J.A. von Windheim

Comparison of the electrical properties of simultaneously deposited homoepitaxial and polycrystalline diamond films

The electrical transport properties of simultaneously deposited, B‐doped homoepitaxial and polycrystalline diamond thin films have been evaluated by Hall‐effect and resistivity measurements over a temperature range of 80–600 K. The same films were later characterized by scanning electron microscopy, secondary‐ion‐mass spectroscopy, and an oxidation defect etch. The study involved four sets of chemical‐vapor‐deposited diamond films with individual B concentrations ranging from 1.5×1017 to 1.5×1020 cm−3. In each of the four cases the mobility of the polycrystalline film was lower than that of the homoepitaxial film by 1–2 orders of magnitude over the entire temperature range. Polycrystalline films also incorporated 2–4 times more B, had 3–5 times higher compensation ratios, and displayed activation energies that were 0.05–0.09 eV lower than in the homoepitaxial films. Hopping conduction was observed in both types of films at low temperatures, but was enhanced in polycrystalline films as evident by higher tr...

Comparison of electronic transport in boron‐doped homoepitaxial, polycrystalline, and natural single‐crystal diamond

Hall‐effect and resistivity measurements were performed on simultaneously deposited B‐doped homoepitaxial and polycrystalline diamond films, as well as a (100)‐oriented type‐IIb natural diamond crystal, over a temperature range of 140–600 K. At 298 K, the respective Hall mobilities for the homoepitaxial and polycrystalline films were 519 and 33 cm2/V s, while the active carrier concentrations were both approximately 2×1014 cm−3. For the natural diamond, a Hall mobility of 564 cm2/V s and a carrier concentration of 2×1013 cm−3 were measured at room temperature. A comparison of the transport behavior of the three specimens indicates that the electronic properties of diamond grown by chemical vapor deposition are potentially of equal or greater quality than natural diamond and that the transport properties of polycrystalline films are severely degraded by the effects of grain boundaries.

Fabrication of diamond thin-film thermistors for high-temperature applications

Abstract CVD diamond thin-film thermistors have been fabricated in various geometries and at different doping levels in an effort to achieve practical resistance- vs. -temperature characteristics. Typical activation energy values reported for polycrystalline films were combined with a targeted resistance range to plot an idealized relationship between resistance and temperature for CVD diamond thin films. The optimum device geometry and boron concentration were subsequently approximated from these ideal plots. Fabricated devices were electrically characterized in air at temperatures ranging from 25°C to over 500°C. Repeatability was demonstrated over two temperature cycles and stability was maintained at 500°C for 9 h. Effects of varying thermistor geometry and boron dopant concentration to achieve useful resistance-temperature relationships will be discussed.

Electrical characterization of semiconducting diamond thin films and single crystals

Naturally occurring semiconducting single crystal (type IIb) diamonds and boron doped polycrystalline thin films were characterized by differential capacitance-voltage and Hall effect measurements, as well as secondary ion mass spectroscopy (SIMS). Results for natural diamonds indicated that the average compensation for a type IIb diamond was >17%. Mobilities for the natural crystals varied between 130 and 564 cm2/V·s at room temperature. The uncompensated dopant concentration obtained by C-V measurements (2.8 ± 0.1 × 1016 cm−3) was consistent with the atomic B concentration measured by SIMS performed on similar samples (3.0 ± 1.5 x 1016 cm−3). Measurement of barrier heights for three different metals (platinum, gold, and aluminum) found essentially the same value of 2.3 ± 0.1 eV in each case, indicating that the Fermi level was pinned at the diamond surface. Polycrystalline semiconducting diamond thin films demonstrated a complex carrier concentration behavior as a function of dopant density. This behavior may be understood in terms of a grain boundary model previously developed for polycrystalline silicon, or by considering a combination of compensation and impurity band conduction effects. The highest mobility measured for a polycrystalline sample was 10 cm2/V·s, indicating that electrical transport in the polycrystalline material was significantly degraded relative to the single crystal samples.

High‐conductance, low‐leakage diamond Schottky diodes

Schottky diodes formed of Al, Au, and Hg on diamond have been characterized as a function of plasma treatment and thermal annealing. Plasmas formed from N2O, H2, or O2 result in high surface leakage, while plasmas formed from N2 or from CF4 with 8.5% O2 result in total leakage <1000 e/s. Annealing the diamond at 660 °C before the Schottky diode is fabricated causes an increase in the forward conductance with n‐factors approaching one. This annealing removes a compensated subsurface layer that often occurs in diamond during normal processing.

Effect of back contact impedance on frequency dependence of capacitance‐voltage measurements on metal/diamond diodes

Differential capacitance‐voltage (C‐V) measurements were performed on Al and Pt rectifying contacts fabricated on natural (type IIb) diamonds. The C‐V data showed frequency dependence, which decreased significantly after reducing the back contact impedance. The frequency dependence seems primarily to be an effect of the contact capacitance, contact resistance, and bulk resistance of diamond. A model which includes these variables has been proposed to explain this frequency dependence using both large and small back contact impedances.

