Search

The growth of thick 2H-SiC layers in Si-Li solution under a continuous CH4flow at atmospheric pressure was investigated. The thickness of the 2H-SiC layer increased linearly with the growth period, reaching as thick as 270-μm with a growth period of 7 hr. The Li concentration in the layers was as low as 1.0 × 1018atoms/cm3, which was less than one-hundredth that observed in a previous study.

Seeded growth of gallium nitride (GaN) crystals on a spontaneously nucleated small GaN by the Na flux method was performed. In this study, we attempted to control the growth habit by changing the flux composition (Ga/Na) and by introducing a small amount of additives (Ca and Li). Our experiment clarified that a low Ga composition was preferred to grow high-crystallinity prismatic GaN crystals with a high growth rate. Furthermore, the transparent GaN single crystals with prism shape could be grown by the addition of Ca and Li.

We measured drain bias stress effects and current collapse in AlGaN/ GaN heterostructure field-effect transistors (HFETs) on a-plane and c-plane GaN substrates. An a-plane AlGaN/GaN HFET (a-HFET) shows small current collapse with a threshold voltage (Vth = -1.8 V). On the other hand, a c-plane HFET (c-HFET) with the same barrier thickness (20 nm) shows a small current collapse, although Vth was negatively large (Vth = -4 V). Current collapse in a-HFET was not large compared with that in conventional c-HFET on GaN. A c-HFET on sapphire was also measured. The current collapses in HFETs on sapphire were larger than that in HFETs on a GaN substrate. The current collapses in the thin-barrier c-HFET (Vth = -1.8 V) were particularly large. Therefore, an a-plane device is promising for a small or positive Vth with small current collapse. (© 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Non-polar GaN substrates have attracted attention because they can improve the performance of GaN devices. Growth of high-quality non-polar GaN single crystals, however, has not been realized. Here, we attempted to grow m-plane GaN crystals using the Na flux method and investigated the effect of additives such as Ca and Li in Na flux on transparency, surface roughness, and crystallinity. We found that the addition of Ca and Li improved the transparency and the surface roughness of the crystals. Moreover, high-temperature growth improved crystallinity and decreased impurities such as Ca, Li, and O in GaN crystals. We also found that the crystallinity did not depend on the concentrations of impurities in GaN crystals. (© 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

In this study, we investigated the relationship between surface morphology and oxygen concentration in GaN layers synthesized from Ga2O vapor and NH3. The surface morphology of GaN layers changed from rough (with large numbers of pits) to (0 0 0 1) smooth planes with the increase in NH3 concentration. SIMS mapping analysis revealed that the oxygen concentration was on the order of 1017 atoms/cm3, the lowest level obtained in GaN layers synthesized from Ga2O vapor and NH3, at the smooth (0 0 0 1) surfaces. By contrast, high oxygen concentration over 1020 atoms/cm3 was detected at the pitted areas. We concluded that GaN films with low oxygen contamination can be obtained by suppressing pit formation and promoting the smoothness of the (0 0 0 1) surface, even when using Ga2O as the Ga source.

We investigated changes in the growth mode and dislocation propagation behavior of Na-flux-grown GaN on GaN (0001) templates with changing flux composition (Ga/Na). Results suggested that deliberate control of the flux composition enables fast growth of thick GaN with a low dislocation density. The growth mode was evaluated by scanning electron microscopy (SEM). The distribution and density of dislocations were investigated using chemical etching. SEM observation revealed that the growth mode changed from a two-dimensional (2D) mode developing (0001) facets to a three-dimensional (3D) mode developing (0001) and {1011} facets as the Ga composition in the flux was increased. A reduction of dislocation density was promoted in the 3D growth mode favored at high Ga compositions, and the minimum dislocation density observed was on the order of 103 cm–2 (280 μm × 200 μm) after growth for 96 h at a Ga composition of 40 mol %. On the other hand, the growth rate was higher in the 2D growth mode favored at low Ga c...

