Table of contents

Solid polymer electrolytes have been prepared by encapsulating plasticized polymer electrolytes based on poly(tetraethylene glycol diacrylate) into the pores of Celgard® microporous membranes. Electrolyte membranes with thicknesses of ∼40 μm have been prepared. The conductivity of the electrolyte is determined by the porosity of the membrane, and a conductivity of at room temperature has been demonstrated. These electrolytes showed good compatibility with Li and an electrochemical stability window spanning the range of 0.0 to ∼4.5 V vs. Li⁺/Li. The excellent chemical and electrochemical stabilities enabling the use of the membranes in rechargeable Li batteries have been confirmed by the discharge/charge cycling of Li/solid polymer cells at room temperature.

The correlation between Li cycling efficiency, Li morphology, Li surface chemistry, and the properties of the Li‐solution interphase was investigated in the THF, 2Me‐THF, 2Me‐Furan (MF), electrolyte system. Surface sensitive FTIR, EDAX‐x‐ray microanalysis, SEM, and impedance spectroscopy were used in conjunction with standard electrochemical techniques. Using THF as a cosolvent to 2Me‐THF decreases the detrimental impact of contaminants such as water as it is more reactive toward lithium than 2Me‐THF. The presence of MF (a few percent) influences the Li surface chemistry since its reduction to surface compounds with alkoxy groups suppresses solvent and salt reduction. However, its effect on Li morphology and cycling efficiency is marginal. It is concluded that the main positive impact of this additive reported in the literature is due to the stabilizing effect of the ethers by its possible reaction with trace Lewis acid contaminants in the solutions. The superiority of as an electrolyte for these solutions is attributed to the precipitation of elementary arsenic and arsenic compounds (e.g., , ) on lithium, which modifies Li deposition to become uniform and smooth.

The power of a lithium battery depends on the mobility of the lithium ion. Since the lithium ion, Li⁺, binds to the mobile and nonmobile molecules in the electrolyte, then the strength of the Li⁺ binding affects the conductivity of the electrolyte. The binding of dimethyl ether, diethyl ether, acetone, ethylene carbonate, and propylene carbonate to lithium ion is calculated using ab initio quantum mechanics techniques. The binding of water and acetaldehyde to Li⁺ has been calculated for higher coordination numbers. Using these energies, coordination numbers are predicted for all the species studied. These energetics also provide the basis for molecular simulations of cationic transport in the electrolyte.

Kinetic properties of a electrode for the electrochemical insertion of lithium ions were studied in the aqueous phase by cyclic voltammetry (CV) and potential step chronoamperometry (PSCA). The cyclic voltammogram showed that the electrochemical insertion reaction of lithium ions proceeds in two steps in the potential range between 0.60 and 0.98 V (vs. saturated calomel electrode). The chemical diffusion coefficients of lithium ions in the manganese oxide were evaluated from the PSCA data using a finite diffusion model. They were in the range of , depending slightly on the applied potential. The activation energy of the diffusion was evaluated as about 44 kJ mol⁻¹ from the temperature dependence of the PSCA. These results suggested that lithium ions diffuse by a hopping mechanism in the spinel phase. The diffusion coefficients, transfer coefficients, and standard rate constants also were estimated from the CV data. The diffusion coefficients agreed reasonably well with those obtained by PSCA.

"Polyveratrole" was synthesized by electrochemical oxidation of 1,2‐dimethoxybenzene in acetonitrile/tetrabutylammonium tetrafluoroborate on a platinum electrode at 1.600 V vs. Ag/Ag>⁺ (0.01M). Instantaneous formation of nuclei occurs followed by a randomly oriented unidimensional growth of green fibrils. The fibril growth that is related to a stacking structure of several "monomer" units of hexamethoxytriphenylene cation radicals associated with counterions. The process of nucleation and fibril growth is a periodic event that leads to the formation of a nest‐like arrangement. As time increases, this arrangement lines up with the electric field, and the product exhibits paramagnetic properties. Ex situreflectance Fourier transform infrared spectroscopy, and solid‐state nuclear magnetic resonance spectra confirm aromatic ring fusion. Calorimetric analysis of the green product shows that polyveratrole sublimes under vacuum at about 250°C.

A cyclic voltammogram for reduction of 1‐adamantanecarbonyl chloride at a glassy carbon or hanging mercury drop electrode in acetonitrile containing 0.10M tetraethylammonium perchlorate displays two irreversible cathodic waves; the first wave is attributable to reductive cleavage of the acyl chloride and the second wave is caused by reduction of 1‐adamantanecarboxaldehyde. Controlled‐potential electrolyses of 1‐adamantanecarbonyl chloride at potentials corresponding to the first wave yield 1‐adamantanecarboxaldehyde in yields that approach 100%. On the basis of the coulometricn value of one, and because 1‐adamantanecarbox(aldehyde‐d) is formed nearly quantitatively when the starting material is electrolyzed in (a hydrogen‐atom donor), we conclude that 1‐adamantanecarbonyl chloride undergoes one‐electron reduction to an acyl radical, which accepts a hydrogen (deuterium) atom from the solvent to afford the aldehyde.

Carbon is one of the best candidate materials for the negative electrode of rechargeable lithium batteries; however, the electrochemical characteristics are not fully understood in terms of the structure of the materials. The relationship linking the volume ration of the graphitic structure of mesocarbon microbeads (MCMBs) and the electrochemical characteristics has been examined, and it was found that the capacity in the range between 0 to 0.25 V (vs. Li/Li⁺) in 1 mol · dm⁻³ (DEE) electrolyte increased with an increase of the of the MCMBs. This result shows that the lithium storage mechanism in this potential range is the lithium‐intercalation reaction into the graphitic layers with the AB or ABC stacking. On the other hand, MCMB heat‐treatment temperature (HTT) 1000°C showed much larger capacity in the range between 0.25 to 1.3 V than higher HTT MCMBs, and it is suggested the interaction among each graphite layer is weaker in nongraphitized carbon than that in well‐graphitized ones.

In situ XANES (x‐ray absorption near‐edge structure) was used to study galvanostatic reduction of the passive film on iron. In a borate buffer, the film appears to be removed in a layer‐by‐layer fashion by reductive dissolution during the first potential arrest. A burst of dissolution takes place at the end of the arrest for slow reduction rates. During the course of reduction, the remaining film shows a composition change from ferric oxide toward . This may indicate that the film has an inner layer of . Galvanostatic reduction of the passive film grown on iron in leads to complete reduction of the passive film to ferrous oxide or hydroxide without any detectable dissolution.

