W.Z. Tang

Large area high quality diamond film deposition by high power DC arc plasma jet operating at gas recycling mode

Abstract Principles of a high power DC arc plasma torch with rotating arc roots and the semi-closed gas recycling system are presented. Recent results on large area high quality free standing diamond film deposition by a high power DC arc plasma jet CVD system operating at gas recycling mode is discussed. It was shown that the uniformity of diamond film deposition over a large substrate area was closely dependent on the uniformity and symmetry of the plasma, which can be adjusted by proper manipulation of the magnetic field and the effects of fluid dynamics. Effect of gas recycling on the quality of diamond films is discussed. By proper design of the whole system and process optimization, thick uniform diamond wafers of Φ60–110 mm in diameter with a thickness up to 2 mm and transparent diamond wafers of Φ30–60 mm in diameter has been deposited successfully. It is demonstrated that gas recycling can be used even for high quality diamond film synthesis. This is of technical and economical importance in the development of high power DC arc plasma jet CVD systems for economical fabrication of high quality diamond wafers.

Adherent diamond coatings on cemented carbide substrates with different cobalt contents

Abstract A new process has been employed for preparing diamond coatings on cemented carbide substrates. Before depositing the diamond coatings, a diffusion barrier interlayer was prepared by solid diffusing boron into the substrates and this interlayer has the function of suppressing catalytic effect of Co on non-diamond formations. It was confirmed that on cemented carbide substrates with Co contents ranging from 3 to 8%, high quality adherent diamond coatings could be deposited. The formation of the boronized interlayer changed the catalytic character of the substrate surface and the interlayer functioned as a diffusion barrier layer against outward Co diffusion. The latter beneficial effect was found to be accompanied by phase transformations in the interlayer, from one composed of mainly WC, CoW2B2 and CoB phases to that of CoWB, WC and Co2B phases, during the high temperature diamond deposition processes.

A comparison in performance of diamond coated cemented carbide cutting tools with and without a boride interlayer

Instead of depleting cobalt from the surface of cemented carbide substrates or introducing a cobalt-free interlayer between the substrates and diamond coatings, a boride interlayer could be employed to modify the catalytic characteristics of cobalt in the substrate surface, preventing cobalt from interfering with the deposition of diamond coatings. This paper compares deposition of diamond coatings and performance of diamond coated cemented carbide cutting tools, with and without such a boride interlayer. We will show that a profound effect of using the interlayer is that it allows the deposition of adherent diamond coatings within a relatively broad temperature range, which improves reproducibility of the coating processes. With the interlayer present, an increase in tool life of 20-fold has been obtained, in comparison with that of uncoated tools. Such an increase in tool life can be attributed to the improved adhesion of the diamond coatings.

Accurate measurement of fracture toughness of free standing diamond films by three-point bending tests with sharp pre-cracked specimens

Abstract The three-point bending test was used for the first time for evaluating plane strain fracture toughness, KIC, of free-standing pre-cracked diamond film specimens. A novel approach to introduce sharp cracks into diamond films was presented. This approach, with a success rate of approximately 80%, allowed the initiation and growth of sharp cracks at the top of the notch in diamond films being observed and controlled in situ under a scanning electron microscope (SEM). This made a more accurate and reliable measurement of KIC in diamond films possible. Fracture toughness values obtained with the sharp pre-cracked specimens were compared with those of simply laser-notched ones. Results showed that with pre-cracked specimens, the measured fracture toughness would be approximately 15% lower than that of laser-notched specimens with notch widths of 0.1–0.12 mm. The method provided for accurate determination of plane strain fracture toughness which may help to clarify the discrepancies and to understand the mechanical properties of CVD diamond films.

Accurate measurement of strength and fracture toughness for miniature-size thick diamond-film samples by three-point bending at constant loading rate

Abstract Fracture strength and fracture toughness (K IC ) of miniature-size thick freestanding diamond film samples were accurately measured by three-point bending at a constant loading rate. A specially designed testing rig, capable of achieving a constant loading rate as small as −1 was used, which is beneficial for accurate measurement of the fracture strength and toughness for brittle materials like diamond films. A novel method was employed to nucleate and accurately control the length of the pre-crack. A specially designed loading fixture was used to ensure that the diamond film sample was tested in a near-perfect vertical position, as required by the ASTM E399 standards. Effects of sample dimensions on the measurement of fracture toughness and fracture strength are discussed; it was shown that miniature-size freestanding diamond film samples can be used for the accurate determination of fracture strength and fracture toughness. However, for accurate measurement of fracture toughness, the use of pre-cracked diamond film samples is absolutely necessary. The effect of loading configuration on the measurement of fracture strength is discussed in detail. In addition to the frequently reported two loading configurations (with the nucleation and growth surface in tension, respectively), a third loading configuration with the edge surface in tension is also discussed. The Weibel modulus was also found to be strongly dependent on the loading configuration. The significance of these findings is discussed.

