Distribution of gold in soils and stream sediments and the use of cyanidation in exploration geochemistry

• Main Author: Delaney, Tracy Anne
• Format: Others
• Language: English
• Published: 2008
• Online Access: http://hdl.handle.net/2429/1342

Obtaining samples that are both representative and small enough to be analyzed efficiently by standard analytical techniques is a common problem in gold exploration. The recent use of cyanide to extract gold from geochemical samples has allowed samples up to 1 kg or larger to be processed. Results of cyanide analyses cannot be interpreted, however, without first understanding the mode of occurrence of gold in the sample media. Therefore, the location of gold in soils and stream sediments over a variety of deposits in different weathering regimes was determined prior to examining the efficiency of cyanidation. Soil samples were collected over gold mineralization in Nevada, Utah and the Yukon Territory, and over tills covering gold deposits in British Columbia and Ontario. Except in Ontario, stream sediments were also collected. After wet sieving into four fractions between 2000 μm (10 mesh, ASTM) and 53 μm (270 mesh, ASTM) and separation of heavy minerals (S.G. 3.3), samples were analyzed for gold by fire assay-atomic absorption spectroscopy (FA-AAS) and cyanide-AAS (CN-AAS). Results indicate that, on average, 70% of the gold in soils and 67% of the gold in stream sediments resides in the finest fraction (-53 μm, -270 ASTM). Furthermore, although gold concentrations are highest in the heavy mineral fractions (HMCs), the percentage of gold is generally higher in the light mineral fractions (LMFs) and -53 μm fractions, particularly in samples from Nevada and the Yukon Territory. Comparison of CN-AAS analyses with those by FA-AAS -- assumed to represent total gold concentrations -- indicate that, on average, 60% of the gold in soils and about 40% of gold in stream sediments was accessible by cyanide solutions. In regard to exploration, a representative 30 g subsample can generally be obtained from wet sieving of a 500 g field sample. Although the -53 μm fraction contains the bulk of the gold -- and the least possibility of erratic values resulting from the nugget effect -- representativity is not greatly reduced in the -212 μm fraction. There appears to be no particular advantage to preparation of HMCs because gold concentrations can generally be detected in combined density fractions. Cyanidation was effective in detecting gold in all six areas. However, there is no advantage in using this analytical method in areas of fine particulate gold where gold concentrations are easily detected using FA-AAS. Cyanidation may be more useful in areas of low gold concentrations where large (i.e. > 100 g) subsamples are required to obtain representativity, or where gold exists in a variety of particle sizes. Sieving to a size fraction below 212 μm is recommended, however, to optimize representativity. === Science, Faculty of === Earth, Ocean and Atmospheric Sciences, Department of === Graduate