Weight recording using a thermobalance type Cahn TG-171 has been applied to study Hot, Filament Enhanced Chemical Vapor Deposition (HFCVD) of carbon films from methane. Changes in the deposition rates during each individual process may indicate four stages of the deposition kinetics: 1- generation of nuclei (slow, linear); 2- growth on nuclei (exponential increase); 3- aggregation of crystals (slowing); 4- growth on the surface of diamond film, completely covering the substrate (linear). An attempt to determine the concentration of graphite in the deposited layer, based on differences in oxidation rates of diamond and graphite has been proposed, as supplementary to X-ray diffraction and Raman spectroscopy.

Experiments to study the temperature in the discharge produced using a dc spiral hollow cathode with CH4-H2 as the feed gas have been carried out during the rowth of diamond. Optical emission from the R branch of the 3d1 Σ v-O-2p Σ v-O rotational vibronic band are used to determine a rotational temperature. Limitations of this method are discussed.

The success of the emerging CVD diamond film industry will depend to a great extent on how well these films can be put to practical applications. Our research is presently focusing on the development of methods for the growth of large thin wall diamond shapes which can be subsequently strongly bonded to other support materials. The shapes such as cylinders, hemispheres, etc. will be designed to take advantage of diamond's superior hardness and thermal conductivity. As these diamond shapes also promise low friction and long wearing surfaces, they have potential uses as bearing surfaces as well as very scratch resistant covers for mirrors and lenses. Our primary research objectives will be concerned with: 1. The development of various apparatus to grow selected shapes of CVD diamond. 2. The development of techniques to transfer and bond the films onto materials other than the original substrates. 3. The physical and chemical characterization of the diamond films, composites, and final products.

A KrF* (248 nm) laser-based process for the deposition of carbon films containing a large fraction of diamond is reported. The experiment consists of focusing the pulsed laser beam at normal incidence onto the surface of a low temperature (20° - 150° C) silicon substrate in the presence of acetic or malonic acid vapor. The process does not require the presence of hydrogen, although an inert buffer gas, e. g. argon, is usually employed. Deposition rates of approximately 1 μm - hour−1 are obtained, and the presence of diamond in the deposited film is inferred from Raman Spectroscopy and Auger Electron Spectroscopy.

We investigated UV laser irradiation as a method to modify the surface and gas phase chemistry in a diamond growth apparatus. In particular, attempts were made to reproduce reported laser-enhanced deposition. The variables included the laser wavelength and intensity, the precursor gas (and hence the gas-phase absorption), the flow rate, and the gas inlet orientation with respect to the filament. The samples were analyzed using optical microscopy, Scanning Electron Microscopy, the Scanning Auger Microprobe, and micro-Raman scattering. In all cases, the laser radiation suppressed or had no effect on diamond deposition in comparison to the adjacent unirradiated regions. The crystals that did grow in the irradiated regions were similar in size and morphology to those from the unirradiated areas, suggesting ablation or nucleation site blockage as possible deposition suppression mechanisms. The results suggest a novel method for diamond film patterning.

We have deposited diamond-shaped particles using a non equilibrium process of laser ablation from a solid graphite target in a hydrogen discharge. The nucleation on the heated silicon surface occurs in small regions of ∼ 0.1 to 0.5 mm covering a small fraction of the surface. The faceting of the crystals observed are mainly octahedral [111] faces. The results of using an eximer laser to ablate a graphite target are described. SEM micrographs show octahedral faceting. Micro Raman spectroscopy on the crystalline features exhibit two bands at 1348 and at ∼ 1600 cm−1. The peak position and large FWHM suggests the existence of disordered sp3 bonding or short range order existing in the film. Another possibility is the existence of a stress state in the crystal due to the non equilibrium nature and the high rate of growth of the crystals. We have also irradiated HFCVD grown diamond film with a XeCl UV- eximer laser. The results indicated that sp2 bonded graphitic and amorphous component are selectively ablated, enhancing the sp3 hybrid bonds in the film.

We have used RBS/Channeling, perturbed angular correlation (PAC) and optical absorption to study the regrowth of disordered layers in diamond produced by implantation with carbon, or with carbon plus boron or indium ions. For C or C plus B implantation doses of 2 ×1015 cm−2 or less, complete recovery of channeling damage occurred after RTA at 1100°C or furnace annealing at 900°C. Optical measurements on samples implanted with high energy carbon ions show better recovery compared to the shallower implantations. PAC results showed that co-implantation with C and In caused a considerable fraction (∼15%) of the In-atoms to occupy well-defined lattice sites characterized by an electric field gradient having its major component along <111>, and a frequency of 116 MHz.

Using selective nucleation at edges or corners of μ m-size dots of Si, periodically arrayed diamond particles have been reproducibly fabricated. The particles are grown from single nuclei and have nearly-single crystal features. The properties of the diamond arrays compared with polycrystalline films have been investigated using cathodoluminescence and Schottky diodes.

The growth of diamond is now known to occur by chemical vapor deposition at low pressure even though diamond is thermodynamically unstable to graphite under these conditions. A new quasi-equilibrium theory is discussed which describes how the rapid multiple regrowth of layers with a large concentration of vacancies may stabilize nucleation and growth of diamond. Accurate theoretical calculations show that vacancies raise the energy of graphite more than diamond thus reversing the usual energy balance and leading to the growth of diamond films. One prediction of this theory is that electron rich conditions enhance both growth and nucleation. Experiments display the enhanced growth and nucleation. Raman measurements on diamond films also suggest the presence of a large concentration of defects in the resulting films.

