The Electrochemical Society (ECS) was founded in 1902 to advance the theory and practice at the forefront of electrochemical and solid state science and technology, and allied subjects.
Find out more about ECS publications
Number 4, 1980
Previous issueNext issue
Open all abstracts, in this issue
F. C. Anson
Abstract not Available.
Roger Parsons
A. N. Dey and R. W. Holmes
An isothermal DTA technique was developed for studying the kinetics of the heterogeneous exothermic reactions involving lithium metal and organic solvents used in the Li/SO2 batteries. The results showed that while film forming agents such as SO2 and PC retard the Li+AN exothermic reaction (the crucial reaction for Li/SO2 cell safety), the organic solvents such as MF, DME, DG, and THF enhance the Li+AN exothermic reaction most probably due to an enhanced solubilizing effect on the Li film. The above solvents by themselves were found to be quite stable with Li.
A. J. Jacobson and S. M. Rich
Amorphous V2S5, prepared by thermal decomposition of (NH4)3VS4, has been found to react readily with n‐butyl lithium to give amorphous compositionsLixV2S5 (x≤6.0). In lithium electrochemical cells, V2S5 cathodes react with 2.5 Li per vanadium above 1.40V on primary discharge at 0.5 mA/cm2. Eighty‐five percent of this capacity is recovered on the first recharge to 2.8V also at 0.5 mA/cm2. On subsequent cycling the capacity eventually falls to 30% of the initial value with an accompanying change in the voltage composition profile.
Robert J. Staniewicz
The use of calcium metal as an anode in SOCl2 electrolytes was investigated. Calcium showed an open‐circuit potential (OCP) of 3.22V vs. a carbon cathode in 1MCa(AlCl4)2‐SOCl2 electrolyte. Hermetically sealed D cells, when discharged at 1 mA/cm2, displayed level discharge plateaus of 2.8V. SEM photographs indicated the rapid formation of a crystalline layer on the anode; the composition of this film was established as CaCl2. Storage of calcium samples in Ca(AlCl4)2‐SOCl2 electrolyte resulted in corrosion of the metal. This corrosion was particularly pronounced at 55°C and it appeared to be a fundamental behavior which was not eliminated by rigid control of electrolyte purity. The following cell chemistry was suggested by chemical analysis of discharged electrolytes and cathodes
2Ca+2SOCl2→SO2+2CaCl2+S
Electrodeposition experiments indicated no evidence for the plating of Ca onto a nickel substrate.
B. MacDougall
The galvanostatic oxidation of nickel electrodes in pH 2.8 Na2SO4 requires the presence of a prior oxide film in order to avoid the active nickel dissolution region. Either the 6–8Å NiO film on electropolished nickel or that formed during the early stages of anodic passivation can serve as an appropriate starting surface for galvanostatic experiments. Anodic charging of such electrodes gives a distinct transition region over a potential range of almost 1V. The current efficiency for nickel dissolution during the potential transient is high (>80%) with only a small portion of the charge contributing to oxide growth. This situation persists even at high charging rates and/or high solutionpH's where it appears that chemical dissolution of the oxide is not playing an important role. The results suggest that the standard high field oxide growth mechanism is not operative during galvanostatic oxidation of nickel. The results are best explained in terms of a defective oxide film where the change of potential during the transient is associated with an increase in the state of film perfection. Anodic charge consumption would be mainly due to a continuous breakdown and repair of the oxide film at defect sites, most of the charge going towards Ni→Ni2++2e during inefficient film repair. The increase in anodic potential with increasing perfection of the film is interpreted in terms of an overvoltage effect, i.e., the area over which breakdown and repair is occurring decreases with increasing film perfection and therefore the anodic current density increases. At potentials > +0.4V, the rate of increase in oxide film perfection decreases and an eventual steady state is reached. The influence of potential of anodization and solution pH on nickel dissolution, oxide repair, and oxide perfection is discussed. The defect model of oxide film development is shown to apply equally well during potentiostatic oxidation of nickel, the NiO film reaching an almost steady thickness very soon after the potential step anodization with the continual long‐term decrease of anodic current with time of anodization being due to a localized increase in oxide film perfection.
