What鈥檚 a smashed and burnt piece of pot doing in the company of the world鈥檚 most ancient gold coins? This fragment of coarse, gritty earthenware, on display in the British Museum鈥檚 coins and medals room, is more valuable than the fabulous golden treasures. It may be a bit of old rubbish retrieved from a dump at Sardis, capital of the legendary kings of Lydia, but hidden in its pores is the story of the world鈥檚 first gold refinery. The discovery of a way to purify gold made Croesus, the last of the Lydian kings, so rich that his name became a byword for wealth.
IT COULD have been worse. In a busy workshop baked by the Mediterranean sun and the fires of half a dozen furnaces, accidents are bound to happen. One day, sometime in the 6th century BC, a craftsman in Sardis stumbled on the uneven floor of the workshop and dropped the cooking pot he was carrying. The pot broke, spilling a mound of salt on the floor. The man carefully gathered up the salt -not because of any superstition, but because buried among the grains were a few handfuls of gold dust. He had been about to place the mixture of gold and salt into the furnace for the start of the refining process, a technique he and his fellow goldworkers had just mastered.
Satisfied that he had retrieved as much of the mixture as he could, the craftsman threw the broken pot onto a rubbish heap. Two and a half thousand years later, in the late-1960s, an American archaeologist from the Harvard-Cornell Archaeological Exploration of Sardis picked up a piece of the pot during excavations of the ancient city, now part of western Turkey. The old bit of pot didn鈥檛 look especially exciting. But then the excavators began to find a lot of small hearths, piles of scorched bricks, little pieces of hammered gold foil and some globules of melted gold. Together these suggested they had uncovered something that was exciting-a gold refinery right in the middle of Sardis, the city that became synonymous with riches and luxury.
Advertisement
The Lydians鈥� wealth came partly from their conquests across Asia Minor, in the form of loot or tribute. But the Lydians also had plenty of gold on their doorstep, with rich deposits in the gravel of the Pactolus river. King Gyges, who lived in the early 7th century BC, found a use for all this gold by introducing coins-the first in the world-and a money-based economy. The earliest coins were made of 鈥渆lectrum鈥�, an alloy of gold and silver, because no one knew how to separate the two.
All native gold contains some silver, and the gold washed from the Pactolus contained between 15 and 40 per cent silver. This variability made it impossible to guarantee the value of coins made from river gold, so Gyges鈥檚 coiners tried to standardise the alloy by adding yet more silver. But by Croesus鈥檚 time, between 561 and 547 BC, the idea of money was really taking off and it became vital to make coins of guaranteed purity-which meant pure silver and pure gold. The refinery at Sardis dates from the time of Croesus, so clearly his craftsmen found a way.
How did they do it? Unfortunately, the earliest account of the refining process was written 500 years after the last goldworker at Sardis laid down his tools. And that account, by a Greek geographer called Agatharchides, had a fatal flaw: the technique didn鈥檛 work. Fortunately, there are plenty of chemical clues in the pots and bricks unearthed at Sardis, enough for metallurgists at the British Museum鈥檚 Department of Scientific Research to piece together the process.
One fragment of pot holds the key to the whole process. This shard is from a 鈥減arting鈥� vessel, a rough cooking pot the size of a saucepan in which the silver was slowly removed from the gold. There are still specks of gold clinging to the surface of the pottery. A few of the specks are of pure, refined gold, indicating that this pot had already been through the process at least once before it was dropped. Other specks are raw gold dust from the Pactolus. These unrefined bits of gold suggest that the pot was on its way back to the furnace with a new batch of unrefined gold when it was broken.
Further chemical analysis turned up signs of salt-probably common sea salt-and enormous quantities of silver chloride throughout the porous body of the pot. These hinted at a simple but highly effective method of refining gold. According to Paul Craddock, who has reconstructed the process, the Lydian goldworkers probably filled ordinary cooking pots with alternate layers of salt and gold. They then plugged the top with clay and placed the pot on a plinth in the furnace.
In the moist atmosphere of a wood-fired furnace, salt reacts with water vapour to give off a mixture of corrosive gases, including chlorine. This starts a reaction with iron salts in the clay of the pot, generating another corrosive vapour-ferric chloride. These vapours then react with silver at the surface of the gold fragments, producing silver chloride, which is surprisingly volatile and travels through the pores of the pot and out into the furnace. The corrosive gases work their way deeper into the gold, moving through the spaces left by the lost silver, until eventually all that is left is gold. This process works most efficiently if the corrosive vapours have a large surface to attack. Gold dust from the river was ideal, but scrap gold and recalled electrum coins had to be hammered into thin foils before processing.
The temperature of the furnace was all-important: it had to be hot enough to create the corrosive vapours, but not so hot that either the salt or the gold-silver alloy melted. That meant keeping the furnace below 800 掳C. The state of quartz crystals in the clay indicates that they were on the point of melting, suggesting that 800掳C was the hottest the pot ever got.
Refining gold this way was a slow business. 鈥淭hey bake it in a kiln for five successive days and as many nights,鈥� wrote Agatharchides. And in that detail, he was probably right. At regular intervals, the goldworkers would take samples to assay, rubbing the metal on a dark touchstone and comparing the colour with that of gold known to be pure. The refinery was a huge success: coins made from its gold contained less than 2 per cent silver, and it probably produced enough gold in a year to mint tens of thousands of gold coins.
All this didn鈥檛 do Croesus much good. Unparalleled wealth wasn鈥檛 enough for him. He decided to start a war with the Persians, and when he consulted the oracle at Delphi about his prospects it foretold the destruction of a great empire. Unfortunately it was his.