Studies of the promising material CoSi2 on GaAs substrates

The thermal stability of CoSi₂ thin films on GaAs substrates has been studied using a variety of techniques. The CoSi₂ thin films were formed by depositing Co(500 Å) and Si(1800 Å) layers on GaAs substrates by electron-beam evaporation followed by annealing processes, where the Si inter-layer was used as a diffusion/reaction barrier at the interface. The resistivity of CoSi₂ thin films formed is about 30 cm. The Schottky barrier height of CoSi₂/n-GaAs is 0.76 eV and the ideality factor is 1.14 after annealing at 750° C for 30 min. The CoSi₂/GaAs interface is determined to be thermally stable and the thin film morphologically uniform on GaAs after 900° C/30 s anneal. The CoSi₂ thin films fulfill the requirements in GaAs self-aligned gate technology.     

Differential evaluation of the Hall effect in silicon with oxygen-related donors

Three kinds of samples were used to form Co suicides by thermal annealing: firstly, a Co film of about 370 Å thick, evaporated on a (100) single crystal Si (Si ^c /Co); secondly, an evaporated boron-containing Si (Si ^e (B)) layer on the top of the first sample (Si ^c /Co/Si ^e (B)). The last sample is in the Co film of the first sample we deposited a Si^e(B) layer (Si ^c /Co/Si ^e (B)/Co). A laterally uniform CoSi₂ layer can be formed from the second and the third samples by annealing at 450 °C. In the first sample, the CoSi₂ can be formed only at temperatures above 500 °C and the disilicide is laterally less uniform than in the second and third samples. The Schottky barrier heights of the three samples derived from the forward and reverse I–V characteristics show that the barrier height is 0.01–0.02eV higher in the uniform case than in the nonuniform case.     

Thermal stability of nanoscale silver metallization in Ag/W/Co/Si(1 0 0) multilayer

In this work, we have studied thermal stability of nanoscale Ag metallization and its contact with CoSi₂ in heat-treated Ag(50 nm)/W(10 nm)/Co(10 nm)/Si(1 0 0) multilayer fabricated by sputtering method. To evaluate thermal stability of the systems, heat-treatment was performed from 300 to 900 °C in an N₂ ambient for 30 min. All the samples were analyzed by four-point-probe sheet resistance measurement (R_s), Rutherford backscattering spectrometry (RBS), X-ray diffractometry (XRD), and atomic force microscopy (AFM). Based on our data analysis, no interdiffiusion, phase formation, and R_s variation was observed up to 500 °C in which the Ag layer showed a (1 1 1) preferred crystallographic orientation with a smooth surface and R_s of about 1 Ω/□. At 600 °C, a sharp increase of R_s value was occurred due to initiation of surface agglomeration, WSi₂ formation, and interdiffusion between the layers. Using XRD spectra, CoSi₂ formed at the Co/Si interface preventing W silicide formation at 750 and 800 °C. Meantime, RBS analysis showed that in this temperature range, the W acts as a cap layer, so that we have obtained a W encapsulated Ag/CoSi₂ contact with a smooth surface. At 900 °C, the CoSi₂ layer decomposed and the layers totally mixed. Therefore, we have shown that in Ag/W/Co/Si(1 0 0) multilayer, the Ag nano-layer is thermally stable up to 500 °C, and formation of W-capped Ag/CoSi₂ contact with R_s of 2 Ω/□ has been occurred at 750-800 °C.     

Silicide formation by furnace annealing of thin Si films on large-grained Ni substrates

Si films with a thickness of approximately 250 nm have been electron-beam evaporated on thick, large-grained Ni substrates (grain size a few mm to 1 cm in diameter). An in situ sputter cleaning procedure has been used to clean the Ni surface before the Si deposition. Thermal annealings have been performed in a vacuum furnace. Ni₂Si is the first phase that grows at temperatures between 240 °C and 300 °C as a laterally uniform interfacial layer with a diffusion-controlled kinetics. The layer thicknessx follows the growth lawx ²=kt, withk=k ₀ exp(-E _a k _B T), wherek ₀=6.3 × 10^–4cm 2/s andE _a=(1-1±0.1) eV. Because of the virtually infinite supply of Ni, annealing at 800 °C for 130min yields a Ni-based solid solution as the final phase. The results are compared with those reported in the literature on suicide formation by the reaction of a thin Ni film on Si substrates, as well as with those for interfacial phase formation in Ni/Zr bilayers.     