David M. Haaland
The catalytic effects of various electrode materials on nonequilibrium oxygen measurements using solid‐electrolyte oxygen sensors have been investigated. Catalytically active electrodes can perturb the measurement of oxygen in nonequilibrium mixtures of oxygen and combustible gases. This paper reports the development of noncatalytic electrodes for accurate measurement of the free oxygen content in nonequilibrium gas mixtures containingO2 and CH4, C3H6, CO, or H2. The effects of Pt, Au, Ag, Ag deposited on Pt, and S‐ or Pb‐poisoned Pt electrodes on the measurement of oxygen in nonequilibrium gas mixtures were determined. These studies were made using solid‐electrolyte oxygen sensors of a new internal‐reference design as well as with traditional air‐reference sensors. Pt electrodes are catalytic in all of the gas mixtures studied. Both Ag electrodes and Ag vapor deposited on Pt were found to be noncatalytic to CH4 oxidation while Au electrodes were only slightly catalytic. Pt electrodes poisoned by S introduced as H2S were also noncatalytic to CH4 oxidation until the sulfur desorbed at ∼800°C. The above electrodes were sufficiently catalytic to CO oxidation that they could not be used for quantitative nonequilibrium oxygen measurements when CO is present. Pt electrodes poisoned by Pb (introduced as PbS) produced the least perturbation of the nonequilibrium mixtures of oxygen with CH4, C3H6, CO, or H2. The Pb‐poisoned Pt electrodes allow quantitative measurement of the bulk nonequilibrium oxygen concentrations at temperatures above 500°–550°C, depending on the combustible gas present. At lower temperatures, even the Pb‐poisoned electrodes are catalytic. This suggests that Pb may poison the platinum activity by decreasing the heat of adsorption of adsorbed reaction species. Sulfur‐ or Pb‐poisoned Pt or Pt with a thin layer (<200Å) of Ag had no adverse effect on the electrochemical O2 pumping capabilities of Pt. Ag and Au electrodes performed poorly as electrochemical O2 pump electrodes. The capability of solid‐electrolyte oxygen sensors to measure the partial pressure of gas phase oxygen at the electrode surface was used to investigate the catalytic reaction mechanism of CO oxidation on Pt electrodes partially poisoned by Pb. The results are best fit by an Eley‐Rideal mechanism where adsorbed O2 reacts with gas phase CO. However, the data are not sufficiently definitive to rule out the Langmuir‐Hinshelwood mechanism where adsorbed O and CO react on the surface. The methods used demonstrate that solid‐electrolyte oxygen sensors can be useful experimental tools for identifying catalytic mechanisms.
R. Accorsi and E. Bergmann
The steam electrolysis between 800° and 950°C has been investigated with a cell consisting of the following elements: electrolyte, 400μ of ytterbia stabilized zirconia; anode, 100μ of vacuum deposited tin doped india; cathode, a sputter deposited cermet of nickel and urania doped ytterbia stabilized zirconia. These cells can be operated below 1.5V at 1 A/cm2 over 1000 hr.
Y. Matsumoto, H. Manabe and E. Sato
Anodic characteristics of La1−xSrxCoO3 (with x=0.2 and 0.4) electrodes were investigated in alkaline solutions. It was found that the catalytic activities for the oxygen evolution reaction are relatively high and that the activity of the electrode with x=0.4 is higher than that of x=0.2. The secondary chemical step, i.e., CoOH+OH−→CoO−+H2O, is rate‐controlling in the oxygen evolution reaction. Therefore, it is concluded that the rate of the secondary chemical step is more rapid at the electrode with the higher value of x because of the beneficial effect of the higher positive charge on the Co cation on this step. La1−xSrxCoO3 electrodes are suitable for the anode material in alkaline solutions, because anodic dissolution scarcely occurs.
M. Elam and E. Gileadi
The mechanism of oxidation of I− in a nonpolar solvent consisting of 1.0MAl2Br6 and 0.80MKBr in an aromatic hydrocarbon was studied. A chemical step following charge transfer, in which I− is regenerated, was observed. It was postulated that bromination of the solvent occurred. Oxidation of Br−by I2 to IBr or Br2 is possible in this system since the standard potentials for different redox couples are much closer than in aqueous solutions. Qualitative mechanistic information obtained from analysis of cyclic voltammograms was consistent with the mechanism derived on the basis of steady‐state current/potential measurements. It was observed that iodine is reduced rapidly in this system if mesitylene is used as the solvent and much more slowly in benzene. The half‐wave potential for the oxidation of I− is about 1.10V vs.RAlE, compared to about 2.2V in aqueous solutions.