High-quality and low cost bulk crystals are needed in the field of group III nitride semiconductors in order to develop optical and electrical devices. There are two approaches for the growth of bulk GaN crystal by the Na flux method. One is to grow thick GaN crystal on a seed GaN crystal grown by vapour phase method. The other one is to grow GaN crystal on the seed GaN crystal obtained by spontaneous nucleation. The both methods show advantages and disadvantages, respectively. We also present the effects of the flux composition on the growth habit and the growth rate for fabrication of boule-shape bulk GaN crystals. (© 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We fabricated and characterized AlGaN/GaN heterostructure growth by MOVPE on vicinal m-plane free-standing GaN substrates prepared by the Na flux method. The miscut angle in the LPE-GaN substrate has a great influence on the surface morphology and crystalline quality of epitaxial GaN and AlGaN films. In particular, AlGaN/GaN on a 4° miscut-angle LPE-GaN substrate shows large step bunching, small-miscut-angle LPE GaN substrate is essential for fabricating on FET structure with a flat surface.

CsB3O5 (CBO) is an excellent nonlinear optical crystal and can be grown by top-seeded solution growth (TSSG) from self-flux solutions. However, there are scattering centers in the as-grown crystal, which decrease the bulk laser-induced damage threshold (LIDT) of the crystal. Postgrowth quenching and vapor transport equilibration (VTE) can reduce the scattering centers effectively. Based on our previous hypothesis and analysis of retrograde solid solution at high temperature in CBO crystal, we adopt selective quenching to 500 °C—fast cooling. Our investigations showed that, after the cooling process, the scattering centers and optical loss of CBO were decreased. It was found that with the increase in cooling rate, the LIDT of CBO crystals was increased. The highest LIDT of fast cooling CBO was 1.5 times that of conventional cooling process; so the crystal quality was improved.

We studied the procedure of cleaning GaN(0001) substrate surfaces by wet etching and subsequent annealing in ultrahigh vacuum for two different types of freestanding GaN wafers: hydride vapor phase epitaxy (HVPE) crystal and Na flux liquid phase epitaxy (LPE) crystal wafers. A flat surface containing GaN(0001)2 × 2 reconstruction was successfully achieved on both HVPE and LPE surfaces by etching in HF and subsequent annealing at ∼ 550 °C but was not achieved by etching in HCl, NaOH, and HNO 3 .

GaN-based optoelectronic and electronic devices such as light-emitting diodes (LEDs), laser, and heterojunction field-effect transistors (HFETs) typically use material grown on foreign substrates such as sapphire, Si, and SiC. However, thermal and lattice mismatch present prevent attainment of quality films deemed necessary by ever increasing demand on device performance. In fact in LEDs intended for solid state lighting, internal quantum efficiencies near 100% might be needed, and further these high efficiencies would have to be retained at very high injection current levels. On the electronic device side, high radio-frequency (RF) power, particularly high-power switching devices, push the material to its limits. Consequently, as has been the case for other successful semiconductor materials systems, native substrates must be developed for the GaN family. In this paper, various approaches such as high-pressure nitrogen solution (HPNS), ammonothermal, and Na flux methods, and an intermediary technique called the hydride vapor phase epitaxy (HVPE; to a lesser extent as there is a review devoted to this technique in this issue) along with their strengths and challenges are discussed.

High-quality and low-cost bulk crystals are needed in the field of group III nitride semiconductors in order to develop optical and electrical devices. Controlling the nucleation phenomenon and solution condition during growth is important for growth of bulk GaN crystal by the Na flux method. The carbon doping into the solution and the forced induction of solution flow are effective to control them. We also present some attempts for improving the quality of seed substrate for fabrication of boule-sized bulk GaN crystals.

We investigated CsLiB 6 O 10 (CLBO) crystal growth in a dry atmosphere and from a stoichiometric melt composition with the aim of obtaining CLBO crystals containing low levels of water as an impurity. We qualitatively compared the amounts of water incorporated into CLBO crystals using a FT-IR spectrometer. It was found that CLBO crystals grown under the above conditions included a lower level of water than those grown in an ambient atmosphere from B 2 O 3 ―poor solution.

In this study, vapor-phase epitaxy (VPE) of GaN oriented-film was performed using Ga 2 O vapor as the Ga source. Ga 2 O vapor was obtained by reducing Ga 2 O 3 powder with H 2 gas at 1000 °C. The Ga 2 O vapor was then reacted with NH 3 on a seed substrate at 1100–1150 °C. A high quality GaN substrate (1 mm thick, with full widths at half maximum of GaN (0 0 0 2) X-ray rocking curve of 107–110 arcsec) prepared by the Na-flux method were used as the seed substrate. After 30 min of growth, a 3-μm flat GaN (0 0 0 1) epitaxial layer was grown on the seed substrate. X-ray diffraction (XRD) measurements showed that the FWHM of the GaN epitaxial layer was 74–111 arcsec, showing high crystallinity. Secondary ion mass spectrometry (SIMS) analysis showed that the oxygen concentration in the epitaxial layer was 1.5×10 18 atoms/cm 3 . Although an oxide was used as the raw material, oxygen concentration close to those in GaN crystal grown by the hydride vapor phase epitaxy (HVPE) and metalorganic vapor phase epitaxy (MOVPE) were achieved. We concluded that the VPE method using Ga 2 O vapor has potential as a simple vapor-phase-growing technique for high-quality GaN films.

