Mg2Si的电子结构和热电输运性质的理论研究

利用全势线性缀加平面波法,对Mg₂Si的几何结构和电子结构进行了计算,得到了稳定的晶格参数以及能带和电子态密度.能带结构表明,Mg₂Si为间接带隙半导体,禁带宽度为020 eV.在此基础上利用玻尔兹曼输运理论和刚性带近似计算了材料的电导率、Seebeck系数和功率因子.结果表明,在温度为700 K时p型和n型掺杂的Mg₂Si功率因子达到最大时的最佳载流子浓度分别为7749×10¹⁹ cm^-3和 关键词： 2Si')" href="#">Mg₂Si 全势线性缀加平面波法 热电输运性质     

Abstracts

Abstract

Isothermal annealing of amorphous Si and Ge has been performed by picosecond pulsed laser irradiation of free-standing films. It is found that the laser induced nucleation rate is about 10²¹-5.10²² cm^?3 s^?1 (Si) and 10²³-10²⁵ cm^?3 s^?1 (Ge) near the melting point. Arrhenius plots of the nucleation rate show that nucleation is thermally activated with an activation energy of about ΔE = 1.8 ± 0.1 eV (Ge) and ΔE = 2.47 ± 0.15 eV (Si).     

An interpretation of dlts spectra of Al/Si3N4/ultrathin SI/GaAs structures — Effect of quantum well or interface states?

We discuss DLTS andC-V measurements on Al/Si₃N₄/Si(2nm)/n-GaAs (≈ 5×10¹⁷ cm^?3) structures. Three discrete deep traps superimposed on a U-shaped interface-state continuum have been identified, with respective thermal energies:E _c?0.53 eV,E _c?0.64 eV, andE _v+0.69 eV. The second one (0.64 eV) is presented as an electric field sensitive level, its enhanced phonon-assisted emission resulting in a rapid shift of the corresponding DLTS peak to lower temperatures, as the applied (negative) reverse bias voltage increases. An interpretation through emission from the quantum well, introduced by means of the intermediate ultrathin Si layer, has failed.     

The influence of the annealing conditions on the photoluminescence of ion-implanted SiO2:Si nanosystem at additional phosphorus implantation

The influence of P ion doping on the photoluminescence (PL) of the system of nanocrystals in SiO₂ matrix (SiO₂:Si) both without annealing and after annealing at various temperatures (provided before and after additional P implantation) is investigated. The Si and P implantation was carried out with ion energies of 150 keV and doses Φ_Si=10¹⁷ cm⁻² and Φ_P=(0.1–300)×10¹⁴ cm⁻² (current density j3 μAcm⁻²). The system after Si implantation was formed at 1000°C and 1100°C (2 h). For the case of SiO₂:Si system as-implanted by P, the intensity of PL was drastically quenched, but partially retained. As for the step-by-step annealing (at progressively increased temperatures) carried out after P implantation, the sign and degree of doping effect change with annealing temperature. The possible mechanisms of these features are discussed.     

Chemical modification of the electrical properties of hydrogenated amorphous carbon films

Semiconducting films of hydrogenated amorphous carbon (a-C:H), prepared via the dc glow discharge decomposition of C₂H₂, have been successfully doped via incorporation of B and P during growth. The doping efficiency achieved was comparable to that achieved in a-Si:H produced in a like manner. For a-C:H films deposited at T_d=250C,

(RT) increased from 10^-12 to 10^-7 ohm^-1 cm^-1 when either 1% PH₃ or 10% B₂H₆ were added to the C₂H₂. A shift of the Fermi level E_F of about 0.7 eV is inferred from changes in the “activation” energy of conduction.     

Great enhancement in remnant polarization of SrBi4Ti4O15 thin films by A-site doping of Fe3+ ions

SrBi4-xFexTi4O15 (SBFT-x) thin films (x = 0.00, 0.05, 0.08, 0.15) have been synthesized on Pt/Ti/SiO2/Si (100) substrates by sol-gel method. This paper finds that Fe-doping does not change the crystal structure of SrBi4Ti4O15 (SBTi). The coercive filed (Ec) and remnant polarisation (Pr) increase at first, then decrease with the increase of Fe doping content. At a maximum applied field of 229 kV/cm, the 2Pr reaches a maximum value of 91.1 μC/cm2 at x = 0.05 and the corresponding Ec is 72 kV/cm. The 2Pr increases by about 260% and the Ec decreases by about 6%, respectively. Obviously, the ferroelectric property of SrBi4Ti4O15 is greatly enhanced by Fe doping. The fatigue-endurance characteristic of the SBFT-0.05 is not improved. After 1.1×109 read/write cycles at a frequency of 50 kHz, the nonvolatile polarisations (Pnv = P*-P∧) decreased about 48% of its initial value.     