Nuclear magnetic resonance spectroscopy has been used to study diamond thin films grown in a low-pressure hot-filament reactor from 13C-labeled acetone. Film quality is assessed, and the origin of deposited carbon determined. Carbon atoms from both bonding environments in the acetone molecule deposit to form only sp 3-bonded material. The relative rate of incorporation of these carbon atoms is determined. Experimental results and equilibrium calculations demonstrate that CO acts as a precursor for diamond growth in the absence of heterogeneous kinetics.

The concept of diamond polytypism was examined and extended beyond the previous work to gain more analytical tools with which to understand and characterize these materials. This paper reports the band positions predicted for the expected infrared active modes of several diamond polytypes and the calculated diffraction patterns for the 21R and 15R polytypes. The infrared active modes were predicted on the basis of a generalized Brlllouin zone similar to that developed for silicon carbide polytypes. The diffraction patterns for the 21R and 15R polytypes were calculated in a manner similar to that used previously for the hexagonal polytypes. These results complete the suite of basic diffraction patterns for diamond polytypes.

In this study we develop a method which utilizes the linewidth of the diamond peak at 1332 cm−1 to determine the diamond domain sizes. We have carried out calculations based on the phonon dispersion curves of diamond and the breakdown of wavevector selection rules in Raman scattering. The linewidths of a series of diamond samples were calculated, and from this theory the dimensions of the diamond regions were obtained. The calculated domain sizes ranged from 57 to 100Å and were compared to values obtained from SEM micrographs which were between 5000 and 25000Å. This is consistent with other results which have found that domain sizes calculated from this theory are in general consistent with sizes determined from electrical conductivity measurements, not from microscopy techniques. In addition, this theory allows for the calculation of the amount of frequency shift due to microcrystallinity, so that the residual stress in the film can be found.

Polycrystalline films and single crystals of diamond produced by microwaveassisted chemical vapour deposition (CVD) have been examined using absorption and cathodoluminescence spectroscopy. Measurements have been carried out on material as-grown, and after radiation damage by 2 MeV electrons.

We have investigated cathodoluminescence (CL) of nitrogen- and boron-doped CVD diamonds compared with type Ib and undoped CVD diamonds. The CL spectrum of nitrogen-doped CVD diamond shows some luminescent centers, which are not observed in undoped one. Especially 2.15 eV (575 nm) and 1.67 eV (GRl) centers are prominent. CVD diamond doped with both nitrogen and boron shows a similar CL spectrum to boron-doped CVD diamond at lower concentration of nitrogen, and shows a spectrum composed of the centers obtained by nitrogen-doping and by boron-doping at higher concentration of nitrogen. Highenergy electron irradiation to nitrogen-doped CVD diamond produces an intense luminescent center having a zero-phonon line at 3.19 eV, and the following annealing make the emission of 2.15 eV center much more intense.

Photoluminescence spectroscopy has been used to characterize polycrystalline diamond films prepared by filament assisted chemical deposition and by combustion (in an oxygenacetylene flame) techniques. The luminescence spectra of the chemical vapor deposited films are dominated by a defect band possibly associated with a neutral vacancy with a strong zero phonon line at 1.68 eV and weak phonon replicas at lower energies. The combustion films exhibit two additional luminescence bands with zero phonon lines at 1.95 and 2.16 eV. The 1.95 eV band has been tentatively assigned to a nitrogen-vacancy pair. We have also observed a strong dependence of the PL spectra radially across a given combustion film and associated this with details of the flame chemistry.

We report efficient CW optical phase conjugation and resonantly enhanced four-wave mixing spectroscopy of the N-V color center in synthetic diamond. An ultra-narrow resonance is observed in the room temperature nearly degenerate four-wave mixing spectrum upon excitation within the vibronic band at 579 nm, furnishing the triplet manifold decay rate.

Optical absorption and luminescence with zero-phonon lines at 1.401 and 1.404 eV is shown to contain fine structure induced by the isotopes of natural Ni. Uniaxial stress and magnetic field perturbations establish that the optical center has trigonal symmetry. A relationship is estimated between the concentration of Ni in these optical centers and the strength of the optical absorption.

Spatially resolved electron energy-loss spectra recorded from very small volumes of diamond containing individual dislocations show extra intensity within the band-gap just below the 1s to bulk conduction band threshold energy, when compared to spectra recorded from neighboring defect free regions. This is interpreted as direct evidence for the presence of vacant defect states associated with the dislocation structure. The contribution of the π* states from the surface layers to this region of the spectra is completely removed by calculating the difference between the spectra recorded on and off the defect. A comparison is drawn between the measured near edge structure and calculations of local density of states reported in the literature.

Intentionally contaminated DC and microwave plasma deposited diamond films were used to evaluate the characterization capability of SIMS, TXRF and RBS for heavy metal measurements. The results showed SIMS is the preferred analytical technique for heavy metal impurities in the diamond films, and TXRF for the top surface analysis of the substrate and pre-processed substrate. In addition TXRF may provide a quick, non-destructive qualitative characterization of the diamond film surface and bulk. The RBS can provide qualitative and quantitative information if the impurity levels are high enough (above 1013 atoms/cm2).

CVD diamond films prepared under varying conditions have been investigated with IR and Raman spectroscopy. Raman spectroscopy was used to characterize the crystal structure, and ir absorption was used to determine the H bonding environments. The growth temperatures were varied in one series, while the CH4/H2 ratio was varied in another. The IR absorption of all the samples showed a weak broad band extending from approximately 2800 to 2970 cm−1 which was attributed to H bonded to sp3 C sites. There was no evidence of ir absorption at 3000 to 3200 cm−1 which would indicate H bonded to sp2 or sp1 C sites. For both series, the Raman measurements showed progression from diamond-like films to diamond films. The ir absorption showed a general trend of less H with more well ordered diamond structures. No sharp transition of the H incorporation vs the growth parameters was observed.