We have determined all the three independent components of quadratic nonlinear-optical coefficients of GaN by highly accurate Maker-fringe measurements on high-quality bulk samples combined with theoretical analysis taking account of the multiple-reflection effects in slightly misoriented optically anisotropic samples. Especially, the d33 coefficient is determined with sufficient accuracies for the first time, to our knowledge, by using (112¯0)-oriented samples. The obtained values of quadratic nonlinear-optical coefficients are d31=2.5±0.1 pm/V, d15=2.5±0.1 pm/V, and d33=−3.8±0.2 pm/V at the fundamental wavelength of 1.064 μm.

Rubrene (5,6,11,12-tetraphenyltetracene) single crystals were grown by using the solution-slow-cooling technique with 1-propanol (PrOH) solvent. Thin hexagonal crystals with an orthorhombic system, typically 300×200×50 μm 3 in size, were obtained. The crystals had flat terraces (∼7 μm) separated by monomolecular steps. Organic field-effect transistors (OFETs) with the single crystals showed carrier mobilities as high as ∼1.6 cm 2 V −1 s −1 . We also obtained thin parallelogram-shaped crystals, typically 300×200×20 μm 3 in size. X-ray diffraction analysis solved that the crystals belong to triclinic system. The crystals showed lower carrier mobilities of ∼0.1 cm 2 V −1 s −1 . This is the first report on the growth and characterization of two polymorphs of rubrene crystal.

It was revealed that the metastable region, in which liquid phase epitaxy (LPE) of GaN single crystals proceeds without the generation of polycrystals, expands with growth temperature in the Na flux method. The metastable region appears when LPE growth proceeds at temperatures above 1073 K, although generation of polycrystals inevitably occurs on a crucible at temperatures less than 1073 K. The highest growth rate of 14 μm/h in a small experimental setup was achieved at a temperature of 1163 K with a nitrogen pressure of 5.5 MPa due to complete suppression of the growth of polycrystals on a crucible, even though the supersaturation at this condition reached a fairly high level. Also, an LPE crystal with a flat surface could easily be obtained under high-temperature conditions.

We report 63Cu-NMR/NQR on the bi-layered high- T c copper oxide Ba 2 CaCu 2 O 4 F 2 , which is expected to be a Mott insulator in an ideal case of a nominal content of fluorine, and less fluorinated Ba 2 CaCu 2 O 4 ( F 1.6 O 0.4 ) to clear the doping mechanism in apical-fluorine multilayered system Ba 2 Ca n - 1 Cu n O 2 n F 2 , where n is the number of the layers in a unit cell. In the case of n = 3 and, especially, n = 4 , the origin of carrier doping has been attributed to a self-doping mechanism based on the nominal stoichiometry; electrons are transferred between the inner CuO 2 plane (IP) and the outer one (OP), keeping Cu + 2 on average. It is revealed that Ba 2 CaCu 2 O 4 F 2 is an underdoped superconductor with hole carriers, doped by an inevitable deviation from the nominal content of apical fluorines. This result suggests that hole carriers are doped in the same manner for the case of n = 3 and 4, and that we should reconsider the self-doping mechanism as the carrier source in the apical-fluorine system.

We have discovered a mechanism which can significantly reduce the dislocation density during the growth of GaN single crystals in the Na flux method. The significant reduction of the dislocation density occurs in the later stage of LPE growth, rather than solely at the seed-LPE interface for which we have already reported evidence indicating the presence of bundling dislocations. The two-step dislocation reduction is the key in achieving extremely low dislocation density using this method. We grew a two-inch GaN crystal with a thickness of 2 mm in order to confirm that the dislocation density decreases as the growth thickness increases. As a result, three-fourths of the surface area exhibited dislocation density of the order 10 2 cm −2 , and the FWHM of the X-ray rocking curve (XRC) measurement in (0 0 0 2) face was 28 arcsec. Here, we report the two-step dislocation reduction mechanism based on LPE growth in the Na flux method.

