Kotaro Hamada

Epitaxially Grown Diamond (001) 2×1/1×2 Surface Investigated by Scanning Tunneling Microscopy in Air

Scanning tunneling microscopy (STM) observation was performed for the surface of diamond epitaxial film which was grown on a diamond (001) substrate by microwave plasma-assisted chemical vapor deposition (CVD). The surface was stable even in air, and it showed a reflection high-energy electron diffraction (RHEED) pattern of the 2×1/1×2 structure. Images of the atomic level corresponding to the RHEED pattern were obtained by STM in air. Significant extension of dimer rows was observed over the entire surface. Strong similarity between Si(001) grown by molecular beam epitaxy (MBE) and diamond (001) grown by CVD was shown.

Energy-filtered imaging in a scanning electron microscope for dopant contrast in InP

We demonstrate that energy-filtered secondary electron (SE) imaging can be used effectively to observe dopant contrast from an InP surface covered with a contamination layer formed by continuous electron irradiation. Although dopant contrast from a surface covered with a contamination layer was almost invisible in a normal SE image, it was still clearly seen in the energy-filtered image. The contrast mechanism is explained in terms of a metal-semiconductor contact charging model and energy shift between the SE distributions across p-type and n-type regions. The results suggest that energy-filtered imaging can reduce the effects of a contamination layer.

Energy-Filtered Secondary-Electron Imaging for Nanoscale Dopant Mapping by Applying a Reverse Bias Voltage

We observed the dopant contrast of an InP structure with magnifications as high as 250,000 by simultaneously applying secondary electron energy-filtering and a reverse bias voltage. The detection modes without energy-filtering and without a bias voltage did not generate a clear contrast. This was because the reverse bias increased the contrast and relatively decreased the sensitivity to the contamination layer while the energy-filtering reduced the influence of the contamination layer, even at a high magnification. This method can be widely used for semiconductor devices and enables practical nanoscale dopant mapping with a high data acquisition rate, and is therefore expected to greatly assist in extending the frontiers of the semiconductor industry.

Minimum detection limit and spatial resolution of thin-sample field-emission electron probe microanalysis

The minimum detection limit and spatial resolution for a thinned semiconductor sample were determined by electron probe microanalysis (EPMA) using a Schottky field emission (FE) electron gun and wavelength dispersive X-ray spectrometry. Comparison of the FE-EPMA results with those obtained using energy dispersive X-ray spectrometry in conjunction with scanning transmission electron microscopy, confirmed that FE-EPMA is largely superior in terms of detection sensitivity. Thin-sample FE-EPMA is demonstrated as a very effective method for high resolution, high sensitivity analysis in a laboratory environment because a high probe current and high signal-to-noise ratio can be achieved.

Facet Passivation of GaInAsP/InP Edge-Emitting Laser Diode by Aluminum Ultrathin Layer Insertion

We investigated the suppression of facet degradation in GaInAsP/InP edge-emitting laser diodes (LDs). And we demonstrated that facet degradation is successfully suppressed using facet passivation by inserting an aluminum thin layer to the interface between the semiconductor facet and the facet coating film. The ultrathin aluminum layer was approximately 20 A thick. From photoelectron spectroscopy analysis of the passivated surface, it was found that this passivation is effective in reducing surface recombination. This reduction effect is caused by the suppression of both phosphorus shortage and oxidation. To investigate the passivation effect, we evaluated the tolerance to facet degradation by high-current injection test up to 600 mA and electrostatic discharge test up to 1.2 kV. Although about half of the tested LDs without passivation were degraded, all the LDs with passivation survived. This facet passivation method can suppress facet degradation markedly and it can be applied to not only GaInAsP/InP LD but also other InP-based LDs.

Analysis of Reverse-Biased Electrostatic-Discharge-Induced Degradation of GaInAsP/InP Buried Heterostructure Laser Diode

We clarified the mechanism of the degradation of reverse-biased electrostatic discharge (ESD). Although the degradation mechanism of forward-biased ESD has been clarified, that of reverse-biased ESD remains unclear. We prepared a 1.31 µm GaInAsP/InP distributed feedback laser diode with a conventional pn-InP buried heterostructure for the analysis of the degradation mechanism of reverse-biased ESD. We used two approaches, namely, the failure analysis and numerical analysis of electric field. We found a defect inside an active layer in the early stage of degradation. Interestingly, such a defect was generated at the periphery of an active layer, but not at the center. On the other hand, we found that a high electric field occurs at an active layer under a reverse-biased condition, particularly at the periphery of an active layer. We also demonstrated the improvement in the tolerance to ESD with a decrease in electric field. From these results, we successfully confirmed that reverse-biased-ESD-induced degradation is caused by the concentration of an electric field.