M. M. Nicholson and F. A. Pizzarello
The anodic oxidation of lutetium diphthalocyanine films on tin oxide in1MKCl was investigated by a galvanostatic transient technique. The reaction was monitored by simultaneous measurement of the optical transmission. At current densities of 0.3–6 mA/cm2, the kinetics was controlled by an ionic space charge in the red oxidation product. A dielectric constant of 10 was estimated for this material, using the solid‐state mobility of the chloride ion determined in a previous moving‐boundary study.
W. E. Triaca, A. M. Castro Luna and A. J. Arvía
The influence of the hydrocarbon partial pressure (10−3 atm≤PE≤1 atm) in the electrooxidation of ethylene on platinized platinum in 1N H2SO4was investigated from potentiodynamic and stationary state data. A positive effect of the ethylene saturation pressure on the electrochemical reaction rate has been observed over the whole pressure range. At PE>5×10−2 atm, the electrochemical reaction rate tends to become pressure independent. The characteristics of the electrochemical reaction can be explained through a reaction pathway involving the electrosorption of ethylene followed by an electron transfer step as rate determining. The present results confirm the recently proposed reaction mechanism for the ethylene potentiodynamic electrooxidation on platinum.
W. M. Vogel, L. J. Bregoli and S. W. Smith
By using meniscus electrodes and shift gases of relatively low CO content, it was possible to study the electrochemical oxidation of H2 in molten carbonate with little interference by CO. Data are presented which indicate that theH2 reaction proceeds with negligible activation polarization on Ni, Au, Pt, and Pt‐Rh alloy. The observed over‐voltages are explained as diffusion over‐voltages for H2, H2O, and CO2. Hydrogen diffusion through metal was appreciable for thin Ni membranes and very fast for Pd wires. The CO oxidation is far more complex and proceeds in stages. At low polarizations the rates are very small but increase greatly above approximately 200 mV. Applied to fuel cell anodes, the results suggest that, in order to utilize large CO content gases efficiently, internal shifting is required.
Akira Fujishima, Yasuhisa Maeda, Kenichi Honda, George H. Brilmyer and Allen J. Bard
During a photoelectrochemical reaction only a portion of the light energy absorbed by the semiconductor (CdS or TiO2 single crystal) is utilized in the electrode reaction. The unused portion of energy is expended through various mechanisms as heat. Therefore by monitoring temperature changes within the photoanode as a function of electrode potential and light intensity, information concerning the efficiency of the process can be obtained. Experimental results are presented and interpreted using a model for the energy balance within the system. This permits the determination of the quantum and energy efficiencies simultaneously without the need to calibrate the light source.
H. H. Strehblow and M. Bettini
The electrochemical reactions of PbTe in acetate buffer of pH 4.9 were compared with those in strongly acidic and alkaline solutions. Rotating ring‐disk studies provided information on the nature of soluble products, while the composition of the surface layers was analyzed by XPS. In the potential range of 0–0.4 VH a Te‐layer of only ∼3Å thickness is formed and an equivalent amount of Pb2+ is dissolved, this Te‐layer being too thin for coating or passivation purposes. At >0.4 VH an oxide layer is formed which contains mostly TeO2(H2TeO3) and ∼30% PbO on the surface. There is a competing process between oxide film growth and dissolution of approximately 50% of these oxides. The reduction of TeO2 and PbO to elemental Te and PbTe occurs at <0 VH and the reduction of Te to soluble tellurides at < −0.5 VH. By appropriate cyclic oxidation and reduction the Pb concentration in the anodic oxide and the reduced layer can be further decreased to ∼15%. The oxide layers are not good insulators (∼1000Å film, anodic current ≥10 mA/cm2 at 1.5 VH). The use of electrochemical processes on PbTe for forming well passivating or insulating surface layers is limited.
