Ancient Chinese craftsmen might have learned to use diamonds to grind and polish ceremonial stone burial axes as long as 6,000 years ago, according to research carried out by US scientists.

Researchers at Harvard University have uncovered strong evidence that ancient Chinese used diamonds with a level of skill difficult to achieve even with modern polishing techniques.

The finding, reported in the February issue of the journal Archaeometry, places this earliest known use of diamond thousands of years earlier than the gem is known to have been used elsewhere in the world.

Scientists had put the earliest use of diamond around 500 BC, for the earliest authors to refer what is likely diamond, Maniliusand Pliny the Elder, lived in Rome during the first century AD.

The latest work also represents the only known prehistoric use of sapphire. The stone worked into polished axes by China's Liangzhu and Sanxingcun cultures around 4000 to 2500 BC has as its most abundant element the mineral corundum, known as ruby in its red form and sapphire in all other colors.

Most other known prehistoric artifacts were fashioned from rocks and minerals no harder than quartz.

"The physics of polishing is poorly understood. It's really more an art than a science," said the first author Peter J. Lu. "Still, it's absolutely remarkable that with the best polishing technologies available today, we couldn't achieve a surface as flat and smooth as was produced 5,000 years ago."

Lu's work may eventually yield new insights into the origins of ancient China's Neolithic artifacts, vast quantities of finely polished jade objects.

Lu studied four ceremonial axes, ranging in size from 13 to 22 centimeters, found at the tombs of wealthy individuals. Three of these axes, dating back to the Sanxingcun culture of 4000 to 3800 BC and the later Liangzhu culture, came from the Nanjing Museum in China; the fourth was discovered at a Liangzhu culture site at Zhejiang Province.

"What's most amazing about these mottled brown and grey stones is that they have been polished to a mirror-like luster," Lu said. "It had been assumed that quartz was used to grind the stones, but it struck me as unlikely that such a fine finish could be the product of polishing with quartz sand."

Lu's subsequent X-ray diffraction, electron microprobe analysis and scanning electron microscopy of the four axes' composition gave more evidence that quartz could not have polished the stones: fully 40 percent corundum, the second-hardest material on earth, the only material that could plausibly have been used to finish them so finely was diamond.

To further test whether diamond might have been used to polish the axes, Lu subjected samples of the fourth axe, 4,500 years old and from the Liangzhu culture, to modern machine polishing with diamond, alumina, and a quartz-based silica abrasive.

Using an atomic force microscope to examine the polished surfaces on a nanometer scale, he determined that the axe's original, exceptionally smooth surface most closely resembled -- although was still superior to -- modern polishing with diamond.

The use of diamond by Liangzhu craftsmen is geologically plausible, as diamond sources exist within 300 km of where the burial axes studied by Lu were found. These ancient workers might have sorted diamonds from gravel using an age-old technique where wet diamond-bearing gravels are run over a greased surface such as a fatty animal hide, and then only the diamonds adhere to the grease.

"I imagine that Neolithic craftsmen were constantly experimenting with new tools, materials and techniques. Without testing things out directly, there is no way anyone would be able to suggest what would work," Lu told Xinhua.

"Although traditionally we tend to think that the ancients had special wisdom to pass down, the original source of that information, however old, had to be determined by trial and error."

However, this ancient technology might have been lost in Chinese history.

"During the Neolithic age, there was no writing. So even though the craftsmen used diamond and sapphire, there would be no way to communicate that information to later generations, without a direct human link, "he said.

"That is, if the knowledge was passed on person-to-person, then the Shang Dynasty (1500 BC) might have been able to know about it and write it down. But there seems to be no such direct human contact between Liangzhu and Shang, and since the Liangzhu people couldn't write anything down, there would be no way to pass on the information."