Quality Assurance/Quality Control (QA/QC)
Sampling is the single most important aspect of an exploration or mining company. Sampling determines the value (or lack there of) of a companies exploration efforts. Since it is the most important part of exploration, it is important to verify the quality and assure the accuracy of results obtained from those samples.
Where sampling errors can come from
Sampling errors come in two basic types, contamination and procedural.
Contamination errors occur at any stage in exploration and is a much larger issue for precious metals. From gold smearing on a saw blade, then being deposited on another sample, to a logger logging core with a gold ring on, any redepositing of a metal or material where it isn’t supposed to be is considered contamination. Rarer examples are cross contamination at the lab level, maybe a crusher between samples wasn’t cleaned properly, maybe a sample got mixed improperly.
Procedural errors are errors resulting from deviations from the sampling procedures. Perhaps a sample tag is mixed up or not recorded properly, maybe the core cutter cut the sample improperly, or maybe the lab didn’t mix the sample properly.
Procedures for testing for errors and verifying results
After a few high profile mining/exploration scams (Bre-X) the Canadian government, and many others have followed suit, demand a higher caliber and standards for reporting sample data. This falls under what is known as NI 43-101 whereby a qualified geoscientist (Professional Geoscientist) must review all aspects of the sampling procedures for accuracy and potential issues.
In addition to strict sample/rock handling guidelines developed by the geologist/engineer to minimize contamination at the exploration site, there are specific methods to verify and increase the validity of sample data at the lab level.
Blanks are material which is guranteed not to contain any of the mineral being assayed for. This will determine if potential contamination at the lab setting is taking place. If a known non-mineralized sample shows a positive result to a mineral – a red flag is raised. Blanks are usually purchased from labs, or simple ground up glass will suffice.
Standards or samples purchased from labs, which have a known grade of mineral to be tested for. They can be high, low or even median grade standards. The purpose of the standards are to test for calibration errors at the lab level. Proper procedure is to have standards of varying grades to best represent close to expected grades in the samples being tested. A standard which is returned with a value outside the accepted range (given by standard source lab) could indicate poor mixing of the sample, or an calibration error of the equipment/personell handling the samples.
Duplicates are taken in two different phases. Core duplicates are a secondary splitting of core resulting in two samples that in theory should be similar in grade. The second, usually done as internal tests in the lab is to retest a sample already tested, again the samples should in theory be similar. A rule of thumb is that all samples above a certain grade are automatically re-sampled for verification. Under regulations some core must remain and be stored for future verification if need be.
If core duplicates at the exploration phase differ by a large amount it could indicate contamination or improper sampling procedures. There is variance in precious metal samples is to nugget effect and can show up in assays.
If lab duplicates come back with variance, it may indicate mixing, procedure or calibration errors. Precious metal duplicates can vary due to the nugget effect. Any company can request any sample to be retested -sometimes by another lab for verification of assays.
It is common practice that all samples above a certain grade be verified by another independent lab.
Insertion of blanks, duplicates (of both types) and standards occur at regular intervals in sampling. For example every 10th sample is a blank, every 25th sample is a standard, every 50th sample is a core duplicate.
These QA/QC samples are inserted and sent to the lab as if they were actual rock samples – thee lab does not know they are blanks/standards or duplicates. In addition labs typically do internal duplicate checks (every 10th sample) and automatically retest all assays above a specific grade.
The entire QA/QC procedure is usually lengthy and covers everything from sample handling at all levels of exploration and processing as well as defined protocols for insertion of standards/blanks and duplicates. Companies are required to keep portions of all samples included in reports for future retesting during an audit if the project advances beyond a simple exploration project.