Bis[4-dimethylamino-N-methyl-4-stilbazolium] terephthanate (BDAS-TP) was newly designed as an organic nonlinear optical crystal. It is derivative of 4-dimethylamino-N-methyl-4-stilbazolium tosylate (DAST), which is well known as a novel terahertz (THz)-wave emitter. We also succeeded in growing single crystals of BDAS-TP for the first time by using a solution-slow-cooling technique. By using a femtosecond laser, we obtained a broadband THz spectrum from BDAS-TP crystal.

To fabricate organic field-effect transistors (OFETs) with high carrier mobility, we attempted to grow 5,6,11,12-tetraphenylnaphthacene (rubrene) single crystals from solution and to improve their quality. Investigations into solvents in which rubrene was highly soluble proved that its solubility depended on the presence or absence of aromatic rings and chloro groups rather than on the polarity of the solvents. Rubrene crystals were grown from aromatic solvents, specifically from toluene, p-xylene, and aniline solvents, as well as from 1,2-dichloroethane (DCE) solvent. As a result, rubrene single crystals larger than 1 mm were obtained. Powder X-ray diffraction (XRD) analysis showed that the crystals obtained from the p-xylene and toluene solvents were rubrene, and 1H-nuclear magnetic resonance (1H-NMR) measurement proved that the crystals had not incorporated the solvent at the detection level. In addition, atomic force microscopy (AFM) confirmed that the rubrene crystals grown from the p-xylene and aniline solvents had flat faces and that the crystal from the p-xylene solvent had monomolecular steps on parts of the surfaces. Rubrene single crystal OFETs with graphite electrodes and parylene as an insulator showed carrier mobilities of ~0.75 cm2 V-1 s-1.

We characterized the surface cleanliness, structure, and composition of GaN(0001) grown by liquid phase epitaxy (LPE) using coaxial impact-collision ion scattering spectroscopy (CAICISS). Contamination on the as-received LPE-GaN(0001) was removed by annealing at 900 °C, but annealing was unable to remove the O atoms from the as-grown surface, indicating that the as-grown sample incorporated O atoms. Although the as-grown sample contained O atoms, its structural perfection was maintained. These facts were explained by localization of O atoms near dislocations. The etched surface exhibited high crystalline quality, which was comparable to that of the sample grown by vapor phase epitaxy (VPE).

We have achieved the first successful growth of 2H-SiC single crystals using the C-Li-Si melt system. Li-Si melt, whose melting point is lower than 1000 oC, was chosen because the 2H-SiC polytype is more stable at lower temperatures than other polytypes such as 3C-, 4H-, and 6H-SiC. Many hexagonal-shaped crystals of approximately 100 m in diameter were observed via a scanning electron microscope (SEM). A high resolution transmission electron microscope (HR-TEM) lattice image of the grown crystals showed a periodical structure with A-B stacking along the direction. These results indicated that the Li-based flux was useful for growing bulk 2H-SiC single crystals.

We succeeded in growing a GaN single crystal substrate with diameter of about two inches using the Na flux method. Our success is due to the development of a new apparatus for growing large GaN single crystals. The crystal grown in this study has a low dislocation density of 2.3×105 (cm-2), The secondary ion mass spectrometry (SIMS) technique demonstrates that the Na element is difficult to be taken in the crystal in both the + and – c directions, resulting in a Na concentration of 4.2 × 1014 (cm-3) in the crystal. Our success in growing a two-inch GaN substrate with a low impurity content and low dislocation density should pave the way for the Na flux method to become a practical application.

We found that a carbon additive could suppress the unfavorable generation of polycrystals in a crucible without reduction in the yield of GaN in the Na flux method. The suppression of polycrystals due to the effect of carbon significantly increased the growth rate of liquid phase epitaxy (LPE), which has been the biggest problem of the Na flux LPE, and enabled an increase in the growth rate above 20 μm/h. A 3 mm-thick 2 in GaN crystal was obtained for the first time. In addition, the carbon additive was found to have another effect in that the nonpolar face could be widely developed. SIMS measurements revealed that carbon added into a Ga–Na mixed melt was hardly taken into LPE crystals, although carbon did have some favorable effects.