Relation between interface states and temperature behavior of the barrier height of silver contacts on clean cleaved n-type silicon

The properties of silver-silicon interfaces formed by cleaving n-type silicon in ultra high vacuum (UHV) in a stream of evaporating silver atoms were studied. The barrier heights of these contacts were measured at different temperatures by using C-V techniques. All measurements were performed in UHV. The dependence of the barrier height upon temperature did not follow the temperature dependence of the Si band gap as it is usually found. The measured temperature behavior depended on the roughness of the Si surface. The temperature behavior can be explained by assuming a specific band structure of the interface states. For Ag contacts on atomically smooth n-type Si, the interface states were found to be arranged in two bands, one band 4 × 10^?3 eV wide with acceptor type states 0.18 eV below the intrinsic level E_i and a density of 10¹⁷ states/cm² eV, and the other 1 eV wide with donor type states with its upper edge 0.28 eV below E_i, and a density of 4 × 10¹⁴ states/cm²eV.     

Hopping conductivity of carbynes modified under high pressures and temperatures: Galvanomagnetic and thermoelectric properties

The conductivity, thermopower, and magnetoresistance of carbynes structurally modified by heating under a high pressure are investigated in the temperature range 1.8–300 K in a magnetic field up to 70 kOe. It is shown that an increase in the synthesis temperature under pressure leads to a transition from 1D hopping conductivity to 2D and then to 3D hopping conductivity. An analysis of transport data at T ≤ 40 K makes it possible to determine the localization radius a ～ (56?140) Å of the wave function and to estimate the density of localized states _g(E _F) for various dimensions d of space: _g(E _F) ≈ 5.8 × 10⁷ eV^?1 cm^?1 (d=1), _g(E _F) ≈5×10¹⁴ eV^?1 cm ^?2 (d=2), and _g(E _F)≈1.1×10²¹ eV^?1 cm_?3 (d=3). A model for hopping conductivity and structure of carbynes is proposed on the basis of clusterization of sp ² bonds in the carbyne matrix on the nanometer scale.     

Silicon negative ion implantation induced vacancy defects in thermally grown SiO2 thin films

ABSTRACT

Thermally grown SiO₂ thin films on a silicon substrate implanted with 100?keV silicon negative ions with fluences varying from 1?×?10¹⁵ to 2?×?10¹⁷ ions cm^?2 have been investigated using Electron spin resonance, Fourier transforms infrared and Photoluminescence techniques. ESR studies revealed the presence of non-bridging oxygen hole centers, E′-centers and P_b-centers at g-values 2.0087, 2.0052 and 2.0010, respectively. These vacancy defects were found to increase with respect to ion fluence. FTIR spectra showed rocking vibration mode, stretching mode, bending vibration mode, and asymmetrical stretching absorption bands at 460, 614, 800 and 1080?cm^?1, respectively. The concentrations of Si–O and Si–Si bonds estimated from the absorption spectra were found to vary between 11.95?×?10²¹ cm^?3 and 5.20?×?10²¹ cm^?3 and between 5.90?×?10²¹ cm^?3 and 3.90?×?10²¹ cm^?3, respectively with an increase in the ion fluence. PL studies revealed the presence of vacancies related to non-bridging oxygen hole centers, which caused the light emission at a wavelength of 720?nm.     

High Fe2+/3+ trap concentration in heavily compensated implanted InP

High Fe concentrations (up to 2×10¹⁹ cm^-3) have been implanted in n-doped InP to compensate the substrate donors. The resulting semi-insulating layers have been investigated by current–voltage (I-V) measurements and photo-induced current transient spectroscopy (PICTS) analyses to characterise the Fe activation process and to study the Fe related deep levels. The activation of the Fe^2+/3+ trap has been assessed by the identification of the deep level located at E_C-0.64 eV. The outcomes of the PICTS measurements have been correlated with the electrically active Fe concentration calculated from a numerical simulation of the I-V characteristics. We observe an increasing linear relation between the electrically active Fe concentration and the substrate doping density, with a maximum active Fe concentration as high as 2×10¹⁸ cm^-3, i.e. more than an order of magnitude above the equilibrium Fe solid solubility. These data are presented and their implications discussed. Received: 4 September 2000 / Accepted: 7 February 2001 / Published online: 23 May 2001     

