, is a good material for sensors. Together with the aerogel substrate the DLC film creates a structure which is very promising for electronics.

The paper deals with multistage interactive - auction system for collective decision making. The system makes it possible to solve some typical decision problems; it is assumed that participants of the "auction" are directly-sometimes financially - interested in the final result of decisions made. One main task of collective decision making is presented. This is connected with the choice of quantity and financing of the public divisible good. Taking into account that in general, procedures used to solve this problem can be considered as realizations of non cooperative N-person games with non-zero sum of payments, different types of equilibrium for such games are discussed. So-called free-rider problem (manipulation) is also analysed. The paper ts completed with the vast list of references on the problems mentioned.

A new method to determine the flat-band voltage VFB in a MOS structure has been proposed. It is based on measuring a voltage which corresponds to a capacitance equal to 0.9 of the oxide capacitance. The method is especially suited for SOI (silicon on insulator) structures, but could be useful in conventional types. In the latter case the error of determination of VFB caused by an uncertainty in doping concentration is smaller than in the classical procedure, especially when the thickness of the dielectric is small.

In conditions of monopolistic position of enterprises, they exhibit unrestrained tendency to increase the prices of their products, without taking care for full utilization of their production capacities. Such a situation may occur in the planned and as well as the free-market national economy. The decision center is able to counteract this tendency by introducing a proper system of income taxes, which results in the optimal prices providing the maximal profit. Simultaneously, due to the high level of the prime costs of small production series, there exists some threshold value in the relation between profit and production rate. If the production rate is less than this value, it is unprofitable. For reasons mentioned, the problem of enterprise profit optimization in the case of production capacities constrained, is a non-trivial mathematical programming problem. It is a mixed discrete-continuous optimization problem: the decision variables connected with selecting ranges of products are of discrete (zero-one) type: the decision variables associated with the choice of the optimal production rate are continuous. The paper presents an optimization algorithm which can be used to solve this mixed discrete-continuous decision problem. The worked out method can be also applied to other socio-economic decision problems of similar type.

Summary The paper is concerned with the development of mathematical models for management of large-scale Research and Development programs. The problem of optimal funding of complex R & D programs consisting of projects, their components and possible technical approaches is considered. The models with completely independent network, shared components and shared technical approaches are analyzed. Optimization algorithms for the models presented as well as the conditions of the existence of optimal solutions are discussed.

The problem of electron beam energy dissipation in the oxide layer of metal-oxide-semiconductor structures is considered. Theoretical calculations indicate that the absorbed dose of an electron beam should depend on the energy of the beam. Experimental results do not confirm this suggestion: the level of radiation damage in the insulating layer of a MOS capacitor was practically energy-independent within the range of 15–50 keV. This disagreement was attributed to the low number of electron scattering events in the dielectric layer of thickness of only several tens of nanometers.

A simple relation for calculating the capacitance of interconnection lines in MOSFET VLSI, including edge effects, is presented, which can be considered as an improved Elmasry equation[1].