Bulk leach extractable gold
Bulk Leach Extractable Gold, more commonly shortened to BLEG is a geochemical sampling/analysis tool used during exploration for gold. It was developed in the early 1980s to address concerns relating to the accurately measuring fine grained gold, and dealing with problems associated with sample heterogeneity.
A large quantity of sample (between 2 and 5 kg) is digested or leached with cold sodium cyanide (NaCN) solution (generally 0.1% NaCN), for one to several days (depending on the assay laboratory involved). The gold is dissolved through its formation of a cyanide complex, which is concentrated through the solvent exchange process into an organic solvent and subsequently analyzed. The use of large sample weights and solvent extraction enables low detection limits, as low as 0.1 ppb. The digest is analyzed for Au (0.01 ppb), Cu (0.01 ppm), Ag (0.5 ppb), and depending on situations, for other elements. However, gold values in BLEG are lower than total assays such as those of fire assays, as it analyzes only the fine grained gold fraction and largely ignores coarser and nuggety gold.
If the need arises, a separate split of the original sample is used for pathfinder elements.
BLEG requires the collection of large samples, generally greater than 2 kg. It is necessary to collect fine sized material – silt to clay - where the fine flakes of gold will reside. Given their shape, these fine flakes of gold don’t act hydro-dynamically like heavy minerals, and will not settle in the same locations in a stream bed. Where possible, attempts should be made to sample flash flood sites (where finer material concentrates) – overbank deposits.
BLEG vs Sieving
BLEG is essentially a chemical sieve, designed to focus on the fine grained gold fraction. While many exploration geologists attest to the value of BLEG, some question its value compared to analyzing a simple sieved-out fine grained fraction. BLEG has been tried and tested over 20 years, applied successfully in different climatic settings, provides robust, excellent repeatability of results and has had an important role in discovery of massive Au deposits. However, some comparisons show that fine-grained sieved samples (<75 μm) show comparable results to BLEG, making the more complex process potentially unnecessary. With the exception of BLEGs applicability over a wider range of climatic conditions, the superiority of one method of the other is, at this point, still largely determined by personal preference.
Disagreement With Much of Above
I completely disagree with this and have been running gold exploration using BLEG for decades - it sounds like it has been derived from another source rather than experience (only one source, old and faulty). I suggest people Google to find out about it (sorry, no time to edit properly).
I would say the advantages are that one can use large samples of the sort of material in which gold is more likely to mechanically not chemically concentrate (sand and gravel rather than clay). Such material needs a large sample because extremely inhomogeneous, and BLEG (cyanide leach) is most appropriate for a large sample. The same issue makes BLEG useful even with coarse gold (e.g. you may have only a few coarse grains in a kilo of gravel or sand, imagine the probability that even one of those few grains would turn up in a small sample, or a sample of clay (not where coarse gold would concentrate) - even if you only partially leach the coarse gold, you will still see an anomaly, so that is not a major issue. Geochemical stream sediment sampling is more dependent on precision in all things than accuracy.
it is common for BLEG samples to give high Au values in places where clay will give low values (often both are anomalous, but sometimes one, sometimes the other - and when only in the clay sample it is usually because the gold is fine, not coarse). Much gold from some deposit types ("orogenic") is coarse and can even have a MEAN size of as much as a few mm in the original souce from which gold is derived - because it is malleable it flattens rather than disintegrates with transport. I have often panned and found all gold at some localities coarser than a mm (decreasing downstream - but you are chasing upstream to the source!). Definitely not always the same pattern, I have surveyed dozens of different goldfields that way using the same method every time and in some you would miss significant anomalies using one approach, in others you would miss using the other. It should not be personal preference (the problem is uninformed people who don't use it, or inexperienced samplers who sample inappropriately - and also cost, but that is usually false economy).
I never sample fine overbank deposits and elsewhere where fines accumulate for BLEG, and I never seive the fines from the same sample. I take the BLEG from the most active channel, the fines for other analysis of Au and metals (many of which adsorb on clays) from within the main stream but where fines accumulate behind fallen logs etc, within metres of the BLEG site. If you sieve from the one sample it means you have fairly inappropriate sample material - you would sieve all day to get enough clay from the huge volume of gravel and sand that would be required to get enough.
Clay samples contain a lot of active carbon in organics so are usually avoided as much as possible in BLEG samples - it can precipitate gold from the BLEG liquor prior to analysis, giving low values in high gold samples.
Some people use Cu and Ag (often a totally different pattern to analysis of fines and too many variables). But it is highly applicable to PGE (platinum group elements) exploration.
I like 5 kg samples but it is heavy to backpack many out of valleys and greatly reduces daily productivity and increases analytical costs (cyanide disposal environmental issues, large leach containers) - 1 kg is fairly standard nowadays.
BLEG sampling is sometimes used for other types of sampling - even coarse gold will dissolve adequately if rolled with cyanide for long enough (that is how we extract gold commercially). Therefore it is also used for analysis of gold in things like quartz veins with coarse "nuggety gold" - no other method will analyse a sufficiently large enough sample (e.g. 50 kg) to get a statistically meaningful result at a single sample site (and sometimes not even then).
No references, but am a geochemist with PhD who has run some small programmes essentially as research exercises, exploring all possibilities and variations.