A dependence of Gallium hydride (GaHx: x = 1, 2, 3) production on the growth rate of GaN crystals grown by the gallium hydride vapor phase epitaxy (GaH-VPE) method was investigated. By this method, GaN crystals are synthesized in a reaction between GaHx and NH3. The growth rate of GaN crystals has been as low as 20 μm/h at a maximum because the poor reactivity of H2 gas with Ga metal limits the production of GaHx. In this study, the interfacial area between Ga metal and H2 gas was enlarged to increase the GaHx synthesis, which thereby enabled an increase in the growth rate to 68 μm/h. We showed clearly that the growth rate of GaN by the GaH-VPE method is limited by the efficiency of the reaction between the Ga metal and H2 gas. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Controlling the nucleation and growth processes is important for growing large, high-quality crystals of oxides, proteins, and nitrides from a solution. We must understand how the solution conditions should be controlled, and determine a universal principle that is applicable to the growth of various high-quality crystals. In this report, we will present the effects of solution stirring and nucleation control on the solution growth of CsLiB 6 O 10 (CLBO), proteins, and GaN crystals with respect to the following issues: (1) effect of solution stirring on the growth of high-quality CLBO crystals, (2) high-quality protein crystal growth by means of forced nucleation induced by laser irradiation and the solution-stirring method, and (3) high-quality 2-in GaN crystal by Na flux liquid-phase epitaxy method.

Growth of bulk 2H–SiC single crystals in a C–Li–Si ternary melt system was successful. The growth of 2H–SiC must be conducted at lower temperatures than that of other SiC poly-types. The lithium flux solved this problem, as it is usable for growth under 1000 °C due to its low melting temperature. Success in the growth of 2H–SiC single crystals in this study may unlock the potential for realizing high-performance electronic devices by fabricating the devices on 2H–SiC single-crystal substrates.

In the liquid phase epitaxy (LPE) growth of GaN single crystals using the Na flux method, we achieved an increase in the growth rate and a change in the growth thickness distribution by generating thermal convection in the Ga–Na melt. The effect of thermal convection on the growth rate and growth thickness distribution became stronger with the strength of the convection, which indicates that thermal convection can produce a high dissolution rate of nitrogen gas into the melt as well as homogenize the nitrogen concentration.

We report systematic Cu-NMR studies on five-layered cuprates from under-doped HgBa2Ca4Cu5O12+δ (Hg-1245(UD)) to slightly overdoped Tl-1245(OVD), and compare with optimally-doped Hg-1245(OPT). In the under-doped Hg-1245(UD), antiferromagnetism (AFM) has been found to take place at TN = 290 K, exhibiting a large antiferromagnetic moment of 0.67–0.69 μB at three inner planes (IP’s). These values are comparable to that reported for non-doped cuprates, suggesting that the IP’s may be in a nearly non-doped regime. Most surprisingly, the AFM order is also detected with MAFM(OP) = 0.1 μB even at two outer planes (OP’s) that are responsible for the onset of superconductivity (SC) with Tc = 72 K. The high-Tc SC at Tc = 72 K can uniformly coexist on a microscopic level with the AFM at OP’s. This is the first microscopic evidence for the uniformly mixed phase of AFM and SC on a single CuO2 plane. Although, the AFM/SC mixed CuO2 planes are significantly separated by three non-doped AFM layers, the onset of AFM does not prevent the occurrence of SC with the high value of Tc = 72 K.

We report the fabrication of a-plane GaN single crystal substrates grown by the Na flux method. In this research, Sr was added into the flux system as additive to control the crystal habit of GaN single crystals. As the amount of Sr in the melt increased, it was found that the GaN crystal shape changed from pyramidal to prismatic crystals elongated parallel to the direction. Additionally, liquid phase epitaxy (LPE) GaN crystals were grown in Sr–Na solution on a-plane GaN templates fabricated by metal organic chemical vapor deposition (MOCVD) on sapphire substrates. ω-scan X-ray diffraction measurements showed that the full width at half maximum (FWHM) of the (11-20) plane of the LPE GaN crystal was smaller (236 arcsec) than that of the a-plane of the GaN template (1152 arcsec).

