Search

In this paper, normal zone in a REBCO single-pancake coil induced by quenches in a low-temperature superconducting outsert coil is investigated in both experimental and numerical methods to discuss the quench protection of REBCO coils. The normal zone is experimentally clarified by measuring a partial voltage in the REBCO coil at temperatures from 30 K to 50 K during quenches of the outsert coil charging 5 T. Using practical fittings for the physical properties of the components of the REBCO coil in ranges of 0-5 T and 10-77 K, the numerical results have good agreement with the experimental results. The simulation reveals the heating distribution with initial temperatures from 10 K to 65 K, which indicates that the maximum temperature in a hot spot becomes higher for the lower initial temperature. Unexpectedly, the temperature distribution becomes rather homogeneous with an initial temperature of 65 K. It suggests that the hot spot at higher temperature has a lower risk of burning out. Finally, possible coil protections are discussed as the operation at the high temperature or the quench heater method to keep the REBCO coil at high temperature when a hot spot appears.

Single photon emission with high degree of circular polarization (DCP) from a positively charged exciton (trion) state in a single quantum dot (QD) is demonstrated under zero magnetic field. Obtained high DCP as large as 0.6 is a clear manifestation of highly effective mutual polarization conversion between single photon and single electron spin. This finding also reveals that the spin relaxation is strongly suppressed during capture and thermalization processes and also in trion ground states in a quantum dot. Moreover, exciton charging state control is shown to be possible by modifying the electron capture probability through the excitation energy tuning, which leads to an exclusive formation of the positive trion in a QD.

Origin of sharp photoluminescence lines observed from an InAlAs quantum dot was identified with the measurements of excitation-power dependences and polarization correlations, together with photon correlation measurements. Single photon emission with high degree of circular polarization (DCP) up to 60% was observed from a positively charged exciton (trion) state in the single quantum dot under non-resonant excitation at zero magnetic field.

In a single self-assembled InAlAs quantum dot, we report the nonlinearity of nuclear spin polarization by optical pumping. Increasing (or decreasing) the excitation power at a particular excitation polarization, nuclear spin polarization shows the abrupt jump (or drop) and a clear hysteresis loop. In addition, the degree of circular polarization of positively charged exciton also exhibits similar bistable behavior and the hysteretic operation has the precise synchronization with the nuclear spin polarization. It suggests that an electron in the positively charged exciton plays a main role to create the abrupt change of the nuclear spin polarization in the observed quantum dot. This abrupt change reveals the sign and magnitude of the electron g -factors in z -direction and can be explained by using a simple model calculation. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Nuclear magnetic resonance on oriented nuclei (NMR-ON) on Sc isotopes (A=44, 44 m, 46, 47) in Fe were performed. Using the known magnetic moments the magnetic hyperfine fields were determined. The hyperfine anomalies of various Sc were deduced; 44Δ47=0.0 (12) %, 44mΔ47=1.2 (4) %, 46Δ47=2.7 (8) %. The measured hyperfine anomalies are briefly discussed based on the shell model.

We investigated optical pumping of nuclear spin polarizations in a single self-assembled In0.75Al0.25As/Al0.3Ga0.7As quantum dot. The nuclear spin polarization exhibits the abrupt jump and hysteresis in the excitation power dependence at a particular excitation polarization. Measurement of circular polarization rate of the photoluminescence reveals that the abrupt change of the nuclear spin polarization is created mainly by the spin flip-flop process between nuclei and an electron of a positive charged exciton in this single quantum dot. Model calculation explains well the experimentally observed bistable behavior in InAlAs quantum dot. By using this abrupt change, the sign and magnitude of electron and hole g-factors in z-direction are verified.