Preparation and performance of diamond coatings on cemented carbide inserts with cobalt boride interlayers

Abstract A novel two-step process has been developed, with a view to improving the reproducibility of techniques for preparing diamond-coated cemented carbide inserts. Firstly, diffusing boron into WC–6% Co inserts produced an interlayer of cobalt boride, and then diamond films were deposited on this interlayer using a DC arcjet plasma system. Investigations showed that the formation of the cobalt boride interlayer favored the growth of high quality diamond films, and the interlayer functioned effectively as a diffusion barrier layer, inhibiting outward Co migrations from the insert interior to the surface. Adhesion of the boronized interlayers as well as the composite coatings was evaluated by using Rockwell A indentation and end-milling tests, respectively. It was demonstrated that both the Co–B interlayer and the diamond coating thereon were very adherent to the substrates, and a 10-fold increase in tool life could be expected with the coated inserts compared with uncoated ones.

Nanocrystalline diamond films produced by direct current arc plasma jet process

Abstract Nanocrystalline diamond films have been prepared by using a direct current arc plasma jet system. Hydrogen flow rate of the chemical vapor deposition process was decreased to induce the transition from micro- to nanocrystalline diamond films. The deposition rate was found to decrease from 20 μm/h for the microcrystalline to 8 μm/h for the nanocrystalline diamond films. It is believed that this proves that the growth of diamond films is somehow facilitated by atomic hydrogen abstractions. On the other hand, the growth of nano diamond films was found to be a quenched process, with a result of formation of colony-featured microstructures. For the nanocrystalline diamond films, the problem of surface roughness remains, not because of faceting of individual diamond grains, as in the case of microcrystalline diamond films, but because of suppressed diffusion of depositing atoms. It is shown that there exists a close relationship between surface roughness of the deposited nanocrystalline diamond films and the substrates, and an improvement in flatness of substrate surfaces could lead to a decrease in roughness of the deposited nanocrystalline diamond films.

CVD diamond nucleation enhanced by ultrasonic pretreatment using diamond and mixture of diamond and TaC powders

Abstract Effects of ultrasonic pretreatment on chemical vapor deposition (CVD) diamond nucleation on Si substrates were systematically studied. Pure 1.5–40 μm-diamond powder and mixtures of 1.5–5 μm-diamond as well as 5–20 μm-Tantalum Carbide (TaC) powder were used in ultrasonic pretreatment. The root-mean-square (Rms) surface roughness of the pretreated substrates, residual diamond and TaC powders left on the substrates were examined using atomic force microscopy (AFM), Raman spectroscopy and X-ray diffraction (XRD), respectively. It was observed that the Rms surface roughness increases with increasing diamond or TaC powder size, and there is some diamond or TaC powder left on the substrates after ultrasonic pretreatment. Diamond films were deposited using microwave plasma chemical vapor deposition (MPCVD) technique and characterized by field emission scanning electron microscopy (FE-SEM). It was found that CVD diamond nucleation density strongly depends on particle size of diamond or TaC powder used, the nucleation density increases with increasing diamond or TaC powder size. A mixture of diamond and TaC powders enhances CVD diamond nucleation much more significantly than that of pure diamond powder. A mixture of 1.5 μm-diamond and 20 μm-TaC powders has an equivalent nucleation enhancement efficiency, which could be caused by pure 40 μm-diamond powder.

Fracture behavior of thick diamond films prepared by DC arc plasma jet method

Abstract The ball-on-ring method was used to investigate the fracture behavior of thick diamond film samples prepared by high-power DC plasma jet CVD method. Both as-grown and polished samples were tested. Great efforts were made to correlate the data obtained to the microstructural features of the tested samples, particularly to the grain size and the pre-existing defects. For as-grown samples, the data were widely scattered such that it was hard to find any grain size dependence on fracture strength and toughness. Whilst for polished samples, a linear relationship of fracture strength to the inverse of the square root of the grain size (the well-known Petch-Hall equation) was found to fit the data reasonably well. This suggests that grain refinement is also an effective way for improving the mechanical properties of CVD diamond films. Pre-existing micro-cracks are also an important factor affecting the mechanical properties of thick CVD diamond films.

Influence of plasma power over growth rate and grain size during diamond deposition using DC arc plasma jet CVD

Abstract Computer-aided mathematical calculations, simulating the mole fractions of C, C2H2, C2H and C2 species, which are found present in the plasma during diamond growth, along the substrate radial length have been presented in this paper. Three different DC arc plasma jet powers, 16, 18 and 20 kW, were employed during this investigation. The purpose of this study was to investigate the effects of these plasma species on crystal size and film growth rate at different plasma powers. The simulations were coupled with experimental work, where polycrystalline diamond films were deposited onto molybdenum substrates using DC arc plasma jet chemical vapor deposition system. The computer-obtained results revealed that there is a noticeable variation in the mole fraction of each plasma species found above and along the radial distance of the substrate. Experimental results show that the average diamond grain size and the growth rate increase with input plasma power. However, the grain size and the growth rate, at any one plasma power, along the substrate radial distance remained constant.