The nature of the passive film grown on iron in a pH 8.4 borate buffer at a high potential [+0.4 V (MSE)] was investigated with high resolution in situ XANES (x‐ray absorption near‐edge spectroscopy). The pre‐edge peak at the base of the Fe K edge showed no evidence of the splitting associated with structures such as , , and which contain iron that is entirely octahedrally coordinated to oxygen. The single pre‐edge peak is consistent with the proposed structure (in which one‐third of the iron ions are tetrahedrally coordinated) or with a disordered structure with distorted coordination polyhedra.

Pitting potentials, , and protection potentials, , have been determined for α‐brass (33% Zn) using a cyclic polarization method. Increasing the F⁻ concentration shifts the critical potential to more active values. The pitting potentials,, and protection potentials, , depend on the logarithmic concentration of F⁻ ions according to the equations: and . Slow strain rate tests (SSRT), at a strain rate of , were performed under open‐circuit and potentiostatic conditions to study the stress corrosion cracking (SCC) characteristics of the α‐brass in solutions of various concentrations (pH 6.8) at 25°C. The minimum concentration of NaF that caused intergranular stress corrosion cracking (IGSCC) was . This concentration, was also the critical level for repassivation, observed in cyclic polarization tests. These results demonstrate a good correlation between the electrical and the mechanical breakdown of the passive film. In the presence of 10⁻¹M the potential range for IGSCC was −150 to −50 mV (SCE). These critical potentials were restricted to the stable passive potential range and also fell within the potential‐pH region where was stable. The formation of a film on the brass after polarization in the passive region was confirmed by x‐ray diffraction (XRD). At more noble potentials and at cathodic potentials below the domain, the failure mode was ductile fracture. These observations of IGSCC of the brass in fluoride solutions support a film rupture‐dissolution mechanism.

A developed strict kinetic model for the growth of porous anodic films was employed to the experimental results obtained at different bath temperatures and current densities of Al anodization. Treatment of the values of parameters involved in the kinetic equation showed that the pore/cell surface density depends only on current density, the pore base diameter depends only on bath temperature, and the equations and accurately apply. By using these equations, mathematical models for other structural features were also formulated describing their dependence on the anodization conditions.

We have examined the electrochemical behavior of tin‐free steel chromium type (TFS‐CT) and polycrystalline iron electrodes in sodium bisulfite containing perchlorate and nitrate solutions. Our experiments indicate that TFS‐CT corrodes on exposure to sodium bisulfite. The extent to which passivation is prevented depends on the supporting electrolyte. A very interesting feature that we have observed is that TFS‐CT exhibits a "memory" of exposure to sodium bisulfite, in that it does not repassivate when subsequently cycled in a bisulfite‐free supporting electrolyte. This "memory effect" and the extent of corrosion depend on the supporting electrolyte used. Polycrystalline iron electrodes on the other hand, corrode on exposure to sodium bisulfite, but repassivate when returned to a supporting electrolyte that is free of bisulfite. The results indicate that bisulfite attacks the protective chrome oxide coating partially or completely, inhibiting subsequent passivation even in bisulfite‐free electrolytes. The results also indicate that reactions of the bisulfite exposed surface with perchlorate or nitrate anions significantly affect corrosion behavior.

Laboratory exposures of copper have been performed at exposure conditions comparable to those in the UN ECE exposure program with respect to air flow conditions, relative humidity, and concentration of the gaseous pollutants sulfur dioxide and nitrogen dioxide. Extrapolation of the weight increases in the laboratory experiments match well those obtained at the test sites with high sulfur dioxide and nitrogen dioxide pollution levels. At these sites, sulfate and nitrate were the dominating surface constituents, as in the laboratory exposures. Additional constituents, detected in the laboratory, but not in the field, were sulfite and nitrite. At test sites with low pollution levels of sulfur dioxide and nitrogen dioxide, weight increases were much higher than in the laboratory exposures. At these sites, sulfate and nitrate were detected, but the relative amount of nitrate was much lower compared to the sites with high levels of sulfur dioxide and nitrogen dioxide. However, the amount of sulfate was practically the same. In addition, was identified as an important compound at some sites. Characteristic of all sites with low pollution levels of nitrogen dioxide are the high levels of ozone. Ozone was not included in the laboratory experiments, which could explain the discrepancy in weight increase. Laboratory experiments, investigating the combined effects of sulfur dioxide and ozone are presented in another publication. Both chloride and ammonia were detected as surface constituents on all field samples, confirming that salts and/or gaseous pollutants, other than sulfur dioxide, nitrogen dioxide and ozone, also are important for the understanding of atmospheric copper corrosion. However, the present investigation does not focus on these pollutants.

Copper samples were exposed for 10 days in synthetic laboratory air at 75% relative humidity. To explore the possible influence of ozone on the atmospheric corrosion rate of copper, various combinations of the gaseous pollutants sulfur dioxide, nitrogen dioxide, and ozone were added. Ozone promotes the oxidation of sulfur dioxide to sulfate more efficiently than nitrogen dioxide does. A synergism between sulfur dioxide and ozone is suggested. This synergism includes both the oxidation of sulfur dioxide by ozone and the capability of ozone to form oxides, hydroxides, or other oxygen‐containing reaction products in the presence of smaller amounts of sulfur dioxide. The synergistic effect possibly can explain the unexpectedly high corrosion rates of copper found at rural sites within the UN ECE exposure program. The rural sites are characterized by low sulfur dioxide and nitrogen dioxide concentrations, and by high ozone concentrations.

The electrical conductivities of sodium polysulfide melts, and , were measured as a function of temperature between 300 and 360°C. Due to the viscous nature of the sodium polysulfide melts, the conductance cells are axisymmetric cylindrical cells with a microelectrode, instead of capillary cells. The values of the Arrhenius activation energy derived from the experimental conductivity data are about 33 kJ/mol. The macroscopic model of sodium polysulfide melts has described the melts as composed of sodium cations, monosulfide anions, and neutral sulfur solvent.¹ For this model, the binary interaction coefficients quantifying the interaction between sodium cations and monosulfide anions were calculated from the experimental conductivity data and literature data for the transference number, diffusion coefficient, activity coefficient, and density.

Chemical‐bath deposition of thin films from solutions has been studied. The effect of various process parameters on the growth and the film quality is presented. A first approach to a mechanistic interpretation of the chemical process, based on the influence of the process parameters on the film growth rate, is reported. The structural, optical, chemical, and electrical properties of the thin films deposited by this method have been studied. The electron diffraction (EDS) analysis shows that the films are microcrystalline with mixed cubic and hexagonal structure. EDS analysis has demonstrated that the films are highly stoichiometric. Scanning electron mircroscopy, atomic force microscopy, and x‐ray photoelectron spectroscoy studies of the thin films deposited by this method show that the films are continuous and homogeneous. Optical measurements have allowed us to detect the presence of the spin‐orbit splitting effect in this material. Electrical conductivity measurements have shown the highly resistive nature of these films .