We report systematic 31P-NMR study on iron (Fe)-based superconductors SrFe2(As_{1-x}P_{x})2 (Sr122AsP), in which a superconducting (SC) transition temperature Tc at x=0.35 increases from Tc=26 K up to 33 K by annealing an as-grown mono-crystalline sample. The present NMR study has unraveled that Tc reaches a highest value of 33 K at x=0.35 around a quantum critical point at which antiferromagnetic (AFM) order disappears. When noting that the SC transition disappears at x=0.6 where the AFM spin fluctuations (SFs) are no longer present, we remark that the onset and increase of Tc are apparently associated with the emergence and enhancement of AFM-SFs, respectively. In the SC state, the residual density of state (RDOS) at the Fermi energy EF in the SC state becomes much smaller for the annealed sample than for the as-grown one, suggesting that some inhomogeneity and/or imperfection for the latter increases RDOS as expected for unconventional SC state with nodal gap. These findings in Sr122AsP are consistent with the unconventional s(+-)-wave Cooper pairing state that is mediated by AFM-SFs. We also discuss other key-ingredients besides the AFM-SFs to increase Tc further.

We report similarities and differences of the transport features in the spin density wave (SDW) and in the field-induced SDW (FISDW) phases of the quasi-one-dimensional compound (TMTSF)2PF6. As temperature decreases below ≈2 K, the resistance in both phases exhibits a maximum and a subsequent strong drop. However, the characteristic temperature of the R(T) maximum and its scaling behavior in different magnetic fields B are evidence that the nonmonotonic R(T) dependences have different origin in SDW and FISDW regions of the phase diagram. We also found that the borderline T0(B, P) which divides the FISDW region of the P-B-T phase diagram into the hysteresis and nonhysteresis domains terminates in the N=1 subphase; the borderline thus has no extension to the SDW N=0 phase.

We report the systematic evolution of the superconducting (SC) characteristics of the heavy-fermion (HF) superconductor CeCoIn 5 via nuclear-quadrupole-resonance (NQR) measurement under pressure ( P ). The application of P significantly suppresses the nuclear spin–lattice relaxation rate 1/ T 1 that is dominated by antiferromagnetic (AFM) spin fluctuations (SFs) specific to a quantum critical point (QCP). It is demonstrated that the marked suppression of AFM SFs leads to a reduction in the SC energy gap or in the coupling strength of the Cooper pair. T c , nevertheless, increases with increasing P due to the increase in HF bandwidth. This is expected to make the lifetime of quasi-particles sufficiently long.

Magnetic properties on Sr2−xCaxRuO4 have been investigated by a microscopic probe of 87Sr-NMR in order to understand the magnetic character on spin-triplet superconductor of Sr2RuO4, which has multibands on the Fermi surface. With substituting Ca for Sr which gives rise to crystal distortion, the Knight shift (K) and the nuclear spin-lattice relaxation rate divided by temperature (1/T1T) increases progressively up to x=1.5. The Korringa relation from K and 1/T1T becomes smaller, indicative of development of ferromagnetic fluctuations with increasing Ca content. This suggests that the q-independent spin fluctuations originating from the 2-dimensional γ band are changed to the ferromagnetic ones by the Ca doping.

Cu-nuclear quadrupole resonance (NQR) measurements have been performed on single crystals Sr 14-x Ca x Cu 24 O 41 (Ca x ) with x =0-11.5, in order to clarify the relation between the magnetic order around x =11.5 and local environments at the Cu sites upon x. Cu-NQR frequency ν Q for the ladders at 4.2 K increases linearly with increasing x, whereas ν Q for the nonmagnetic Zhang-Rice (ZR) sites in the chains does not. Detailed site assignment of the spectra, which are different from those in the previous paper, is presented.

High-quality and low cost bulk crystals are needed in the field of group III nitride semiconductors in order to develop optical and electrical devices. There are two approaches for the growth of bulk GaN crystal by the Na flux method. One is to grow thick GaN crystal on a large seed GaN crystal grown by vapor phase method. The other is to grow GaN crystal on a small seed GaN crystal. 3 in diameter GaN crystals were grown on the large GaN seed crystal. In the case of the growth on a small GaN seed, we obtained bulk crystal with a pyramidal shape and its height and diameter were 15 mm and >20 mm, respectively. We also present the effects of the impurity in the solution on the property and growth habit.

