Search

PN junctions are basic building blocks of many electronic devices and their performance depends on the structural properties of the component layers and on the type and the amount of the doping impurities incorporated. Magnesium is the common p-type dopant for nitride semiconductors while silicon and more recently germanium are the n-dopants of choice. In this paper, therefore we analyze the quantitative limits for Mg and Ge incorporation on GaN and InGaN with high In content. We also discuss the challenges posed by the growth and characterization of InGaN pn-junctions and we discuss the properties of large area, long wavelength nanocolumn LEDs grown on silicon (1 1 1) by PA-MBE.

In-situ Auger electron spectroscopy was used to determine the 1:1 flux ratios of Ga and N during growth of GaN by molecular beam epitaxy at low substrate temperatures. By linearly ramping the Ga-flux, while keeping the N-flux constant, and simultaneously measuring the chemical composition by monitoring N and Ga Auger peaks, the time of deviation from stoichiometry could be determined. The method was applied at very low substrate temperatures where reflection high-energy electron diffraction does not reveal clear growth mode changes. The importance of the N- vs Ga-rich conditions were confirmed with transmission electron microscopy which showed a distinct change in crystallinity between material at the top and bottom of the film, which are in agreement with previous findings.

Highly mismatched GaN 1− x Sb x alloys were grown under N-rich conditions at low substrate temperatures (325–550 °C) at a growth rates of ~0.09 μm/hr on sapphire. The alloys ranged in Sb composition from 0% to 16%, with the bandgap shifting from 3.3 to 1.6 eV in accordance with the band anticrossing (BAC) model. We compare these results to growths from another chamber, having a different N source, and using a faster growth rate (~0.24 μm/hr), much lower substrate temperatures (as low as 80 °C), different III/V ratios and absolute fluxes. Despite the range of morphologies obtained, all alloys follow the predictions of the BAC model with the bandgap only depending on the Sb composition.

The electronic structure of the conduction and valence bands of highly mismatched alloys (HMAs) such as GaN 1− x Sb x are well described by the band anticrossing model. The properties of this alloy, which has a large band gap range and controllable valence band positions, make it a candidate material for efficient solar energy conversion devices. We have examined the growth and structural properties of amorphous and crystalline GaN 1− x Sb x . These HMAs were grown by low temperature molecular beam epitaxy (MBE) under Ga-rich conditions. While there is a monotonic linear increase of Sb incorporation with Sb overpressure, there was no obvious dependence of Sb incorporation with growth temperature for the range of 10–470 °C. At growth temperatures lower than 100 °C, GaN 1− x Sb x HMAs lose crystallinity and become amorphous for Sb compositions at or exceeding ~5%. Ga-rich growth resulted in strong absorption at energies as low as 1 eV for GaN 1− x Sb x alloys of all compositions. The strong low energy absorption may arise from a Ga-related defect band.

We report the growth and characterization of the lattice-matched Zn 1- x Cd x Te 1- y O y (ZnCdTeO) layers on ZnTe substrates by radio frequency plasma-assisted molecular beam epitaxy technique. The Cd composition increases linearly with increasing Cd/(Zn+Cd) flux ratio, indicating a controllability of Cd composition by Cd flux. Introduction of O radical during the growth of ZnCdTe resulted in the formation of ZnCdTeO layer. At particular O and Cd compositions lattice-matched ZnCdTeO epilayers on ZnTe substrate were obtained. Photoreflectance (PR) spectroscopy on the lattice-matched ZnCdTeO layer revealed two distinct PR features in the energy regions at 2.2–2.5 eV and 1.5–1.8 eV, which can be attributed to transitions from the valence band to the two conduction subbands, E + and E − , respectively.

Using low temperature MBE, we have shown that it is possible to grow amorphous GaN1-xAsx layers with a variable As content (0 < x < 0.8) on both crystalline (sapphire and silicon) and amorphous (glass and Pyrex glass) substrates. Despite the fact that the samples with high As content are amorphous, we observe a gradual continuous decrease of bandgap from similar to 3.4 to similar to 0.8 eV with increase in As content. To the best of our knowledge this is the first demonstration of homogeneous amorphous GaN-based alloys over a wide composition range. The large band gap range of the amorphous phase of GaNAs covers much of the solar spectrum. The amorphous nature of the GaNAs alloys is particularly advantageous since low cost substrates such as glass and Pyrex glass can be used for solar cell fabrication.

We have studied the low-temperature growth of gallium nitride arsenide (GaN)As layers on sapphire substrates by plasma-assisted molecular beam epitaxy. We have succeeded in achieving GaN1−xAsx alloys over a large composition range by growing the films much below the normal GaN growth temperatures with increasing the As2 flux as well as Ga:N flux ratio. We found that alloys with high As content x>0.1 are amorphous and those with x

CdSeTe epilayers were grown by molecular beam epitaxy on (1 0 0)-oriented vicinal silicon wafers. By controlling the growth conditions, either multiple-phase or single-phase epilayers could be deposited. In this paper we present a detailed investigation of the structure of a single-phase CdSeTe epilayer using X-ray diffraction, Rutherford back scattering, and transmission electron microscopy. Photoreflectance and photocurrent results are also presented.

We have performed systematic studies of the effects of high-energy particle irradiation on the properties of InGaN alloys. In agreement with the amphoteric defect model, irradiation of InN produces donor-like defects. The electron concentration increases with increasing radiation dose and saturates at 4 x 10 20 cm -3 at very high doses. We find that the increase of the electron concentration causes a large blue-shift of the absorption edge, which is well explained by the Burstein-Moss effect. The maximum electron concentration decreases with increasing Ga fraction in irradiated In 1-x Ga x N alloys as the conduction band edge approaches the Fermi level stabilization energy (E FS ). For x>0.66 the conduction band edge moves above E FS and the irradiation of n-type films produces acceptor-like defects, resulting in a reduced free electron concentration. An analysis of the concentration dependence of the electron mobility in InN indicates that the dominant defects in irradiated InN are triply charged donors. Finally, we show that InN films doped with Mg acceptors behave like undoped films above a threshold radiation dose.

The optical properties and electronic structure of molecular-beam epitaxy grown InN and In-rich group III-nitride alloy films are studied. The band gap of InN is determined to be 0.7 eV by optical absorption, photoluminescence, and photo-modulated reflectance. The band gap exhibits weaker temperature and pressure dependencies than those of GaN and AlN. The narrow band gap leads to a strong k·p interaction, resulting in a non-parabolic conduction band, which is studied by the free electron concentration dependence of the electron effective mass. Highly n-type InN exhibits a large Burstein-Moss shift in the optical absorption edge; this effect may be responsible for the 1.9 eV band gap reported previously for some degenerately doped InN films. The band gap bowing parameters of the InGaN and InAlN alloy systems are determined. The band offset of InN with other group III-nitrides is presented and its effect on p-type doping is discussed.

The amphoteric native defect model is applied to the understanding of the variations in the dopant activation efficiency in II–VI compounds. It is shown that the location of the common energy reference, the Fermi level stabilization energy, relative to the band edges can be used to determine the doping induced reduction of the formation energy and the enhancement of the concentration of compensating native defects. The model is applied to the most extensively studied compound semiconductors as well as to ternary and quaternary alloys.