Visual observations of the back side of a Zn electrode in a single cell indicate that the patterns of Zn formed during charge, and the patterns of formed during discharge, are both very reproducible. Comparison of these patterns after various early cycles shows identical shapes, indicating that these patterns are established during the initial formation. Potential oscillations of the Zn electrode, during the middle of the charge period, indicate that the electrode oscillates between the Zn‐ and the dominating states. It is proposed that the formation, which is obtained at very low current density, is at least partially responsible for nonuniform distributions of Zn and , which are then maintained throughout the cycling of the cell. It is suggested to investigate the effect of the formation procedure on the shape change of the Zn electrode. It is quite possible that a fast initial formation step might improve the uniformity of the Zn electrode and its shape change later on during cycling.

Thin films of and have been galvanostatically electroplated onto a platinum rotating disk electrode from simple sulfate baths containing 0.5M of the more noble metal sulfate and 0.1M of the less noble metal sulfate. The experimental results are compared to those of previous studies of codeposition in order to study the anomalous codeposition behavior of the binary iron‐group alloys. Comparison of the electrodeposition results indicates that codeposition of these binary alloys is not totally analogous. It was found that codepositions of and show more mass‐transfer effects than does deposition within the range of current densities studied. A model of anomalous codeposition put forth previously for was applied to the electrodeposition of and to determine the extensibility of the model, which assumes metal monohydroxides, , are the important charge‐transfer species. This model was unable to characterize fully either or electrodeposition. However, with minor changes to the hydrolysis constants used in the model, the model predictions were found to agree with the data for codeposition and greatly improve the fit for the results.

In this study, we report on the selective electrochemical oxidation of coal at room temperature in an alkaline slurry at 1.0 V (vs. SCE) with little or no oxygen production. Electrode activities and selectivities toward coal oxidation and oxygen evolution were determined by monitoring and analyzing anodic and cathodic gas products. The activities of platinum and various nickel surfaces were compared. Results indicate that coal reacts predominantly with OH radicals during the first 300 to 400 C/g of coal at an initial rate of 0.23 A/cm² in a slurry containing 14.3 g of coal per liter. After the 400 C/g level is achieved, the rate of coal oxidation begins to decrease, and oxygen evolution becomes appreciable. Coal oxidation continues to greater than 900 C/g coal. The 900 C/g reaction level corresponds to approximately one electron per six carbon atoms in the coal substrate and exceeds that expected for any type of surface passivation/functionalization. The rate‐controlling steps in the coal reaction sequence appear to be OH radical formation on the electrode surface during the first 400 C/g of reaction and contacting of unreacted coal particles with the electrode thereafter. Coal oxidation competes with oxygen evolution, but the latter becomes significant only after the coal substrate has been depleted.

A transient agglomerate model for simulation and analysis of experimental data, obtained by current interruption on porous molten carbonate fuel cell cathodes, is presented. The initial fast change of the potential after current interruption on a polarized electrode is due to the closed‐circuit potential distribution in the electrode. Conventional estimation of the iR corrected overvoltage by current interruption on porous electrodes, with finite electronic conductivity in the solid phase and a finite ionic conductivity of the pore electrolyte, leads to an overcompensation of the external potential drop and an underestimation of the total steady‐state overvoltage due to the internal currents passing in the electrode after interruption. The overcompensation of the external potential drop is directly proportional to the geometric current density and to the thickness of the electrode and inversely proportional to the sum of the effective conductivities in the electrode matrix and the pore electrolyte.

Palladium is a preeminent material for the preparation of sensors for hydrogen and hydrogen‐evolving compounds. Conducting polyaniline can be chemically or electrochemically functionalized by the incorporation of palladium clusters. Different interfaces in a three‐dimensional matrix for hydrogen adsorption, desorption, and evolution were synthesized and characterized. Dispersions of palladium clusters in the polymer film were formed by various preparation routes, which can be classified as one‐ or two‐step processes. In the one‐step process, the composite material was obtained during the electrosynthesis of polyaniline film. In the two‐step processes, Pd aggregated into the polyaniline modified electrode. Electrochemical examination, x‐ray photoelectron spectroscopy, and Auger electron spectroscopy have been employed to characterize the composite materials in view of the hydrogen sorption and evolution as well as the binding energy state and the spatial distribution of the palladium clusters in polyaniline film.

Oxidation of water on an illuminated electrode was investigated using a rotating ring‐disk electrode, focusing on the role of the reducing agent, , added in solution. The disk was illuminated with a chopped‐light source, and the corresponding ring response at the Pt‐ring, , was recorded. Although oxidation products were expected to be produced on the disk surface and carried to the ring electrode, was found to be negative in dilute solution even at large negative potentials, e.g., −0.8 V vs. SCE. This phenomenon was observed in neutral and basic solutions. It is proposed that radical is formed at an illuminated ‐disk and subsequently initiates a homogeneous free‐radical chain oxidation of sulfite ion. This chain reaction consumes oxygen to be supplied from the solution via the disk to the ring, reducing the ring current associated with the reduction of oxygen. As a result, the ring current is lower under illumination than in the dark.

The polarographic and cyclic voltammetric reductions of the three isomeric nitrobenzaldehydes in molten acetamide are investigated. Nitrobenzaldehydes exist in two forms in this melt. The aldehyde group is solvated by acetamide, and there exists an equilibrium between the free nitrobenzaldehyde and the acetamide‐solvated nitrobenzaldehyde. The kinetics of solvation is dependent on the acid‐base conditions of the melt, temperature of study, and position of the aldehyde group.

The electro‐oxidation of ethanol on thermally prepared ruthenium oxide in alkaline solution was investigated by rotating‐disk electrode (RDE) and polarization techniques. The oxidation of ethanol was mediated by perruthenate electrogenerated on the electrode surface. The following mechanism of ethanol oxidation on ruthenium oxide was proposed Heterogeneous reactions

Homogeneous reaction

The kinetic parameters also were obtained by comparing the developed kinetic model and the experimental results.

In situ Ni K‐edge and La L_m‐edge x‐ray absorption fine structure (XAFS) measurements of electrodes supported on hydrophobic carbon black‐polytetrafluoroethylene (PTFE) layers were carried out in strongly alkaline electrolytes as a function of the state of charge. The Ni K‐edge x‐ray absorption near‐edge structure (XANES) for the material in the discharged (and uncharged) and charged states were essentially identical to those reported earlier for crystalline before and after exposure to gas‐phase hydrogen, respectively. The Ni K‐edge extended x‐ray absorption fine structure (EXAFS) analysis of the most prominent shell yielded average Ni‐Ni nearest neighbor distances of for the charged compared to for the discharged and uncharged electrodes. For the La L_m‐edge XANES, the white line was found to be significantly more intense for the charged than for the uncharged , an effect consistent with an increase in the density of empty d‐like states just above the Fermi level on hydrogen injection.