Epitaxial nickel and cobalt suicide formation by rapid thermal annealing

Thin films of epitaxial NiSi₂ and CoSi₂ were formed by short-duration incoherent light exposure of evaporated Ni or Co films on <111> Si single crystals. The crystalline quality of these suicides is comparable to what has been obtained for long-duration furnace annealed suicides, as deduced from channeling measurements. NiSi₂ is of high crystalline quality at all temperatures at which it is formed whereas the CoSi₂ films recrystallize at a temperature of 980°C.     

Comparison of profile tailing in SIMS analyses of various impurities in silicon using nitrogen,oxygen, and neon ion beams at near-normal incidence

We have performed a systematic SIMS study into the effect of (i) the chemical nature and (ii) the energy of the primary ions on the decay length which characterizes the exponential fall-off of impurity sputter profiles. The samples consisted of low resistivity, p-type Si covered with thin metallic overlayers. Bombardment was carried out at 2° off normal. Aspect (i) was investigated for tracers of Cu and Ga using N ₂ ⁺ , O ₂ ⁺ , and Ne⁺ primary ions at an energy of 5 keV/atom. The effect of the beam energy, aspect (ii), was studied for eight different tracer species and N ₂ ⁺ primary ions at energies between 2 and 5 keV/atom. In the case of Ga, was found to be shorter with N ₂ ⁺ or O ₂ ⁺ primary ions (=7.0 and 7.5 nm, respectively) than with Ne⁺ (=12 nm). This effect is attributed to beam induced formation of Si₃N₄ or SiO₂ layers, whereby the effective width of the internal distribution of intermixed Ga impurities in the Si subsystem is reduced significantly. In contrast to Ga, the decay length for Cu is smallest under bombardment with Ne⁺ (=16 nm), quite large with N ₂ ⁺ (26 nm) and extremely large with O ₂ ⁺ (2.2 m). Segregation of Cu atoms at the Si₃N₄/Si and the SiO₂/Si interface, respectively, is responsible for this depressed impurity removal rate. Within experimental accuracy the observed variation of the decay length with N ₂ ⁺ energy E [keV/atom] can be written in the form =kE ^p, where k and p are element specific parameters which range from k=1.2 nm for Pb to 10 nm for Cu and from p=0.6 for Cu and Ag to 1.0 for Pb. The results are discussed with reference to conceivable shapes of the distribution of intermixed impurity atoms.On leave from NTT Applied Electronics Laboratories, 3-9-11, Midori-cho, Musashino-shi, Tokyo 180, Japan     

Ion mixing in Al,Si, and their oxides

The redistribution of thin metallic markers due to ion irradiation was studied by backscattering spectrometry in Al, Al₂O₃, Si, and SiO₂. Marker species were selected for their similar masses and different chemical reactivities with the host media and included Ti, Fe, W, Pt, and Au. It was found that the marker signals are Gaussian and that the variance ² of the marker atom distributions increases linearly with the dose of the irradiation, is insensitive to the temperature of irradiation in the range of 80–300 K, and depends linearly on the nuclear stopping power of the incident ions. The absolute values of ² for Ti, Fe, W, Pt, and Au markers in Al and Al₂O₃, W, and Pt in SiO₂ and W in Si is, within±50 %, of 6.5×10³Å² for 300 keV, 8×10¹⁵ Xe ions/cm². These observations suggest that collisional cascade mixing is a dominant mechanism in this type of impurity-matrix combinations. Only Au and Pt in Si mix at a larger rate: ² for Pt is about 3 and for Au about 5 times larger than ² for all other markers. Lower threshold displacement energies and/or the contribution of processes other than cascade mixing are possible considered reasons. In polycrystalline Al, a rapid migration of Au and Pt atoms throughout the Al layer, similar to grain boundary diffusion, is observed.     

Structure transition of single-texture CoSi2 nanolayer grown by refractory-interlayer-mediated epitaxy method

In this investigation, the crystalline structure of a nanometric CoSi₂ layer, formed in heat treated Co/W_xTa_(1−x)/Si(1 0 0) systems, has been studied by XRD analysis. Careful measurements of the diffraction intensities revealed that temporary formation of a metastable diamond cubic structure of CoSi₂ phase, rather than its usual CaF₂ structure, was occurred. It has been shown that formation of this metastable structure depends on the kind of the applied interlayer in addition to the annealing temperature. Among the studied systems with x = 0, 0.25, 0.5, 0.75 and 1, the second and the last systems resulted in growing a (1 0 0) single-texture CoSi₂ layer with the preferred usual CaF₂ structure, a strained lattice parameter, and the best thermal stability (900-1000 °C).     

