Dissolution kinetics and diffusivity of silver in glassy layers for hybrid microelectronics

Silver and palladium/silver compositions are widely used in hybrid microelectronics, as electrodes for dielectric layers and multilayers, terminations of thick film resistors and interconnections. Interactions between Ag and the adjacent films are known to affect the microcircuit performances. The present study is aimed at collecting data on the behavior of Ag-based films in contact with glassy layers. Most experiments were performed with a glass with composition 68.2 PbO : 30.5 SiO₂ : 1.3 Al₂O₃ wt %. Two different systems were analyzed. The first system consists of thick films prepared from a paste containing glass and either 3 or 15 wt % silver particles; both fine (spherical grains, 0.5–1 m diameter) and coarse (flakes, 2–5 m, <1 m thick) Ag powders were used for these pastes. The distribution of Ag in the film was studied with X-ray diffraction, scanning electron microscopy and fluorescence analysis. The results show that Ag floats on the glassy layer. Diffraction of X-rays generated by a synchrotron radiation source allowed us to study the kinetics of silver dissolution in the glass; this phenomenon is consistent with the Avrami theory, with an apparent activation energy E _dis=0.69±0.04 eV. The second system analyzed, Ag-based terminations of glass layers fired at various peak temperatures, enabled us to obtain quantitative values for both Ag solid solubility (about 2.5 wt %) and Ag diffusion coefficients D _Ag(T ). Typical values of D _Ag(850 °C) are 30.3±11.9 10^–8 cm²/s; an apparent activation energy of the diffusion process is E _a=0.6±0.1 eV.     

Aqueous precipitation and electrical properties of In2S3: characterization of the In2S3/polyaniline and In2S3/polypyrrole heterojunctions

The stability of indium sulphide aqueous supersaturated solutions at pH 2.50 and 25°C was investigated. Spontaneous precipitation proceeded at rates proportional to the solution supersaturation via a polynuclear mechanism and the phase formed was identified as In₂S₃. The absorption spectrum of In₂S₃ was measured from 190–800 nm and from the absorption threshold the optical energy gap was estimated to be E ₀ = (1.8 ± 0.1) eV. The thermal energy gap E _t = (1.6 ± 0.2)eV was determined from resistivity against temperature measurements and a thermopower coefficient S = –100VK^–1 at room temperature was found. Finally, In₂S₃/polyaniline and In₂S₃/polypyrrole heterojunctions were prepared and from the investigation of their I–V characteristics, the values of the ideality factor, n, the saturation current density, J ₀, and the effective barrier height, _B, were determined to be n = (15 ± 2), J ₀ = (38 ± 7) A m^–2 and _B = 0.56 eV for the polyaniline and n = (64 ± 8), J ₀ = (13 ± 2) A m^–2 and _B = 0.59 eV for the polypyrrole heterojunction.     

Ionic transport studies of the glassy silver vanadomolybdate system

Ionic conductivity of the Ag₂O-MoO₃-V₂0₅ system has been studied over a wide range of frequency, temperature and composition. A narrower glass forming region has been found in comparison to the corresponding Ag₂O-MoO₃-P₂O₅ and Ag₂O-B₂O₃-P₂O₅ systems. The highest conductivity at room temperature, _rt, = 3.21 × 10^–6–1 cm^–1 (d.c.) with an activation energy,E _act, of 0.466 eV, was observed for the glass former's ratio of unity. Further, it reached a maximum value of 2.2 × 10^–2¨-1 cm^–1 withE _act = 0.153 eV when the oxide-base glass was dissolved with Agl. D.c. conductivity, hopping rate and relaxation time in the present system have been found to be characterized by the same activation energy.     

AES and XPS study on the tarnishing of silver in alkaline sulphide solutions

Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) were used to follow the changes that occur on the surfaces of silver sheets when reacting with sulphide solutions of pH 12. The results obtained enabled conclusions to be drawn about the nature of the Ag₂S films. Immersion in dilute sulphide solutions ( 3 × 10^–4 M) resulted in highly irregular films that were interrupted by shallow depressions (cavities). Inside these cavities AES analysis proved the existence of massive carbon species, the presence of which delayed healing of the cavities by the formation of Ag₂S film, thus allowing the silver metal interface to continue in contact with the solution oxygen. This explains why silver electrodes maintain the Ag/Ag₂O potential while immersed in dilute sulphide solutions. Equilibration of silver with more concentrated sulphide solutions (5 × 10^–4 to 5 × 10^–3 M) effected healing of the cavities through the formation of a continuous Ag₂S film. Once this continuous film is formed, the metal interface is no longer accessible to the solution oxygen. At this point the transition of potential from Ag/Ag₂O to Ag/Ag₂S occurs.     