Electrochemical techniques (potentiodynamic polarization curves and electrochemical impedance spectroscoy (EIS) were used to study the effect of dibenzylsulfoxide adsorption on the dissolution of iron in 5% hydrochloric acid. To discuss different proposed mechanisms of dibenzylsulfoxide adsorption at the iron surface a quantum chemical topological method was used, which is suitable for studying the interaction of organic molecules with metal surfaces. Based on the results of the topological modeling, adsorption followed by reduction of dibenzylsulfoxide to sulfide is most favorable. EIS shows the possible effect dibenzylsulfoxide has on the electrochemical desorption of adsorbed hydrogen. Topological studies indicate strengthening of the bond of the adsorbed hydrogen to the metal due to adsorption of the inhibitor.

Thin films of nanocrystallites have been surface‐modified with thionine, methylene blue, and oxazine 170 by adsorption from the corresponding dye solutions. The strong electrostatic interaction between the cationic dye and the negatively charged semiconductor nanocrystallites results in close packing of the dye on the semiconductor surface. These closely packed H‐aggregates of the adsorbed dye are active both electrochemically and photoelectrochemically. Electron transfer from semiconductor nanocrystallites into the adsorbed dye aggregates leads to bleaching of the colored film. The extent of dye bleaching which is readily controlled by the applied potential, has been probed by spectroelectrochemical measurements. The photocurrent action spectra of these dye‐modified films indicate charge injection from excited dye aggregate into the semiconductor nanocrystallites with an incident photon‐to‐photocurrent efficiency of <1%.

We report here the effects of systematic cation and dissolved salt variation on photoelectrochemistry. It is shown that Ag⁺, highly soluble in concentrated iodide electrolytes, enhances photopotential and photoelectrochemical cell power. The variation of four highly soluble iodide salts containing Na⁺, K⁺, Li⁺, or Zn²⁺ cations are discussed in conjunction with activity and conductivity. Zinc iodide permits the nominal dissolution of up to 27 m I⁻ into the photoelectrolyte. Lithium polyiodide electrolytes enhance observed photocurrent and appear to be the preferred cation at conventional illumination. The effects of other cations, Mg²⁺, NH₄⁺, H⁺, Cd²⁺, and Ca²⁺ on illuminated in polyiodide are also reported, and the passivation by certain electrolytes and the decomposition of certain electrolytes discussed.

Aqueous polyiodide solutions were investigated with UV/visible spectroscopy, and equilibria and speciation of solutions relevant to photoelectrochemistry were determined. At 1 molal or greater iodide concentrations the spectral variation is consistent with assignment of a new 359 nm absorbance peak and an equilibrium constant of 0.2 associated with are shown to be at significant concentration in these electrolytes. At light intensities over 10 suns, photoetching and photocorrosion occur in mass‐transport limited 1 molal polyiodide electrolytes. However, regenerative polyiodide oxidation is sustained in more concentrated electrolytes.

Two resistance‐capacitance (RC) time constants are detectable in the open‐circuit impedance of illuminated (bare and Ag‐treated) in contact with solution. The dependence of the time constants on illumination intensity and concentration of V³⁺ and V²⁺ is used to identify processes related to recombination and electrochemical charge transfer, respectively. Charge transfer (CT) rates at metal‐treated electrodes [(Ag)] turn out to be somewhat smaller than at metallic Ag electrodes. For 0.3M, CT at is sufficiently fast for solar cell operation under solar irradiation intensities (exchange current density ). CT at bare p‐InP is much slower . Impedance data are fitted with two transformable equivalent circuits (differential models) which both yield similar values for recombination and CT elements. Thereby charging/discharging processes at the interface can be attributed either to a charging/ discharging of the differential Helmholtz capacity or an interfacial defect capacity. Measured values for either element are equal or smaller than the value of bulk metal electrodes [about 2 to 20 μF cm⁻² at , < 1 μF cm⁻² at bare].

A robust small volume (0.2 ml) electrochemical cell has been designed which enables simultaneous rapid scan voltam‐metric (linear diffusion conditions) and electron spin resonance (ESR) experiments to be undertaken directly in the cavity of the spectrometer over a wide temperature range in low (e.g., dichloromethane) or high (e.g., acetonitrile) dielectric solvents. Additionally, the design facilitates the study of electrochemically generated highly oxygen and/or moisture sensitive radicals. The close proximity of the specially designed reference electrode to the working electrode and the thick layer solution conditions enables small cell (RC) time constants to be achieved. These conditions allow the ESR signal at constant field to be measured at fast scan rates. Alternatively, ensemble averaging of the complete ESR spectrum at constant potential enables high quality ESR spectra to be obtained for reactive electrochemically generated radicals. Data on the oxidation of and reduction of  [, acac = acetylacetone], tetramethrin, and 5,5‐dipropylpyridine‐2,6‐dicarbothioate are presented to demonstrate the features of the newly designed cell. Lowering the temperature down to 130 K does not lead to destruction of the cell, so that frozen solution spectra of electrochemically generated radicals may be obtained.

and glasses have been synthesized. The effect of incorporation of fluorine in the structure of phosphate glasses has been characterized by using x‐ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy, differential scanning calorimetry, x‐ray diffraction, and electrical conductivity studies. Results provide evidence for fluorine participation in the phosphate network by forming P—F bonds indicating the presence of and species. A systematic change in , the ratio of bridging oxygen and nonbridging oxygen, and the concentration of different structural species such as , , and P—F with the fluorine content were observed. DC conductivity and activation energy data were obtained. The conductivity mechanism is discussed in terms of the effect of modification of phosphate structure by the incorporation of fluorine.

The composite solid electrolytes mole percent (m/o) prepared by conventional powder metallurgy and solution casting processes, have been investigated by impedance analysis, x‐ray diffraction (XRD), differential thermal analysis (DTA), and scanning electron microscopy (SEM). The composites prepared by the former method exhibit about an order of magnitude lower conductivity than those prepared by the latter technique. The XRD and DTA results show that no phases other than and are present in the composites. SEM results show that grains are interspersed with particles. The conductivity of dispersed systems depends on the processing, particle size, and concentration of . The mechanism of enhanced conductivity is proposed to be the generation of excess Rb⁺ ion vacancies in the host matrix at the matrix‐particle interface. Macroscopically the results are explained satisfactorily on the basis of random resistor network model by assuming the formation of high conducting space charge layer along the matrix‐particle interface.