Photoluminescence properties of Er-doped β-FeSi2 grown by ion implantation

Photoluminescence (PL) properties of Er-doped β-FeSi₂ (β-FeSi₂:Er) and Er-doped Si (Si:Er) grown by ion implantation were investigated. In PL measurements at 4.2 K, the β-FeSi₂:Er showed the 1.54 μm PL due to the intra-4f shell transition of ⁴I_13/2→⁴I_15/2 in Er³⁺ ions without a defect-related PL observed in Si:Er. In the dependence of the PL intensity on excitation photon flux density, the obtained optical excitation cross-section σ in β-FeSi₂:Er (σ=7×10⁻¹⁷ cm²) is smaller than that in Si:Er (σ=1×10^-15 cm²). In the time-resolved PL and the temperature dependence of the PL intensity, the 1.54 μm PL in β-FeSi₂:Er showed a longer lifetime and larger activation energies for non-radiative recombination (NR) processes than Si:Er. These results revealed that NR centers induced by ion implantation damage were suppressed in β-FeSi₂:Er, but the energy back transfer from Er³⁺ to β-FeSi₂ was larger than Si:Er.     

Quenching of Si nanocrystal photoluminescence by doping with gold or phosphorous

Si nanocrystals embedded in SiO₂ doped with P and Au at concentrations in the range of 1×10¹⁸-3×10²⁰ cm⁻³ exhibit photoluminescence quenching. Upon increasing the Au concentration, a gradual decrease in nanocrystal photoluminescence intensity is observed. Using a statistical model for luminescence quenching, we derive a typical radius of ∼3 nm for nanocrystals luminescing around 800 nm. Au doping also leads to a luminescence lifetime reduction, which is attributed to energy transfer between adjacent Si nanocrystals, possibly mediated by the presence of Au in the form of ions or nanocrystals. Doping with P at concentrations up to 3×10¹⁹ cm⁻³ leads to a luminescence enhancement, most likely due to passivation of the nanocrystal-SiO₂ interfaces. Upon further P doping the nanocrystal luminescence gradually decreases, with little change in luminescence lifetime.     

The average structure of 5.10-Dihydro-5.10-Diethylphenazinium-Iodide (E2P·I1.6): A new linear chain Polyiodide compound

Upon oxidation of 5.10-dihydro-5.10-diethylphenazine (E₂P) with iodine golden-green lustrous crystals of a compound with stoichiometry E₂P.I_1.6 were isolated. The compound crystallizes in the tetragonal space group D₄² with $\text{a = 12.321(2)}\text{A}\text{?}$ and $\text{c = 5.330(2)}\text{A}\text{?}$. The E₂P and I form interpenetrating incommensurate sublattices along c, with an iodine repeat distance of 9.7 Å. Static susceptibility measurements at room temperature give χ_g = + 0.994 × 10^?6g^?1 × cm³. This corresponds to one unpaired electron spin per two formular units. Single-crystal EPR indicates that the paramagnetism is associated with weakly interacting E₂P⁺ cation radicals. The 300K-d.c. conductivity of 3×10^?2Ω^?1cm^?1 and activation energy of 0.17±0.02eV for single crystals is consequently associated with the polyiodide chains, and not with the E₂P⁺ cation radicals.     

注氮GaAs中的深能级及其对自由载流子的补偿

详细研究了注氮n型GaAs中深的和浅的杂质缺陷的电学性质。深能级瞬态谱(DLTS)技术测量表明,能量为140keV和剂量为1×10¹³cm^-2的氮离子注入并经800℃退火30min的GaAs中存在四个电子陷阱,E₁(0.111),E₂(0.234),E₃(0.415),E₄(0.669)和一个空穴陷阱H(0.545),而在能量为20keV和剂量为5×10^{14关键词：}     

High pressure tuning of optical transitions in Mg[Pt(CN)4]·7H2O

Mg[Pt(CN)₄]·7H₂O belongs to the class of tetracyanoplatinates(II) which crystallize in columnar structures. In different M_x[Pt(CN)₄]·yH₂O (MCP) single crystals the in-chain Pt-Pt-distance R varies between 3.67 Å (NaCP) and 3.15 Å (MgCP). Two optical transitions can be observed in polarized emission with the electric field vector E either parallel or perpendicular to the columnar (c)-axis. Polarized emission spectra of MgCP are recorded under hydrostatic pressure up to p ≈ 18 kbar (at 295 K). The transition energy v?₆ can be tuned from 17,600 cm^-1 to about 12,000 cm^-1 (2.18-1.48 eV). The pressure induced red shift for the two transitions is: E 6 c: dv?₆/dp = -320±20 cm^-1/kbar, E ⊥ c: dv?_⊥/dp = -270±20 cm^-1/kbar. These values are discussed in the context of the known functional relationship (for ambient conditions) between v? and R.     