We study the exciton states of vertically coupled InAlAs/AlGaAs self-assembled quantum dot (QD) pairs which were designed to have different wetting-layer edges in two QD layers. From a single QD pair, micro-photoluminescence demonstrates four exciton s levels as a result of interdot coupling. Consistent with theory, such a level can be split into four sub-levels. In general, 2-fold splitting is exhibited by the doublet under magnetic field. The doublet is of near-circular polarization in one side but of near-linear polarization in the other side. This abnormal behavior might be understood by heavy hole–light hole mixing. The other 2-fold splitting is observed by photoluminescence excitation measurement without magnetic field. This splitting is suggested resulting from the spatial asymmetry of the coupled QD pair. It is possible that these two split states are robust against magnetic field, and are selectively excited by using pump energy associated with different wetting layers.

Nonresonant carrier tunneling has been studied as a function of temperature in GaAs/AlGaAs double quantum wells (DQWs). Time-resolved pump and probe reflectance measurements allow the direct observation of tunneling at any temperature between 15 K and room temperature. We found that for two DQWs with different barrier thicknesses, the tunneling time abruptly decreases above a critical temperature while it remains almost constant below the critical temperature. This critical temperature is shown to correspond to the exciton binding energy. Rate equation analysis explains this behavior as the thermalization of excitons into free electrons that have a faster tunneling time than excitons.

We report optical pumpingof nuclear spinpolarizationinasingle self-assembledIn0.75Al0.25As/Al0.3Ga0.7As quantum dot where the exciton has a fine structure. Through the change of state mixing that is induced by the anisotropic exchange interaction and the applied magnetic field, the relationship between the magnitude of the Overhauser shift and the optically injected electron spin polarization is discussed. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We have investigated the influence of postgrowth thermal annealing on the optical properties of InAs/GaAs self-assembled quantum dots. Upon annealing, the emission from InAs QDs shifts toward higher energies with a narrowing of the photoluminescence spectra of QD ensemble. The time-integrated PL spectra show also the nonradiative recombination rate does not increase by annealing process. In addition, from the timeresolved PL, the exciton radiative recombination rate increases with increasing the annealing temperature and agrees well with the atomic-like radiative recombination. These results indicate that interdiffusion of Ga and In atoms into and out of the QDs leads to an increase of the average dot size and concurrently a decrease in the confinement potentials of the QDs preserving their optical quality and zero-dimensional density of states after interdiffusion in the studied annealing temperature range. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Triggered single-photon generation from InAlAs quantum dot (QD) was demonstrated for the first time. Emitted photon energy coincides with high detection efficiency range of Si single-photon detectors, which is highly suitable for free-space communication. Single-QD spectroscopy and crossed photon correlation measurements unambiguously revealed that several emitting lines observed in a single mesa structure originated from the identical QD, and two temporary competing decay processes associated with neutral states and charged states were identified. Presence of the competing process is also inferred from an analysis of steady-state photoluminescence intensities. Formation process of charged exciton in QD is also discussed.

Two types of improved core confinement were observed during centrally focused electron cyclotron heating (ECH) into plasmas sustained by counter (CNTR) and Co neutral beam injections (NBI) in the Large Helical Device. The CNTR NBI plasma displayed transition phenomena to the high-electron-temperature state and had a clear electron internal transport barrier, while the Co NBI plasma did not show a clear transition or an ECH power threshold but showed broad high temperature profiles with moderate temperature gradient. This indicated that the Co NBI plasma with additional ECH also had an improved core confinement. The electron heat transport characteristics of these plasmas were directly investigated using heat pulse propagation excited by modulated ECH. These effects appear to be related to the m/n = 2/1 rational surface or the island induced by NBI beam-driven current.

