Hardness Testing Of Gemstones: A Re-evaluation Of The Values

(via Gemmology Queensland, Vol 3, No.1, Jan 2002/IGC Conference Madrid 2001) Michael Gray writes:

Hardness testing is one of the methods that a mineralogist uses to determine the identity of a mineral sample, but which has rarely been used by the gemologist. A mineralogist can usually find some inconspicuous face or broken area to conduct a scratch test, but the gemologist rarely has had a surface on a gemstone to do a comparable test without visible damage to the polish of the stone. However, the ability to choose a random surface may give a false value when a range of hardness exist, as in the case of kyanite, which has a documented variation from 4.5 to 5 on the Moh’s scale of hardness. While a species with such a diverse range can be easy to identify and test, there are many other species that have smaller variations that have not been measured. We are preparing a method of testing hardness during the faceting process. Our present system was developed by Friedrich Mohs in 1822, whereby he set relative values of hardness to ten easily obtainable species for comparison to each other, and measured by making scratches on surfaces of the unknown material with these ten species. This scale is still used as the standard, but variations in hardness has been known since at least 1844 using a device called sclerometer, and researchers were able to deduce objective values, although some variations in these scales are noted, depending upon the method and preparation used. In fact, some researchers determined that topaz is less hard than quartz when the sclerometer was used on powders of these two species.

Lapidaries have always noticed some of these variations in hardness in some species, especially noticeable in heat treated sapphires and rubies. A good example of this variation is the species fresnoite. In the references, the published hardness varies from “3 to 4” to a value of 4.7. A lapidary must rely upon these figures to determine which cutting and polishing laps and media to use in fashioning a gemstone. A great variation in hardness was noticed in the grinding and polishing of the stone, with only one direction being noticeably ‘soft’ as compared to the other facets. There is only one direction being in the range of 4 in hardness, with all other directions being noticeably harder.

These are hardly the only examples to cite; there are many materials, such as spodumene, sillimanite, elbaite, and many of the garnets that exhibit some noticeable variation in hardness during faceting. Diamond cutters know of the variations in their specialty, and use that knowledge in the fashioning of those gems.

The argument could be made that Moh’s scale has worked for almost two centuries, and that there is n need to change this system. That is not the purpose of this study. There are a number of purposes why this study has been initiated to determine the actual hardness of a species, such as fresnoite, where the original description may be inaccurate to document the variation of hardness, as well ad ‘normal’ hardness, within a species. To determine if a variation of hardness may be caused by a man-induced treatment, such as heat treatment of corundum, which then can be used as a way of detecting these treatments. This might also be taken further to determine origin, as well, depending upon impurities, coloring agents, and variations in chemical composition to help clear up conflicting values in different publications to relate these values to the existing Moh’s scale of hardness so that the numbers are understandable to academics and lay persons alike.

Equipment has been developed recently that makes it much easier to conduct hardness testing. The development of ultrasonic indenters that can be calibrated based on the Rockwell and Vickers scales, two of the glass and metal industry’s standards, should eliminate most of the danger of stone breakage during hardness testing. Our testing will be done during the cutting stages, so any noticeable marks made on the softer stones can be polished out.

The ten species used in the Moh’s scale will have been measured, resulting in a set of numbers that will set the parameters for the decimal places between species. Stones are being faceted of these materials to make note of any noticeable variations during cutting, and then measurements taken of these materials. After these values are established, checked, and documented, other species will be faceted, with noticeable hard and soft areas marked for testing. On stones with variable hardnesses, at least four hardness tests will be performed, one each on the ‘softest’ and ‘hardest’ directions, since these normally occur only in one direction, and several tests to establish the ‘normal’ hardness, the hardness over the rest of the material. Therefore, while a range may be established, a third number may represent the general hardness of the material.

The practicality of using gemstones for hardness testing is the general ‘purity’ of the sample being used. Many of the published hardness values were obtained using tiny pieces of the type specimens when they were first discovered and therefore were probably unable to get much accuracy, especially minerals described prior to the 1960s. Unfortunately, once a mineral is named and documented, the values may come up dependent upon the locality and/or if the material has been treated, and all of this information will be documented, as well, whenever possible. Ultimately, we should be able to document the crystal orientation that these variable hardnesses will occur.

There is also an important side benefit to this testing. While hardness testing is only used in a miniscule number of gem identifications, we hope to show that this test can be used in gem labs, using the modern equipment available today. The marks made by a skilled and knowledgeable worker using this modern equipment can be so tiny as to be undetectable by the naked eye, and can be placed on an inconspicuous spot on the girdle or pavilion of the stone. With the values obtained through this research, there will be one more test available for questionable gems. It should also be possible to test mounted stones as well, where refractive indices and specific gravities may not be able to be obtained without removal of the stone from the mounting.