Highly reproducible secondary electron imaging under electron irradiation using high-pass energy filtering in low-voltage scanning electron microscopy.

The reproducibility of contrast in secondary electron (SE) imaging during continuous electron irradiation, which caused surface contamination, was investigated using SE high-pass energy filtering in low-voltage scanning electron microscopy (SEM). According to high-pass energy-filtered imaging, dopant contrast in an indium phosphide remained remarkably stable during continuous electron irradiation although the contrast in unfiltered SE images decreased rapidly as a contamination layer was formed. Charge neutralization and the SE energy distributions indicate that the contamination layer induces a positive charge. This results in a decrease of low-energy SE emissions and reduced dopant contrast in unfiltered SE images. The retention of contrast was also observed in high-pass energy-filtered images of a gold surface. These results suggest that this imaging method can be widely used when SE intensities decrease under continuous electron irradiation in unfiltered SE images. Thus, high-pass energy-filtered SE imaging will be of a great assistance for SEM users in the reproducibility of contrast such as a quantitative dopant mapping in semiconductors.

Electrostatic-discharge-induced degradation of 1.3μm AlGaInAs∕InP buried heterostructure laser diodes

Degradation of 1.3μm AlGaInAs buried heterostructure laser diodes due to electrostatic discharge (ESD) is studied. The degradation mechanism of this material has not previously been clear and so the ESD tolerance was evaluated. Degradation occurred at 0.5 and 2.5kV for forward and reverse polarities, respectively. Because that ESD tolerance for forward polarity is insufficient for practical applications, we focused on it in analyzing the degradation mechanism. Elliptically shaped melted regions are observed in the active layer of the facet. Such regions developed inside a cavity under the application of ESD pulses. These results indicate that degradation is caused by melting due to optical absorption.

Electrostatic-Discharge-Induced Degradation Caused by Argon Ion Bombardment in Facet-Coating Process of GaInAsP/InP Laser Diode

Electrostatic discharge (ESD)-induced degradation is one of the most important reliability issues of GaInAsP/InP laser diodes. We investigated the relation between ion irradiation in the facet-coating process and ESD-induced degradation. We used electron-beam evaporation with two types of argon ion irradiation for facet-coating. One type of irradiation is used for facet cleaning, in which argon ions bombard the facet directly. Although the ion energy was as low as 40 eV, a lack of phosphorus and enhanced oxidation were found on the ion-irradiated surface. Furthermore, an increase in the surface recombination current and the enhancement of ESD-induced degradation were observed by extending the ion irradiation time. The other type of irradiation is used to promote evaporation. Here, argon ions do not bombard the facet directly. Thus, it had little effect on ESD-induced degradation. From these results, we successfully confirmed that direct ion irradiation increases surface recombination and accelerates ESD-induced degradation even if the ion energy is as low as 40 eV.

Combination of high spatial resolution and low minimum detection limit using thinned specimens in cutting-edge electron probe microanalysis.

The effect of sample thickness on the spatial resolution and minimum detection limit (MDL) has been investigated for field-emission electron probe microanalysis with wavelength dispersive X-ray spectroscopy (FE-EPMA-WDX). Indium gallium phosphide samples thinned to thicknesses of about 100, 130, 210, 310, and 430 nm provided effective thin-sample FE-EPMA-WDX in the resolution range of 40-350 nm and MDL range of 13,000-600 ppm (mass). A comparison of the FE-EPMA results for thin and bulk samples demonstrated that thin-sample FE-EPMA can achieve both higher sensitivity and better spatial resolution than is possible using bulk samples. Most of the X-rays that determine the MDL are generated in a surface region of the sample with a depth of approximately 300 nm. The spatial resolution and MDL can be tuned by the sample thickness. Furthermore, analysis of small amounts of Cl in SiO2 indicated that thin-sample FE-EPMA can realize a spatial resolution and MDL of 41 nm and 446 ppm at Iprob=50 nA, respectively, whereas bulk-sample FE-EPMA offers a resolution of only 348 nm and MDL of 426 ppm.