We report the results of nuclear spin–lattice relaxation time, T 1 of 11 B in MgB 2 , Al-doped MgB 2 and NbB 2 . A T 1 T =constant behavior was observed in the normal state, indicating the absence of strong magnetic correlation. In the superconducting (SC) state, 1/ T 1 shows a tiny coherence peak just below T c and decreases exponentially, demonstrating an s-wave superconductivity. The magnitude of SC gap is estimated as 2 Δ / k B T c ∼5, which is quite larger than the weak-coupling value in the BCS theory, 2 Δ / k B T c =3.5. This rules out a possibility for the multiple SC gaps that was suggested in many literatures. The anisotropy in the upper critical field was also corroborated by the T 1 measurement on a bulk polycrystalline sample. In NbB 2 with a low value of T c =5 K, the T 1 measurement revealed the distinct coherence peak just below T c , followed by an exponential decrease with a magnitude of SC gap 2 Δ / k B T c =3.1. These data are consistent with the weak-coupling BCS theory. As Al 3+ is substituted for Mg 2+ , 1/ T 1 T , which is proportional to the square of the density of states at the Fermi level N ( E F ), decreases. The variation in T c against the relative changes in N ( E F ) deduced from 1/ T 1 T is well fitted in terms of the McMillan equation by assuming a characteristic phonon frequency ω ∼ 700 K and an electron–phonon coupling constant λ ∼ 0.87. Thus obtained values are in good agreement with the values suggested by the theoretical works. The high- T c superconductivity in MgB 2 is shown to occur through the strong coupling with high frequency phonons.

The isothermal equation of state of CeCu2Ge2 at 10 K has been studied up to 20 GPa in a diamond-anvil cell using angle-dispersive X-ray diffraction technique with a synchrotron source and an imaging plate detector. At 15.0±1.6 GPa, an anomalous contraction of the volume was observed while the crystal structure remained unchanged. Relevance of the result with the enhancement of the superconducting transition temperature reported in this vicinity is discussed.

We demonstrate highly efficient frequency doubling of a 160 mW-AlGaAs three-section tunable distributed Bragg reflector (DBR) laser diode with real refractive-index-guided structure in a quasi-phase-matched second harmonic generation (QPM-SHG) ridge-type waveguide device. We realized high coupling efficiency of the DBR laser diode and the QPM-SHG waveguide device using a planar-type butt-coupling configuration. High-power SHG blue violet laser was fabricated, whose volume is 0.3 cc. This SHG blue-violet laser satisfied the several specifications for optical disk systems, and the pulsed blue-violet light (410 nm) power of 62 mW was generated.

31P-NMR study on LaFe(As1–x P x )(O0.86F0.14) unraveled that antiferromagnetic spin fluctuations (AFMSFs) emerge significantly at x=0.4 where the Tc is markedly enhanced, indicating that the AFMSFs are one of the important factors for raising Tc . From extensive comparison over wide compositions for 0 ≤ x ≤1 and 0 ≤ y ≤ 0.14, we revealed that there are two different types in temperature evolution of the AFMSFs: One is enhanced particularly at low energies that evolves only at low temperatures, which mainly derives from the two orbitals of dxz/yz . The other is distributed broadly at finite energies that appears up to high temperatures, which derives from three orbitals of dxy and dxz/yz . The highest Tc (=27 K) state in the present compositions appears at (x, y)=(0.4, 0.1) where two characteristics of AFMSFs merge, suggesting the contribution of the AFMSFs over wide energies to the onset of SC. The nonmonotonic variation of Tc in LaFe(As1–x P x )(O1–y F y ) is attributed to the AFMSFs from degenarated multiple-3d-orbitals on iron-pnictide superconductors.

We optimized the initial GaN growths of nonpolar m- and a-plane GaN grown on liquid phase epitaxy (LPE-) GaN substrates. Such nonpolar LPE-GaN substrates were produced by cutting slices from c-plane LPE-GaN bulk single crystals grown by the Na-flux method. We investigated the qualities of epitaxially grown GaN layers on the LPE-GaN substrates, with different polishing methods, which are chemical mechanical polishing and plasma dry etch polishing. We found that the crystalline quality of epitaxial GaN on m-plane GaN is very sensitive to the surface polishing even the minute unevenness in GaN substrate. Our experiments also indicated that a low initial growth rate was necessary to obtain high-crystalline-quality epitaxial m-plane GaN. In contrast, high-crystalline-quality a-plane GaN was obtained even with a high initial growth rate, indicating that the crystalline quality of a-plane GaN is not sensitive to surface roughness. (© 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

