We report bulk GaInNAs p − i − n photodiodes lattice-matched to GaAs substrates, grown by solid source molecular beam epitaxy with photoresponses out to ϳ1.3 m. The as-grown samples were characterized optically, structurally, and electrically. A low background doping concentration in the range of 10 14 -10 15 cm −3 was obtained in the samples. One of the samples with a 0.5 m thick GaInNAs absorbing layer gave a responsivity of 0.11 A/W for a band edge of 1.28 m at reverse bias of 2 V.

We report on the properties of GaNAsP/GaP lasers which offer a potential route to producing lasers monolithically on silicon. Lasing has been observed over a wide temperature range with pulsed threshold current density of 2.5 kA/ cm 2 at 80 K ͑ = 890 nm͒. Temperature dependence measurements show that the radiative component of the threshold is relatively temperature stable while the overall threshold current is temperature sensitive. A sublinear variation of spontaneous emission versus current coupled with a decrease in external quantum efficiency with increasing temperature and an increase in threshold current with hydrostatic pressure implies that a carrier leakage path is the dominant carrier recombination mechanism. © 2008 American Institute of Physics. ͓DOI: 10.1063/1.2975845͔The dominance of silicon ͑Si͒ for electronic and microelectronic circuit applications has lead to the search for monolithic optoelectronic integrated circuits ͑OEICs͒ on Si substrates. One of the key components of OEICs is a laser material for efficient light emission. However, the indirect band gap of Si has meant efficient light emission and gain have been difficult to achieve. Several strategies for producing lasers on silicon have been proposed, such as the "hybrid" laser, 1 whereby an InP-based active region is wafer fused onto a silicon/silica based waveguide or utilizing the Raman effect with external optical pumping. 2 However monolithic growth on a silicon substrate coupled with electrical injection has remained challenging. Growth of conventional direct III-V compound semiconductors directly onto Si is very difficult due to the formation of threading dislocations as a result of the large lattice mismatch. However, it has been shown that GaP can be grown without dislocations on Si due to the relatively small difference in lattice constant ͓Ͻ0.4% at room temperature ͑RT͔͒. 3 GaP is itself an indirect band gap semiconductor, but a GaNAsP alloy with high As fractions and dilute N fractions ͑of ϳ4%͒ can form a direct band gap material approximately lattice matched to GaP and Si. 4 Hence the realization of a GaP-based direct band gap semiconductor laser material on a silicon substrate can provide a realistic route toward monolithic laser sources for silicon-based OEICs. This is also another example of the potential for dilute nitride based materials in optoelectronic components.In this letter, we investigate the properties of GaNAsP lasers grown on GaP substrates. Using high pressure and low temperature techniques we have probed the extent to which a two level band anticrossing ͑BAC͒ ͑Ref. 5͒ model may be used to describe this material and have investigated the degree to which different carrier recombination processes govern laser behavior.The samples studied were grown by metal organic vapor phase epitaxy ͑MOVPE͒ on a GaP substrate. They consist of a single 6 nm GaN 0.04 As 0.8 P 0.16 2.5% compressively strained quantum well ͑SQW͒ within two undoped 150 nm GaP barrier/separate confinement layers. Optical confinement is provided by...

GaNP / GaP is promising for yellow-amber-red light emitting diodes ͑LEDs͒. In this study, pressure and temperature dependent electroluminescence and photocurrent measurements on bulk GaP / GaN 0.006 P 0.994 / GaP LED structures are presented. Below ϳ110 K, emission is observed from several localized nitrogen states. At room temperature, the band-edge energy increases weakly with pressure at a rate of +1.6 meV/ kbar, substantially lower than the ⌫ band gap of GaP ͑+9.5 meV/ kbar͒. Thus, despite the multiplicity of nitrogen levels, the band anticrossing model reasonably describes this system based on an average of the nitrogen states. Furthermore, carrier leakage into the X minima of GaP reduces the efficiency in GaNP-LEDs with increasing pressure. For a number of years, nitrogen has been added to the indirect semiconductor GaP to act as a radiative defect center to produce light emitting diodes ͑LEDs͒.1 Recently, there has been renewed interest in III-V-N semiconductors due to the large band-gap bowing that is observed in such materials when only a few percent of nitrogen is added. The interaction between the localized N states and the conduction band ͑CB͒ minimum has been effectively modeled with both the band anticrossing ͑BAC͒ model 2 and the empirical pseudopotential method.3 This has recently led to activity in dilute nitrides based on GaP where the fraction of nitrogen can be carefully controlled via molecular beam epitaxy ͑MBE͒ growth. Additionally, yellow-amber-red GaNP LEDs may offer better device characteristics than current AlInGaP devices due to a weaker dependence of the band gap 4 and lower thermal resistivity. 5 In this letter, we investigate the validity of a simple BAC model in GaNP-based materials and consider the factors limiting the electro-optic efficiency of LEDs based on GaNP.The devices studied here were grown by MBE and utilize simplified chip processing by one-step growth on transparent 350 m thick n-type GaP ͑100͒ substrates. The epitaxial layers consist of a 0.3 m thick n-GaP layer ͑Si doped͒, 0.15/ 0.1/ 0.15 m thick undoped GaP / GaN 0.006 P 0.994 / GaP active region, and 0.8 m thick p-GaP ͑Be doped͒ contact layer. LED chips were fabricated using Ge/ Au/ Ni/ Au and AuZn metallization for n-and p-type contacts, respectively.We employ temperature dependent measurements performed with a standard closed cycle cryostat setup over the temperature range of 70-300 K. This gives temperature and current dependencies of the electroluminescence ͑EL͒ spectra. The application of high hydrostatic pressure causes an increase in the direct band gap, and a reduction in the indirect X minima, thereby allowing investigations of the band structure and leakage effects into the indirect minima of GaP. Pressure dependent measurements were performed over the range of 0 -10 kbar using gaseous helium as the pressure medium. In the dilute nitrides, high pressure can also be used to tune the interaction between the nitrogen level͑s͒ and the conduction band of the host material, forming a useful means of investiga...

We use high pressure techniques to investigate the properties of two classes of “dilute‐nitride‐phosphide”‐based devices; GaNP/GaP light emitting diodes for yellow–amber–red display applications and GaNAsP/GaP lasers, a potential route to producing lasers monolithically on silicon. Based upon high pressure electroluminescence measurements we find that the band anti‐crossing (BAC) model reasonably describes the GaN(As)P system based on an average of the nitrogen states. In terms of device characteristics, we find that carrier leakage into the X‐minima of GaP reduces the efficiency of GaNP/ GaP LEDs with increasing pressure. Lasing has been observed in GaNAsP/GaP devices with a pulsed threshold current density of 2.5 kA/cm2 at 80 K (λ = 890 nm). A weak increase in threshold current with hydrostatic pressure indicates that a carrier leakage path that does not involve the GaP X ‐minima is the dominant carrier recombination mechanism in these devices, in contrast to the LEDs. (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Abstract-In this paper the authors present a comprehensive study of the threshold current and its temperature dependence in novel direct band-gap Ga(NAsP)/GaP QW lasers which provide a potential route to lattice matched monolithic integration of long term stable semiconductor lasers on silicon. It is found that near room temperature, the threshold current is dominated by nonradiative recombination accounting for ~87% of the total threshold current density. A strong increase in threshold current with hydrostatic pressure implies that a carrier leakage path is the dominant carrier recombination mechanism.