Hot electron noise and g-r noise in short-channel JFETs

Hot electron noise in short-channel JFETs is measured above 150°K and generation-recombination noise due to donors is measured below 120°K, where it predominates over the first. The hot electron noise increases with increasing V_GS - V_P for a given temperature T and increasing with decreasing T at a given V_GS - V_P. The g-r noise in donors increases very rapidly with decreasing temperature and is found to be governed by an activation energy between E₀ and 2E₀, where E₀ is the activation energy of the donors. The theories of these effects can qualitatively explain the data; for the g-r noise it must be taken into account that the activation energy of the donors decreases with increasing electric field (Poole-Frenkel effect).     

Structural,opto-thermal and electrical properties of ZnO:Mo sprayed thin films

Zinc oxide (ZnO) thin films doped with molybdenum (Mo) have been prepared by the spray pyrolysis technique. X-ray analysis shows that ZnO:Mo thin films crystallize in hexagonal structure with a preferred orientation of the crystallites along (002) direction. The surface topography of these films was performed by the atomic force microscopy. The dispersion of the refractive index was discussed in terms of the single oscillator model proposed by Wemple and DiDomenico. The single oscillator energy (E₀) as well as the dispersion energy (E_d) were therefore calculated. Finally, the electric conductivity was investigated depending on the effect of temperature. The activation energy (E_a) was found to range from 0.63 to 0.94 eV; the electrical behavior can be correlated with Mo-doping.     

The small signal admittance of carbon implanted p-n diodes

In this paper we consider, in detail, how the introduction of radiation damage centres, produced by the implanation of carbon ions, affects the small signal admittance of silicon p-n diodes. Thermally stimulated capacitance measurements are used to obtain the charge states and activation energies of the damage centres. For carbon doses between 1 × 10¹¹ cm^?2 and 1 × 10¹² cm^?2 two trapping levels are observed with activation energies of E_t?E_v=0·31 eV and E_c?E_t=0·37 eV, and for doses between 5 × 10¹² cm^?2 and 5 × 10¹³ cm^?2 an extra level appears with an energy of E_c?E_t=0·25 eV. A study is made of the effects of these damage centres on the small signal capacitance and conductance of the diodes under forward bias. The results are interpreted in terms of a conductivity modulation effect, and it is proposed that this technique yields valuable information on the profile of the damage centres.     

Participation of the electron subsystem of a crystal in the reactions of defect-complex decomposition in semiconductors

The reactions of defect-complex decomposition in semiconductors are considered. The contribution from the electron subsystem to the reaction rate is taken into account by adding a change in the electron-subsystem energy of a crystal (as a result the reaction) to the energy barrier of the reaction. The theoretical and experimental data are compared by the example of the reactions of E-center decomposition in the n-type phosphorus-doped silicon. The dependence of temperature of isochronous annealing of E centers on a donor-impurity concentration is explained. The first stage of annealing (T _ann≈400 K) in a low-resistivity silicon is caused by the decomposition of the E center and can be explained using the model of a vacancy as the double acceptor center with a negative correlation energy and values of vacancy charge-exchange levels E _V (0/-)=E _c -0.99 eV, E _V (-/—)=E _c =?0.39 eV. From the comparison between calculated and experimental data, the dissociation energy of E center and the degeneracy factor are obtained to be U _a0≈0.96 eV and $g_E^ - /g_E^0 = 1/16$ , respectively.     

The physical significance of the T0 anomalies in Schottky barriers

Levine's suggestion of an effective exponential energy density of interface states to explain the T₀ anomalies in Schottky barriers is compared with the Bardeen interface model with a parabolic density of states. Except for cases in which the characteristic energy E₀ is less than kT, the two models are shown to be equivalent with the Bardeen model revealing more of the physical situation. When E₀ is less than kT and the T₀ anomaly is still observed, the interface state model is physically untenable, but an alternative interpretation in terms of an appropriate doping profile and conversion of semiconductor type near the metal contact is a possible explanation. Except when E₀ < kT it is impossible to distinguish between the two models from current-voltage characteristics alone.     