Enhanced field emission from Si doped nanocrystalline AlN thin films

Si doped and undoped nanocrystalline aluminum nitride thin films were deposited on various substrates by direct current sputtering technique. X-ray diffraction analysis confirmed the formation of phase pure hexagonal aluminum nitride with a single peak corresponding to (1 0 0) reflection of AlN with lattice constants, a = 0.3114 nm and c = 0.4986 nm. Energy dispersive analysis of X-rays confirmed the presence of Si in the doped AlN films. Atomic force microscopic studies showed that the average particle size of the film prepared at substrate temperature 200 °C was 9.5 nm, but when 5 at.% Si was incorporated the average particle size increased to ∼21 nm. Field emission study indicated that, with increasing Si doping concentration, the emission characteristics have been improved. The turn-on field (E_to) was 15.0 (±0.7) V/μm, 8.0 (±0.4) V/μm and 7.8 (±0.5) V/μm for undoped, 3 at.% and 5 at.% Si doped AlN films respectively and the maximum current density of 0.27 μA/cm² has been observed for 5 at.% Si doped nanocrystalline AlN film. It was also found that the dielectric properties were highly dependent on Si doping.     

Formation of nanosize silicides films on the Si(111) and Si(100) surfaces by low-energy ion implantation

The formation of nanosize silicides films by implantation of B, P, Ba, and alkali metal atoms in Si(111) and Si(100) followed by thermal annealing is studied by electron spectroscopy and slow-electron diffraction methods. It is shown that implantation of ions with a large dose D > 10¹⁶ cm^?2 and short-term heating lead to the formation of thin silicides films with new surface superstructures: \(Si(111) - (\sqrt 3 \times \sqrt 3 )R30^ \circ - B\) , Si(100)-2 × 2Ba, Si(111)-1 × 1P, etc.     

Photodesorption from ultra-thin metal films – a comparison of SO<Subscript>2</Subscript> and NO<Subscript>2</Subscript> on Ag/Si(100)

The photochemistry of SO₂ on thin epitaxial Ag films (5–60 nm) deposited on Si(100) has been studied using laser light with the wavelengths of 266, 355, and 532 nm. SO₂ desorbs with cross sections of 1.7×10^-19,1.7×10^-20 and 2.9×10^-21 cm², respectively. The average translation energy, 〈E_trans/2k〉, is 440 K for 266 and 355 nm light, and 270 K for 532 nm light. Cross sections for a 60 nm thick Ag film are practically identical to the ones for Ag(111) as the substrate. An increase by a factor of ∼3.5 is observed when the film thickness is reduced to 5 nm for 266 and 355 nm light. No significant change is observed for 532 nm excitation. The film thickness has no significant influence on the translational energy of the photodesorbed molecules. The data are discussed in connection with the change of absorptivity of the metal film–semiconductor system. A model is put forward which takes into account the light absorption in the Si substrate and the reduced relaxation of excited electrons in Si. Modelling indicates that electrons excited in the Si substrate with energies and parallel momenta not allowed in Ag contribute to the surface chemistry after crossing the gap in the projected band structure of Ag(111). PACS 82.45.MP; 73.63.-b; 82.50.Bc     

Silicon doping of MBE-grown GaAs films

Two concentration ranges of silicon doping in MBE-grown GaAs films have been investigated in some detail. In lightly doped films, with a free-electron concentration of ≈10¹⁶cm^?3, low-temperature photoluminescence spectra have been analysed to develop a model to account for spectral features previously attributed to Ge and Si acceptor levels. In heavily doped films, a maximum free-electron concentration of ≈7×10¹⁸ cm^?3 has been obtained, which is only rather weakly dependent on growth conditions and the nature of the arsenic species (As₂ or As₄). Transmission electron microscopy has shown that no significant precipitation effects occur when higher Si fluxes are used but there is evidence for autocompensation. The maximum PL intensity (300 K) is found at a lower free electron concentration then with Sn-doped films, and is more sharply peaked, but there is no evidence for an anomalous Moss-Burstein shift.     

Optical spectroscopy, 1.5 μm emission and up-conversion properties of Er-doped Li3Ba2Gd3(MoO4)8 crystal

Er³⁺:Li₃Ba₂Gd₃(MoO₄)₈ crystal has been grown from a melt of Li₂MoO₄ by the top seeded solution growth method (TSSG). The polarized spectral properties of Er³⁺:Li₃Ba₂Gd₃(MoO₄)₈ crystal were investigated and the spectroscopic parameters were calculated and analyzed based on the Judd-Ofelt (J-O) theory. The emission cross-sections were calculated by the Fuchtbauer-Ladenburg (F-L) equation and the peak values of the emission band at 1535 nm were 9.7×10⁻²¹, 7.9×10⁻²¹ and 8.4×10⁻²¹ cm² for E∥b, E∥D1 and E∥D2, respectively. Under 977 nm excitation five up-conversion fluorescence bands around 490, 530, 550, 660 and 800 nm were observed, and the possible up-conversion mechanisms were proposed.