D. Pavlov and Z. Dinev
The electrode system Pb/PbOlayer/PbSO4membrane/H2SO4 solution forms when a lead electrode immersed in an H2SO4 solution is in the potential range between −400 and +960 mV (vs. the Hg/HgSO4 electrode). When the electrode potential is raised above +960 mV the PbO and PbSO4 layers are oxidized to PbO2. The mechanism of this oxidation is determined by the electric properties and the crystal structure of PbO and PbSO4. The paper deals with the electrochemical behavior of the electrode during potential sweeps from the PbO into the PbO2 potential ranges. It was found that first PbO is oxidized via solid‐phase reactions to nonstoichiometric PbOn without a change of its crystal structure. This takes place via the electrochemical reaction
kPbO+m2 H2O=kPbOn+mH++me−
which proceeds at the PbO/solution interface. Since PbOn is a semiconductor, the oxidation of the Pb2+ ions in its crystal lattice probably takes place through a mechanism in which surface states and mobile acceptors take part. Such are O− and O, evolved during the reaction of oxygen evolution. When nreaches a critical value n=d, α‐PbO2 nucleation commences in the pores of the PbSO4 membrane. When this phase reaches the H2SO4 solution the oxidation of PbSO4 crystals to β‐PbO2 begins. This takes place according to the mechanism
dissolution of PbSO4→diffusion of Pb2+ ions→oxidation of Pb2+ at the PbO2 surface→precipitation of β‐PbO2
The dissolution of PbSO4 crystals begins at the apexes of the crystals and continues at their edges and faces.
P. H. Melville
A technique using integrals of Bessel functions has been used to calculate the effect of an anodic current distribution of the form ja=jm(β0+β1r2/a2 +β2r4/a4) within the pit on the potential distribution in the plane of the electrode surface, and analytic solutions have been obtained for situations where linear polarization kinetics may be assumed. If the radius of the pit is small compared to the Wagner polarization parameter Lc for the unpitted electrode, the potential distribution is independent of the polarization parameter, and thus insensitive to the assumption of linear kinetics. The potential within the pit depends mainly on the value of the total current flowing from the pit, and is less sensitive to the form of the current distribution. The potential outside the pit on the surface of the electrode depends on the mean current density and is virtually unaffected by the form of the current distribution.
M. Skyllas Kazacos and B. Miller
The reaction sequence for the reduction of H2SeO3 in 1MH2SO4 at various surfaces and the formation of selenide films in this medium with and without metal ions present have been investigated by rotating disk and ring‐disk electrode techniques. The system is complicated by competitive paths leading to selenium formation from Se(−2)‐Se(+4) chemical reactions and the result is sensitive both to the nature of the substrate with the possibility of metal selenide formation and to the concentration of solution species. The implications for cathodic deposition of semiconductors like CdSe, from voltammetric and photospectral data, are discussed.
A. Damjanovic, L. ‐S. R. Yeh and J. F. Wolf
The growth of anodic oxide films at platinum in sulfuric acid solution was studied at different temperatures using galvanostatic transients. The activation energy and preexponential factor for the rate of growth were evaluated. In the plane in which the rate‐determining step occurs the surface density of ions, or ion pairs participating in the growth process, is about 1015 cm−2. This is close to the density of atoms in the surface of the metal. The analysis appears to support the process at the metal/oxide interface being the rate‐determining step. However, the same value for N is compatible with, the field‐assisted place exchange process as the rate‐determining step at the metal/ oxide interface or within the oxide phase. Because the growth follows the same rate equation from the very early stages of oxide formation, the process at the metal/oxide interface seems to be the rate‐determining step.
Y. Matsuda and H. Satake
F. Simchock, K. Ramachandran, R. Haynes and F. J. Jannett
Carlos H. Contreras‐Ortega and Peter A. Rock
D. W. Rice, P. B. P. Phipps and R. Tremoureux
P. Doig and P. E. J. Flewitt
Alfred Milch and Pedro Tasaico
The effect of extended shelf life on the information storage capability of tellurium films on polymethylmethacrylate (PMMA) was investigated. Aging tests were performed for as long as fourteen months under various temperature and humidity conditions. The rate of change of sensitivity to the writing of information (hole formation by laser machining) was chosen as the basis on which to estimate stability quantitatively. It is suggested that aging is caused by oxidation, and a theoretical model for film oxidation as a function of temperature and relative humidity is proposed. Preliminary experimental support for the model is presented, based on the comparison of predicted and observed accelerated decay.
D. W. Rice, R. J. Cappell, W. Kinsolving and J. J. Laskowski
The indoor corrosion rates of copper, silver, nickel, cobalt, and iron were measured at eight locations in the United States. Concurrent pollutant and relative humidity measurements were also made at six sites. The rates are shown to be lognormally distributed over the sample population. A statistical comparison of outdoor and indoor rates shows that Ag corrodes approximately the same both indoors and outdoors. Cu and Ni corrode 100 times faster outdoors, and Fe corrodes 2000 times faster outdoors. The corrosion rate sensitivity to relative humidity is speculated to explain these differences. Pollutant concentrations are shown to be generally attenuated indoors except for NH3and reduced forms of sulfur. The indoor corrosion rates of copper and silver are reasonably well correlated to the measured reduced sulfur concentrations. Nickel rates correlate best to the acid chlorine containing gases.