Two-electron tin centers with negative correlation energy in lead chalcogenides. Determination of the Hubbard energy

It is shown using Mössbauer spectroscopy on the isotope ¹¹⁹Sn that an isovalent tin impurity in the solid solutions PbS_1?z Se _z is a two-electron donor with negative correlation energy. The energy levels associated with the tin impurity are found in the lower half of the band gap for z<0.7 and are found among the allowed states of the valence band for z>0.7. We have demonstrated electron transfer between neutral and ionized tin centers. The activation energy E _o of this process decreases monotonically as z is increased, which reflects the fact that the tin energy levels move toward the top of the valence band, and for z>0.7 the activation energy is E _o =U/2, where U is the Hubbard energy.     

Physical investigations on Sb2S3 sprayed thin film for optoelectronic applications

Sb₂S₃ thin films have been obtained at 250 °C on glass substrates using the spray pyrolysis techniques. The structural study by means of XRD analysis shows that Sb₂S₃ thin film crystallized in the orthorhombic phase. The discussion of some structural constants has been made by means of both XRD and AFM investigations. Moreover, the optical analysis via the transmittance and the reflectance measurements reveals that Sb₂S₃ sprayed thin film has a direct transition with the band gap energy E_g equal to 1.72 eV. The analysis in 300–2500 nm domain of the refractive index data through Wemple–DiDomenico model leads to the single oscillator energy (E₀=2.32 eV), and the dispersion energy (E_d=10.03 eV). The electrical study leads to the dc activation energy is of the order of 0.72 eV and the maximum barrier high is W_M=0.87 eV. From the power exponent variation in terms of the heated temperature, it is found that the mechanism of conduction matches well the correlated barrier hopping CBH model.     

Detailed study about influence of oxygen on trap properties in SiOxNy by the thermally stimulated current and maximum entropy method

The trap amount depending on trap energy levels [N_t(E_t)] in various silicon oxynitride films were investigated. Using the thermally stimulated current and the maximum entropy method, we determined N_t(E_t) with very high energy resolution. In N_t(E_t), many E_t were observed between 1.2 and 1.6 eV. Interestingly, their amounts significantly depended on the film compositions. The influence of oxygen on N_t(E_t) is also discussed.     

Comparative Study of Shallow Acceptor Levels in Unintentionally Doped p-Type GaSe Crystals Prepared by the Bridgman and Liquid Phase Solution Growth Methods

GaSe crystals were grown by liquid phase solution growth, using the temperature difference method under controlled vapor pressure. Temperature-dependent Hall effect measurements were used to show the different thermal activation energies of the acceptor levels of the commercially available Bridgman-grown GaSe crystals and our liquid phase solution-grown crystals. Unintentionally doped GaSe crystals showed p-type conduction; also, the thermal activation energies of the acceptor levels of the Bridgman-grown crystals were 30 meV + E _v and 50 meV + E _v while that of liquid phase-grown crystals was 15 meV + E _v (where E _v is the energy of the valence band). We conclude that low temperature growth and stoichiometry control can effectively inhibit native point defect formation.     

Electron trap behaviour in Te-doped GaAs0.6P0.4

An electron trap spectrum has been obtained in Te-doped GaAsP by DLTs and transient capacitance measurements. The two traps identified display non-exponential emission and capture characteristics, the capture rate depending on temperature. The dominant trap A has an activation energy, E_a = 0.20 ± 0.02 eV and a constant concentration in the epilayer of typically 0.1N_d, trap B has an activation energy, E_a = 0.4 eV.The defect is donor related and characterised by non-radiative capture and lattice-relaxation multiphonon emission. Photocapacitance measurements provide the electron photoionization cross-section of the centre, and in agreement, a phonon broadened lineshape theory gave a threshold of 0.62 eV supporting the large lattice relaxation model. Evidence for persistent photoconductivity is also presented.     

Measure the barrier height of manganite p–n heterojunction by activation energy measurement

The barrier height of the manganite based p–n heterojunction is identified from the activation energy. The La_0.35Pr_0.32Ca_0.33MnO₃/Nb-doped SrTiO₃ p–n heterojunction is fabricated by the pulse laser deposition technology. The junction shows good rectifying behavior which can be well described by the Shockley equation. A satisfactorily logarithmic linear dependence of resistance on temperature is observed, and also the relation between bias and activation energy (E_A) deduced from the R−1/T curves is linear. As a result, the interfacial barrier of the heterojunction is obtained by extrapolating the Bias –E_A plot to Y axis, which is 0.95 eV.     