Systematic P-NMR studies on LaFe(As_{1-x}P_x)(O_{1-y}F_y) with y=0.05 and 0.1 have revealed that the antiferromagnetic spin fluctuations (AFMSFs) at low energies are markedly enhanced around x=0.6 and 0.4, respectively, and as a result, Tc exhibits respective peaks at 24 K and 27 K against the P-substitution for As. This result demonstrates that the AFMSFs are responsible for the increase in Tc for LaFe(As_{1-x}P_x)(O_{1-y}F_y) as a primary mediator of the Cooper pairing. From a systematic comparison of AFMSFs with a series of (La_{1-z}Y_z)FeAsO_{\delta} compounds in which Tc reaches 50 K for z=0.95, we remark that a moderate development of AFMSFs causes the Tc to increase up to 50 K under the condition that the local lattice parameters of FeAs tetrahedron approaches those of the regular tetrahedron. We propose that the T_c of Fe-pnictides exceeding 50 K is maximized under an intimate collaboration of the AFMSFs and other factors originating from the optimization of the local structure.

63 Cu -NMR measurement has been performed to investigate the coexistence of superconductivity and ferromagnetism in a hybrid ruthenium (Ru)–copper (Cu) oxide superconductor RuSr2YCu2O8 (RuY1212). Broadening of the Cu-NMR signal below TM demonstrates that Cu nuclei in CuO2 planes feel a dipolar field from ordered Ru moments in RuO2 planes. On the other hand, the opening of the SC gap is confirmed from the observation of a decrease in the nuclear spin–lattice relaxation rate, 63 1/T 1 , below Tconset=45 K. These NMR results provide evidence that the superconductivity coexists with the (field induced) ferromagnetism at a microscopic scale. We also found that the T dependence and the magnitude of 1/T1T are quite similar to those in YBa2Cu3O7 (Y1237), indicative of the same spin-fluctuation character in both compounds. This suggests that the hole content in RuY1212 is comparable to that in Y1237.

We find contrastive response of the pseudogap (PG) to high magnetic fields up to 28.5 T based on 63 Cu NMR measurements. In the slightly overdoped TlSr 2 CaCu 2 O 6.8 , the PG is strongly field dependent and shown to be due to the superconducting fluctuations. By contrast, the PG in the underdoped YBa 2 Cu 4 O 8 does not depend on magnetic fields up to 28.5 T. These results imply that there exists a field-insensitive PG up to a certain doping level beyond which it is taken over by the superconducting fluctuations-induced one.

From Ru- and Cu-NMR studies, we present evidence for coexistence of superconductivity and ferromagnetism in a cuprate superconductor ${\mathrm{RuSr}}_{2}{\mathrm{YCu}}_{2}{\mathrm{O}}_{8}$ (RuY1212). The observation of a large enhancement of a radio-frequency field for the Ru-NMR signal at zero field reveals the existence of a ferromagnetic (FM) component in the ordered ${\mathrm{RuO}}_{2}$ plane below a Curie temperature of ${T}_{M}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}150\mathrm{K}$. Just below the onset temperature of superconductivity ${T}_{c}^{\mathrm{onset}}\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}45\mathrm{K}$, a remarkable decrease of the nuclear spin-lattice relaxation rate ${1/T}_{1}$ was observed within the ordered ${\mathrm{RuO}}_{2}$ plane as well as the ${\mathrm{CuO}}_{2}$ plane, revealing that the superconducting gap coexists with the FM component in the ${\mathrm{RuO}}_{2}$ plane on a microscopic scale. In addition, from the observation of a sharp peak in ${}^{101}({1/T}_{1})$ at ${T}_{c}^{\mathrm{zero}}\ensuremath{\sim}23\mathrm{K}$ where the resistivity becomes zero, we suggest that the motion of self-induced vortices originating from fluctuations of the FM component induces the resistivity between ${T}_{c}^{\mathrm{onset}}$ and ${T}_{c}^{\mathrm{zero}}$ in RuY1212.

We report on the basis of Cu-NQR measurements that superconductivity (SC) and antiferromagnetism (AF) coexist on a microscopic level in CeCu_{2}(Si_{1-x}Ge_{x})_{2}, once a tiny amount of 1%Ge (x = 0.01) is substituted for Si. This coexistence arises because Ge substitution expands the unit-cell volume in nearly homogeneous CeCu2Si2 where the SC coexists with slowly fluctuating magnetic waves. We propose that the underlying exotic phases of SC and AF in either nearly homogeneous or slightly Ge substituted CeCu2Si2 are accountable based on the SO(5) theory that unifies the SC and AF. We suggest that the mechanism of the SC and AF is common in CeCu2Si2.