Characteristics of MHD instabilities and their impacts on plasma confinement are studied in current free plasmas of the Large Helical Device. Spontaneous L?H transition is often observed in high beta plasmas close to 2% at low toroidal fields (Bt ? 0.75?T). The stored energy starts to rise rapidly just after the transition accompanying the clear rise in the electron density but quickly saturates due to the growth of the m = 2/n = 3 mode (m and n: poloidal and toroidal mode numbers), the rational surface of which is located in the edge barrier region, and edge localized mode (ELM) like activities having fairly small amplitude but high repetition frequency. Even in low beta plasmas without L?H transitions, ELM-like activities are sometimes induced in high performance plasmas with a steep edge pressure gradient and transiently reduce the stored energy up to 10%. Energetic ion driven MHD modes such as Alfv?n eigenmodes (AEs) are studied in a very wide range of characteristic parameters (the averaged beta of energetic ions, ?b?, and the ratio of energetic ion velocity to the Alfv?n velocity, Vb?/VA), of which range includes all tokamak data. In addition to the observation of toroidicity induced AEs (TAEs), coherent magnetic fluctuations of helicity induced AEs (HAEs) have been detected for the first time in NBI heated plasmas. The transition of a core-localized TAE to a global AE (GAE) is also observed in a discharge with temporal evolution of the rotational transform profile, having a similarity to the phenomenon observed in a reversed shear tokamak. At low magnetic fields, bursting TAEs transiently induce a significant loss of energetic ions, up to 40% of injected beams, but on the other hand play an important role in triggering the formation of transport barriers in the core and edge regions.

Standard diagnostics for fundamental plasma parameters and for plasma physics are routinely utilized for daily operation and physics studies in the large helical device (LHD) with high reliability. Diagnostics for steady-state plasma are under intensive development, especially for Te, ne (yttrium–aluminium garnet (YAG) laser Thomson, CO2 laser polarimeter), data acquisition in steady-state and heat-resistant probes. To clarify the plasma properties of the helical structure, two- or three-dimensional diagnostics are being aggressively developed: tangential cameras (fast SX TV, photon counting CCD, Hα TV); tomography (tangential SX CCD, bolometer); imaging (bolometer, ECE, reflectometer). Divertor and edge physics are important key issues for steady-state operation. Diagnostics for neutral flux (Hα array, Zeeman spectroscopy) and ne (fast scanning probe, Li beam probe, pulsed radar reflectometer) are also in advanced stages of development. In addition to these, advanced diagnostics are being intensively developed in LHD through domestic and international collaborations.

Significant progress has been made with ion cyclotron range of frequencies (ICRF) heating in the Large Helical Device. This is mainly due to better confinement of the helically trapped particles and less accumulation of impurities in the region of the plasma core. During the past two years, ICRF heating power has been increased from 1.35 to 2.7 MW. Various wave-mode tests were carried out using minority-ion heating, second-harmonic heating, slow-wave heating and high-density fast-wave heating at the fundamental cyclotron frequency. This fundamental heating mode extended the plasma density range of effective ICRF heating to a value of 1×1020 m−3. This use of the heating mode was its first successful application in large fusion devices. Using the minority-ion mode gave the best performance, and the stored energy reached 240 kJ using ICRF alone. This was obtained for the inward-shifted magnetic axis configuration. The improvement associated with the axis-shift was common for both bulk plasma and highly accelerated particles. For the minority-ion mode, high-energy ions up to 500 keV were observed by concentrating the heating power near the plasma axis. The confinement properties of high-energy particles were studied for different magnetic axis configurations, using the power-modulation technique. It confirmed that with the inward-shifted configuration the confinement of high-energy particles was better than with the normal configuration. By increasing the distance of the plasma to the vessel wall to about 2 cm, the impurity influx was sufficiently reduced to allow sustainment of the plasma with ICRF heating alone for more than 2 min.

Internal transport barriers with respect to electron thermal transport (eITB) were observed in the large helical device, when the electron cyclotron resonance heating (ECH) power was highly localized on the centre of a plasma sustained by neutral beam injection. The eITB is characterized by a high central electron temperature of 6–8 keV with an extremely steep gradient, as high as 55 keV m−1 and a low electron thermal diffusivity within a normalized average radius ρ≈0.3 as well as by the existence of clear thresholds for the ECH power and plasma collisionality.