Activation energies for the different electromigration mechanisms in aluminum

The activation energies concerning all electromigration contributions in aluminum are theoretically determined starting from the activation energy for self-diffusion. For grain boundary electromigration the activation energy was found to be E_g = 0.4–0.5 ± 0.04 eV, whereas all other contributions except the surface electromigration are described by higher activation energies. For comparing with experimental results the grain boundary activation energy was determined by direct measurement of aluminum drift velocities. It is shown that the known measuring method is valid only in combination with some modifications. The measured activation energies agree well with the estimated ones.     

The effects of gate length variation and trapping effects on the transient response of AlGaN/GaN HEMT’s on SiC substrates

In AlGaN/GaN heterostructure field-effect transistors, the surface defects and dislocations may serve as trapping centers and affect the device performance via leakage current. In this paper we report results of our investigation of the trapping characteristics of Al_0.25Ga_0.75N/GaN HEMT using the Conductance Deep Level Transient Spectroscopy (CDLTS). Two deep level electronic defects were observed labeled E₁ and HL₁, with activation energies E_a1 = 1.36 eV and E_a2 = 0.63 eV. The hole-trap HL₁ is characterized for the first time in our studies. We identified the characteristics of the traps at the AlGaN/GaN interface adjoining the channel and the surface along the ungated region between the gate and the drain, as well as the effects of the surface traps.     

Temperature dependence of residual stress in epitaxial GaAs/Si(100) films determined from photoreflectance spectroscopy data

In the temperature range T=10–300 K, photoreflectance spectroscopy was used to study the temperature dependence of residual stress in epitaxial n-GaAs films (1–5 μm thick, electron concentration of 10¹⁶–10¹⁷ cm⁻³) grown on Si(100) substrates. A qualitative analysis showed that the photoreflectance spectra measured in the energy region of the E ₀ transition in GaAs had two components. They consisted of the electromodulation component caused by the valence subband |3/2; ±1/2〉-conduction band transition and the low-energy excitonic component. The magnitude of stress was determined from the value of the strain-induced energy shift of the fundamental transition from the subband |3/2; ±1/2〉 with respect to the band gap of the unstressed material E ₀(T)-E ₀ ^{|3/2; ±1/2〉} (T). The increase in the energy shift E ₀-E ₀ ^{|3/2; ±1/2〉} from 22 ± 3 meV at 296 K to 29 ± 3 meV at 10 K, which was found in the experiments, gives evidence of an increase in biaxial stress with decreasing temperature. __________ Translated from Fizika i Tekhnika Poluprovodnikov, Vol. 34, No. 1, 2000, pp. 73–80. Original Russian Text Copyright ? 2000 by Kuz’menko, Ganzha, Bochurova, Domashevskaya, Schreiber, Hildebrandt, Mo, Peiner, Schlachetzki.     

The ESD protection characteristic and low-frequency noise analysis of GaN Schottky barrier diode with fluorine-based plasma treatment

In this work, the electrostatic discharge (ESD) protection of Schottky diode with fluorine-based plasma treatment is proven by transmission-line pulse (TLP) measurement. And the low-frequency noise (LFN) is used to determine the activation energy (E_a) of a GaN Schottky diode with plasma treatment. This experiment compares the Schottky diode with CF₄ and CHF₃ plasma treatment, respectively with a standard Schottky diode to determine the characteristics and to assess the low-frequency noise and the ESD protection ability.     

Electrical and optical properties of AgInS2

Crystal sizes of AgInS₂ grown by a directional freezing depend on sulphur pressures at the preparation. The conductivity is only n-type and nominally undoped AgInS₂ has the resistivity of 25 Ω-cm and the Hall mobility of 64 cm²/V sec. Sulphur vacancies of AgInS₂ become electron-trapping levels in the forbidden band. It is obtained from the measurements of thermally stimulated current that the levels lie at E_T1 = 0·19±0·01 eV and E_T2 = 0·40±0·01 eV, and the concentrations depend on sulphur pressures at the crystal preparations. AuAgInS₂ contacts operate as a Schottky barrier diode and the barrier height is 0·97 eV. AgInS₂ has a dichroism because of its uniaxial lattice structure. The transition is direct for $\text{E}\text{⊥c}$ and indirect for $\text{E}\text{∥c}$, and the values for the energy gap are E_g^⊥ = 1·88±0·01 eV and E_g^∥ = 1·77±0·01 eV, respectively.     