H. F. Bittner, J. T. Bell and J. F. Land
Effects of manganese and carbon on oxidation properties of simulated Incoloy 800 were determined. Iron‐base alloys containing 32% nickel and 23% chromium were prepared with 0, 1, or 2% manganese and residual (0.002%) or 0.1% carbon. Samples were exposed at 660°C to 0.94 atm steam for 2, 48, 528 hr, and resulting oxides were examined by optical microscopy and analyzed using Secondary Ion Mass Spectrometry(SIMS) as well as x‐ray diffraction techniques. Carbon additions in all cases led to thicker oxides with higher αFe2O3 to Cr2O3 ratios. Manganese was incorporated into the spinel phase, forming Mn(Fe1−xCrx)2O4. Manganese additions to alloys containing carbon tended to mitigate the effects of carbon additions, leading to thinner oxides with lower αFe2O3 to Cr2O3 ratios.
K. Kumar
85 Ag‐15 Cu flex leads, used as current‐carrying conductors in inertial instruments, corrode in a BTFE (bromotrifluoroethylene) fluid environment The corrosion scale has a nodular appearance similar to that of the Ag‐rich particles which are present in the flex lead matrix. The Ag‐rich phase is the predominant phase in the material and exists as discrete particles surrounded by the Cu‐rich phase located at what appears as a quasi‐continuous network of grain boundaries. Corrosion proceeds along the Cu‐rich phase resulting in separation of the Ag‐rich metallic particles from the matrix and incorporation into the corrosion debris. After buildup of a certain thickness, the corrosion layer detaches from the flex lead (contaminating the fluid) and exposes fresh metal surface for more attack. When the chemical etching behavior of a good lead (one that showed negligible corrosion) was compared to that of a poor lead (where substantial corrosion effects were observed), the good lead was found to etch more slowly, by at least a factor of six. The differences in etching behavior have been attributed mainly to differences in prior thermal history, and partly to possible local variations in overall composition. The observed corrosion has been explained by mechanisms involving chemical reaction between the metallic phases and the fluid, and those involving the formation of an electrochemical corrosion cell.
Fumio Shimura
The pyramidal hillocks bounded by four convergent {111} planes have appeared on annealed (001) silicon wafers. Observations by TEM (transmission electron microscope) of an annealed (001) wafer with oxidation‐induced stacking faults show that the pyramidal hillock is not an extraneous substance such as SiO2 precipitate but an intrinsic part of the substrate silicon crystal. A possible mechanism of hillock formation discussed in detail verifies the formation of pyramidal hillocks on annealed (001) silicon wafers. By using the ultrasonic agitating etching technique, a flat (001) surface without pyramidal hillocks can be obtained.
Jean‐Paul Chané
It is shown that the homogeneity in thickness of a large area GaAs thin layer obtained by VPE is a function of the composition of the gaseous phase, its homogeneity, and the hydrodynamics of its flow in the vicinity of the substrate. The influence of each of these factors has been investigated. Modifications have consequently been introduced into the low temperature growth technique usingAsCl3/H2/GaAs which lead to very uniform layers with a pinch‐off voltage variation of ±5% for FET.
N. Miyata, K. Miyake, K. Koga and T. Fukushima
Transparent conducting films of cadmium‐tin oxide (CTO) were prepared by rf sputtering from a sintered CdO‐SnO2 target in an Ar or an Ar‐O2 atmosphere. The electrical and optical properties of the films were found to depend upon the oxygen concentration in the gas mixture. The films with the lowest resistivity of 4.82×10−4 Ω‐cm and high optical transmission of 85% over the visible region were obtained without post‐heat‐treatment.
V. K. Jain, C. L. Gupta, R. K. Jain and Prem Swarup
Efforts have been made to study the electret and photoelectret effects in pure and doped negative photoresist KTFR. In KTFR, dipoles are available to form electret and photoelectret effect due to the photopolymerization. This term can be applied to cross‐linking of polymers. Details are not available regarding the initial steps of the photoreaction but it is clear that N2is split off and nitrene is formed as a highly reactive intermediate, which can subsequently add to double bonds or insert into labile‐CH or ‐NH groups in polymeric materials. Doping of impurities created certain new energy levels or molecular states. More investigations are required for the optimization to change the sensitivity of this material on other wavelengths without changing its basic properties.
