R. Treichler

Mg diffusion during metalorganic vapor phase epitaxy of InP

The diffusion mechanism of Mg has been studied during low‐pressure metalorganic vapor phase epitaxy of InP. The Mg dopant profiles were measured by secondary‐ion mass spectroscopy. The analysis reveals that abrupt Mg dopant profiles are possible; the Mg diffusion, however, strongly depends on the Mg concentration in the crystal lattice. Simultaneous doping with Si leads to a distinct decrease of the Mg diffusion. This behavior is consistent with a model assuming that Mg diffuses as a complex involving a deep donor.

Growth of GaInAs and GaInAsP lattice matched to InP by metalorganic MBE

Abstract This study reports on the growth of GaInAs and GaInAsP single layers and heterostructures by metalorganic MBE (MOMBE) using trimethylindium (TMI), triethylgallium (TEG), arsine (AsH 3 ) and phosphine (PH 3 ) as starting materials. The growth parameters were optimized for a temperature range where also high quality InP is available. Ternary and quaternary layers exhibiting excellent uniformity across a wafer diameter of 3 inches were obtained with regard to layer thickness (≤1.5%) and material composition. SIMS measurements on GaInAs/InP and GaInAsP/InP double heterostructures revealed even for the problematic element As an abrupt modulation of about three orders of magnitude or more depending on the structure.

Comparison of Zn and Mg incorporation in MOVPE InP/GaInAsP laser structures

The objective of this work is to study the incorporation process of Zn in InP and related ternary and quaternary layers for long wavelength laser applications in comparison with the alternative acceptor Mg. In InP above a critical concentration of (1–2)×1018 cm−3 a sudden onset of dopant diffusion during growth is observed for Zn and for Mg as well. This diffusion during growth can be markedly reduced by counter-doping with Si (Fermi level effect). Below the critical concentration Zn dopant profiles exhibit the same steep flanks as Mg dopant profiles suggesting the same low diffusion coefficients. However, Zn appears to be more suitable forp-type doping of InP, GaInAs and GaInAsP, because an accurate control of the dopant level in the epitaxial layers is easier to achieve with Zn than with Mg.

Improvements in silicon doping of InP and GaInAs in metalorganic molecular beam epitaxy

Abstract We have investigated the Si doping of InP and GaInAs in metalorganic molecular beam epitaxy (MOMBE) by using a conventional Si effusion cell. In order to reduce the formation of SiC promoted by the background gases in MOMBE, we introduced a liquid nitrogen cooled baffle between the cell and the mechanical shutter. The results show that the passivating reaction can be substantially suppressed by a proper treatment of the source cell. The doping efficiency remains constant over a long period of operation corresponding to a large total layer thickness (>100 μm). The comparison of SIMS analysis with Hall data reveals an electrical activation of Si in InP up to 100% and about 65% for Si in GaInAs. These results and the investigations on doping profiles show that Si is a suitable donor in InP and GaInAs in the MOMBE process.

Calibration of the Fe2+ intracenter absorption in InP

A study is presented on the assessment of the Fe2+ concentration in Fe‐doped InP by the quantitative evaluation of the Fe2+ intracenter absorption at 2800–3200 cm−1. Based on a comparison with secondary‐ion mass spectrometry data from n‐type samples, a calibration factor is derived for the oscillator strength of the zero‐phonon lines at 10 K. The detection limit for 500 μm samples is ≊5×1013 cm−3.

Mg incorporation process during LP-MOVPE of InP, GaInAs and GaInAsP

Abstract The Mg incorporation process has been studied during low pressure metalorganic vapor phase epitaxy of InP, GaInAs and GaInAsP layers. In InP a cubic increase of the carrier concentration versus Mg flow in the gas phase is observed; a maximum carrier concentration of 2 × 10 18 cm −3 can be obtained. Results from Hall measurements and low temperature photoluminescence suggest that above a critical level of 10 18 atoms/cm 3 Mg is incorporated in InP both as a substitutional acceptor and as an interstitial donor. In ternary layers and in quaternary layers with a rather high Ga content the carrier concentration exhibits a square increase versus Mg flow; a saturation at a level of approximately 1 × 10 19 cm -3 is reached. The data can be explained assuming that in Ga containing compounds, such as ternary and quaternary layers, the concentration of interstitial Mg is much lower than in InP leading to a markedly reduced Mg diffusion during growth even at high concentrations.

