L. Diederich

Field emission from diamond, diamond-like and nanostructured carbon films

Abstract We have deposited nanotube films on silicon via a chemical vapor deposition (CVD) growth process known from the deposition of diamond. We used a metallic catalyst which was deposited onto the silicon surface prior to the CVD deposition. The films are very pure, adhere well and are very well suited for electron field emission. We measured emission at 2.6 V/μm (for 1 nA emission current) and an emission site density reaching 104/cm2 at 3–4 V/μm as measured on a phosphor screen. Electrons originate at the Fermi level and the high local fields at the emission site is produced by the geometry of the nanotube. The results obtained on these films are comparable to those from differently prepared CVD diamond films. So far, we have no evidence that electron injection occurs. The emission process is governed by field amplification at protrusions and tips. In a second experiment we have measured emission from a metallic micrometer sized grain fixed on a diamond (100) surface, with different surface termination (hydrogen, oxygen, sp2 carbon). The field emitted electron energy distribution (FEED) spectra show large energy shifts which are due to the surface resistivity and not due to injection of electrons in the conduction band. Hence, energy shifts in FEED spectra do not necessarily reflect an injection mechanism.

Photoemission from the negative electron affinity (100) natural hydrogen terminated diamond surface

Abstract We reported recently about the cleaning and polishing of natural doped (100) and (111) diamond surfaces in a hydrogen plasma at 870°C and 40 mbar [Kuttel et al., Surf. Sci. 337 (1995) L812]. This smooth (100) surface shows a sharp negative electron affinity (NEA) peak for the 2 × 1 monohydride terminated surface upon annealing to 300°C, experimentally observed by ultraviolet photoemission spectroscopy (UPS). The effect of annealing the crystal up to 1000°C in an UHV environment ( p −9 mbar) results in a positive electron affinity (PEA) whereas a subsequent atomic hydrogen adsorption from a heated filament leads to an 1 × 1 reconstruction and to the reappearance of the NEA peak with a lower intensity than the plasma exposed surface. Upon annealing the surface up to 1000°C at a pressure of 5 × 10 −8 mbar, a NEA is observed which probably is caused by remaining hydrogen at the surface. The hydride terminated surface seems to be responsible for the NEA property. The link between the NEA property and the band bending of the (100) surface are elucidated by photoemission spectroscopy of the core level and the valence band and we measured the spatial distribution of the NEA peak for the plasma exposed surface by angle resolved UPS. Finally, we show photoelectron current measurements and the influence of the bias applied to the surface.

The preparation and characterization of low surface roughness (111) and (100) natural diamonds by hydrogen plasma

Abstract Natural doped (100) and (111) diamond surfaces were polished in a hydrogen plasma at 870°C and 40 mbar and analyzed by AFM, LEED and X-ray photoelectron diffraction (XPD). The initial surface roughness of 7 nm (rms) is decreased to 1 nm on the (111) surface and a 2 × 1 LEED pattern is observed after annealing at 1000°. After etching the (100) surface has a roughness of 0.8 nm and shows a sharp 2 × 1 LEED pattern which is even stable in air without annealing. XPD measurements indicate that the quality of the top surface layer (30 A) is increased considerably by the etching and annealing process. Exposing the surfaces to atomic hydrogen produced by a heated filament leads to an increase of the surface roughness comparable to what was observed on the as received sample. The often reported 1 × 1 reconstruction seems to be a consequence of a large surface roughness and is absent on smooth surfaces.

Photoelectron emission from the negative electron affinity caesiated natural diamond (100) surface

Abstract The unique property of negative electron affinity (NEA) makes diamond surfaces very interesting for technical applications like cold cathodes and and field emitters. We investigated the hydrogen-terminated and hydrogen-free, caesiated diamond (100) surfaces by means of low-energy electron diffraction, X-ray photoelectron spectroscopy and ultraviolet photoelectron spectroscopy. The photoemission measurements permit us to determine the electron affinity and to draw the band diagrams for these differently terminated surfaces. The NEA peak reappears after caesiation of the hydrogen-free surface, whereas it vanishes after exposure to molecular oxygen. From the shift of the C ls core level, an increasing downward band-bending of 1.2 eV was inferred following caesiation of the hydrogen-free surface and 0.6 eV following caesiation of the hydrogen-terminated surface.

Electron field emission from a cesiated NEA diamond (100) surface: an activation concept

Abstract Electron field emission (FE) was measured from a natural doped diamond (100) surface prepared to show a negative electron affinity (NEA). While the bare surface showed no FE, emission was observed from the cesiated surface after an activation phase. A model for FE from a cesiated surface is presented. The assumption of activation also proved to be a valuable concept when looking at FE from CVD films. The nature of this activation may consist of an increase in the emitting surface area, which would account for the higher FE current by a breakdown between anode and cathode. At the same time morphological changes were observed by AFM, perhaps showing amorphization.

NEA peak of the differently terminated and oriented diamond surfaces

The negative electron affinity (NEA) peak of differently terminated and oriented diamond surfaces is investigated by means of ultraviolet photoemission spectroscopy. Electron emission measurements in the range below the conduction band minimum (CBM) up to the vacuum level Evac permit the quantitative calculation of the upper limit of the NEA value. The inelastic scattering at the surface to the vacuum interface and the emission of electrons from the unoccupied surface states, situated in the band gap, are the mechanisms responsible for explaining this below CBM emission. All the H-terminated diamond surfaces present NEA. However, the characteristic NEA peak observed in the spectra is only detected for the (100)-(2×1):H surface and to a lesser extent for the (110)-(1×1):H surface, while it is absent for the (111)-(1×1):H surface because of the k||-conservation in the photoemission process.

Surface-state dispersion of hydrogenated and hydrogen-free diamond (100) surfaces determined by angle-resolved photoemission

Abstract We have investigated the smooth monohydride and the hydrogen-free diamond (100)−(2 × 1) surfaces by means of angle-resolved ultraviolet photoelectron spectroscopy along the main symmetry axes of the surface Brillouin zone. We discriminate the two differently terminated diamond (100)−(2 × 1) surfaces by their surface-state band dispersion. Furthermore, our experimental results are in good agreement with recent theoretical calculations [J. Furthmuller, J. Hafner, G. Kresse, Phys. Rev. B 53 (1996) 7334].