ÐÓ°ÉÔ­´´

The top of the Mohs scale of hardness, used by engineers, is defined by diamond, with a rating of 10. (Talcum powder, at the bottom of the scale, scores 1.) But materials scientists now believe that it should be possible to make substances that will extend the scale. Such ‘superhard’ materials could replace rare and expensive diamonds in applications for which they are presently indispensable.

A material’s hardness is a measure of its resistance to both elastic and plastic deformation: it makes something more difficult to crush, stretch, scratch or dent. As it depends on a combination of properties, hardness is difficult to quantify precisely; engineers use the Mohs scale as a rule of thumb.

It is not only diamond’s hardness that has made it the material of choice for so many technological applications such as cutting, abrading and polishing: its high thermal conductivity at room temperature also dissipates heat from the working surfaces. Furthermore, diamond is transparent to visible and infrared light, can withstand ionising radiation and resist chemical attack from most acids and oxidising agents. But at 700 °C it burns in air to form carbon dioxide, ruling it out for cutting steel.

For years the search has been for a material to improve on that. A harder substance with greater resistance to oxidation would be useful in chemically hostile environments. It might also withstand higher temperatures than diamond, allowing it to be used as a conductor or semiconductor to replace silicon in high-temperature electronics. Something with good thermal conductivity would be even better, as it would dissipate heat more effectively. Superhardness would come to the fore where materials are needed to replace conventional lubricants and even diamond films cannot be used – such as in high-speed cutting tools, food processing equipment and magnetic storage devices such as computer discs.

Oiling the wheels of progress

Lubricants such as oil cannot be used on engine parts operating at the very high temperatures and pressures found in environments such as rocket fuel pumps. But a smooth, thin coating of superhard material could ideally reduce friction between moving parts. The surface atoms of a hard material are tightly bound to the rest of the material, reducing their tendency to stick momentarily, causing friction, to other materials they touch.

Similarly, superhard materials made of light elements could be very lightweight, and replace steel in components, such as ball bearings, used in high-speed machinery. Rotational forces often tear bearings apart, so the lighter the material the more durable the component should be. All sorts of manufacturers could benefit, beginning with the aerospace sector but eventually reaching more mundane machinery such as photocopiers.

ÐÓ°ÉÔ­´´s have turned to theory to help them design superhard materials. Hardness can be expressed more quantitatively than in Moh’s scale with the bulk modulus, K, which measures the proportion by which a material shrinks under uniform pressure from all directions. (The exact definition means K has units of pressure.) On the atomic scale, a material’s hardness is determined by the packing density of the constituent atoms and the strength of the chemical bonds between them: the denser and stronger, the better. Microscopic faults in the crystal reduce the bulk modulus, because the mean applied pressure is unevenly distributed. On the macroscopic scale, K is also affected by porosity and crystal grain size.

An idealised diamond consists of an infinite array of carbon atoms, each covalently bonded to four neighbours arranged at the corners of a tetrahedron around it. As covalent bonds are highly resistant to deformation, the crystal lattice is very rigid – helping give diamond its hardness.

From experimental measurements, Marvin Cohen and Amy Liu of the University of California at Berkeley derived a scaling equation for calculating K values in solids with tetrahedral crystals such as diamond. It says that K = (1971-220)d ^-3.5 where the value of (which can vary from 0 to 2) indicates how covalent the bonds are, and d is the bond length in angstroms (1 angstrom equals 10 ^-10 metres). For diamond, d is roughly 1.53, is 0 and K is 443 gigapascals.

Ralf Riedel and his research team at the Max Planck Institute for Metallurgy in Stuttgart have concentrated on synthesising novel ceramic materials containing boron, carbon and nitrogen. The boron carbonitrides are not expected to be superhard but might have other interesting electronic properties. The related boron nitrides, which lack the carbon atoms, are the most promising materials. In the cubic form of boron nitride, the units comprising the crystal lattice are arranged with their axes at right angles to each other. However, Cohen and Liu dashed early hopes for this crystal by calculating (on the basis of their equation) that its bulk modulus should be only 367 gigapascals, less than that of diamond. Researchers later established the modulus experimentally as 369 gigapascals. Still, while falling short on hardness, it does share many of diamond’s other properties such as high thermal conductivity and stability.

