= GaN blue LED development
{c}
<Isamu Akasaki> chose <metalorganic vapour-phase epitaxy> and <sapphire>, which tolerated hot <ammonia>. Large <lattice mismatch> hindered film growth. In 1985, <Hiroshi Amano> and <Isamu Akasaki>'s group developed an <aluminium nitride> <epitaxial buffer layer> deposited before hotter <gallium nitride> growth. The buffer supplies initial growth sites; growing islands join into a continuous film. It improves growth without making the lattices identical. <Isamu Akasaki> separately credits <Shuji Nakamura>'s 1991 <gallium nitride> buffer.https://www.kapitza.ras.ru/~glazkov/students/Nobel-akasaki-lecture.pdf[§§3.2–4.2, pp. 13–20, Figs. 7–11: growth sequence and microscopy]{ref}
The 1986 growth paper reports a vertical <metalorganic vapour-phase epitaxy> reactor, source gases mixed just before delivery to slanted <sapphire>, and subsequent <gallium nitride> growth over <aluminium nitride>. Surface microscopy, <X-ray diffraction> and <photoluminescence> tested different aspects of the result. Buffered and unbuffered growth supplied the comparison. Its reported preparation conditions should not be replaced by one universal buffer <temperature>: the later lecture's account of the first low-temperature trial and the paper's parameter study describe different stages of development.<Metalorganic vapor phase epitaxial growth of a high quality GaN film using an AlN buffer layer by Hiroshi Amano and others>[p. 353, Fig. 1 and Table I; p. 354, surface and diffraction comparisons]{ref}
A <magnesium> <acceptor impurity> in <gallium nitride> accepts a <valence band> <Electron>, leaving a mobile <electron hole>. <Hydrogen passivation of magnesium acceptors in gallium nitride> suppresses this activity. Post-growth activation restores existing impurities' electrical activity; ionization of active <acceptor impurities> supplies mobile <electron holes>. Adding <magnesium> alone is insufficient.<Hole Compensation Mechanism of P-Type GaN Films by Shuji Nakamura and others>[p. 1258, abstract and §1: passivation and activation]{ref}
In 1989, <Hiroshi Amano>, <Masahiro Kito>, <Kazumasa Hiramatsu> and <Isamu Akasaki> obtained <p-type semiconductor> conduction in <magnesium>-doped <gallium nitride> using low-energy <Electron> beam irradiation. <Hiroshi Amano> recalls initially resistive samples, an unreliable hot-probe test, then <Masahiro Kito>'s <Hall effect> confirmation. A junction irradiated in an M-shaped region emitted in that pattern. Treatment, carrier-sign measurement and spatial emission supplied different evidence. Earlier <zinc>-doped samples brightened under irradiation without becoming <p-type semiconductors>: brightness alone did not prove the carrier sign.https://journals.aps.org/rmp/pdf/10.1103/RevModPhys.87.1133[§V, pp. 1135–1136, Figs. 3–4; authors of 1989 paper in references, p. 1137]{ref}
<Shuji Nakamura>, <Takashi Mukai>, <Masayuki Senoh> and <Naruhito Iwasa>'s 1992 thermal-activation paper established another route using <annealing> in <nitrogen>. Their companion study compared already conducting films heated in <nitrogen> with films heated in <ammonia>: the latter became highly resistive again. Its <hydrogen>-complex explanation connected this atmosphere dependence to <acceptor impurity> passivation. These are thermal treatment and atmosphere comparisons, distinct from the earlier <Electron> beam intervention; neither observation alone images an individual <hydrogen> complex.<Thermal Annealing Effects on P-Type Mg-Doped GaN Films by Shuji Nakamura and others>[abstract: nitrogen annealing above 700 degrees Celsius and electrical changes]{ref}<Hole Compensation Mechanism of P-Type GaN Films by Shuji Nakamura and others>[p. 1258, abstract and §§1–2: activation history, nitrogen/ammonia comparison and hydrogenation hypothesis]{ref}
<Shuji Nakamura>'s two-flow <metalorganic chemical vapor deposition> reactor, <gallium nitride> buffer, thermal activation and improved <indium gallium nitride> growth enabled bright <double heterostructure> devices, followed by thinner <quantum well> devices.https://journals.aps.org/rmp/pdf/10.1103/RevModPhys.87.1139[§§III–IV.A, pp. 1140–1144; Figs. 9–11]{ref}
The original 1994 paper by <Shuji Nakamura>, <Takashi Mukai> and <Masayuki Senoh> describes a <sapphire>/<gallium nitride>-buffer device with <zinc>-doped <indium gallium nitride> emission and <aluminium gallium nitride> confinement. After <nitrogen> <annealing>, etching exposed the n-side for a separate contact. At room <temperature>, varied forward <electric currents> gave emission <spectra> and optical-output curves alongside a <current-voltage characteristic>. At $20\,\mathrm{mA}$: $1.5\,\mathrm{mW}$ output, $450\,\mathrm{nm}$ peak <wavelength>, $3.6\,\mathrm V$ terminal <electric potential difference>. The peak stayed nearly fixed across the current series. The paper omits a complete calibrated collection apparatus; comparisons with earlier devices change several preparation details.<Candela-class high-brightness InGaN⁄AlGaN double-heterostructure blue-light-emitting diodes by Shuji Nakamura and others>[pp. 1687–1688, fabrication and Figs. 1–4]{ref}
Back to article page