Observation of a structural phase transition in a CoSi2 layer buried in 〈111〉 Si

Single crystalline multilayered structures of Si/CoSi₂/Si111 made by high dose implantation of Co in a Si wafer were investigated with⁵⁷Co source Mössbauer spectroscopy, channeling measurements and X-ray diffraction. The results point to a structural phase transition in the CoSi₂ buried layer between 180 and 220 K.     

Thermal oxidation of cobalt disilicide

The thermal oxidation kinetics of cobalt disilicide on Si substrates have been investigated in the temperature range of 650–1100 °C in dry oxygen and wet oxygen. A surface layer of SiO₂ grows parabolically with time. The growth rate is independent of the substrate orientation (111 or 100) and thickness of the CoSi₂ layer. We surmize that the oxidation mechanism is dominated by the diffusion of an oxidant through the growing SiO₂. Activation energies for the dry and wet oxidation are 1.49±0.05 eV and 1.05±0.05 eV, respectively. The kinetics is exactly the same as for NiSi₂ oxidation which suggest that the same mechanism controls the oxidation of these two similar suicides.     

Measurement of the πmeson polarizabilities via the γp → γπ+n reaction

An experiment on the radiative ^{+}-meson photoproduction from the proton ( p ^{+}n) was carried out at the Mainz Microtron MAMI in the kinematic region 537MeV < E < 817MeV, 140^° 180^°. The ^{+}-meson polarizabilities have been determined from a comparison of the data with the predictions of two different theoretical models, the first one being based on an effective pole model with pseudoscalar coupling while the second one is based on diagrams describing both resonant and nonresonant contributions. The validity of the models has been verified by comparing the predictions with the present experimental data in the kinematic region where the pion polarizability contribution is negligible ( s₁ < 5m²) and where the difference between the predictions of the two models does not exceed 3%. In the region, where the pion polarizability contribution is substantial ( 5 < s₁/m² < 15, -12 < t/m² < - 2), the difference of the electric () and the magnetic () polarizabilities has been determined. As a result we find . This result is at variance with recent calculations in the framework of chiral perturbation theory.     

Chemical bonding and interface analysis of ultrathin silicon-nitride layers produced by ion implantation and Electron Beam Rapid Thermal Annealing (EB-RTA)

¹⁵N ₂ ⁺ ions were implanted into c-Si with an energy of 5 keV/atom and fluences ranging from 5×10¹⁶ to 2×10¹⁷ atoms/cm² at RT to form ultrathin silicon-nitride layers (SiN _x ) with different N/Si ratios depending on the fluences (up to an overstoichiometric N/Si ratio of 1.65). The ¹⁵N depth distributions were analysed by the resonant nuclear reaction ¹⁵N(p, )¹²C(E _res=429 keV). The implanted samples were processed by Electron Beam Rapid Thermal Annealing (EB-RTA) at 1150° C for 15 s (ramping up and down 5° C/s). The chemical structure of the ¹⁵N implantation into Si was investigated by EXAFS and NEXAFS. Channeling-RBS (⁴He⁺, E ₀=1.5 MeV) measurements were performed to observe the transition region (disordered-Si layer, d-Si) being underneath of the SiN _x layer (typical values of layer thicknesses:SiN _x 24 nm, d-Si 6 nm).     

Rates of change in high temperature electrical resistivity and oxygen diffusion coefficient in Ba2YCu3Ox

The change in electrical resistance with time for bulk, thick-film, and thin-film Ba₂YCu₃O_x at atmospheric pressure is described as a function of the oxygen partial pressure (100 to 0.001%) and temperature (320°–750°C). The potential usefulness of these materials as oxygen sensors is demonstrated. The rate of equilibration is faster during oxygen uptake than during its loss. Time constants to reach equilibration (1/e remaining), qualitatively scale with sample dimensions. For a 1m film at 600° C, <1 s for the range of P_O2 (O2 being a shorthand for O₂) from 100% to 0.001%. The rate increases markedly with increasing P_O2. The actual resistance decreases with P_O2 at a rate of log/log P_O2 = 0.4 at 700° C showing adequate sensitivity for sensor purposes. Times for the transient resistance change in the sample where used to estimate the oxygen diffusion coefficient in the ceramic. The diffusivities obtained are 4·10^–11–1·10^–12 cm²/s in the 435°–320° C range, with an activation energy of 27 kcal/mole.     