The determination of Gibbs energy of the exchange reaction of sulphides using beta-alumina solid electrolyte

EMF measurements of the cell Na (l) |-alumina | Na₂S + A9₂S + Ag were carried out in the temperature range of 463.4 to 843.7 K. The results were represented by the equationE/V (±0.00057) = 1.75793(±0.00086) – 4.452(±0.014) × 10^–4(T) where 463.4 < T < 737.7 K. The Gibbs energy of the exchange reaction 1/2Ag₂S(s) + Na(l) 1/2Na₂S(s) + Ag(s) was determined from the EMFs. The standard Gibbs energy of formation of Na₂S in the reaction 2Na(l) + 1/2S₂(g) = Na₂S(s) was obtained from the values of EMF and in the literature as (J mol^–1 ± 1160) = –427940 + 124.92T where 463.4 < T < 737.7 K. The eutectic temperature of the Ag₂S-Na₂S system was estimated to be 738 ± 5 K.     

Specific heat of normal and superfluid3He

Extensive measurements of the heat capacity of liquid ³ He in the normal and superfluid phases are reported. The experiments range from 0.8 to 10 mK and cover pressures from 0 to 32.5 bar in zero magnetic field. The phase diagram of ³ He, based on the platinum NMR temperature scale, is presented. In the normal liquid at low pressures and near the superfluid transitionT _c an excess specific heat is found. The effective massm* of³He is at all pressures about 30% smaller than the values reported earlier. The calculated Fermi liquid parameters F₀ and F₁ are reduced asm*/m, while the spin alignment factor (1 + Z₀/4)^–1 is enhanced from 3.1–3.8 to 4.3–5.3, depending on pressure. The specific heat discontinuity C/C atT _c is forP = 0 close to the BCS value 1.43, whereas at 32.5 bar C/C is 1.90±0.03 in the B phase and 2.04±0.03 in the A phase, revealing distinctly the pressure dependence of strong coupling effects. The temperature dependence of the specific heat in the B phase agrees with a model calculation of Serene and Rainer. The latent heatL at the AB transition is 1.14±0.02 µJ/mole forP = 32.5 bar and decreases quickly as the polycritical point is approached; at 23.0 bar,L = 0.03 ± 0.02 µJ/mole.Work supported by the Academy of Finland.     

Fast ion transport in the new mixed system (CuI)x·[(Ag2O)2·(V2O5)]100−x (40≤x≤80)

The new mixed system (Cul) _x ·[(Ag₂O)₂·V₂O₅)]_100–x where x=40, 45, 50, 55, 60, 65, 70, 75 and 80 mol% was investigated as a possible glassy fast ion conductor by preparing molten mixtures and quenching them to low temperatures. The analysis of their composition was carried out using differential scanning calorimetry (DSC) and powder X-ray diffraction (XRD) techniques. These studies have confirmed the formation of new substances. Formation of AgI in some samples was also revealed by XRD analysis and by the occurrence of a characteristic phase transition temperature around 420 K identified through DSC experiments as well. Detailed temperature-dependent a.c. electrical conductivity studies were carried out on the new materials by a.c. impedance analysis in the frequency range 65.5 kHz-1 Hz and over the temperature range 293 to 398 K. It has been found that the highest electrical conductivity of 3.64×10^–3 S cm^–1 at 305 K due to the migration of Ag⁺ ions and the lowest activation energy of 0.1 eV in the above temperature range of investigation could be realized for the composition 40CuI-40Ag₂O-20V₂O₅ in the mixed system.     

The growth of non-stoichiometric apatites using the constant composition method

The kinetics of the crystal growth of calcium-deficient hydroxyapatites with different stoichiometry (Ca₅-(HPO₄)(PO₄)₃-(OH)₁-) have been investigated at 37°C using the constant composition method. The growth was performed in solutions supersaturated only with respect to Ca₅(PO₄)₃(OH) (HAp) by inoculating with well-characterized seed crystals. The stoichiometry of the grown apatites was consistent with values of 00.185. The deviation from HAp stoichiometry of the growing apatite increased with increasing supersaturation degree (S). The constant composition method also provides relevant information about the solubility behaviour of the growing phase with a definite composition. From the decrease of the normalized growth rate j with decreasing S, an estimate could be made of the composition of the solution for which the growth ceases. The determined solubility product of the grown apatite (4.28×10^-54 M⁹) was higher than the value obtained by the equilibration of the seed material. The results were interpreted on the basis of differences in crystal lattice perfection.This paper was accepted for publication after the 1995 Conference of the European Society of Biomaterials, Oporto, Portugal, 10–13 September.     