The properties of a transparent oxide formed on zirconium by anodic oxidation in carbonate buffer are described. The transparent oxide has all the properties of an ideal valve‐metal oxide. The film grows via a high field conduction mechanism. The potential and film thickness both increase linearly with time under constant‐current growth conditions to well over 100 V with no sign of electronic leakage or breakdown. When the oxide is grown at a current density of 93 μA/cm², the electric field in the oxide is 4.1 MV/cm, and the oxide is anisotropic with parallel and perpendicular to the field. The relative permittivity is 38.3 at the anodizing field, and the oxide shows electrostrictive effects similar to those exhibited by other valve‐metal oxides with high relative permittivities. Both the refractive index and the relative permittivity increase when the field is removed, and there is a corresponding decrease in film thickness of just under 1%. Cathodic reduction inserts hydrogen into the oxide to a limited depth, and the outer layer thus formed is optically absorbing. Subsequent anodic oxidation removes the hydrogen and returns the film to its initial transparent state.

Radio frequency Ar and plasma jets generated in a hollow electrode terminated by a small size Ti nozzle were used for deposition of Ti and films. The regime with low content of reactive gas resulted in an extreme enhancement of deposition rate, more than one order higher than that of Ti. Abrupt increase of the nozzle temperature to >1350°C in this regime is caused by additional thermal energy imparted to the nozzle. Abrupt decrease of the cathode dc self‐bias and simultaneous steep increase of the optical emission from Ti and from ions Ti⁺, , and Ar⁺ indicate the transition of the discharge into the RPJ‐arc. This reflects in an enhanced production of Ti and Ti⁺ which leads to a high rate (≈μm/min) growth of films. The microhardness of films at low nitrogen content is and the film resistivity is 80 μΩ cm. Geometry of the particle transport and the distribution of the, film thickness on substrates confirm a hollow profile of the particle density distribution in the plasma jet.

Due to the continuing trend toward higher integration scales and smaller structural dimensions in semiconductor technology, a new class of detrimental contaminants arises: volatile organics. Recently it has been shown that, among others, they cause severe problems in epitaxial growth, in oxidation kinetics, and in all wet chemical treatments of the wafer. In order to select appropriate polymeric materials for future boxes, minienvironments, or other equipment, we investigated the outgassing characteristics of polypropylene (PP: natural, antistatic, and blue, respectively), polycarbonate (PC), periluoroalkoxy polymer (PFA), polyvinylidenefluoride (PVDF), acrylonitrile butadiene styrene copolymer (ABS), and polytetrafluoroethylene (PTFE). All samples were taken from commercial products for semiconductor technology. Each polymer was heated up from 60 to 20°C below its softening temperature with continuous monitoring of the amount of outgassing. Additionally, the outgassing components were identified separately. PTFE and PFA showed the lowest amounts of outgassings over the entire temperature range. If only selected temperature ranges are of interest, other polymers like PVDF or PC are almost as good. From the knowledge about particular outgassings and the understanding of the chemistry of the polymer, measures can be derived on how to improve certain polymers.

Thermal oxidation is performed in a copper‐contaminated ambient on Czochralski and float‐zone silicon, and on samples with damaged and undamaged surfaces, in order to evaluate the role of oxygen supersaturation in the starting material and the influence of formation of oxidation‐induced stacking faults on the microstructure and electrical properties of the interface underlayers. The microstructure is controlled using transmission electron microscopy and secondary ion mass spectrometry, and the minority carrier diffusion length is analyzed by the electron beam‐induced current technique. It is shown that the thermal oxidation of Czochralski silicon induces the precipitation of large copper colonies associated with oxidation stacking faults decorated with oxygen at the interface. In float‐zone silicon the oxidation stacking faults are themselves nucleation sites of copper precipitates. When there is no oxidation stacking fault formation (undamaged initial surface), copper precipitation may occur on dislocation nets. The extended defects are highly recombinant when they are associated to copper precipitates. The diffusion length is decreased in regions free of copper colonies—which indicates the growth of point‐like defects during the thermal process—and partially restored by hydrogenation treatment.

This paper reports on the efficiency of a less critical chemical cleaning process on the removal of particulates and metal contamination from the silicon surface. Having completed the entire RCA‐based cleaning process in either nonfiltered or point‐of‐use filtered distilled instead of deionized water, a final dip in diluted HF followed by an immersion of the wafer in boiling isopropyl alcohol (IPA) is shown to be effective in reducing particulate levels on its surface without increasing the metal contamination content with regard to previous cleaning steps. It is also shown that early breakdowns of MOS capacitors are predominantly governed by the particulate content on the silicon surface. In addition, it is shown that sulfur can remain on the silicon surface after a dip in diluted HF solution and that after a following hot IPA rinsing, sodium, potassium, calcium, and copper can also be detected on the silicon surface.

Spin coating of a polymer film on a substrate with topography is modeled. Nonnewtonian fluid behavior, solvent diffusivity within coating, and coating leveling are related using this model. The dependence of the coating step height on initial coating concentration, spin speed, feature radial position, and feature dimensions is investigated. For a large feature, the coating profile is dominated by centrifugal force, while for small features the coating profile is dominated by capillary force and the coating is more level. A glassy polymer skin is formed during spinning. The coating profile is then controlled by shrinkage after the skin is formed. The predicted step height is within 10% of that measured experimentally.

Excellent shallow p⁺n junctions have been formed by implanting ions into thin polycrystalline Si films and subsequent annealing. The samples implanted at 50 keV to a dose show a leakage of 1 nA/cm² and a junction depth of about 0.05 μm after a 800°C annealing. To reduce the series resistances of the junctions, silicidation with different cobalt thickness was used to drive the as‐implanted dopants in the polysilicon films into the resultant junctions of the silicon substrates. For the low energy implantation at 50 keV at all dosages, silicidation can result in poor electrical characteristics due to the confinement of the dopants by the silicidation process. On the other hand, the electrical characteristics can be retained when a higher implantation energy of 100 keV with a dosage higher than was used. In addition, the samples implanted at 125 keV show poor electrical characteristics for the nonsilicided junctions but good characteristics after the silicidation. It is attributed to the enhanced defect annihilation by the formed silicide. Furthermore, silicided implanted through poly‐Si junctions with excellent characteristics can be fabricated after a low temperature (600°C) annealing if the samples are implanted at 100 keV with a dosage higher than .

