, , , ,

Some alchemical recipes, whilst appearing to be entirely practical on the surface, are probably not so underneath. One highlighted by Larry Principe in his “The secrets of alchemy”, is the sublimation of mercury with vitriol and saltpetre as carried out in 14th century Europe, from page 65:

“John first describes a series of sublimations of mercury with vitriol and saltpeter, followed by various digestions and distillations. Despite the apparently clear directions, however, his first step will not work in a modern laboratory if followed verbatim. The sublimate “white as snow” that John describes making is undoubteldy mercuric chloride; therefore, the starting mixture must have included common salt, but this substance is not mentioned in the list of ingredients. There are two possible explanations. First, John’s saltpeter might have been quite impure and contained a large quantity of common salt. In fact his De Confectione contains an annotation towards the end that notes how crude saltpeter ordinarily contains salt, and gives a method for purifying it by fractional crystallisation. The second possibility is that John intentionally left out the crucial ingredient as a way of preserving secrecy.”

Now, when laid out like that, it looks like there was a deliberate mistake, but then the question is also what mention of saltpetre and it’s purification is made in other alchemical works, especially those which John may have had access to? Certainly the “Of the invention of verity or perfection”, a pseudo-geberian work of the 14th century mentions the purification of common salt, salt-peter and others, indicating that other people thought they were different substances and so you would be expected to be aware of the differences. There is also an early 13th century work attributed to Michael Scot which tells you how to distinguish between different salts by how they react on a burning coal.
The problem is that such methods wouldn’t tell you about a small proportion of salt in the saltpetre, analytical chemistry being several hundred years in the future.

The question that has arisen in my mind though, is why would you expect much salt, sodium chloride, in the saltpetre anyway? The answer appears to be in the animal dung and urine that is used, with ash, to make potassium nitrate; which salt is mentioned by Lazarus Ercker, the late 16th century expert on mineralogy, as leading to crystals of salt forming in the liquid used to extract it from the earth. The salt crystals are then skimmed out. Of course in the mid-14th century they simply weren’t as good at that sort of thing, so not been able to remove them, or the potassium nitrate might well have been imported, adulterated, then sold on again at great price. So it is possible that the saltpetre would have salt in it, and Principe’s explanation is correct.

But then Principe writes:
“If this is the case, then it is significant that the end of De confectione includes a rather out of place paragraph describing the general importance of table salt (sal cibi, or ‘salt of food’), it’s ubiquity, its use in purifying metals, and so forth, and then states that ‘the whole secret is in salt’. Is this an example of the dispesion of knowledge?”

I do have a suspicion about the bit regarding salt, and Principe agrees in his notes and references at the back of the book, the relevant one saying “Because we lack a critical edition of John’s work, I hesitate to assert that the paragraphs about salt are his; they might have been added by a later follower who recognised the need for salt.” Which is good scholarship on his part, but this mention of common salt contradicts the specific instructions in the text for purifying saltpetre! Maybe they were added later as well?

I happen to have a couple of thoughts relating to it. Firstly, you can apparently get white mercuric compounds by reaction with nitrates or sulphuric acid, forming mercurous nitrate or mercuric sulphate. Of course the commonly used methods are rather different from those given in the above recipe; nevertheless I have my suspicions about what would be formed given that sublimation involves the species being hot and in a vapour state, ideal for reactions. This is definitely worth testing in an experiment.
Secondly, there are two kinds of recipes in later medieval alchemy that sublime substances with mercury. One, which features in “Misticall wordes and Names infinite” by Locke, compiled around 1571, uses earlier works including later medieval ones. Chapter 4 is taken from the MS work “Scoller and master”, which appears in the 15th century and was very popular in England. It clearly states to grind salt and vitriol together, dry it and grind it together with mercury, then carry out the sublimation. There’s also the late medieval/ post medieval Compound of Compounds, which uses salt. Even better, the Invention of Verity or Perfection, a pseudo-Geberian work of probably also the 14th century, says to use a pound of reddened vitriol, 2lbs of calcined alum, 1 pound of common salt and a fourth part of saltpetre, with the mercury. Now that will produce mercuric chloride, and I note the reddened vitriol means it has been calcined, so nothing much will happen from that.

For some colour, here’s the yellow-white mixture of green vitriol, saltpetre and ammonium chloride mixed together:
acid making mixture

Now, the Testamentum later attributed to Raymond Lull, according to the French translation which I have, specifically states to use vitriol and saltpetre when subliming your mercury. The Testamentum dates from around 1332, written in Catalan, and as such doesn’t seem to have been so popular for a while, nevertheless, it agrees with John of Rupescissa’s text as to the substances to sublimate with mercury. So we have recipes which use vitriol and saltpetre, and ones which say to use salt, all in the 14th century. The precise justification will vary from author to author, and I need to learn more Latin so I can understand what they wrote, and this is potentially an interesting thing to investigate.

But then if we turn to chapter 5 of Locke’s treatise, also taken from Scoller and Master, we find instructions to grind mercury with common salt, wash it, and then carry out a sublimation with salt and vitriol. Why specify common salt once, but not use the term in the previous recipe? A mistake? Carelessness? Or which type of salt should you use? Or maybe it’s just an artefact of the recipes having been copied out by Simon Forman and he didn’t feel the need to keep writing ‘common salt’. You can see how this is getting complicated!

Certainly alchemists used to use many different salts, with al-Rhazi having a recipe which used 7 of them, and by the 13th century European alchemists knew of several, and of course, salt-nitre was one of them.

So it occcured to me that at some point the terms for salt, common salt, sal-nitre or saltpetre, all got mixed up or improperly copied, or deliberately confused. Sal, Sal-common, sal-nitre, sal-petre, natron, etc, it is a fertile field for someone to say “No, he actually meant this” or for someone’s handwiting to be illegible.

How can we tell the reasons for this confusion apart? Obviously we could pin down the improperly copied one if we have manuscripts which vary slightly, and that has been suggested as an answer for a number of issues in alchemical work.
But then how can we tell if something has been deliberately changed or left out? I don’t see that we can prove it one way or another; Principe’s example works if the only white substance you can possibly get from subliming a mixture of green vitriol, saltpetre and mercury is mercuric chloride, but as I wrote, I feel like experimenting on it to see what I get.