Comparison of the electric properties of single-crystal and polycrystalline diamond by hall effect and capacitance-voltage measurements

Abstract The transport properties of type IIb (naturally B-doped), single-crystal diamond were investigated by differential capacitance-voltage ( C - V ) measurements, Hall effect measurements and resistivity measurements. The results for the Hall effect and resistivity measurements on single-crystal samples were compared with similar measurements on polycrystalline samples grown by microwave-assisted chemical vapor deposition. The C - V and Hall effect measurements on single crystal, type IIb diamond samples showed consistent results. Temperature-dependent resistivity measurements verified an activation energy of 0.36 eV, while room temperature Hall effect measurements on a (100)-oriented, type IIb single crystal indicated that the active carrier concentration was 2.1 × 10 13 cm −3 . This result was consistent with total B concentrations measured by both C - V and secondary ion mass spectroscopy (SIMS). The transport properties for polycrystalline samples were not as good as those for single-crystal samples. At room temperature, the mobilities were 3 and 10 cm 2 V −1 s −1 for two in situ doped polycrystalline thin films, as compared with 325 cm 2 V −1 s −1 for the single-crystal sample. For one in situ doped sample, the activation energy was measured to be 0.05 eV and the room temperature carrier concentration was 1.8 × 10 16 cm −3 , while SIMS indicated that the total atomic B concentration was 5 × 10 18 cm −3 . The relatively low carrier concentrations measured in the polycrystalline samples may be indicative of a high level of compensation (about 65%) or trapping of charge at the grain boundaries, leading to a depletion of carriers in the crystallites.

Combustion growth of large diamond crystals

This paper reports the successful growth of optically transparent, individual diamond crystals up to millimeter diameters on silicon substrates by oxygen-acetylene combustion flames at atmospheric pressure. The growth process consisted of three steps: (i) achieve a suitable nucleation density by pretreating the as-received Si substrate in an acetylene-rich flame (oxygen-to-acetylene ratio Rf = 0.95) for about 30 min at a downstream position (7–10 mm away from the tip of the flame inner cone); (ii) grow crystals up to ∼ 200 μm in diameter in an annular area on the substrate at Rf = 0.98 and a substrate-to-tip of the flame inner cone distance of 2 mm; (iii) move the preferred crystals from the annular region into either the central core region of the flame feather or near the edge of the flame feather for further growth up to millimeter diameters under carefully controlled conditions. The final step of moving the crystals into a different growth region was necessary to avoid extensive secondary nucleation and structural defects. The key factor for diamond crystals to grow up to millimeter diameters was to maintain the growth conditions at the growing surface constant throughout the process. It was found that the crystal surface temperature, which was the most sensitive and also one of the most critical parameters, could be effectively maintained constant by decreasing the total gas flow rate as growth continued. Both the crystal growth orientations and the amount of nitrogen impurity incorporated in the diamond lattice were closely related to the crystal surface temperature. It is believed that the gas flow dynamics, or more specifically, the boundary layer thickness, played an important role in the growth and morphological development of large diamond crystals.

Hall effect measurements on CdTe electrodeposited from tri-n-butylphosphine telluride

Abstract The carrier transport properties of CdTe electrodeposited from tri- n -butylphosphine telluride have been studied by resistivity and Hall effect measurements as a function of temperature. Argon annealed electrodeposited CdTe was found to be consistently p -type, with resistivity values typically 10 6 –10 7 Ω·cm. Various donor and acceptor dopants (Cd, Te, Cu, Ag, In and O 2 ) were incorporated into polycrystalline CdTe films by three methods: electrochemical co-deposition, electromigration and vapour techniques. The carrier concentration of p -type CdTe could be systematically increased by the increasing the current density for the electromigration of copper. The increase in carrier concentration was accompanied by a decrease in resistivity and a decrease in mobility. P → n conversion of CdTe was achieved by diffusion of Cd at high temperature. The activity of the dopant was dependent on the method that was used for incorporation. The effect of dopant density on the resistivity of the polycrystalline cadmium telluride films can consistently be described by the grain boundary model. In this model charging of grain boundary states decreases the carrier density and results in a barrier. According to the model, dopants accumulated at grain boundaries do not generate carriers and do not affect the density of interface states. Comparison of theory with results indicated that a significant portion of copper dopant was not active. For cadmium dopant, the barrier energy varied from 0.04 to 0.09 eV for crystallite sizes from 14 to 77 nm.

Photoelectrochemical Deposition of Cadmium Telluride Using Tri‐N‐Butylphosphine Telluride

Thin film cadmium telluride has been deposited cathodically on titanium in a photoelectrochemical cell (PEC) using a propylene carbonate solution of Cd(II) and tri-n-butylphosphine telluride (BPT) at 100/sup 0/C. Illumination of the cathode enhances the cathodic current relative to that observed during the dark process. Furthermore, the current under illumination decreases more slowly than it does for the dark process, providing thicker films in a shorter time. Under illumination, the Te/Cd ratio of the film is independent of light intensity and applied potential in the range, -0.8 to -1.4V (Ag/AgCl). This ratio and the deposition current depend on the concentration of cadmium ion relative to that for BPT. Thicknesses of ca. 0.5 ..mu..m are obtained in 15 min for films that have a Te/Cd ratio of ca. 1.0. Cyclic voltammetry (CV) indicates that Cd (II) and BPT form a complex that is involved in the electron transfer process. A mechanism consistent with these results is proposed.