Seven kinds of commercially available organic spin‐on glass (SOG) were evaluated and analyzed to reveal a relationship between film properties and their chemical composition. Film properties evaluated were shrinkage, refractive index, wet/dry etching rates, and adhesion strength. The structure of SOG solute was analyzed by ²⁹Si‐nuclear magnetic resource (NMR) spectroscopy. Composition of the solution was assayed in terms of GC‐MS as well as ¹H‐ and ¹³C‐NMR. The properties were widely distributed among seven SOG films, which could be explained on the basis of the concentration of Si‐C direct bond in the solute calculated from ²⁹Si‐NMR spectroscopy. Important parameters corresponding to the film quality such as film shrinkages, wet and dry etching rates decreased, with an increase of Si‐C direct bonding. From this point of view, SOG solute having a high organic content is favorable for forming film with high cracking resistance and good planarization of the underlayer topology. The composition of the solvent appears to affect the viscosity of the solution, which dominates the properties during coating such as occurrence of striation or film thickness per coat.

There is lingering disagreement over the relative oxidation kinetics of and in the regime of passive oxidation. Various oxidation enthalpies have been reported for these materials, in some cases with suggestions of an increase at higher temperatures. The result is uncertainty over the respective diffusion mechanisms that limit oxidation rates in these materials. It is suspected that this confusion may be due, in large part, to the effects of impurities from the oxidation environment. Accordingly, this study was aimed at clarifying some of the issues by undertaking the simultaneous oxidation of CVD and within a furnace tube of fused silica known to be very pure. We discuss the results obtained in the interval 1200 to 1500°C, and their implications on the respective rate‐limiting mechanisms.

The density of traps generated inside of thin has been calculated by measuring the transient discharge of the traps and applying the tunneling front model to these discharges. The number of traps created by the high voltage stress was proportional to the cube root of the fluence that flowed through the oxide during the stress. The trap generation rate varied as the . As was the case with interface trap generation, there was a voltage dependence to the bulk trap generation. The bulk trap generation was comparable in magnitude to the interface trap generation. The trap generation results were used to generate simulated time‐dependent dielectric breakdown data using a model of dielectric breakdown based on random generation of traps inside the oxide. The simulated breakdown distributions agreed well with measured distributions.

We studied the thermal decomposition properties of methylhydrazine (MH) which is used in the chemical vapor deposition (CVD) of titanium‐nitride . The decomposition was carried out on the surfaces of silicon dioxide and titanium (Ti) at low pressures. The by‐products were detected by the gas chromatography mass‐spectrometry (GCMS), and surfaces were analyzed by electron spectroscopy for chemical analysis and Auger electron spectroscopy. MH decomposed on surface above 400°C, producing methane , ammonia , and monomethylamine (MMA). Above 650°C, and were predominantly generated instead of MH. The decomposition temperature on the Ti substrate was 200°C and was even lower than the temperature on . As the substrate temperature increased, Ti was converted to approaching a stoichiometric around 500°C. As a result, MMA was observed around 400°C. Due to the formation of and its poor capability for donating electrons rather than Ti, the MH decomposition rate was reduced as temperature increased above 400°C.

A direct liquid coinjection system has been applied to the chemical vapor deposition of copper using the commercially available Cu(I) precursor , where  = 1,1,1,5,5,5‐hexafluoroacetylacetonate and  = trimethylvinylsilane. Precursor delivery was enhanced through the use of a coinjection system wherein additional TMVS was mixed with the copper precursor before injection into the vaporization chamber. The results reported here demonstrate the capability of depositing blanket copper of high purity (on the order of 99.99% copper) and low resistivity . These copper films have been deposited at rates up to and exceeding 1500 Å/min. The effects of temperature and carrier gas on deposition rate and resistivity are examined. The as‐deposited films demonstrate a dependence of grain size with thickness and little structural or morphological change with annealing. This study suggests that liquid coinjection is an effective method for enhancing deposition rates and for producing high quality copper films from copper(I) precursors.

Excimer laser pulses with wavelengths of 248 and 308 nm were used to selectively seed Pd on polyimide (PI) surfaces, making them suitable for electroless plating. This novel seeding process for insulating materials is accomplished with the sample immersed in the seeding solution, occurs only on the areas of the substrate that are illuminated (through the liquid) by the laser light, and does not require prior treatment of the surface. The seeding solution is transparent to the laser light and the metal deposition occurs as a consequence of the photoabsorption in the solid. This leads to electron transfer from the solid film into the solution and reduction of the Pd ions in contact with the surface. The Pd content of the seeded samples increased with the number of pulses, but was independent of repetition rate. The deposition rate of Pd did not exhibit a significant dependence on wavelength, in agreement with UV absorption spectra of PI and a single photon absorption process for electron excitation to allowed unoccupied states. As for the PD distribution, the deposits consisted of islands with distributions that depended on surface properties as well as on laser‐material interactions. Sufficient PD seeds for uniform electroless plating of Cu and Co were attained after 3000 pulses at fluences ≃30 mJ/cm². Although these fluences are much lower than those used for ablation of PI under water, distinct kinds of surface roughness were observed depending on the laser light and on the different types of PI.

A study has been carried out to prepare ‐coated phosphors utilizing a sol‐gel technique with mixed solutions of titanium alkoxides and barium alkoxides and acetate in various solvents. The precursor gel of prepared by the hydrolysis of iso‐propoxides of barium and titanium in iso‐propylalcohol transformed to crystalline at a low temperature of 550°C. Coating was carried out by heating the precursor gel containing phosphors in air at 450°C for 5 h, followed by heating in nitrogen‐flowing atmosphere at 570°C for 10 h in order to prevent phosphors from being oxidized. However, the heat‐treatment in nitrogen flowing atmosphere depressed the electroluminescent intensity because of oxygen defects in . It was found that annealing in air at 400°C for 5 h gives rise to the brightest phosphors.

The use of a potentiometric sensor to monitor the state of an catalyst during oxidation at 550°C was investigated. The technique of solid electrolyte potentiometry was employed to monitor in situ the oxygen activity (the oxidation state) of the catalyst. The reaction proceeded through an ionic transport redox mechanism where the reoxidation of the reduced sites occurs by lattice diffusion of oxygen rather than by reaction with gas‐phase oxygen. The rate was accordingly a function of the catalyst oxygen activity and the gas‐phase oxygen (reoxidation determined reaction) or carbon monoxide (reduction determined reaction) partial pressure. Hence, the EMF of the sensor is necessary to predict the reaction rate for this system. Such a sensor could find wider application for reactor control, particularly for partial oxidation processes.

Polycrystalline, terbium‐doped yttrium oxide samples were sintered under various conditions and examined using positron lifetime spectroscopy and cathodoluminescence. Rapid grain growth was found to occur at sintering temperatures between 1500 and 1600°C, as shown by both the positron lifetimes and the luminescent intensities. The correlation between the results demonstrates the utility of positron lifetime measurements for investigating microstructural properties of materials and the importance of these properties to the luminescence.