High-Pressure Phase Transitions of M2O5(M = V, Nb, Ta) and Thermal Stability of New Polymorphs

The stability regions of various M₂O₅(M = V, Nb, Ta) polymorphs were studied by quenching samples from 600–1300°C at pressures in the range 5.0–8.0 GPa. Above 7.0 GPa, Nb₂O₅and Ta₂O₅were found to have a new polymorph (Zphase) in which the metal atoms are in sevenfold coordination. In addition, the samples contained the high-pressure polymorph B-M₂O₅ with the rutile-related structure. Differential thermal analysis at atmospheric pressure showed a weak, broad exotherm, indicating that the transformation of Z-M₂O₅into other phases begins at 100–150°C and reaches completion at 400°C in Nb₂O₅and 550°C in Ta₂O₅. At p 8.0 GPa and t> 750°C, a new high-pressure phase B-V₂O₅, isostructural with B-Nb₂O₅, was identified (a= 11.9640(6) Å, b= 4.6986(3) Å, c= 5.3249(3) Å, = 104.338(4)°, V= 290.01 ų, Z= 4, sp. gr. C2/c). At atmospheric pressure, B-V₂O₅transforms into -V₂O₅, with two strong exothermic peaks at 230 and 270°C. Experiments in the pressure range 5.0–8.0 GPa confirmed that the high-pressure phase -V₂O₅has a broad stability region.     

Spontaneous nuclear ferromagnetic ordering of in nuclei in AuIn2 Part I. Nuclear specific heat and nuclear susceptibility

We have measured the nuclear specific heat C_n and nuclear susceptibility _n of In nuclei (I=9/2, =5.5 _n) in the cubic intermetallic compound AuIn₂ (Korringa constant =0.11 Ksec) in the normal conducting state at 30K10mK and 2mTB115 mT. Our data show a positive nuclear Weiss temperature =+ 43 K and that the In nuclei undergo a nuclear ferromagnetic transition at T_c=35 K. The In nuclei experience an internal field of about 10 mT (obtained from C_n at T>T_c ). The nuclear ordering temperature T_c and the internal field increase with applied magnetic field. From the data we deduce exchange constants for the investigated system. The critical entropy reduction S(T_c)/S_max=8.6% and critical enthalpy E=0.28 RT_c are in reasonable agreement with the measured ordering temperature T_c,applying the Heisenberg model for a simple cubic I=9/2 spin system. The nuclear spin relaxation time calculated from the real and imaginary parts of _n is 10 msec at T>50 K, but drops to <1msec at T_c.This is the first observation of a spontaneous nuclear magnetic ordering transition in a not-hyperfine-enhanced metal at thermal equilibrium, i.e. at T _nuclear =T _electron .     

Growth and Properties of V-Doped HgSe Crystals

HgSeV crystals with a vanadium concentration ranging from 10²⁴to 10²⁶m^–3were grown, and their transport properties were investigated between 77 and 400 K in magnetic fields of up to 12.73 kA/m. The effect of long-term annealing in Se vapor on the electron concentration and mobility in the crystals was studied. The n(N _V), (T), and R _H(T) data obtained before and after annealing suggest that the V dopant produces a resonance donor level in the conduction band of HgSe at E _V 0.250–0.260 eV.     

Relationship between thermal conductivity,speed of sound,and isobaric heat capacity of liquids

Using two liquids-water and toluene — as an example, the author determines the dependence of the coefficient of thermal conductivity on the speed of sound and isobaric molar heat capacity for high state parameters.Notation coefficient of thermal conductivity - u speed of sound - _S same for a saturated liquid - c isobaric molar heat capacity - density - p_s same, for a saturated liquid - p pressure - p_s same, for a saturated liquid - R gas constant - T absolute temperature - x coefficient of thermal activity - x _s same, for a saturated liquid - L, constants in Eqs. (1) and (2) Translated from Inzhenerno-Fizicheskii Zhurnal, Vol. 39, No. 2, pp. 311–314, August, 1980.     