The behaviour of high-energy ions accelerated by an ion cyclotron range of frequency (ICRF) electric field in the large helical device (LHD) is discussed. A better confinement performance of high-energy ions in the inward-shifted magnetic axis configuration was experimentally verified by measuring their energy spectrum and comparing it with the effective temperature determined by an electron slowing down process. In the standard magnetic axis configuration a saturation of the measured tail temperature was observed as the effective temperature was increased. The ratio between these two quantities is a measure of the quality of transfer efficiency from high-energy ions to a bulk plasma; when this efficiency was compared with Monte Carlo simulations the results agreed fairly well. The ratio of the stored energy of the high-energy ions to that of the bulk plasma was measured using an ICRF heating power modulation method; it was deduced from phase differences between total and bulk plasma stored energies and the modulated ICRF heating power. The measured high energy fraction agreed with that calculated using the injected ICRF heating power, the transfer efficiency determined in the experiment and the confinement scaling of the LHD plasma.

Ion temperature at the plasma centre has been measured from Doppler broadening of Ti XXI (2.61 A) and Ar XVII (3.95 A) x-ray lines using a newly installed crystal spectrometer with CCD detector in ECH, NBI and ICRF plasmas of Large Helical Device (LHD). The ion temperature obtained in a range of 0.6 and 3.5 keV was analysed with electron density and compared with electron temperature. A new parameter range of Ti>Te was found in low-density (ne

Intrinsic impurity behaviour in neutral beam heated LHD long pulse discharges with a discharge duration of 10 s was investigated. Spectroscopic and bolometric measurements showed a remarkable temporal increase of core radiation due to metallic impurities. Central impurity accumulation was found in a narrow plasma density region (e = 1–3×1019 m−3) for both density ramp-up discharges and constant density discharges. In the density ramp-up discharges, high-Z impurities were accumulated in the low-density region and diffused out from the core plasma in the high-density region. The impurity behaviour was compared with neoclassical predictions.

The improvement in the particle confinement of pellet shots on the Large Helical Device (LHD) was found after methane (CH4) mixed hydrogen gas-puff discharges. Only four discharges introducing ∼20 Pa m 3 of CH4 caused the reduction in the radiation loss and the level of metal impurities, together with the enhanced recycling, which is expected as the real time carbonization effect. The decay rate of the electron density was mitigated in the pellet shot after CH4 discharges. Transport analysis shows 60% reduction in the particle transport coefficient at half the averaged minor radius.

The experimentally optimized magnetic field configuration of the Large Helical Device [A. Iiyoshi et al., Nucl. Fusion 39, 1245 (1999)], where the magnetic axis is shifted inward by 15 cm from the early theoretical prediction, reveals 50% better global energy confinement than the prediction of the scaling law. This configuration has been investigated further from the viewpoints of high energy particle confinement and magnetohydrodynamic (MHD) stability. The confinement of high energy ions is improved as expected. The minority heating of ion cyclotron range of frequency was successful and the heating efficiency was improved by the inward shift. The confinement of passing particles by neutral beam injection was also improved under low magnetic field strength, and there could be obtained an almost steady high beta discharge up to 3% in volume average. This was a surprising result because the observed pressure gradient exceeded the Mercier unstable limit. The observed MHD activities became as high as beta but they did not grow enough to deteriorate the confinement of high energy ions or the performance of the bulk plasma, which was still 50% better than the scaling. According to these favorable results, better performance would be expected by increasing the heating power because the neoclassical transport can also be improved there.

By using an appropriately oriented seed from the TiAl–Si system, the TiAl/Ti 3 Al lamellar structure was aligned parallel to the growth direction for ingots having composition of Ti–46Al–0.5Si–0.5X (X=Re, W and Mo) and Ti–47.5Al–0.5Re (at.%). The seeded and directionally solidified quaternary alloys containing either Re, W or Mo exhibit a nice balance of low- and high-temperature mechanical properties. Tensile elongation more than 20% is noted for the W- and Mo-containing alloys. The creep resistance for the three quaternary alloys is more than an order of magnitude better than other relevant TiAl alloys produced by conventional ingot metallurgy methods, with the best creep properties being obtained for the Re-containing alloy. The method to predict alloy compositions appropriate for aligning the lamellar structure of two-phase TiAl alloys of multi-component by directional solidification is proposed based on the assignment of ‘Al-equivalent’ for each of the alloying elements.

