G. C. Hudson

Direct deposition of polycrystalline diamond films on Si(100) without surface pretreatment

Dense nucleation of small‐grain polycrystalline diamond films on Si(100) substrates has been accomplished without the use of any surface pretreatment such as abrasive diamond scratching, surface oil treatments, or diamond‐like carbon predeposition. Diamond depositions occurred in a low‐pressure rf plasma‐assisted chemical vapor deposition system using mixtures of CF4 and H2. Films deposited at 5 Torr and 850 °C on as‐received silicon wafers show dense nucleation, well‐defined facets, and crystallites which ranged in size from 500 to 10 000 A. X‐ray photoelectron spectroscopy and electron energy loss show the films to be diamond with no major impurity and no detectable graphitic component. Raman spectroscopy shows a pronounced 1332 cm−1 line accompanied with a broad band centered about 1500 cm−1.

The role of atomic hydrogen and its influence on the enhancement of secondary electron emission from C(001) surfaces

The role of chemisorbed hydrogen in the enhancement of low-energy electron emission from natural type IIb C(001) diamond surfaces has been investigated. A hydrogen induced low-energy emission peak, whose intensity was found to be a linear function of surface coverage, was observed. The direct observation of emission from vacuum level states in the photoemission spectra has determined a negative electron affinity of ∼0.4 eV for the hydrogenated C(001)-1×1 surface. Constant initial states photoemission has unambiguously identified the electron emission process with the escape of electrons from bulk electron states at the conduction-band minimum.

SECONDARY ELECTRON EMISSION ENHANCEMENT AND DEFECT CONTRAST FROM DIAMOND FOLLOWING EXPOSURE TO ATOMIC HYDROGEN

Polished nominal (100) surfaces of four types of diamonds were exposed to atomic hydrogen by hot filament cracking of H2 gas or by immersion in a H2 plasma discharge. Both types IIa and IIb (100) diamond surfaces exhibited the following characteristic changes: (a) secondary electron (SE) yield increased by a factor of ∼30 as measured in a scanning electron microscope (SEM), (b) near‐surface, nontopographical defects were observable directly using the conventional SE mode of the SEM, (c) surface conductance increased by up to 10 orders of magnitude. These changes were observed only weakly in nitrogen‐containing types Ia and Ib diamonds.

Method of fabricating a free‐standing diamond single crystal using growth from the vapor phase

By combining a low temperature (600 °C) chemical vapor deposition process for homoepitaxial diamond and conventional ion implantation, we have made and lifted off a synthetic diamond single crystal plate. Before growth, a type Ia C(100) crystal was exposed to a self implant of 190 keV energy and dose of 1×1016 cm−2. Homoepitaxial diamond growth conditions were used that are based on water‐alcohol source chemistries. To achieve layer separation (‘‘lift‐off’’), samples were annealed to a temperature sufficient to graphitize the buried implant‐damaged region. Contactless electrochemical etching was found to remove the graphite, and a transparent synthetic (100) single crystal diamond plate of 17.5 μm thickness was lifted off. This free‐standing diamond single crystal plate was characterized and found to be comparable to homoepitaxial films grown on unimplanted single crystal diamond.

Demonstration of a method to fabricate a large-area diamond single crystal

A multi-step process to fabricate a diamond single crystal that is larger than the original, natural, commercially-obtained crystals is described. Starting with 3.0 mm × 3.0 mm × 0.25 mm, natural, type Ia C(100) crystals that have had their edges oriented to (010) and (001), we have successfully bonded two to a Si substrate in close proximity to each other. Subsequent diamond homoepitaxy using plasma-enhanced chemical vapor deposition of up to

Influence Of Surface Terminating Species On Electron Emission From Diamond Surfaces

75 μm thickness has enabled epitaxial overgrowth to join the two diamonds. The topography was excellent, and microRaman spectroscopy indicated only a 0.6 cm−1 line broadening (crystal degradation) at the joint. The creation of etch pits (via oxidizing flame) on the joined diamond surface indicated a higher defect density at the joint, but this more-defective region was constrained to within the dimensions of the original gap between the diamond crystals. These results indicate that this process of epitaxial joining of diamond single crystals has the potential to be scaled up to larger area in order to fabricate a diamond single crystal of desired area and reasonable crystal perfection.

Etch-delineation of defects in diamond by exposure to an oxidizing flame

Changes in electron affinity on the C(001) surface of type Ifb diamonds have been studied using a variety of surface analytical techniques, including ultraviolet photoemission spectroscopy, secondary electron emission spectroscopy and constant initial states photoemission. Following H-plasma exposure, an intense low-energy emission peak was observed with all spectroscopies. The emission intensity associated with the chemisorbed hydrogen was found to be a linear function of surface hydrogen coverage. The proposed mechanism for the hydrogen induced changes in electron affinity is the creation of a dipole on the surface by the addition of hydrogen which opposes the surface potential of the bare surface. A total change in electron affinity of 2.2 eV was measured upon hydrogen termination of the clean 2x1 surface. Constant initial states photoemission demonstrates that the intense low-energy electron emission observed arises from electrons emitted from bulk states at the conduction band edge. Oxygen, as an electronegative species, was found to have the opposite effect and the electron affinity was increased by ∼3.7 eV upon oxygen termination relative to the clean 2x I surface.

Enhancement of Diamond Nucleation by Graphite Fibers Local to Substrate Surfaces in H2 - CH4 rf Discharges.

An experimental study of the etching properties of defects in diamond using propane flame exposure in air is presented. Both natural diamond crystals and polycrystalline diamond films were exposed to a flame for an optimum time of 3 ‐ 4 s. This process topographically delineates defects in diamond via an accelerated etch rate at defect sites. Using transmission electron microscopy (TEM) to determine the exact nature and density of defects present in the diamond, we have found a direct correlation between topographical delineation observed by scanning electron microscopy (SEM) and the defect structure observed by TEM.

Dense Nucleation of Polycrystalline Diamond Films Using CF4-H2 Low Pressure rf Discharges

Abstract A method has been discovered for enhancing diamond nucleation without using any mechanical treatment to the surface. We have observed that the presence of graphite fibers tangential to a substrate surface greatly enhances the nucleation of diamond crystals immediately underneath the fiber. Diamond growth has been observed along lines and in small clusters replicating the graphite fiber pattern following exposures of unscratched silicon, nickel, and fused quartz substrates to 2% CH 4 in H 2 rf-discharges.

Fabrication of a Free-Standing, Synthetic, Single Crystal Diamond Plate Using Ion Implantation and Plasma-Enhanced Chemical Vapor Deposition

Abstract Dense nucleation of small-grain polycrystalline diamond films on Si(100) substrates has been accomplished without the use of any surface pretreatment such as abrasive diamond scratching, surface oil treatments, or diamond-like carbon predeposition. The depositions occurred in a low pressure rf plasma assisted chemical vapor deposition system using mixtures of CF 4 and H 2 . Films deposited at 5 Torr and 850°C on as-received silicon wafers show dense nucleation, well-defined facets, and crystallites which ranged in size from 500 to 10,000 a. X-ray photoelectron spectroscopy and electron energy loss show the films to be diamond with no major impurity and no detectable graphitic component. Raman spectroscopy shows a pronounced 1332 cm- 1 line accompanied with a broad band centered about 1500 cm- 1 . Preliminary data indicates that this process is applicable to substrates other than silicon. This process will have important applications in areas where surface pre-treatments, such as diamond polishing, are not viable.