Fire Assay

Fire Assay is a technique which analyzes the amount of precious metal (gold, silver, palladium and platinum) in a sample of ore or scrap. It is generally regarded as the most accurate (7-10 ppb or 0.1g/t), economical, and consistent method for gold analysis, though it is time consuming. It works with all forms of gold, and there is no such thing "un-assayable" gold, provided the method is done correctly

It is mentioned in the Doomsday Book and was invented as far back as 2000 BC, when very pure silver was produced from lead ore in Asia Minor. The earliest detailed written record known is De Rey Metallica by Georgius Agricola in 1556, which served as the standard book on the subject for about two centuries.

The process can be split into three stages: fusion, cupellation, and analysis.

Fusion: A sample of 10-30 g is blended into a fine powder. The most common sample is 30 g, which is roughly equivalent to a traditional "assay tonne". Up to 150 g can be used for particularly low concentrations.

This is added into a crucible along with lithard (PbO), borax (sodium borate), silica (sodium dioxide) and soda ash (sodium carbonate) to form a melt, and heated to 1000-1200° C. The exact temperature and composition of the ingredients depends on what trace elements are present in the sample. For instance, a siliceous ore requires a basic flux, and a basic ore requires an acid flux; soda ash is good at removing sulfur; etc. The details are adjusted to suit the ore, using an approach based on long empirical experience. The whole process from start to finish has a high degree of skill and benefits from having an assayer with knowledge and intuition.

The main purpose of these extra ingredients are to create a "flux" which mixes with the sample. This flux lowers the melting point of the mixture (mainly required for the metal oxide impurities) and imparts a homogeneous fluidity.

A small amount of carbon is added causing some of the lithard to be oxidized to lead and sink to the bottom of the melt along with any silver. Gold platinum and palladium in the melt will then drift into the lead-silver alloy at the bottom, while the impurities are collected into the liquid slag. The reactions here are complex and much of it has not been theoretically mapped out, but the basic principle is that the hot PbO is very oxidizing, and the slag is very soluble to metal oxides, so anything not a precious metal will tend to be oxidized and drift into the slag. In addition, most precious metals exhibit a good affinity for lead by themselves, and even the ones which do not will be dense enough to sink into the pool at the bottom, forming a mechanical mixture.

The melt is added to a mold and allowed to cool, then the glassy slag separates cleanly by tapping with a small hammer, leaving a lead button.

Cupellation:  The lead button is placed inside a cupel, which is a crucible made of pourous bone ash or magnesia. When heated to about 800-1000° C in air, the molten lead is oxidized back to PbO which then melts and is drawn into the cupel by capilliary action, leaving a small bead (called a dore bead) of precious metal. Cupels are rated by the grams of lead oxide they will absorb.

Analysis: The dore bead can be analyzed directly, or the silver can first be separated using nitric acid, dissolving the silver as silver nitrate. Separating the metals allows gravimetric analysis. The most common analytical method used today is AA (Atomic absorption spectroscropy). Other methods used are ICP (Inductively coupled plasma atomic emission spectroscopy), ICP/MS (Inductively coupled plasma mass spectroscopy) or for very low concentrations, INAA (Instrumental Neutron Activation Analysis).

Stable Cyclic Carbenes

By "stable" we mean carbenes which can exist at room temperature without being bonded to a metal ion. Stable cyclic carbenes are classified into five families: NHCs (N-heterocyclic carbenes), Thiazolylidenes, PHCs (P-heterocyclic carbenes), Cyclopropenylidenes and CAACs (Cyclic alkyl amino carbenes).

The neutral carbene exists in either a singlet or triplet state:

This depends on what is larger, the pairing energy or the difference between the p and sp orbitals. From 1990-2010, stable singlet carbenes have been isolated at room temperature, while stable triplets have been observed cold.

The stable singlet carbenes are more important, as they considered extremely useful as catalysts when complexed to a transition metal.

NHCs are by far the largest group with hundreds of publications, so review articles tend to focus on these.

We can see the singlet-triplet gaps of the five families (kcal/mol):

This suggests that NHCs are mostly likely to produce singlets, which might explain why they are so popular with researchers.

σ-electron withdrawing groups favor the singlet state, σ donating groups favor triplet state. But the most important factor in stablizing singlets is the presence of a π-donor (an element or a pi bond) next to the carbene carbon. Bulky substituents can also offer kinetic stability and force the molecule into a shape which encourages π-donation onto the carbene carbon. The more full the empty carbene p-orbital is, the higher the singlet-triplet gap, the more stable the carbene.

The carbene also gains stability by linking its side atoms into a 3-5 membered ring, like in the examples above. First, because the shape of the ring can force s-character into the carbene lone-pair orbital. Second because it forces π-orbitals of π-donors to align with the empty carbene p orbital, increasing overlap/donation.

The first stable NHCs was made in 1991 by Arduengo. Who grew crystals of this thing, now known as Arduengo's Carbene:

Those side groups are called adamantane groups. They tend to be shorthanded as "Ad". This molecule has all the stabilization attributes mentioned above.

The most well known NHC is the 2nd generation Grubbs catalyst used for olefin metathesis, which won him the nobel prize in 2005.

Another feature of NHCs is that some are used as organic catalysts in their own right, without being complexed to a metal.

Going through the other four classes, stable thiazolylidenes came in 1997. The inferior singlet-triplet gap is from the inferior pi-donating capacity of sulfur compared to nitrogen. They tend to require very bulky side groups. They have some utility in metal complexes, but no highly effective thiazolylidene-based catalysts have been reported.

The first stable P-heterocyclic carbenes was isolated in 2005:

Mes* = 2,4,6-tri-(t-butyl)phenyl.

Only a few other PHCs are known. They all require enormous groups to be stable.

Also in 2005 were the first cyclic (alkyl) (amino) carbenes. These also have low orbital splitting, partly from having only one heteroatom next to the carbene carbon, and partly from the sigma donating character of the sp³ carbon. The orbital splitting of CAACs and PHCs are similar, suggesting that one animo group has about the same effect as two phospino groups with bulky substituents.

The adjacent carbon has to be quaternary to prevent a 1,3 hydride shift.

CAACs are being heavily researched and arguably have the most potential uses, since the sp³ carbon can produce sigma-donor effects on the carbene carbon more powerful than all the other cyclic carbenes. The sp³ carbon also has more potential for steric bulk protecting the carbene atom.

The relatively empty p orbital in CAACs also give this class some stranger reactivity and unique side-reactions. They form stable aminoketenes when exposed to carbon monoxide, and split H₂ and NH₃ under normal laboratory conditions.

We have the final class produced in 2006, Cyclopropenylidenes. People were surprised to see a stable carbene without even one pi-donating heteroatom adjacent to the carbene carbon. Only one has ever been isolated:

The strained flat shape of the ring puts the empty p orbital of the carbene planar with the double bond, so both adjacent carbons act as pi donors. This is enhanced by the isopropylamino groups, for they can donate their lone pairs into the pi* antibonding orbital, which has the right symmetry to overlap with the empty carbene p orbital. Finally the carbene lone pair is forced into a sp² shape, which enhances the sigma-pi split even further.