During the first two years of the LHD experiment the following results have been achieved: (i) higher Te (Te(0) = 4.4 keV at ne = 5.3 × 1018 m-3 and Pabs = 1.8 MW); (ii) higher confinement (τE = 0.3 s, Te(0) = 1.1 keV at ne = 6.5 × 1019 m-3 and Pabs = 2.0 MW); (iii) higher stored energy, Wpdia = 880 kJ at B = 2.75 T. High performance plasmas have been realized in the inward shifted magnetic axis configuration (R = 3.6 m) where helical symmetry is recovered and the particle orbit properties are improved by a trade-off of MHD stability properties due to the appearance of a magnetic hill. Energy confinement was systematically higher than that predicted by the International Stellarator Scaling 95 by up to a factor of 1.6 and was comparable with the ELMy H mode confinement capability in tokamaks. This confinement improvement is attributed to configuration control (inward shift of the magnetic axis) and to the formation of a high edge temperature. The average beta value achieved reached 2.4% at B = 1.3 T, the highest beta value ever obtained in a helical device, and so far no degradation of confinement by MHD phenomena has been observed. The inward shifted configuration has also led to successful ICRF minority ion heating. ICRF powers up to 1.3 MW were reliably injected into the plasma without significant impurity contamination, and a plasma with a stored energy of 200 kJ was sustained for 5 s by ICRF alone. As another important result, long pulse discharges of more than 1 min were successfully achieved separately with an NBI heating of 0.5 MW and with an ICRF heating of 0.85 MW.

The mechanical properties of TiAl-based alloys with lamellar microstructure are extremely anisotropic. However, if the lamellar microstructure can be aligned parallel to the growth direction, the resulting material should possess a good combination of mechanical properties. Unfortunately, simple casting operations often lead to a solidification texture with the lamellar boundaries perpendicular to the heat flow direction. This difficulty can be overcome by directionally solidifying TiAl-based alloys. We have been performing directional solidification experiments with and without using a seeding technique. The current status of directional solidification of TiAl-based alloys is reviewed.

The magnetic hyperfine splitting frequencies of (i) $^{57}\mathrm{Ni}$ in Fe ${(}^{57}$NiFe), (ii) $^{57}\mathrm{Ni}$ in Ni ${(}^{57}$NiNi), and (iii) $^{59}\mathrm{Fe}$ in Fe ${(}^{59}$FeFe) were measured with the technique of nuclear magnetic resonance on oriented nuclei (NMR-ON) at low temperatures. The NMR-ON spectra were observed by detecting the \ensuremath{\beta}-ray asymmetries. The resonance frequencies at zero external magnetic field were deduced from the extrapolation of the external field dependence of the resonance frequencies: \ensuremath{\nu}${(}^{57}$NiFe)=94.23(2) MHz, \ensuremath{\nu}${(}^{57}$NiNi)=30.02(4) MHz, and \ensuremath{\nu}${(}^{59}$FeFe)=57.719(11) MHz. With the known hyperfine fields of ${\mathit{B}}_{\mathrm{HF}}$(NiNi)=-7.406(8) T and ${\mathit{B}}_{\mathrm{HF}}$(FeFe)=-33.82(3) T, the magnetic moments of $^{57}\mathrm{Ni}$ and $^{59}\mathrm{Fe}$ were determined to be \ensuremath{\mu}${(}^{57}$Ni)=-0.7975(14)${\mathrm{\ensuremath{\mu}}}_{\mathit{N}}$ and \ensuremath{\mu}${(}^{59}$Fe)=-0.3358(4)${\mathrm{\ensuremath{\mu}}}_{\mathit{N}}$. The precise value of the hyperfine field of ${\mathit{B}}_{\mathrm{HF}}$${(}^{57}$NiFe) was also deduced as -23.25(5) T. The measured values of the magnetic moment are compared with those based on the shell model. \textcopyright{} 1996 The American Physical Society.