Temperature dependent preferential sputtering in CoSi2 and NbSi2

Sputtering of CoSi₂ and NbSi₂ has been carried out by Xe ion bombardment at room temperature, as well as at elevated temperatures putting these systems in their radiation-enhanced diffusion regimes. The range of the Xe ions (at 200–260 keV) was appreciably less than the thickness of the silicides. The samples were analyzed by 2 MeV He⁺ backscattering spectrometry, x-ray diffraction and optical microscopy. The ratio of the sputtering yield of Si to that of the metal (i.e., Co or Nb) always exceeds the stoichiometric ratio 21, leading to Si depleted surface layers. The amount of the sputtered species increases almost linearly with dose until intermixing of the silicide with the underlying Si becomes appreciable. This happens at lower doses in the radiation-enhanced diffusion regime than at room temperature. Irradiation of CoSi₂ samples at high temperature leads to a broadening of the implanted Xe profile compared to the room temperature profile. No such phenomenon has been found in NbSi₂. The effect of Xe broadening on the sputtering yields is discussed.     

The Debye-Waller factor and the diffusion process for hydrogen in Nb,Ta, and V single crystals investigated by neutron spectroscopy

Quasielastic scattering of slow neutrons on hydrogen diffusing in the-phase of NbH_0.02, TaH_0.13 and VH_0.07 single crystals was investigated in a wide range of temperatures and scattering vectorsQ (0.5Q2.5 Å^–1). The incoherent scattering lawS _d (Q,) for four different diffusion models was consistently compared with the measured lineshapes. At elevated temperatures one had to introduce correlated jumps to describe the experimental data, whereas at room temperature a model with jumps between adjacent sites is sufficient. The integrated quasielastic intensityI(Q) obtained from the fit ofS _d (Q,) with the measured spectra follows an isotropic Debye-Waller factor with mean square amplitudes u ²=0.02–0.04 Ų for H in Ta (20°C–500°C), and u ²=0.03–0.04 Ų for H in Nb (20°C–300°C). For H in V,I(Q) obtained from the analysis of the quasielastic scattering deviates from a normal Debye-Waller factor behaviour. This effect is assumed to be due to the flight process between the interstitial sites. On the other hand, a normal Debye-Waller factor was obtained from theQ-dependence of the inelastic scattering of the band modes, with values of u ²=0.02–0.04 Ų (135°C–500°C). The observed values of u ² were compared with theoretical calculations.     

CEMS Investigations of Fe-Silicide Phases Formed by the Method of Concentration Controlled Phase Selection

Conversion electron Mössbauer spectroscopy (CEMS) measurements have been made on Fe-silcide samples formed using the method of concentration controlled phase selection. To prepare the samples a 10 nm layer of Fe₃₀M₇₀ (M=Cr, Ni) was evaporated onto Si(100) surfaces, followed by evaporation of a 60 nm Fe layer. Diffusion of the Fe into the Si substrate and the formation of different Fe–Si phases was achieved by subjecting the evaporated samples to a series of heating stages, which consisted of (a) a 10 min anneal at 800°C plus etch of the residual surface layer, (b) a further 3 hr anneal at 800°C, (c) a 60 mJ excimer laser anneal to an energy density of 0.8 J/cm², and (d) a final 3 hr anneal at 800°C. CEMS measurements were used to track the Fe-silicide phases formed. The CEMS spectra consisted of doublets which, based on established hyperfine parameters, could be assigned to - or -FeSi₂ or cubic FeSi. The spectra showed that -FeSi₂ had formed already at the first annealing stage. Excimer laser annealing resulted in the formation of a phase with hyperfine parameters consistent with those of -FeSi₂. A further 3 hr anneal at 800°C resulted in complete reversal to the semiconducting -FeSi₂ phase.     

Formation of Ohmic contacts to n-GaAs using ion beam mixing of Tellurium

Ohmic contacts were formed on n-GaAs using thin evaporated layers of Te followed by bombardment of 100 keV Ar⁺ ions. The specific contact resistance _c showed a strong dependence on the ion dose in the range 10¹⁴ to 10¹⁶ ions cm^–2, with higher doses leading to progressively lower specific contact resistance. The substrate temperature during ion bombardment was varied in the range from 25 to 200° C and was found to have only a minor effect on the resultant values of _c. Elevated temperature aging of the Ohmic contacts at 200° C resulted in a progressive increase in the specific contact resistance, independent of either the ion dose or the substrate temperature used for ion beam mixing. Rutherford backscattering studies (RBS) indicate that the Ohmic contact behaviour was due to the in-diffusion of Te and subsequent formation of a heavily doped n ⁺ layer at the Te-GaAs interface.     