Characterizations of InzGa1?z As1?x?yN xSbyP-i-N structures grown on GaAs by molecular beam epitaxy

The GaAs-based double-heterojunction P-i-N structures using In_zGa_1–zAs_1–x–yN_xSb_y as the i-layer is investigated for the first time using solid source molecular beam epitaxy. High quality coherent bulk InGaAsN_3.45%Sb (0.5 m) with a lattice-mismatch of 2.6 × 10^–3 can be achieved due to the surfactant properties of antimony, while the bulk InGaAsN_2% at 0.5 m with 1.06× 10^–3 mismatch is fully relaxed. Rapid thermal annealing (RTA) resulted in 25 meV low temperature (LT) photoluminescence (PL) full-width half-maximums (FWHM) for the bulk InGaAsNSb layers. InGaAsNSb experiences 30% less blueshifting with subjected under the same annealing conditions as InGaAsN with similar N content. Room temperature (RT) absorption measurements are in the range of 0.9–1.4 eV. The absorption edge of 1.41 m is achieved for sample D4.     

Inelastic Interaction of an Electron with a C60 Cluster

The method of intersecting beams of C₆₀ fullerene clusters and of electrons is used to investigate the production of C ₆₀ ⁺ and C ₆₀ ^– ions and the radiation in the wavelength range from 300 to 800 nm for the electron energy E _e ranging from zero to 100 eV. The absolute values of the ionization and electron-attachment cross sections [⁺(E _e ) and ^–(E _e ), respectively] are determined. A maximum of ⁺(E _e ) of 0.53 nm²is observed at E _e = 52 eV. For a C₆₀ cluster excited by electron impact, delayed (150 s) ionization initiated by collision with the surface and ionization due to thermionic emission (its characteristic time at E _e = 60 eV is 6 s) is observed. It is found that, for E _e < 0.4 eV, the formation of C ₆₀ ^– is defined by the polarization capture of an s-electron (^– E _e ^–1 ), and a formula is suggested for determining ^–(E _e ). In the region of E _e from 1 to 6 eV, the cross section ^–(E _e ) shows only slight variations about ^–(E _e ) = 0.36 ± 0.03 nm². For E _e > 7.5 eV, C ₆₀ ^– ions proved to be unstable to electron autodetachment. In the region of intersection of C₆₀ and electron beams, radiation of a quasicontinuous spectrum described by a modified Planck formula for the thermal emission of spherical particles of diameter d is recorded. For E _e > 47 eV, the brightness temperature proved to be 3150 ± 50 K. It is found that this radiation is emitted predominantly by hot C ₆₀ ^+* ions produced as a result of thermionic emission from C ₆₀ ^* . The rate of radiation loss of energy by a C ₆₀ ^+* ion and the cross section for the formation of radiating C ₆₀ ^+* ions are found to be, respectively, 5.5 × 10⁵ eV/s at T = 3150 K and 2 × 10^–2 nm² at E _e = 60 eV.     

Critical light scattering in4He near the gas-liquid critical point

Scattered intensities of light were measured near the gas-liquid critical point of ⁴ He at scattering angles of 30, 60, and 90° as functions of the reduced temperature =|T–T _c |/T _c along the critical isochore (T>T _c ) and the coexistence curve (T>T _c ). The temperature range was 3×10 ^–5 <<1.5×10 ^–2 . Critical exponents and coefficients describing divergence of the generalized susceptibility and the correlation length are obtained as (T>T _c )=1.31±0.02, v(T>T _c )=0.66±0.02, ₀ (T>T _c )=4.2±0.6 Å, (T>T _c )=1.32±0.02, v(T>T _c )=0.68±0.02, _± (T>T _c )=2.6±0.7 Å, =0.06±0.06(T>T _c ), 0.05±0.05(T>T _c ), and ₀(T>T _c )/x_± (T>T _c )=3.6±0.4. It is pointed out that the quantal nature of ⁴ He has remarkable influence on the critical behavior of ⁴ He in the above-mentioned temperature region.     