Nuclear magnetic resonance on oriented206BiFe and182ReFe has been measured using recoil implanted samples. Clearly resolved satellite structures were found for both systems. By annealing at an appropriate temperature, the resonance width was reduced. The satellite line structure for182ReFe was studied with various annealing temperatures. The ratio of the resonance strength changed with the annealing temperature. The resonances for182ReFe were also measured in external magnetic fields of 0.1, 0.2 and 0.4 T. The resonance frequencies for the satellite structure at an external magnetic field of 0.2 T were determined:v1=691.7(3),v2=684.4(2),v3=687.2(2) MHz for206BiFe;v1=231.15(5),v2=230.20(5) for182ReFe. The origin of the satellite structure is discussed. The effective relaxation time of182ReFe at 8 mK and external magnetic field of 0.2 T was determined to be 18.6(6) s using a single-exponential fit.

The negative differential resistance properties of resonant tunnelling barrier diodes with InGaAs quantum wells developed by us for new transistor applications are reviewed. The emphasis is on the peak current density and the peak-to-valley ratio. The tunnel barriers of InAlAs, pseudomorphic AlAs, AlAsSb and GaAsSb are studied. A simple theoretical calculation which includes the mid-gap band non-parabolicity of tunnelling electrons is described. These theoretical calculations show excellent agreement with the observed peak current density at 77 K and 300 K and a reasonable agreement with the valley current density at 300 K.

The magnetic hyperfine splitting frequencies of90NbNi and93mMoNi in an external magnetic field of 0.2 T have been determined by the NMR-ON method to be 18.52(7) and23.73(10) MHz, respectively. With the assumption of Knight shift factorK=0 and with the knowng-factors, the hyperfine fields of90NbNi and93mMoNi were deduced asBHF(90NbNi)=-4.118(16) T andBHF(93mMoNi)=-3.491(33) T. The rather long spin-lattice relaxation time of 32(5) min was observed for90NbNi at an external magnetic field of 0.2T and8 mK.

Nuclear magnetic resonance on oriented76,77,82BrFe has been measured using recoil-implanted samples. The magnetic hyperfine splitting frequency of82BrFe in a zero external magnetic field has been determined to be 201.90(3) MHz. The resonances of76BrFe and77BrFe were also observed in an external magnetic field of 0.2 T asv(76BrFe)=340.9(3) MHz andv(77BrFe)=403.5(2) MHz. With the known values of theg-factors, the hyperfine fields have been deduced:B HF(82BrFe)=81.397(27) T,B HF(76BrFe)=81.38(7) T. Theg-factor of77Br was determined to be |0.6487(4)|.

A magnetic property of Pm as an impurity in PrNi single crystal was investigated by means of the low-temperature nuclear orientation of143,144Pm. The angular distribution of γ-ray anisotropies revealed that the direction of hyperfine field experienced by the nuclei lie in the (a, c) plane andmade an angle of 60° (5) or 240° (5) with respect to the crystallinec-axis. From the temperature dependence of the anisotropies the strength of the hyperfine field of Pm in PrNi was deduced to be 185(22)T. The hyperfine field of Co at the Ni site of PrNi was found to be less than 4T.

The magnetic hyperfine splitting frequencies of123INi,124INi and131INi in a zero external magnetic field have been determined by the NMR-ON method as 258.9(1), 165.9(1) and 179.5(2) MHz, respectively. With the known values of the magnetic moments, the magnetic hyperfine fields have been deduced:BHF(123INi)=30.17(5) T,BHF(124INi)=30.14(9) T,BHF(131INi)=30.06(4) T; the weighted average isBHF(INi)=30.11(4) T. The small difference of theBHF(131INi) with those of123INi and124INi is discussed comparing with results of the hyperfine splitting frequency of iodine in iron host.

