An electrochromic mirror electrode based on reversible uptake of hydrogen in nickel magnesium alloy films is reported. Thin, magnesium-rich Ni-Mg films prepared on glass substrates by cosputtering from Ni and Mg targets are mirror-like in appearance and have low visible transmittance. Upon exposure to hydrogen gas or on cathodic polarization in alkaline electrolyte, the films take up hydrogen and become transparent. When hydrogen is removed, the mirror properties are recovered. The transition is believed to result from reversible formation of Mg 2 NiH 4 and MgH 2 . A thin overlayer of palladium was found to enhance the kinetics of hydrogen insertion and extraction, and to protect the metal surface against oxidation.Devices capable of switching between mirror and transparent states may find applications in architectural and transportation energy conservation, lighting and displays, aerospace insolation control, and optical communications systems. Switchable mirrors based on rare earth hydrides were discovered by Huiberts et al., 1 who observed a reversible metal-to-insulator transition when a thin film (150 to 500 nm) of yttrium or lanthanum coated with a thin layer of palladium was exposed to hydrogen gas. The transition accompanies conversion of a metallic dihydride phase to a semiconducting trihydride. Rare earth-magnesium alloy films 2 were subsequently found to be superior to the pure lanthanides in maximum transparency and mirror-state reflectivity. Phase separation appears to occur when these alloys take up hydrogen, giving transparent MgH 2 and LnH 2-3 , both of which may participate in the switching mechanism. 3 Because the rare earths are highly vulnerable to oxidation, a Pd overlayer at least 5 nm thick is required for films exposed to air or to an alkaline electrolyte. Although the Pd catalyzes the uptake and removal of hydrogen, it limits the maximum transparency of the composite film to about 50%. 4 The influence of the Pd layer and of a Pd/AlO x composite cap layer on hydrogen uptake by rare earth-based films have been studied extensively by van Gogh et al. 5 and by van der Molen et al. In the case of nickel, a stoichiometric alloy phase, Mg 2 Ni, with the same Mg-Ni ratio as in the hydride, can be prepared from the elements. In Mg 2 Ni (Fig. 1a), there are Ni-Ni bonds (shown as hollow rods), two types of Ni-Mg bonds (solid rods) and three types of Mg-Mg bonds (not shown).10 The Ni-Ni bonds and Mg-Mg bonds are shorter in the alloy than in the pure elements. The alloy absorbs hydrogen without structural rearrangement up to a composition of Mg 2 NiH 0.3 .7 This phase has metallic properties similar to those of the pure alloy. Further introduction of hydrogen produces Mg 2 NiH 4 (Fig. 1b) Ni-Mg films were deposited by DC magnetron co-sputtering from 2 in diameter Ni and Mg (99.98%) targets onto glass substrates with and without transparent conductive coatings. The base pressure was 1.4 x 10 -7 Torr, process pressure 2 mTorr, Ni power 20 W, Mg power 22 W, target-tosubstrate distance 7.5 cm. Dep...

InN films with free electron concentrations ranging from mid-1017 to mid-1020 cm−3 have been studied using optical absorption, Hall effect, and secondary ion mass spectrometry. The optical absorption edge covers a wide energy range from the intrinsic band gap of InN of about 0.7 to about 1.7 eV which is close to the previously accepted band gap of InN. The electron concentration dependence of the optical absorption edge energy is fully accounted for by the Burstein–Moss shift. Results of secondary ion mass spectrometry measurements indicate that O and H impurities cannot fully account for the free electron concentration in the films.

GaN layers have been grown on m-plane sapphire by metal organic vapor phase epitaxy using low-temperature AlN nucleation layers. Depending on substrate nitridation and annealing treatments prior to depositing the nucleation layer, the crystal orientation of the resulting GaN layer may be either (11−22) or (1−100) (m plane). For suitably controlled conditions, GaN epilayers with a single m-plane orientation are reproducibly obtained as confirmed by x-ray diffraction. There is a 90° in-plane rotation of the epilayer such that the GaN a axis is parallel to the sapphire c axis.

Carbon-doped GaN layers grown by molecular-beam epitaxy are studied with photoluminescence and positron annihilation spectroscopy. Semi-insulating layers doped with >1018 cm−3 carbon show a strong luminescence band centered at ∼2.2 eV (yellow luminescence). The absolute intensity of the 2.2 eV band is compared with the gallium vacancy concentration determined by positron annihilation spectroscopy. The results indicate that a high concentration of gallium vacancies is not necessary for yellow luminescence and that there is in fact a causal relationship between carbon and the 2.2 eV band. Markedly different deep-level ionization energies are found for the high-temperature quenching of the 2.2 eV photoluminescence in carbon-doped and reference samples. We propose that while the model of Neugebauer and Van de Walle [Appl. Phys. Lett. 69, 503 (1996)] applies for GaN of low carbon concentration, a different yellow luminescence mechanism is involved when the interstitial carbon concentration is comparable to or exceeds the gallium vacancy concentration.

The luminescence of InGaN single quantum wells grown by molecular-beam epitaxy under fixed conditions over a series of c-axis GaN nanowire arrays with different geometrical parameters was studied. For arrays with variable GaN average wire diameters and fixed wire densities, the InGaN luminescence peak shifted to higher energy with decreasing wire diameter. It is shown that this trend cannot be attributed to lateral quantum confinement or diameter-dependent InGaN strain. For arrays with variable wire densities and fixed average diameters, the InGaN emission appeared as two distinct bands of different colours, the relative intensities of which depended on the wire density. By optimizing both the GaN wire density and InGaN growth conditions, the colours of the two different bands were combined to realize phosphor-free white light-emitting diodes. The mechanisms for the dependence of the InGaN luminescence on the geometrical parameters of the GaN nanowire array are discussed.

The properties of a broad 2.86 eV photoluminescence band in carbon-doped GaN were studied as a function of C-doping level, temperature, and excitation density. The results are consistent with a C Ga -C N deep donor-deep acceptor recombination mechanism as proposed by Seager et al. For GaN:C grown by molecular-beam epitaxy (MBE) the 2.86 eV band is observed in Si co-doped layers exhibiting high n-type conductivity as well as in semi-insulating material. For low excitation density (4 W/cm 2 ) the 2.86 eV band intensity decreases as a function of cw-laser exposure time over a period of many minutes. The transient behavior is consistent with a model based on carrier diffusion and charge trapping-induced Coulomb barriers. The temperature dependence of the blue luminescence below 150 K was different for carbon-contaminated GaN grown by metalorganic vapor phase epitaxy (MOVPE) compared to C-doped MBE GaN.

Gallium nitride is grown by plasma-assisted molecular-beam epitaxy on (111) and (001) silicon substrates using sputter-deposited hafnium nitride buffer layers. Wurtzite GaN epitaxial layers are obtained on both the (111) and (001)

No abstract