Hiroshi Kimura

Carbon in semi‐insulating, liquid encapsulated Czochralski GaAs

We have investigated undoped, semi‐insulating GaAs, grown by the liquid encapsulated Czochralski technique in pyrolytic BN crucibles. The concentration of carbon, the principal shallow acceptor, depends on the water content of the B2O3 encapsulant. Dry B2O3 (100–150 ppm H2O) results in carbon concentrations as high as 8×1015 cm−3, while wet B2O3 (500 ppm H2O) results in carbon concentrations below the detection limit of 3–5×1014 cm−3. In samples with carbon below 5×1014 cm−3, the compensation (acceptors minus donors shallower than EL2) is also low, and the room‐temperature carrier concentration can exceed 1011 cm−3. For samples with carbon above 5×1014 cm−3, the room‐temperature carrier concentration is below 5×107 cm−3 for all samples measured. The compensation is less than the carbon concentration in high carbon samples, indicating that donors shallower than EL2 are significant in determining the value of the compensation in these samples.

Magnetic field effects on float-zone Si crystal growth

Abstract A transverse magnetic field of 1800 G was applied to the float-zone growth of Ga-doped Si to assess its effects on the distribution of Ga in the crystal. The crystals were grown at a constant average growth rate of 4 mm/min and crystal rotation rates of 12, 6 and 0 rpm. Both application of a magnetic field and rotation of the crystal lead to reduction in the frequency of the fine non-rotational striations. The magnetic field also curtails complex structure (“feathering”) in the striations. The application of a magnetic field also reduces the amplitude of axial spreading resistance fluctuations at the center of the crystal for rotation rates of 6 and 12 rpm.

Low-dislocation indium-alloyed GaAs

Abstract We have grown crystals of In-alloyed GaAs, In x Ga 1 - x As, by the LEC method from pyrolytic BN crucibles without the intentional addition of any donor or acceptor impurities. The values of x ranged from 0.0028 to 0.023, depending on the initial melt composition and the extent of solidification of the original melt when the sample crystallized. The values of x were determined by energy-dispersive X-ray analysis, Zeeman atomic absorption analysis, and photoluminescence measurement of the bandgap shift, the last method calibrated by neutron activation analysis. In the range of In content studied, the effective distribution coefficient of In between crystal and melt is 0.118. Dislocation density, revealed by KOH etching, decreases as x increases, approaching zero in some regions of the crystal cross section. The influence of metal-As stoichiometry is similar to the case of undoped and unalloyed GaAs, and operates through the same mechanisms. With sufficient As in the melt the alloy crystals are semi-insulating, but if the As content of the melt is too low initially or drops too low during growth, the crystal is low-resistivity p-type.

Growth and characterization of GaAs alloyed with In and P

Abstract Indium- and phosphorus-alloyed GaAs crystals were grown by the liquid encapsulated Czochralski (LEC) method. Phosphorus was found to complement indium, which has relatively low concentration at the seed end, in providing the lattice hardening. the dislocation density was low even at the seed end of the crystals. Lattice-matched wafers exhibited near zero band gap shift in the crystal where the ratio of phosphorus to indium ranged from 1 to 1.3. The distribution coefficient of phosphorus was determined to be 3.0±0.3 for the two crystals grown at 2 and 4 mm/h. The crystals exhibited high room-temperature electron mobility of 000 cm2/V·s and resistivities ranging from low 107 to low 108 ω cm. Temperature-dependent Hall measurements revealed no extraneous levels attributable to indium or phosphorus in the band gap.

Rapid evaluation processes for candidate CO and HC sensor materials; examination of SnO2, Co3O4, and CuxMn3−xO4 (1 < x ≤ 1.5)

Abstract To overcome the difficulty in determining which new metal-oxide materials may be useful as carbon monoxide or hydrocarbon sensors, a rapid evaluation process using differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) has been developed. Three basic response modes, absorptive, electrochemical, and thermistor, are herein described for metal-oxide materials. Data from the evaluation experiments can provide evidence that one or more of these response modes is active for a given material. For illustration, the well-known sensor material SnO2 has been examined, along with Co3O4 and a new sensor candidate, CuxMn3−xO4 (1

The effect of isolated dislocations in substrate and device properties in low dislocation czochralski GaAs

Czochralski GaAs grown with In incorporated into the melt has large regions with fewer than 100 cm-2 dislocations. We have examined the effect of these dislocations on substrate and device properties. Infrared transmission images reveal dark filaments of high EL2 concentration a few tens of microns in diameter surrounding dislocations, Cathodo and photoluminescence images show orders of magnitude contrast in band-edge luminescence intensity near dislocations. Single dislocations appear to be surrounded by bright rings ˜200 μm in diameter in luminescence images, with dark spots 50 to 75 μm across centered on the dislocation. More complex luminescence structures with larger dark regions (˜150 μ across) and central bright spots are centered on small dislocation clusters. Differences in lifetime of photogenerated electrons or holes are the most likely cause of the luminescence contrast. Anneals typical of our post-implant processing substantially lower the luminescence contrast, suggesting the defect lowering the lifetime is removed by annealing. This may partially explain why we do not observe any effect of dislocation proximity on the properties of devices made in the material, in spite of the enormous luminescence contrast observed near dislocations.

Chapter 2 InAs-Alloyed GaAs Substrates for Direct Ion Implantation

Publisher Summary Semi-insulating GaAs substrates for direct ion implantation fabrication of field effect transistors and integrated circuits have become increasingly available in the past few years. The generally preferred process for preparing semi-insulating GaAs is liquid encapsulated Czochralski (LEC) growth without the intentional addition of impurities, and many commercial suppliers are offering substrate wafers made from LEC material. The chapter discusses an approach to better uniformity based on eliminating dislocations from GaAs substrate crystals by replacing a small fraction of the Ga atoms with the isoelectronic substituent. The chapter explores the most useful technique for reducing or eliminating dislocations in GaAs while retaining the advantages of LEC growth is the substitution of In for part of the Ga. The resulting material is formally an alloy of InAs and GaAs, In x Gal 1-x As, with x usually smaller than 0.01. It seems best to designate these dilute alloys as InAs-alloyed GaAs to emphasize that the ratio of In + Ga to As is essentially unity. On the InAs-alloyed GaAs, the chapter discusses its growth, metallurgical and crystallographic behavior, electrical and optical properties, and effectiveness as a substrate for achieving uniform device properties.

Improved Uniformity In Float Zone Si:Ga

The performance of an extrinsic silicon detector array depends on achieving uniform distribution of both the major dopant and compensating impurities in the array. While in some cases uniformity can be achieved during device processing (for example, by neutron transmutation doping to control compensation), usually the uniformity must be grown in. We report the influence of changing the pull rate and rotation rate in the float zone growth of gallium-doped silicon on the resulting distribution of gallium in the crystal. The gallium distribution was monitored by etching studies and by spreading resistance and four-point probe measurements. High rotation rates and low pull rates favor higher uniformity. These results can be understood in terms of steady-state segregation theory. We also examined the effects of prolonged diffusion on originally nonuniform Si:Ga by monitoring spreading resistance and measuring Hall effect vs. temperature. Observed improvements in uniformity were consistent with diffusion theory; 16 days of diffusion at 1300°C produced marked improvement in Ga uniformity.