NMR-ON measurements have been performed for60CoFe−Si. The samples of Fe−Si (6 at. %) single crystal dish with the (110) surface were used. Rosonance measurements have been carried out with the magnetization direction along the and the axes. Five prominent resonances were found at 166, 162, 151, 147 and 135 Mllz. The resonance at 166 Mllz has been known to be due to60Co without neighbourt Si atoms. The resonances at 151 and 135 Mllz are due to60Co nuclei with one and two Si atom(s) in the 1st neighbour site, respectively. The resonances at 162 and 147 Mllz would be due to the contribution of the 3rd and the magnetization axes, the differences of the resonance width have been qualitatively explained using the dipolar field.

$\mathrm{In}\mathrm{As}∕\mathrm{Ga}\mathrm{As}$ quantum dots (QD's) are formed by postgrowth annealing of an InAs wetting layer thinner than the critical thickness for the transition from two- (2D) to three-dimensional (3D) growth mode. Reflection high energy electron diffraction is used to monitor the QD formation. Based on a mean-field theory [Phys. Rev. Lett. 79, 897 (1997)], the time evolution of total QD's volume, first increasing and finally saturating, is well explained by precursors forming during wetting layer growth and converting into nucleated QD's after growth stop. Both the saturation QD's volume and the QD nucleation rate depend exponentially on the InAs coverage. These behaviors and their temperature and InAs growth rate dependences are essentially understandable in the frame of the mean-field theory. Similar analysis to conventional QD growth suggests that the often observed significant mass transport from wetting layer to QD's can be ascribed to the precursors existing before 2D-3D growth mode transition.

Ordered and modulated structures in GaAsSb grown on [110]InP substrates by molecular beam epitaxy, has been evaluated by transmission electron microscopy. In the electron diffraction pattern from the GaAsSb, superstructure spots associated with CuAu-I type ordered structure are found. When the offset angle of the substrates increases, the ordering becomes stronger. The ordering becomes also stronger with increasing growth temperature in the range 470-530/spl deg/C. In high resolution images of the crystal, areas with doubling in periodicity of 220 and 200 lattice fringes are observed locally, which is due to CuAu-I type ordering, indicating that the ordering is not perfect and that ordered regions are likely to be microdomains. Modulated structure are also found in both the and directions. Coarse and fine modulated structures are clearly observed, and the periodicity of the coarse structure in the direction is much longer than that in the direction. These findings lead us to conclude that atomic ordering and spinodal decomposition of the crystal are competing on the growth surface via surface diffusion of deposited atoms.

In LHD discharges a significant enhancement of the global energy confinement has been achieved for the first time in a helical device with an edge thermal barrier, which exhibits a sharp gradient at the edge of the temperature profile. Key features associated with the barrier are quite different from those seen in tokamaks: (i) almost no change in particle (including impurity) transport, (ii) a gradual formation of the barrier, (iii) a very high ratio of the edge temperature to the average temperature, and (iv) no edge relaxation phenomenon. These features are very attractive in applying the thermal barrier to future reactor grade devices.

High degree of excitonic spin-state preservation during energy relaxation is demonstrated in a single quantum dot. Optical-phonon resonances and corresponding suppression of spin relaxation are clearly identified as dip structures in photoluminescence excitation spectra probed by the positive trion emission. Consequently, under the longitudinal optical-phonon resonant excitation, a distinguishably high degree of circular polarization up to $\ensuremath{\sim}0.85$ is achieved. Extended spin-relaxation time in the positive trion ground state compared to that in neutral exciton ground state is also revealed.

The oscillations of excitonic and nuclear spin polarizations in an optically pumped single self-assembled In0.8Al0.2As/Al0.35Ga0.65As quantum dot (QD) were clearly observed under the excitation of a wetting layer edge at B=5 T. This indicates that an exciton pair with opposite spins is alternatively created via the half-localized state only by changing the delay time between cross-linearly polarized pulse pair. Furthermore, periodic modulation of Zeeman energy synchronizes the degree of circular polarization of photoluminescence from a single QD, indicating that the Overhauser field follows the optically created electron spin polarization in half-localized states of a QD, and the half-localized state in a QD consists of a confined electron in a discrete state and hole in the continuum state.

