Science

December 1990

by Ruth Levy Guyer and Daniel E. Koshland, Jr.

Diamond: Glittering Prize for Materials Science

Its combination of properties, like its appearance, is absolutely dazzling. Diamond is the hardest substance known. It is inert to chemical corrosion and can withstand compressive forces and radiation. It conducts heat better than any other material, has extremely high electrical resistance, and is transparent to visible light, x-rays, ultraviolet radiation, and much of the infrared spectrum. And, with respect to most of these features, diamond is superior to all other known materials.

Because of these outstanding properties, synthetic diamond materials -- both crystals and thin films -- that could be made cheaply would have great potential research and commercial applications. Can they be produced? 1990 saw the start of the era in which this possibility could become a reality.

Diamonds in the rough. Before it was feasible to make synthetic diamonds, diamonds could only be obtained through mining, and never has there been what could be called a ready supply. Today, less than 20 tons of natural diamonds are mined each year throughout the world. Brazil, India, and South Africa have been, at different times, the world's major diamond-mining nations, though diamonds are also found in many other countries; today the Kaapvaal craton of southern Africa is one of the world's most productive diamond-mining centers.

Natural diamonds form in the earth's mantle in regions of high temperature and high pressure. Volcanic eruptions that originate from such regions bring diamonds to upper portions of the earth's crust in rocks known as kimberlites. Diamonds are mined from the conduits of the volcanoes and from nearby placer deposits in stream beds and beaches.

The switch to synthetic. As new technologies have been developed for the production of artificial diamonds, the quest for diamonds has shifted more and more from the mine to the laboratory. The number of potential uses for diamond-based materials and the enormous profits anticipated have engendered an international race for high-quality artificial diamond production. Technologic breakthroughs for growing diamond production. Technologic breakthroughs for growing diamond materials and diamond films have come fast and furious in 1990. In addition, use of purer starting materials made possible the production of isotopically pure diamond films that have properties superior even to those of natural diamonds: the most exceptional of these is the exceptional of these is the extraordinary ability of the pure films to conduct heat. Although the cost of making synthetic diamond films with state-of-the-art chemical vapor deposition (CVD) methods is still high--estimated at around $100 per carat--the price could drop significantly with the optimization of CVD technology.

Applications. A few diamond-based and diamond-coated products are already in use commercially--x-ray windows in electron microscopes, strong abrasion-resistant industrial tools, and diaphragms for tweeters in stereo speakers--but these represent only a tiny fraction of the anticipated applications. For hard-to-service, hard-to-reach environments where high pressures and temperatures, intense radiation, high salt content, and other adverse conditions can destroy materials (places like the ocean, space, engines, and nuclear reactors), fabrication of diamond materials and devices may be justified already, even at the currently high costs of production.

In both adverse and more standard settings, diamond substrates for semiconductors will be able to efficiently transport heat from electronic circuitry, obviating the need for cumbersome cooling systems. Because CVD diamond films have both high thermal conductivity and high electrical resistivity, the jewel in the crown of diamond film technology may well be superfast integrated diamond circuits. Diamond diodes (the building block of transistors, which, in turn, are the building blocks of integrated circuits) have recently been made. If successful doping of diamond can be accomplished routinely, diamond devices could someday replace silicon semiconductors. Whereas silicon chips can withstand temperatures up to 300 °C, one estimate is that diamond chips might be able to withstand temperatures as high as 5000 °C.

Doped single-crystal diamond films are needed for diamond semiconductors; for other applications polycrystalline diamond films are adequate. For example, abrasion-resistant tools are coated with this type of film. Industry faces a different sort of challenge with regard to these tools, namely determining what would be an equitable price to charge for saws and knives that never need sharpening or replacement.

Diamond thin films can be put on windows and lenses to make them scratch-proof, nonreflecting, and permeable to light. Because diamond films are wear-resistant, they might be fashioned into efficient, low-friction, unlubricated bearings for machinery and prosthetic devices. A megaproject that may be in the offing is the production of diamond films for use as high-speed detectors for the superconducting supercollider; it is predicted to involve more than a million carats of diamond film.