C. Lawrence Anderson, K. V. Vaidyanathan, H. L. Dunlap and G. S. Kamath
A novel capless technique for the annealing of ion‐implanted GaAs is described. The apparatus employed consists of a large solution of GaAs in Ga, into which is immersed a suitable sample holder containing the water to be annealed. Se‐implanted layers annealed using this technique behave electrically in the manner expected for well‐annealed group VI dopant implants into GaAs. The electrical activity of implanted Si, however, decreases with increasing anneal temperature in the range from 800° to 900°C. Possible explanations for the behavior of the implanted Si are discussed.
K. Tokunaga and D. W. Hess
Etch rates of aluminum and native aluminum oxide films in carbon tetrachloride plasmas were investigated as functions of substrate temperature and electrode material in a parallel plate plasma etching system. Use of stainless steel electrodes resulted in the formation of a carbon‐chlorine polymer which precluded aluminum etching. Anodized aluminum electrodes minimized polymer formation and etching was achieved. Several inherent differences in the electrode materials were considered to account for this effect. Analysis of the effluent from the reactor suggested possible chemical reactions occurring in the carbon tetrachloride plasma, as well as a precursor to polymer formation.
A. K. M. Zakzouk
Localized conducting regions of silicon dioxide are a major source of failure of integrated circuits. By using nematic liquid crystals in contact with the oxide, and simultaneously applying an electric field across such a sandwich structure, defective regions can be nondestructively identified. By examining the number of defects as a function of the magnitude and polarity of the applied voltage, time, and oxide thickness, it is possible to obtain information on the cause and nature of these localized regions of high conduction. Measurements have been performed on oxide conduction in thin layers (1000–2000Å) of silicon dioxide, similar to those commonly employed as gate dielectrics in MOS devices, grown in dry oxygen. The liquid crystal technique has been shown to be a valuable nondestructive technique for studying the behavior of oxide defects as a function of the magnitude and polarity of the applied voltage, time, and oxide thickness.
Fumitaka Kitagawa, Tomoyoshi Mishima and Kiyoshi Takahashi
The electrical and optical properties of ZnTe and ZnSe prepared by molecular beam epitaxy have been studied. Epitaxial growth was found to take place at ≳ 250°C for ZnTe on GaAs and ZnTe and at ≳ 300°C for ZnSe on GaAs; the epitaxial growth ceased at 430°∼470°C for ZnTe and at 450°∼500°Cfor ZnSe because of the reevaporation of compound and/or impinging molecules. The observed growth rate dependence on substrate temperature and on impinging rate ratio has been explained on the basis of a model in which the number of incorporated and reevaporated molecules are strongly coverage dependent.ZnSe on GaAs, and ZnTe on GaAs and ZnTe had good mirror smooth surfaces. ZnTe on InAs and ZnS, however, had a rather rough surface. Ga was found to diffuse, due to a large number of misfit defects at GaAs‐ZnTeinterface, from the GaAs substrate into ZnTe influencing the electrical and optical properties of grown layers considerably. The resistivities of undoped layers were ∼5 Ω‐cm for ZnTe on ZnTe and ∼104 Ω‐cm for ZnSe on GaAs. Sb was used to increase the hole concentration of ZnTe on ZnTe up to ∼1018cm−3. The resistivities of ZnSe on GaAs were found to decrease to ∼0.07 Ω‐cm by Ga doping. The results of photoluminescence measurement suggest that the higher substrate temperature makes the better quality layers.
Inderjeet Kaur, D. K. Pandya and K. L. Chopra
The kinetics of growth for chemical deposition of CdS films from alkaline solutions of cadmium salts has been studied with respect to temperature of deposition and the relative concentrations of the various reactants in the solution. It has been established that the growth of the film takes place either by ion‐by‐ion condensation of Cd+2 and S−2 ions or by adsorption of colloidal particles of CdS formed in the solution, depending on the various deposition parameters and the method of preparation. The former process of growth results in thin, hard, adherent and specularly reflecting films, whereas the latter results in thick, powdery and diffusely reflecting films. Occurrence of different polymorphic phases of CdS (hexagonal and cubic) has been observed under different growth conditions. A model for growth mechanism has also been proposed.
