Diamond-like carbon

Diamond-like carbon is a class of amorphous carbon materials that display some of the typical properties of diamond. DLC is usually applied as coatings to other materials that could benefit from such properties.
DLC exists in seven different forms. All seven contain significant amounts of sp³ hybridized carbon atoms. The reason that there are different types is that even diamond can be found in two crystalline polytypes. The more common one uses a cubic lattice, while the less common one, lonsdaleite, has a hexagonal lattice. By mixing these polytypes at the nanoscale, DLC coatings can be made that at the same time are amorphous, flexible, and yet purely sp³ bonded "diamond". The hardest, strongest, and slickest is tetrahedral amorphous carbon. Ta-C can be considered to be the "pure" form of DLC, since it consists almost entirely of sp³ bonded carbon atoms. Other materials notably hydrogen and graphitic sp² carbon are present in the other 6 forms usually as a result of the specific production process. The various forms can also be doped with elements such as Silicon to impart other desirable properties.
The various forms of DLC can be applied to almost any material that is compatible with a vacuum environment.

History

S. Aisenberg and R. Chabot are widely recognized as pioneers in the field of diamond-like carbon research. In a seminal 1971 paper published in the Journal of Applied Physics, they detailed a novel ion-beam deposition technique used to create amorphous carbon films with properties remarkably similar to natural diamond.
In 2006, the market for outsourced DLC coatings was estimated as about €30,000,000 in the European Union.
In 2011, researchers at Stanford University announced a super-hard amorphous diamond under conditions of ultrahigh pressure. The diamond lacks the crystalline structure of diamond but has the light weight characteristic of carbon.
In 2021, Chinese researchers announced AM-III, a super-hard, fullerene-based form of amorphous carbon. It is also a semi-conductor with a bandgap range of 1.5 to 2.2 eV. The material demonstrated a hardness of 113 GPa on a Vickers hardness test vs diamonds rate at around 70 to 100 GPa. It was hard enough to scratch the surface of a diamond.

Distinction from natural and synthetic diamond

Naturally occurring diamond is almost always found in the crystalline form with a purely cubic orientation of sp³ bonded carbon atoms. Sometimes there are lattice defects or inclusions of atoms of other elements that give color to the stone, but the lattice arrangement of the carbons remains cubic and bonding is purely sp³. The internal energy of the cubic polytype is slightly lower than that of the hexagonal form and growth rates from molten material in both natural and bulk synthetic diamond production methods are slow enough that the lattice structure has time to grow in the lowest energy form that is possible for sp³ bonding of carbon atoms. In contrast, DLC is typically produced by processes in which high energy precursive carbons are rapidly cooled or quenched on relatively cold surfaces. In those cases cubic and hexagonal lattices can be randomly intermixed, layer by atomic layer, because there is no time available for one of the crystalline geometries to grow at the expense of the other before the atoms are "frozen" in place in the material. Amorphous DLC coatings can result in materials that have no long-range crystalline order. Without long range order there are no brittle fracture planes, so such coatings are flexible and conformal to the underlying shape being coated, while still being as hard as diamond. In fact this property has been exploited to study atom-by-atom wear at the nanoscale in DLC.

Production

There are several methods for producing DLC, which rely on the lower density of sp² than sp³ carbon. So the application of pressure, impact, catalysis, or some combination of these at the atomic scale can force sp² bonded carbon atoms closer together into sp³ bonds. This must be done vigorously enough that the atoms cannot simply spring back apart into separations characteristic of sp² bonds. Usually techniques either combine such a compression with a push of the new cluster of sp³ bonded carbon deeper into the coating so that there is no room for expansion back to separations needed for sp² bonding; or the new cluster is buried by the arrival of new carbon destined for the next cycle of impacts. It is reasonable to envisage the process as a "hail" of projectiles that produce localized, faster, nanoscale versions of the classic combinations of heat and pressure that produce natural and synthetic diamond. Because they occur independently at many places across the surface of a growing film or coating, they tend to produce an analog of a cobblestone street with the cobbles being nodules or clusters of sp³ bonded carbon. Depending upon the particular "recipe" being used, there are cycles of deposition of carbon and impact or continuous proportions of new carbon arriving and projectiles conveying the impacts needed to force the formation of the sp³ bonds. As a result, ta-C may have the structure of a cobblestone street, or the nodules may "melt together" to make something more like a sponge or the cobbles may be so small as to be nearly invisible to imaging. A classic "medium" morphology for a ta-C film is shown in the figure.
Diamond-like carbon can also be made from the Preceramic polymer poly through pyrolysis at atmospheric pressure, or through physical vapor deposition methods. Poly can be synthesized from haloforms in multiple ways, including the reduction of bromoform with NaK, or the electroinitiated polymerization of chloroform, with yields of 46-85% and 30-40% respectively. The electroinitiated polymerization method can also be generalized for other polycarbynes as well.