Zn doping of InPGaInAsP device structures in metalorganic molecular beam epitaxy using diethylzinc

Diethylzinc was used as a gaseous p-type dopant source for growth of InP/GaInAsP layers in metalorganic molecular beam epitaxy. In InP layers a significant effect of growth temperature on Zn incorporation and on electrical activation has been found. Additionally, data from a variation of the dopant cracker cell temperature suggest that the dopant molecules should not be fully decomposed for an efficient dopant incorporation. A comparison of Hall data with data from secondary ion mass spectrometry (SIMS) reveals that in InP up to 60% of the acceptors appear to be electrically active under optimized experimental parameters, in GaInAs the activation is above 90%. The SIMS data show that dopant profiles with steep flanks can be obtained in InP/GaInAsP structures. However, a dopant redistribution occurs, which is more pronounced in InP layers. This effect is correlated with the dopant incorporation behaviour on substitutional and interstitial sites. The dopant incorporation process is discussed in detail, and the implications for growth of InP/GaInAsP device structures are outlined.

Silicon migration during MOVPE of AlGaAs/GaAs laser structures

Abstract Silicon migration during MOVPE growth of AlGaAs/GaAs laser structures has been studied by secondary ion mass spectroscopy (SIMS). The migration process has been found to depend mainly on the silison concentration in the AlGaAs layer, both for silane and for disilane as the doping gas. Above a critical concentration of about 3×10 18 cm −3 silicon migrates into the nominally undoped GaAs layer. This shift in silicon front becomes even more pronounced, when the GaAs layer is grown at a lower rate than the AlGaAs layer. The silicon depth profile exhibits the same steepness as the aluminum depth profile, even in layers with a large shift in silicon front. The migration process appears to be preferential towards the growth front. It is concluded that the process is not only determined by diffusion, but also by surface kinetics. The influence of silicon migration on threshold current density of broad area lasers is only significant for a large shift of the silicon front into the active GaAs layer.

Interaction of the dopants Mg and Si in AlxGa1−x As/GaAs heterolayers (MOVPE): application to DQW laser structures

Abstract The interaction of Mg and Si has been studied in GaAs/Al x Ga 1− x As DQW laser structures with a 50 nm Si diffusion barrier. The samples have been exposed to capless heat treatments at 860°C and under Si/SiO 2 and Si 3 N 4 cap layers, and were analysed by SIMS. The Mg diffusion is highly dependent on the surface conditions during heating. A Si barrier is effective for temperature treatments under H 2 /AsH 3 and Si/SiO 2 . It is not effective under a Si 3 N 4 cap where we detected very fast Mg diffusion. The Mg diffusion behaviour is discussed in terms of Si-Mg interaction and the influence of crystal defects.

Ion implanted confinement layer for an InP/InGaAs/InP:Fe heterojunction field‐effect transistor

A p‐doped confinement layer was fabricated by Be implantation for an InP/InGaAs/InP:Fe heterojunction field‐effect transistor (HFET). The epitaxial layers were grown by metalorganic vapor phase epitaxy and were suitable for the integration with a pin photodiode. The pn junction of the gate was formed by Zn diffusion that is not influenced by the preceding ion implantation. In the output characteristics of the HFET the pinch‐off voltage changes from Vp=−7 V without to Vp=−3.5 V with confinement layer. No degradation of the maximum transconductance was observed. The gate leakage current is 80 nA at a gate source voltage of Vgs=−5 V.