We investigated an optically created nuclear field in a single InAlAs quantum dot and demonstrated that the nuclear field can be used to discriminate whether photoluminescence lines originate from the same dot or a different dot. Since the nonlinear response of the nuclear field is sensitive to the electron g factor and correlation time of a coupled electron-nuclear spin system, the resultant Overhauser shift is their good measure for individual quantum dots. This method provides a simple and convenient alternative to the standard photon cross-correlation method.

We have observed a chopping frequency dependence of the photoreflectance (PR) spectra of the MQW structure : GaAs/AlAs multiple-quantum-well (MQW)/GaAs buffer/semi-insulating GaAs substrate. The PR spectra obtained from the GaAs buffer layer and the MQW reveal different frequency dependences. It is found that the buffer/MQW interface states or deep levels in the buffer control the frequency dependence of the modulating field that exist in the buffer. It is finally shown that the PR technique is a powerful tool for characterizing deep levels in the MQW or SL structures grown on semi-insulating materials as compared to the deep level transient spectroscopy (DLTS) method which cannot be applied in these cases.

Photoemission partial yield speetroscopy (PYS) has been used to study As-stabilized GaAs(100) (2 × 4)-c (2 × 8) and Al x Ga 1− x As(100) surfaces. Doublet features corresponding to transitions from Ga 3d core levels to empty surface states are observed. This serves as an evidence for the existence of Ga dangling bonds which is consistent with the missing-dimer-row model for the (2 × 4)-c(2 × 8) reconstruction. No such PYS structures out of As 3d appear in the spectra. The doublet features are smeared after the ion-bomburdment-and-unnealing treatment, indicating that, in such cases, the surfaces are most probably not well ordered although they are Ga-rich.

Raman scattering spectra were measured in superionic glasses AgIAg 2 OMoO 3 [(AgI) x (Ag 2 MoO 4 ) 1− x ] ( x =0.60, 0.72) in the temperature range from 29 to 302 K. In these spectra, three bands were found at 318, 800 and 875 cm −1 , which are characteristic of MoO 4 2− ions. Below 250 cm −1 , two bands were found at 40 and 115 cm −1 . Comparing the spectra measured at 29 and 73 K, these bands are found to be caused by the first order Raman scattering. Their integrated intensity increases in rough proportion to the ratio of AgI to Ag 2 MoO 4 in the glass. Further, the shape of the Raman spectrum below 180 cm −1 is very similar to that of α-AgI crystal. These facts suggest the existence of small clusters of AgI in the glass.

We used reflection electron microscopy (REM) to observe GaAs (001) surfaces grown at 700 °C by molecular beam epitaxy. The REM images revealed such interesting features as high step density and large undulations that occurred even though we interrupted epitaxial growth for 3 min, in an attempt to produce a flat surface. A cross section of the large undulation showed a structure consisting almost entirely of plateaus, 0.1–2 μm wide and several monolayers high but less than 30 A. Small steps distributed uniformly across the tops of these plateaus appeared comparatively straight in the [110] direction but appeared to ‘‘zigzag’’ in the [110] direction—a phenomenon we think may have been caused by Ga surface migration anisotropy.

Laser dye (Rhodamine 6G, Rhodamine B or Oxazine 4) was doped in a matrix of alumina gel glass by the sol-gel technique using aqueous sol solution. The undesirable dye aggregation usually observed in aqueous solution is largely reduced in the film, even at a concentration as high as 10 −2 M. When this film was irradiated with a N 2 laser, the film gave laser emission with a beam divergence of about 0.1 rad. The width of the dye laser emission is 10 nm which is typical of a dye laser. The calculated conversion efficiency is 2.1%. A decrease of power is linearly proportional to the shot number, which indicates a one-photon process.