Brian D. Thoms

Production and characterization of smooth, hydrogen‐terminated diamond C(100)

We report the production of smooth and well‐ordered C(100) surfaces by exposure to a pure hydrogen plasma. A two domain 2×1 surface reconstruction is observed by low energy electron diffraction with half‐order spots visible using incident electrons with energies as low as 13 eV. High‐resolution electron energy loss spectroscopy reveals a large enhancement in specular reflectivity of low energy electrons following plasma treatment. The hydrogenated surface is stable in air and free of adsorbed hydrocarbons upon insertion into ultrahigh vacuum.

HREELS and LEED of HC(100): the 2 × 1 monohydride dimer row reconstruction

High resolution electron energy loss spectroscopy (HREELS) and low energy electron diffraction (LEED) have been used to investigate the hydrogen (deuterium) atom exposed diamond (100) surface. HREEL spectra indicate only monohydride species after ex situ exposure of the surface at 800°C to a hydrogen plasma. Likewise, no dihydride is observed after in situ exposure of the surface to hydrogen atoms produced by a hot tungsten filament. In addition, a sharp two-domain 2 × 1 reconstruction is observed by LEED for the hydrogen (deuterium) saturated surface. These data indicate that hydrogenated C(100) assumes the monohydride dimer row reconstruction. A two-domain 2 × 1 LEED pattern is also observed from C(100) following heating to approximately 1050°C and HREEL spectra show no features due to adsorbed hydrogen (deuterium). These data are consistent with the π-bonded dimer row reconstruction for bare C(100). HREEL spectra of the bare surface show peaks at 710, 1060, and 1280 cm−1.

A vibrational study of the adsorption and desorption of hydrogen on polycrystalline diamond

The adsorption and desorption of hydrogen from diamond films were studied in ultrahigh vacuum using high resolution electron energy loss spectroscopy as a probe of surface vibrations. Auger electron and energy loss spectroscopies were also used to characterize the diamond surface. The samples studied were boron‐doped polycrystalline diamond films with chiefly (111) oriented facets. We attribute the observed spectral features to a monohydride species and local sp3 bonding on the diamond surface exposed to atomic hydrogen. A significant fraction of the hydrogen desorbs from the surface between 950 and 1000 °C, although some hydrogen persists even after heating to 1050 °C.

Adsorption and abstraction of hydrogen on polycrystalline diamond

The processes of atomic hydrogen adsorption and abstraction on a diamond surface determine the fraction of sites available for reaction with carbon containing species during growth. The relative efficiencies of hydrogen atom adsorption and abstraction on a polycrystalline diamond surface were determined at surface temperatures of 80 and 600 °C using high resolution electron energy loss spectroscopy. Abstraction is seen to occur 0.05 times as efficiently as adsorption on a diamond surface at 80 °C. This is interpreted to indicate that the activation barrier to abstraction is higher than in analogous gas phase reactions. No change in either the adsorption or abstraction rate is seen for a diamond surface at 600 °C indicating that hydrogen atoms do not accommodate the surface during the reaction. Thus, abstraction proceeds via a generalized Eley–Rideal mechanism.

Hydrogen on polycrystalline diamond films: Studies of isothermal desorption and atomic deuterium abstraction

Studies of hydrogen isothermal desorption and abstraction from polycrystalline diamond surfaces are presented. The surface H and D coverages (θH and θD) are measured in real time by mass analyzing the recoiled ions generated in a time‐of‐flight scattering and recoil spectroscopy (TOF‐SARS) experiment. For surface temperatures (Ts) from 825 and 920 °C, isothermal H2 desorption is 1st order in θH with a measured activation energy, ET, of 69±6 kcal/mol and a pre‐exponential factor, ν, of 1010.5±0.9 s−1. For H2 desorption from diamond, the estimated ΔET based on bond energy calculations is ≊88 kcal/mol, substantially higher than the experimentally measured ET. This difference suggests π‐bonding of the surface after H2 desorption is involved. Using a simple bond order argument, the π‐bonding contribution is estimated to be ≊21 kcal/mol. The abstraction and replacement of absorbed H by atomic deuterium (Dat) is explained by three first‐order reactions. Under a constant Dat flux, the rate of abstraction of adsor...

Characterization of Cl2/Ar high density plasmas for semiconductor etching

Chlorine-based high density plasmas, commonly used in the etching of elemental and compound semiconductors, are characterized using mass spectrometry, optical emission spectroscopy, and electrostatic probes. Plasma fluxes are characterized by three-dimensional Langmuir probe measurements and optical emission spectroscopy. The flux is further characterized at the substrate platen by mass spectrometry to determine its makeup in terms of charged or neutral species and atomic or molecular species. Langmuir probe investigations show variations in electron temperature (2–6 eV), plasma density (1×1010 to 1×1012 cm−3), and plasma potential (5–25 V) as process conditions (microwave power, total pressure, and fraction of Cl2 in Ar) and measurement location are varied. Concurrent optical emission spectroscopy measurements of ionized species are in general agreement with Langmuir probe results. Further, optical emission spectroscopy of neutral and ionized species provides global insight into the variation of atomic/m...