The interactions of copper and copper oxide layers with 1,1,1,5,5,5‐hexafluoro‐2,4‐pentanedione (H⁺hfac) have been investigated. The results provide supporting evidence for proposed reactions of copper with H⁺hfac which are thought to be responsible for the vapor‐phase etching of copper. Specific reaction products depend on the chemical state of the copper in the film. Reaction of H⁺hfac with Cu²⁺ yields volatile reaction products of and , while Cu¹⁺ yields the same products with a change in the chemical state of the surface to Cu⁽⁰⁾. No reaction is observable between H⁺hfac and Cu⁽⁰⁾ at temperatures studied in these experiments.

The diameter of particles which adversely affects the yield has been shrinking as ULSI devices are more and more miniaturized. Ultrafine particles with diameters of 0.1 μm or less have become important recently. Ultrafine particles of this type are expected to be difficult to remove. This study has established a method to evaluate ultrafine particle removal efficiency. Ultrafine metallic particles with diameters of several to several hundreds of nanometers were deposited on the Si surface using a gas deposition method. The removal efficiency of the ultrafine particles using various cleaning solutions was investigated. APM cleaning can remove 150 nm Au particles, but cannot remove ultrafine Au particles with a diameter less than several tens of nanometers. In addition, the Si surface becomes rougher when a cleaning is performed to remove Au ultrafine particles. This is believed to be because noble metals such as Au, Ag, and Cu, which feature a higher electronegativity than Si, attract electrons from Si facilitating Si oxidation.

The minimization of particle contamination during wet processing of Si wafers, such as from particles in solutions with surfactants, was studied. It was demonstrated that particle deposition on wafer surfaces in solutions is determined by an attractive or a repulsive electrostatic force through the zeta potentials of particles and substrates. In practice, wafer surfaces consist of materials such as Si, poly‐Si, , , Al‐alloy, and photoresist, and all these materials exhibit different zeta potentials in wet chemical solutions. In solutions with a , the Si surface exhibits a negative zeta potential while and surfaces have positive zeta potentials. This means that particle deposition will always occur either on Si surfaces or on and surfaces in acidic solutions that contain particles. Therefore, in order to suppress particle deposition on the wafer surfaces, it is essential, that the wafer surface and the particles must have the same polarity of the zeta potential. This can be achieved by adding an anionic or cationic surfactant, where the wafer surfaces and the particles exhibit the same polarity of the zeta potential.

The effects of injecting carriers at different gate insulator fields on observed threshold voltage shifts , during optically assisted electron injection, to quantify charged and neutral defect densities in unirradiated and, for the first time, x‐ray irradiated [2.4 Mrads ] devices were examined. In unirradiated devices, where intrinisic fixed positive charge (FPC) densities are below measurement limits, accompanying an injection fluence of [the fluence normally used to fill large cross section neutral electron traps (NETs)] in the range of gate insulator fields between 1.5 and 7 MV/cm was found to decrease from positive to negative values. This was interpreted as indicating a decrease in large NET cross sections with increasing gate insulator field above 1.5 MV/cm, accompanied by the net generation of FPCs, at , as evidenced by dropping below its initial value. Based on two stage injections, the first injection for, followed by reinjection at , it was concluded also that for , large cross section NETs and fixed negative charges, in addition to the FPCs mentioned above, were generated, and that field alone does not result in defect generation. In irradiated devices, [2.4 Mrads ] examined in a similar manner, the high‐field behavior was somewhat different. At an injection fluence of , (the fluence normally used to fill FPCs) the accompanying first decreased to around zero which was attributed to an FPC cross section decrease, and then increased slightly with increasing oxide fields, attributed to additional large angle scattering. At an additional injection fluence of (again, the fluence normally used to fill NETs), the accompanying first increased, as expected due to the filling of both NETs and any unfilled FPCs, and then decreased at . The decrease in at the higher fields is similar to that seen in the unirradiated devices, indicating again that the NET cross section decreased. However, in this case never dropped below its initial value and , measured during the subsequent low field reinjection for FPCs, showed no increase in the number of FPCs above that initially present, so there is no direct evidence of net FPC generation. In these devices using two stage injection sequences, it was determined also that large cross section NETs were not generated. The effects of the device being "on" or "off" during injection, depending on the applied substrate bias, were also examined using "continuous" injections. Evidence is presented which suggests strongly that to obtain the proper cross sections and/or densities of defects, the devices must be on.

Zn‐diffusion experiments were performed in an open furnace with and using and as source materials, respectively. A reproducible diffusion, as expected from the literature, was observed with InP in the temperature range between 400 and 600°C and with at 550°C. Area‐selective diffusion was performed through evaporated masks. Anomalously enhanced lateral diffusion under the mask observed with was sup‐pressed by a 200 nm thick cap layer between the ternary layer and the mask. A dark current density of as low as 40 pA/μm² at a reverse bias of 5 V, measured with lateral diodes, shows that a nonleaky lateral pn junction was formed, which is effective in separation of optically generated free carriers. The response time at 15 V reverse bias was 35 ps for a photodetector with an electrode separation of 6 μm. Simultaneously, low RC charging times in a 50 Ω load of 10 ps were found for the same bias conditions indicating that the transient behavior of the lateral pn photodiode is dominated by electron drift.

The conduction current in silicon nitride increases even at constant electric field as the nitride thickness is reduced to less than 5 nm in oxide equivalent thickness . In order to analyze the charge transport in the ultrathin nitrides less than 5 nm , we measured the thickness and temperature dependence of conduction current through nitrides of 3.4 to 10.2 nm, in the temperature range from 77 to 398 K. Current increase was observed in both the tunnel emission component, which is thickness dependent, and in the temperature‐dependent component. The temperature‐dependent current component was dominant at high temperatures and low fields in the ultrathin nitride. The method of separating the electron and hole currents was used for both n‐ and p‐channel metal‐nitride‐silicon transistors, to study the charge transport in nitrides from 3.8 to 8.6 nm , at 296 and 398 K. The increase in the number of electrons injected into the nitride was larger than the increase in the number of holes injected into the nitride when the nitride thickness was reduced. The increase in electron current flowing out of the nitride was also large compared with the increase in hole current flowing out of the nitride. We claim that the contribution of electrons to the total charge transport is increased with the reduction in nitride thickness. Finally, we discussed the dependence of the breakdown field on nitride thickness in oxide/nitride/oxide structures. We claim that top and bottom oxides should be as thin as possible to obtain the high breakdown field.