Yorimitsu Nishitani, Kenzo Akita, Akio Yamaguchi and Tsuyoshi Kotani
An improved LPE method in which the solid surface is always covered by a solution during step‐graded growth of GaAs1−xSbx on GaAs substrates produces etch pit densities of 3–6×105 cm−2. In contrast, layers grown by conventional step or continuously graded LPE have EPD of 6×106−2×107 cm−2. The composition of the top layer is approximately GaAs0.9Sb0.1 in all cases. The reduction of EPD by the improved LPE method is attributed to limiting the nucleation of small islands to the first buffer layer, under the conditions that the grown surface is always covered by the solution subsequent to the growth of the first buffer layer.
O. L. Russo
It is shown that the reflectance of modulated unpolarized incident radiation at the pseudo‐polarizing angle can be used to accurately determine the optical constants of absorbing media in the visible region. The approximation that the principal angle of incidence occurs at the pseudo‐polarizing angle is assumed to contribute negligible error. Expressions for the refractive index,n, and the extinction coefficient, K, are given in terms of the pseudo‐polarizing angle and the reflectance ratio of the parallel (to the plane of incidence) reflected wave vector intensity to the normal reflected wave vector intensity. The experimental values for n and K are determined at wavelengths in the visible spectrum between 4250 and 5750Å, for boron‐doped p‐type highly polished silicon of 0.01 and 95 Ω‐cm resistivity. These values correspond to impurity concentrations of about 10−3–10−6%, respectively. The 0.01 Ω‐cm Kcurve indicates the possibility that direct optical transitions between acceptor levels and the conduction band can be resolved and that this occurs at about 5250Å (≃2.4 eV).
M. Druminski
The deposition of epitaxial silicon films on (012) Czochralski sapphire from the silane/hydrogen system was investigated for growth temperatures from 880° to 980°C and for growth rates from 0.2 to 6.0 μm/min. The film quality was evaluated by optical absorption measurements. The temperature range, in which epitaxial film quality was obtained, shifts with increasing growth rates to higher deposition temperatures. This epitaxial temperature range is rather broad, e.g., 915°–965°C for 0.2–0.5 μm/min and 935°–985°C for 5.3–6.0 μm/min. A region for the optimum epitaxial deposition conditions, very suitable for MOS applications, is defined by the minimum values obtained from the optical absorption measurements. At temperatures below 915°C, decreasing crystalline quality was observed due to an increasing fraction of polycrystalline silicon. In addition the influence of the H2O content on the crystalline quality of the silicon films was investigated; for the production of a good epitaxial silicon film the H2O impurity fraction must be smaller than 1 ppm, especially in the low temperature region.
B. Leroy and C. Plougonven
High temperature processing of Czochralski grown silicon wafers can create temperature gradients high enough to generate slip. The generation of slip and the slip patterns have been found to depend on three factors: the temperature and the temperature gradient, the amount and form of the precipitated oxygen, and the direction of the initial bow and the wafer diameter over thickness ratio. A model is presented to fit experimental data of critical stress in silicon, temperature gradients, and wafer curvature to predict the critical temperature above which warpage will occur. An important result is that increased wafer diameter over thickness ratio makes the wafers more sensitive to warpage. When this is the case the difference in stresses on both sides of the wafer, because of the bending, makes the area affected by slip on the concave side much larger than on the convex side.
Eiji Takayama and Noboru Kimizuka
The composition of the wustite phase field under various oxygen fugacities was determined in great detail by means of the thermogravimetric method in the temperature range of 1100°–1300°C. There is a breaking point, which means a second or possibly higher order transition in the wustite phase, in every isothermal line of oxygen fugacity vs. composition of wustite. The phase diagram containing the subphase boundary was obtained. Some thermodynamic functions were also determined and compared with the previously published data.
A. L. Dawar, O. P. Taneja, K. V. Krishna and P. C. Mathur
G. Blasse and G. J. Dirksen
Bruce E. Deal
T. W. Sigmon, V. R. Deline, C. A. Evans Jr. and W. M. Katz
G. F. Neumark, B. J. Fitzpatrick, P. M. Harnack, S. P. Herko, K. Kosai and R. N. Bhargava
Marc J. Madou, Karl W. Frese Jr. and S. Roy Morrison
H. S. Gamble, N. E. Evans and S. H. Raza
E. A. Taft