The effect of hydrostatic pressure on transverse strength of glass and carbon fibre-epoxy composites

An attempt was made to measure indirectly the transverse tensile strengths of uniaxially aligned fibre pultrusions by the diametral compression of disc-shaped samples using concave loading anvils. Two types of composite were investigated, containing 60%V _f of either type AS carbon fibres (CFRP) or S glass fibres (GRP), both in an epoxy resin matrix. Testing was carried out at atmospheric and under superposed hydrostatic pressures, –H, extending to 300 M Pa. The resultant principal stresses at the disc centre were ₁ = _A +H; ₂ =H; ₃ = –3_A +H, where _A = 2P/dt for a disc of diameter,d, and thickness,t, subjected to a loadP. Deviations from linearity in the load-deflection response were detected throughout the pressure range at ã70% and ã90% of the failure load for CFRP and GRP, respectively, and these were associated with resin yielding. The pressure dependence of _A, approximately –0.1H, was consistent with a two-parameter yielding criterion predicting hypothetical yield stresses in simple tension and compression of ã81 and –109 MPa, respectively, for both matrix materials. Irrespective of pressureeventual fractures took place along the loading diameter, but in the CFRP specimens tested under pressure initial cracks at the disc centres were at ã45 ° to the loading axis, i.e. on the plane of maximum shear stress. Fractographic observations were consistent with transverse failure taking place by fibre-matrix decohesion in GRP and by resin fracture in CFRP. Other than the atmospheric datum point for CFRP, the pressure dependence of _A for failure, _F, was also approximately –0.1H. Of the various stress, strain and strain energy criteria for failure examined, only critical tensile strain was found consistent with this pressure dependence.     

Analysis of structural defect annealing in copper-base alloys exhibiting the shape memory effect

The process of martensite stabilization is investigated by resistometric method and positron annihilation spectroscopy in Cu-Al-Zn and Cu-Al-Ni alloys exhibiting the shape memory effect. A strong stabilization of martensite observed in Cu-Al-Zn alloys is explained by diffusion of divacancies and monovacancies formed during quenching with the following values of the Arrhenius relation E _2V ^M = (0.69 ± 0.02) eV, K ₀₄ = 6.4 × 10^8.0±0.1s^–1 and E _1V ^M = (0.84 ± 0.02) eV, K ₀₃ = 6 × 10^9.00±0.15s^–1, respectively. It was found that the energy of vacancy formation in Cu-8.5Al-14.5Zn(%wt.) alloy was 0.56 eV. The proposed model has been confirmed by applying the positron annihilation technique. In Cu-Al-Ni alloys the disappearance of quenched-in vacancies is described by one recovery process only with the diffusion energy of monovacancy 0.74 eV.     

Photogeneration of Donor Pairs in ZnSe{Ag} Crystals

Experimental evidence is presented that photoinduced defect reactions in ZnSe{Ag} crystals give rise to the formation of (Ag⁰ _i)₂ interstitial donor pairs. As a result, the induced impurity photoconductivity spectrum shows, along with the band at 0.29 eV due to isolated Ag⁰ _i atoms, a series of narrow bands in the range 0.3–0.6 eV (quasi-line spectrum). The energy positions and relative intensities of these bands are accounted for in terms of different pair separations in (Ag⁰ _i)₂. The charge instability of Ag_i under band-gap excitation, associated with multiple electron trapping processes, leads to atomic instability, which gives rise to ionization-enhanced diffusion and the formation of (Ag⁰ _i)₂ donor pairs.     

The effect of long range order on the activation energy for atomic migration in NiAl alloys; resistivity study

The influence of the degree of long range order (S) on the migration energy in NiAl was investigated by electrical resistivity and X-ray diffraction. Iron (12at.%) was over-alloyed to the stoichiometric NiAl in order to decrease the order-disorder transition temperature.(NiAl12Fe) The degree of LRO in NiAl12Fe decreased gradually as temperature increased. The activation energies for migration of iron in NiAl12Fe (Q _A ^Fe* that were measured from the temperature dependence of jumping frequency (1/_Fe*) were 1.42 ± 0.19 eV, 1.35 ± 0.18 eV, 1.16 ± 0.06 eV for the alloys quenched from 1073 K, 1273 K, and 1473 K respectively. The activation energies for migration of excess vacancy (Q _A ^v* in NiAl single crystal, were 1.55 ± 0.02 eV for the specimen quenched from 1473 K and 1.40 ± 0.11 eV for one quenched from 1673 K. The degree of LRO (S) in single crystalline NiAl calculated from the result of Hughes Lautensclager, Cohen and Brittain, J. Appl. Phys. 42 (1971) 3705 seems to be consistent with this decreasing tendency in the activation energy. The reduction of the degree of LRO (S) can explain these variations in the migration energy as the Girifalco's model.