Reduction of optical absorption at a wavelength of around 0.8 μm in LPE garnet (TmBiCa)3(FeGaPt)5O12

Liquid-phase epitaxy (LPE) garnet films (TmBiCa)₃(FeGaPt)₅O₁₂ have been grown using only Bi₂O₃ as the flux so that the film containing Bi gives high specific Faraday rotation. The film does not contain Pb, which may affect optical absorption. The optical absorption coefficient at 810 nm has been effectively reduced by doping Ca in the melt. Our data show that a minimum level of and of the anisotropy constantK _u and also the maximum of the electrical resistivity are achieved when Ca²⁺ replaces Fe²⁺. Fe²⁺ results from Pt⁴⁺ incorporation in the film due to a Bi₂O₃ flux attack on a Pt crucible. Using a compensated film, of 58 cm^–1 and a figure of merit of 9deg/dB were obtained.     

Nitric acid oxidation of Si (NAOS) method for low temperature fabrication of SiO2/Si and SiO2/SiC structures

We have developed low temperature formation methods of SiO₂/Si and SiO₂/SiC structures by use of nitric acid, i.e., nitric acid oxidation of Si (or SiC) (NAOS) methods. By use of the azeotropic NAOS method (i.e., immersion in 68 wt% HNO₃ aqueous solutions at 120 °C), an ultrathin (i.e., 1.3-1.4 nm) SiO₂ layer with a low leakage current density can be formed on Si. The leakage current density can be further decreased by post-metallization anneal (PMA) at 200 °C in hydrogen atmosphere, and consequently the leakage current density at the gate bias voltage of 1 V becomes 1/4-1/20 of that of an ultrathin (i.e., 1.5 nm) thermal oxide layer usually formed at temperatures between 800 and 900 °C. The low leakage current density is attributable to (i) low interface state density, (ii) low SiO₂ gap-state density, and (iii) high band discontinuity energy at the SiO₂/Si interface arising from the high atomic density of the NAOS SiO₂ layer.For the formation of a relatively thick (i.e., ≥10 nm) SiO₂ layer, we have developed the two-step NAOS method in which the initial and subsequent oxidation is performed by immersion in ∼40 wt% HNO₃ and azeotropic HNO₃ aqueous solutions, respectively. In this case, the SiO₂ formation rate does not depend on the Si surface orientation. Using the two-step NAOS method, a uniform thickness SiO₂ layer can be formed even on the rough surface of poly-crystalline Si thin films. The atomic density of the two-step NAOS SiO₂ layer is slightly higher than that for thermal oxide. When PMA at 250 °C in hydrogen is performed on the two-step NAOS SiO₂ layer, the current-voltage and capacitance-voltage characteristics become as good as those for thermal oxide formed at 900 °C.A relatively thick (i.e., ≥10 nm) SiO₂ layer can also be formed on SiC at 120 °C by use of the two-step NAOS method. With no treatment before the NAOS method, the leakage current density is very high, but by heat treatment at 400 °C in pure hydrogen, the leakage current density is decreased by approximately seven orders of magnitude. The hydrogen treatment greatly smoothens the SiC surface, and the subsequent NAOS method results in the formation of an atomically smooth SiO₂/SiC interface and a uniform thickness SiO₂.     

Mössbauer Optimization of the Direct Synthesis of β-FeSi2 by Ion Beam Mixing of Fe/Si Bilayers

At present, there is an increasing interest in the iron di-silicide phase -FeSi₂, which is supposed to be a direct band gap semiconductor and one of the most promising materials for silicon-based optoelectronics, e.g., light-emitting devices, solar cells, and photo detectors. But this phase is very difficult to be produced. Here, the successful direct synthesis of this phase by ion beam mixing of Fe/Si bilayers at temperatures in the range of 400 to 600°C is reported. The aim of the experiments was to achieve a complete reaction of the deposited Fe layer with the Si substrate that results in the formation of a pure, single-phased -FeSi₂ surface layer. The obtained silicide layers, their structure and composition are investigated by conversion electron Mössbauer spectroscopy (CEMS), Rutherford backscattering spectrometry (RBS), and X-ray diffraction (XRD). The fraction of the -FeSi₂ formed is determined by CEMS as function of ion species, energy, fluence and temperature. Complete growth and formation of a single-phased -FeSi₂ layer was achieved by 205 keV Xe ion irradiation at a fluence of 2×10¹⁶ ions/cm² at 600°C.