COMMENTARY — Quality control, part 1

-scale quality control (QC) procedures to monitor the sampling and analyses of geological materials have only recently been used in mineral exploration.

At one time, only a few companies had an established, corporate-wide policy of monitoring sampling and analytical procedures — from the outcrop and drill core through to the warehouse, where a representative portion of sample is archived. The Bre-X scandal is changing this attitude.

With a bit of forethought, planning and money, a well-designed QC program can detect contamination, salting, sampling inconsistencies, laboratory biases, analytical procedural errors, sample misnumbering or misordering, degraded analytical detection limits and changes in mineralogy. Most importantly, it can quantify both the accuracy and precision of the entire sampling and analytical process. Knowledge of these two quantities provides an estimate of the potential risk associated with calculating an average grade of a drilled mineral deposit.

QC procedures in mineral exploration are necessary to quantify accuracy, precision and detection limit, and to detect contamination of samples.

* Accuracy can be defined as the degree to which an analysis, or the mean of a set of analyses, approaches the true concentration.

* Precision is the percent relative variation at the two standard deviation confidence (95%) level. In this definition, the larger the precision number, the less precise the sampling and analysis.

* Detection limit is defined as the point at which the precision of a sampling and analytical procedure equals 100% at the two standard deviation confidence level.

The early stages of exploration are primarily concerned with precision and contamination. Accuracy is of lesser importance. The main concern is the reproducibility of results (precision), the repetition of patterns and the detection of possible contamination.

Reconnaissance and preliminary exploration should include the preparation of blank (barren) samples to monitor contamination, and a routine for inclusion of field duplicate samples within a sample stream to begin to monitor precision.

Usually, 40 to 50 kg of blank material are needed (so that the same blank can be used throughout the program). The blank sample must be inserted into the drill sample stream in such a way as to be invisible to the laboratory. If the blank sample yields a detectable concentration of the element being sought, contamination of the sample batch has likely occurred.

Field duplicate samples are splits of drill core, reverse-Circulation cuttings or outcrop samples from the same sampling interval. Both halves of a split or sawn diamond drill core can be used, but, usually, a field duplicate of core consists of a sampling interval, which is quartered and placed in separate sample bags with separate sample numbers.

Reverse-Circulation drill duplicates must be obtained directly from the drill sample discharge as drilling is progressing. These field duplicate samples must also be blind to the laboratory and treated like normal samples. For gold exploration, at least one field duplicate sample for every 20 samples should be inserted in the sample stream. It is important that the duplicates be taken randomly, with no regard to rock type or geographical position. Any bias in the duplicate sampling program will negate the validity of subsequent precision and detection limit calculations.

These field duplicates contain the cumulative uncertainties associated with the entire sampling and analytical process. In addition, the exploration company should instruct the laboratory to split the coarse, jaw-Crushed samples into two preparation duplicates, which are then pulverized and analyzed separately, and to report the pulp duplicate analysis. By compiling the field, prep and pulp duplicate types, the analyst can ensure that the overall precision, as well as the source of the greatest uncertainty in the sampling and analytical process, is quantified with recently developed statistical procedures.

Second-phase drill programs are concerned not only with contamination and precision but must determine the accuracy of the sampling and analytical procedures. Accuracy can only be determined by using standard samples submitted from the field in such a way as to be blind to the lab.

There are different types of standards used during the processing of geological material. Labs use artificial aqueous standards to calibrate instruments. Some use artificial solid standards, though most labs manufacture standards from different geological media.

Although those procedures are necessary for the laboratory to monitor itself, the use of these standards does not aid the exploration company whatsoever.

If a company does not submit its own set of standards with its samples, no quantitative estimate of accuracy can be attributed to the analysis of samples submitted during an advanced drill program.

— The preceding is from a presentation given by the author, a consulting geologist, to mining analysts, brokers and media.