Clearly, to achieve the glittering prize of superhardness, a material must have very short, very covalent bonds. Liu and Cohen calculated that the then theoretical compound of carbon and nitrogen with the formula C₃N₄ should fit the bill: the crystal form known as -C₃N₄ was their target. Extrapolating from measurements on a known silicon compound, -Si₃N₄, they reckoned that in the carbon nitride would be 0.5 and d would be between 1.47 and 1.49. From this they calculated it would have a bulk modulus of between 461 and 483 gigapascals. To support their result the researchers calculated the bulk modulus again, this time working from first principles and helped in their calculations by a supercomputer. They obtained a value of 427 gigapascals. So theoretically, at least, -C₃N₄ should have a compressibility comparable to diamond’s.

However, synthesising the compound proved difficult. In 1990 Michael Wixom of KMS Fusion, a company based in Ann Arbor, Michigan, reported that he had prepared a solid containing carbon and nitrogen from a heat-treated resin of melamine and formaldehyde resin, using shock compression. This process applies sudden, explosive high temperature and pressure, inducing an ‘instantaneous’ crystal formation. However, Wixom could not find any tetrahedrally bonded -C₃N₄.

Last year, Yip-Wah Chung and his research team at Northwestern University in Evanston, Illinois, appeared to make more progress. Using a technique that is relatively economical (compared to those of Wixom) known as direct current magnetron sputtering, Chung and his colleagues prepared thin films, just tens of microns thick and containing only carbon and nitrogen atoms. The team grew a series of films under different conditions on various materials, including polycrystalline zirconium, sodium chloride and special glasses.

For the hardness tests, films were deposited on silicon, and a diamond point (known as a ‘nanoindentor’) used to try to poke holes in the material’s surface. Usually the diamond tip makes a dent. But in Chung’s test, it left no impression. There are two possible reasons: either the new film was harder than diamond, or it was superelastic, with an exceptionally high yield strength, and recovered its shape after being poked with the diamond indentor. Chung’s team also measured the material’s friction against steel, and obtained values close to those for diamond.

One problem Chung faces in verifying whether his material is superhard is that of producing a single-phase layer consisting only of crystalline -C₃N₄. The material produced so far is an amorphous mixture containing only a few per cent of crystalline material. Chung says he would expect the amorphous matrix to be softer than the crystalline phase; ‘however, having superhard particles in a softer matrix should give the coating enormous strength’, he adds. That could be just as useful as having a pure superhard material.

Searching for a breakthrough

Chung is reluctant to predict when a genuine superhard material will be found though his research team is forming an industrial consortium to look at commercial applications.

But some researchers, such as Steve Bull of the materials development division at Harwell Atomic Energy Authority, doubt whether superhard materials will ever be commercially viable. He sees ‘more mileage’ in investigating ways of processing diamond and cubic boron nitride.

John Evetts and Rob Somech of the materials science department at the University of Cambridge are less sceptical. They hope to take up the superhard pursuit, and are currently seeking financial support for their proposals. Evetts believes the work is ‘high risk’, but says that -C₃N₄ seems very interesting.

A number of researchers, among them Riedel and Chung, think boron-based compounds could also be worth looking at, as elemental boron and diamond have very similar molar volumes – which can equate with hardness. (Boron’s is 4.6 cubic centimetres, against diamond’s 3.4 cubic centimetres. They are investigating various combinations of carbon, boron and nitrogen, as well as the incorporation of oxygen atoms and metal components to form ceramic alloys.

Despite all this activity, carbon, the stuff of diamonds, might yet retain its crown. Rodney Ruoff of IBM’s research laboratories in Yorktown Heights, New York, and Arthur Ruoff of Cornell University have predicted that the now ubiquitous buckminsterfullerene, or buckyball (carbon in a 60-atom sphere), might crystallise at about 20 000 times atmospheric pressure to form a material with a bulk modulus even higher than that of the hypothetically superhard -C₃N₄. Clearly, taking diamond’s crown away will be an extremely hard job.

David Bradley is a science writer and editor.