Properties

As implied by the name, diamond-like carbon, the value of such coatings accrues from their ability to provide some of the properties of diamond to surfaces of almost any material. The primary desirable qualities are hardness, wear resistance, and slickness. DLC properties highly depends on plasma processing deposition parameters, like effect of bias voltage, DLC coating thickness, interlayer thickness, etc. Moreover, the heat treatment also change the coating properties such as hardness, toughness and wear rate.
However, which properties are added to a surface and to what degree depends upon which of the 7 forms are applied, and further upon the amounts and types of diluents added to reduce the cost of production. In 2006 the Association of German Engineers, VDI, the largest engineering association in Western Europe issued an authoritative report VDI2840 in order to clarify the existing multiplicity of confusing terms and trade names. It provides a unique classification and nomenclature for diamond-like-carbon and diamond films. It succeeded in reporting all information necessary to identify and to compare different DLC films which are offered on the market. Quoting from that document:

These bonds can occur not only with crystals - in other words, in solids with long-range order - but also in amorphous solids where the atoms are in a random arrangement. In this case there will be bonding only between a few individual atoms and not in a long-range order extending over a large number of atoms. The bond types have a considerable influence on the material properties of amorphous carbon films. If the sp² type is predominant the film will be softer, if the sp³ type is predominant the film will be harder.

A secondary determinant of quality was found to be the fractional content of hydrogen. Some of the production methods involve hydrogen or methane as a catalyst and a considerable percentage of hydrogen can remain in the finished DLC material. When it is recalled that the soft plastic, polyethylene is made from carbon that is bonded purely by the diamond-like sp³ bonds, but also includes chemically bonded hydrogen, it is not surprising to learn that fractions of hydrogen remaining in DLC films degrade them almost as much as do residues of sp² bonded carbon. The VDI2840 report confirmed the utility of locating a particular DLC material onto a 2-dimensional map on which the X-axis described the fraction of hydrogen in the material and the Y-axis described the fraction of sp³ bonded carbon atoms. The highest quality of diamond-like properties was affirmed to be correlated with the proximity of the map point plotting the coordinates of a particular material to the upper left corner at, namely 0% hydrogen and 100% sp³ bonding. That "pure" DLC material is ta-C and others are approximations that are degraded by diluents such as hydrogen, sp² bonded carbon, and metals. Valuable properties of materials that are ta-C, or nearly ta-C follow.

Hardness

Within the "cobblestones", nodules, clusters, or "sponges" bond angles may be distorted from those found in either pure cubic or hexagonal lattices because of intermixing of the two. The result is internal stress that can appear to add to the hardness measured for a sample of DLC. Hardness is often measured by nanoindentation methods in which a finely pointed stylus of natural diamond is forced into the surface of a specimen. If the sample is so thin that there is only a single layer of nodules, then the stylus may enter the DLC layer between the hard cobblestones and push them apart without sensing the hardness of the sp³ bonded volumes. Measurements would be low. Conversely, if the probing stylus enters a film thick enough to have several layers of nodules so it cannot be spread laterally, or if it enters on top of a cobblestone in a single layer, then it will measure not only the real hardness of the diamond bonding, but an apparent hardness even greater because the internal compressive stress in those nodules would provide further resistance to penetration of the material by the stylus. Nanoindentation measurements have reported hardness as great as 50% more than values for natural crystalline diamond. Since the stylus is blunted in such cases or even broken, actual numbers for hardness that exceed that of natural diamond are meaningless. They only show that the hard parts of an optimal ta-C material will break natural diamond rather than the inverse. Nevertheless, from a practical viewpoint it does not matter how the resistance of a DLC material is developed, it can be harder than natural diamond in usage. One method of testing the coating hardness is by means of the Persoz pendulum.
In a microhardness test of a DLC coating, a case-hardened 9310 bearing steel was tested using a diamond-tipped indenter tool supplied by Fisher Scientific International. The tool used a comparison of force applied to indentation depth, similar to the Rockwell Scale hardness measurement method. Microhardness testing of uncoated steel was limited to an indentation depth of approximately 1.2 microns. This same bearing steel was then coated with a 2.0 micron thickness DLC coating. Microhardness testing on the coated steel was then conducted, limiting indentation of the coating to a depth of approximately 0.15 microns, or 7.5 percent of the coating thickness. Measurements were repeated five times on uncoated steel and 12 times on coated steel. As a reference, the uncoated bearing steel had a hardness of Rockwell C 60. The average microhardness measured was 7,133 N/mm² for the uncoated steel and 9,571 N/mm² for the coated steel, suggesting the coating had a microhardness of approximately 34 percent harder than Rockwell C 60. A measurement of the plastic deformation, or permanent indentation scar, caused by the micro-indenter, indicated an elasticity of 35 percent for steel and 86 percent for the DLC. Measurement of plastic deformation is used for Vickers hardness measurements. As expected, the greater "closing" of the indentation scar for the coating suggested much higher Vickers hardness, in a ratio of greater than two times that of the uncoated steel, and therefore rendering Vickers hardness calculations not meaningful.