Small defects observed by IR light‐scattering tomography after short‐time annealing were shown to be related to the nuclei of ring‐distributed oxidation‐induced stacking faults (ring‐OSF) in Czochralski‐silicon crystals grown with growth rates of about 0.8 mm/min. The behavior of these defects was studied under different annealing conditions, and it was found that they were already present in the as‐grown crystal below the detection limit of available techniques. Preannealing in nitrogen at 1150°C introduced a stacking‐fault‐free region with subsequent oxidation. The reason for this was due neither to the annihilation of the ring‐OSF nuclei nor to the formation of punched out dislocation loops from precipitates during nitrogen ambient annealing, but probably due to a change in the nature of the OSF nuclei occurring, making them incapable of OSF nucleation during subsequent oxidation. Various small defects were observed to exist in the as‐grown crystal, and the radial position of these defects depended on their critical size. The distribution of as‐grown defects within the crystal reflects the differences in the thermal stability of the defect nuclei due to point defect variations in the crystal during growth.

P‐doped film is grown by low pressure chemical vapor deposition using the gas system. The stoichiometric is grown on an Si substrate at a temperature of 973 K (700°C). gas is injected during growth and a highly P‐doped (∼10²¹ atom/cm³) film is obtained. Decreased growth temperature suppresses generation of defects in the Si substrate and makes a smooth interface between the Si and the crystals. Such improvements in hetero structure, fabricated at a growth temperature of 973 K, allow hetero‐junctions with low leakage current.

This paper is a result of several years of work on the silicon floating zone (FZ) process. Starting with numerical calculations of RF current density distribution, heat transfer, and thermal stresses, theoretical results are compared with FZ experiments. The simplified numerical models show qualitatively and quantitatively useful results for designing RF inductors under the assumption of steady‐state growth and for sufficiently low disturbances by the field of the inductor slot. Above all, the thermal influence of modifications of the inductor profile can be predicted more accurately.

We have analyzed oxide charging and interface trap generation induced by high‐field tunnel injection in thin (200 to 250 Å) thermal oxides of silicon prepared using rapid thermal oxidation (RTO). Constant‐current Fowler‐Nordheim tunnel (FNT) injection was used to stress aluminum gate MOS capacitors and to generate oxide and interface trapped charge. Analysis of both the flatband voltage and the gate voltage necessary to maintain constant current allows us to separate oxide charge generation from charge generated near and at the interface. The analysis reveals, in addition to background electron trapping at water‐related trapping centers, the trapping of holes near the interface as well as the generation of interface charge and anomalous positive charge (APC) near the interface. The appearance of these charge components is consistent with a comprehensive model of oxide and interface charging proposed recently by DiMaria and co‐workers, ad we make use of this model in discussing the extraction of information on oxide charging from FNT charging curves. In particular, we show that under simplifying assumptions consistent with device processing on a particular oxide, useful information concerning generation of specific charge species can be obtained.

Charge trapping and interface state generation in rapid thermal processed (RTP) oxide and ‐nitrided oxide metal oxide semiconductor capacitors have been investigated by internal photoemission (IPE) electron injection from both the Al gate and the Si substrate. It is found that the generation rate of the interface state by IPE electron photoinjection from the Si substrate is one order of magnitude higher than that from the Al gate for both RTP oxide and nitrided oxide samples at low electron injection fluence, . The polarity dependence of interface state generation is discussed by means of the hydrogen species induced generation model. Photocurrent voltage (photo‐ IV) technique has been used to determine the centrpid and density of trapped charge after IPE electron photoinjection into the dielectrics from both electrodes.

We report photoelectrochemical measurements of diffusion lengths in p‐type multicrystalline silicon used for photovoltaic solar cells. We show that, under some conditions, this method provides a very practical means for rendering an account of the quality nature of this material through diffusion length variations. The results obtained during two years of use of this method in the industrial processing of photovoltaic cells are presented and discussed.

A custom‐designed microelectromechanical system (MEMS) force transducer, with a volume less than 1 mm³, is being fabricated to measure force development in isolated cardiac muscle cells to elucidate the physiology of muscle contraction. A single heart cell is attached to flexible, hinged polysilicon plates submerged in a nutrient saline solution. As the cell contracts, the plates bend, and the contractile force can be measured based on the known spring constant of the plate. The amount of deflection is measured by piezoresistive, ion‐implanted strain gauges placed at the base of the plates. Prototype structures have been fabricated and have been mechanically tested using probes. We have demonstrated that living rat heart cells can be attached to polysilicon using a silicone sealant. Polysilicon is an inert material when exposed to cardiac cells and their saline environment and has no effect on the cells themselves.

Growth mechanisms of three different orientations Si water oxidized in have been investigated in this study. A thickness crossover phenomenon in oxidation rates was found for orientations (110) and (111) at a critical oxide thickness 150 Å. From our results, this phenomenon is closely related with the initial native oxide before oxidation.

A dry etching technique for subquarter‐micron Cu interconnects has been developed. A self‐aligned passivation film on sidewalls of Cu lines is formed during etching. The thickness of the sidewall film composed of can be controlled precisely by the composition of the etching gas mixture. The addition of to the etching gas mixture of results in a sidewall film free of chlorine. Therefore, the sidewall film acts to protect Cu from oxidation and corrosion, and sustains the multilayered structure without film peeling during the fabrication process.

The adsorption of dye monolayers on nanocrystalline titanium dioxide surfaces leads to high photoelectrochemical yields in the visible range, which gives a renewal of interest in the use of this system in solar energy. A high resonance effect in these dyes enabled the attainment of the Raman spectra of the adsorbed molecules; further, a new series of Raman peaks appeared during the polarization of modified by Ru‐bipyridinium compounds in the presence of an electron donor. They correspond to a molecular configuration in which an electron is exchanged during the measurement between a particular ligand and a substrate titanium atom. This configuration appears then to be similar to the short‐lived excited state of the metal ligand charge‐transfer complex (MLCT).

A preliminary investigation of the electrochemical deposition of Pt, Pb, and Hg adlayers on conductive diamond thin‐film surfaces has been made using cyclic voltammetry and scanning electron microscopy. The diamond thin films employed were polycrystalline, grown on conductive Si substrates (1 cm²) to a thickness of ca. 14 μm, and doped with boron at a nominal atomic concentration ranging between 10¹⁹ and 10²⁰ cm⁻³. The cyclic volammetric measurements were performed both in a conventional glass electrochemical cell and in a thin‐layer flow cell. The results demonstrate that metallization of diamond film surfaces electrochemically is feasible, opening the door for the development of novel catalytic electrodes, sensors, and detectors using this advanced material.

Abstract not Available.

Abstract not Available.