In addition to the production of diamond films and coatings, free-standing diamond materials are being fabricated. Diamond nozzles have been cast for use in diesel engines, and diamond sheets, domes, and tubes have been prepared on metal preforms. Because the template is etched away, full advantage can be taken of diamond's extreme properties in the free standing constructs.

The technological spadework. Interest in the production of artificial diamonds was expressed at the turn of the century, but it was not until 1958 that a method was patented in the United States for preparing diamond materials from methane at high pressures and high temperatures (1600 K and about 55 kilobars). However, as the methane burned, graphite was also deposited, severely limiting the speed of diamond deposition and therefore the success of the process. (Both diamond and graphite are pure carbon materials, but the way that carbon atoms are organized in them differs: diamond is a rigid, dense, and essentially incompressible crystal in which tetrahedrally coordinated carbon atoms are linked in a cubic crystal lattice by covalent bonds; graphite is a soft material in which two types of bonds form to create a macrostructure of parallel sheets with hexagonal symmetry.)

In 1977, researchers in the Soviet Union found that deposition of the troublesome graphite could be prevented if excess atomic hydrogen were added to the reaction chamber. Hydrogen may both suppress formation of graphite nuclei and contribute to the creation of free radical sites. By 1981 the Russian scientists reported that they were able to form both single-crystal diamond films on diamond substrates and multiple diamond crystals on metal substrates.

The film industry runs fast-forward. With a solution to the graphite problem at hand, Japanese and other researchers began in the early 1980s to develop low-pressure CVD methods; these methods yielded high-quality single-crystal and polycrystalline films. With CVD, hydrogen gas is heated with a simple hydrocarbon compound such as methane (referred to in some of the popular accounts of the achievement as swamp gas, vodka, and sake) to temperatures of 2200 °C. The carbon atoms are atomized and ionized and then rearrange and condensed out onto the substrate. Diamond films made with CVD methods have proven to be both smoother and larger than those that could be made under high-pressure and high-temperature conditions.

In July of this year, scientists in the United States reported that isotopically pure diamond films (containing 99.9% carbon-12 and not the 1% carbon-13 that is present in natural diamonds) had been grown. The pure films not only conducted hear 50% better than the best natural diamonds but also withstood damage by laser radiation ten times more effectively than natural diamond.

Vapor deposition methodology now appears to be in an exponential phase of growth. Diamond films can be grown at pressures ranging from tens of torrs to 1 atmosphere. Film growth rates of 1 millimeter per hour are possible. Diverse volatilization methods have become available, including microwave discharges, hot filaments, plasma torches, and ion beams. The deposition of films at lower-than-normal temperatures (around 300 °C instead of the standard 700 to 1100 °C) has been accomplished through the addition of halogens to reaction mixtures; this is an important step if diamond is to be deposited on temperature-sensitive substrates. All of these variations on the basic CVD theme are making possible faster production of better materials with diverse morphologies.

For some purposes, diamond-like carbon films (which contain less than 1% hydrogen) or diamond-like hydrocarbon films (those with 20 to 60% hydrogen) may be as good or better than diamond films. In general, these films can be deposited at lower temperatures than can pure diamond films. The development of similar materials, such as the boron nitrides, may also benefit from the diamond technology boom.

The many facets of diamond film technology. Much has been accomplished with CVD technology even in the absence of a clear understanding of how and why this process transmutes hydrocarbon into diamond. Experimental and theoretical approaches are now coming together to provide an understanding of the kinds of intermediates that form during the deposition process, the specific molecular species that promote the growth of the crystal lattice, the types of atoms that bind to the crystal's edges, the impurities (nitrogen, boron, and others) that disrupt the formation of diamond crystals, the ways in which reaction conditions affect the speed of film deposition and film thickness and shape, and the means by which specific properties of films enhance their usefulness for various applications. Most materials--metals, ceramics, plastics, polymers, and paper--are considered desirable substrates for some application of diamond films, and so an understanding of what promotes bonding is critical. For some applications, epitaxial growth is required, whereas in other cases the substrate serves only as a form on which a film is fabricated.

It has taken a quarter of a century for artificial diamond film production to get under way, but, now that it has, the future for diamond-based materials is likely to be a gem.

3