Effects of Bonding Wires and Epoxy Molding Compound on Gold and Copper Ball Bonds Intermetallic Growth Kinetics in Electronic Packaging

This paper discusses the influence of bonding wires and epoxy mold compounds (EMC) on intermetallic compound (IMC) diffusion kinetics and apparent activation energies (E _aa) of CuAl and AuAl IMCs in a fineline ball grid array package. The objective of this study is to study the CuAl and AuAl IMC growth rates with different epoxy mold compounds and to determine the apparent activation energies of different combination of package bills of materials. IMC thickness measurement has been carried out to estimate the coefficient of diffusion (D _o) and E _aa various aging conditions of different EMCs and bonding wires. Apparent activation energies (E _aa) of both wire types were investigated after high temperature storage life tests (HTSL) for both molding compounds. Au bonds were identified to have faster IMC formation, compared to slower IMC growth of Cu. The E _aa obtained for CuAl IMC diffusion kinetics are 1.08 and 1.04 eV with EMC A and EMC B, respectively. For AuAl IMC diffusion kinetics, the E _aa obtained are 1.04 and 0.98 eV, respectively, on EMC A and EMC B. These values are close to previous HTSL studies conducted on Au and Cu ball bonds and are in agreement to the theory of HTSL performance of Au and Cu bonding wires.Overall, EMC B shows slightly lower apparent activation energy (E _aa) valueas in CuAl and AuAl IMCs. This proves that the different types of epoxy mold compounds have some influence on IMC growth rates.     

Autocorrelation measurements of CdSxSe1-x thin film noise

Autocorrelation functions of generation-recombination noise of CdS_xSe_1-x thin films illuminated by white light are presented. These functions can be described by a sum of exponential terms corresponding to the Lorentzians that represent the spectral density function. For some positions of the quasi-Fermi level E_Fn, however, a periodic part of the autocorrelation is observed whose properties depend strongly on the Fermi energy E_Fn. The Wiener-Khintchine transformation of the periodic autocorrelation functions reveals a narrow peak in the spectral density function. A simple condition for the presence of a harmonic autocorrelation is derived that leads to an inverted occupancy of a multilevel trap system.     

An indirect measurement of energy gap for silicon and germanium

Using temperature characteristics obtained from the reverse saturation current of p-n junctions and the forward-bias voltage, the energy gaps of Si and Ge can be indirectly measured at 0 K by means of linear extrapolation. Considering the different components of the saturation currents of p-n junctions for Si and Ge, we have forsaken the methods prevalent in the usual CAD of electronic circuits and, instead, have derived two distinct equations to measure the E_g0 of Si and Ge. The paper shows, after including band-gap narrowing effects, that the extrapolated intrinsic E_g0 of the sample Si approaches the generally accepted values, with an error within 0.8–1.0% for Si. The E_g0 for doped Ge is 0.67–0.75 eV approximately, when using, a small power transistor as sample. The model of physics used in the test has been shown to be clear-cut and simple, thus providing a simple way for testing the parameter E_g0, both practically and pedagogically.     

Effects of grain-boundary trapping-state energy distribution on the activation energy of resistivity of polycrystalline-silicon films

The trapping model assuming a δ-function energy distribution of trapping states has been accepted as an effective first-order approximation for modeling the electrical properties of polysilicon films. However, it predicts that as the doping concentration N is smaller than a critical level $\text{N}{}^1$, the activation energy of resistivity E_a is independent of N. This is inconsistent with experimental observations. In this paper a trapping model using a Gaussian energy distribution of trapping states is introduced to calculate E_a vs N. The results demonstrate a good agreement with the experimental data of boron-doped polysilicon films. The physical bases of such an improvement and the existence of a Gaussian energy distribution of trapping states have been addressed.