Surface oxidation chemistry of β-SiC

Well-ordered, Si- and C-terminated β-SiC (100) surfaces were oxidized at low oxygen pressure and high temperature to elucidate the mechanism of oxidation. The samples were studied by high resolution electron energy loss spectroscopy (HREELS), Auger electron and electron loss spectroscopies, and low energy electron diffraction. Si-terminated surfaces exposed at Tsub=1050 °C to either O2 or hot filament-activated oxygen (O+O2*) and then cooled without oxygen were C-terminated and exhibited similar oxidation rates, surface composition, and HREELS spectra. The change from Si- to C-termination slightly shifted the surface optical phonon. Activated oxygen played no discernible role in the removal of the Si overlayer at this temperature. Samples oxidized under similar conditions but cooled under oxygen were Si-terminated, indicating that the surface lost carbon as CO and CO2 during cooldown. The conversion from Si- to C-termination during oxidation did not occur because of self-limiting etching of a Si monolayer, but rather because the oxidation products of silicon volatilized more rapidly than those of carbon. Oxidation at room temperature resulted in formation of a Si oxide layer.Well-ordered, Si- and C-terminated β-SiC (100) surfaces were oxidized at low oxygen pressure and high temperature to elucidate the mechanism of oxidation. The samples were studied by high resolution electron energy loss spectroscopy (HREELS), Auger electron and electron loss spectroscopies, and low energy electron diffraction. Si-terminated surfaces exposed at Tsub=1050 °C to either O2 or hot filament-activated oxygen (O+O2*) and then cooled without oxygen were C-terminated and exhibited similar oxidation rates, surface composition, and HREELS spectra. The change from Si- to C-termination slightly shifted the surface optical phonon. Activated oxygen played no discernible role in the removal of the Si overlayer at this temperature. Samples oxidized under similar conditions but cooled under oxygen were Si-terminated, indicating that the surface lost carbon as CO and CO2 during cooldown. The conversion from Si- to C-termination during oxidation did not occur because of self-limiting etching of a Si monolayer...

Isothermal desorption of hydrogen from polycrystalline diamond films

The kinetics of isothermal H2 desorption from polycrystalline diamond are studied in real time. The surface H coverage (θH) is measured by mass analyzing the recoiled H+ ion signal during the desorption. We find that the H2 desorption is 1st order in θH with an activation energy of 69 ± 6 kcalmol and a prefactor of 1010.5 ± 0.9 s−1. We suggest that formation of a CC π-bond on the clean surface plays a key role in H2 desorption from diamond, a view consistent with previous theoretical calculations of H2 desorption from diamond.

CHARACTERIZATION OF HIGH DENSITY CH4/H2/AR PLASMAS FOR COMPOUND SEMICONDUCTOR ETCHING

High density plasmas generated using gas mixtures of methane, hydrogen, and argon are characterized using mass spectrometry, optical emission spectroscopy, and three dimensional Langmuir probing. Such plasmas are commonly used to etch compound semiconductors. In this work we examine the chemical and electrical properties of the flux to the region where substrates are placed during processing. The dominant species in the flux are identified as H, H+, CH3, CH3+, Ar, and ArH+. Plasma parameters in the source region include electron temperatures of 4–9 eV, plasma densities of 1–5×1011 cm−3, and plasma potentials of 24–44 V as process conditions are varied. These parameters are considerably reduced in the process region of the plasma to electron temperatures of 2–6 eV, plasma densities of 1×109 to 2.5×1010 cm−3, and plasma potentials of 3–14 V. Mass and optical emission spectral data are correlated to Langmuir probe results and the effects of varying process conditions on plasma properties are presented and di...

Carrier concentration and surface electron accumulation in indium nitride layers grown by high pressure chemical vapor deposition

Electronic and structural properties of InN layer grown by high pressure chemical vapor deposition have been studied by high-resolution electron energy loss spectroscopy (HREELS) and room temperature infrared reflection measurements. HREEL spectra after atomic hydrogen cleaning exhibit N–H bending and stretching vibrations with no indications of an indium overlayer or droplet formation. Broad conduction band plasmon excitations are observed centered at 3100–4200cm−1 at various locations across the surface in HREEL spectra acquired with 25eV incident electron energy. The plasmon excitations are shifted about 300cm−1 higher in spectra acquired using 7eV electrons due to higher plasma frequency and carrier concentration at the surface than in the bulk which indicates surface electron accumulation. Infrared reflectance data acquired at various spots across the surface showed a similar variation in bulk plasma frequency. A three phase thin film reflection model fitted to the infrared data yielded carrier conce...