Objective identification of mineralogy using reflected light spectroscopy has become commonplace in the mineral resources industry. Primarily using visible and short-wave infra-red light, the technique is useful for highlighting mineral phases that can be notoriously difficult to identify using conventional visual logging. Examples include considering fine-grained samples where individual mineral grains are impossible to see under a hand lens or distinguishing between the different types of white mica or chlorites.
ALS can collect high quality reflected light spectra from samples in the visible and short- wave infra-red wavelength ranges, using a TerraSpec® 4 HR spectrometer. Crushed sample material (e.g. coarse reject) or drill chips from RC or AC drilling are the recommended sample type, and these can be provided in bags or chip trays. It is common for each sample to represent 1m of drilled material, though larger composites are also feasible. Samples can also be collected from outcrop.
The output from the TRSPEC-20 method is either a raw spectral file in ASD format, or an ASCII file. Data can then be imported into several software packages designed to process spectral data to develop interpretations and derived outputs to aid project objectives.
Some geoscientists have the requisite skills and experience to interpret spectral data. Alternatively, raw spectral data can be interpreted by a third-party. ALS has partnered with AusSpec/IMDEX to offer the aiSIRIS cloud-based automated mineral interpretation service. This system uses artificial intelligence to interpret spectra, outputting mineral assemblages and spectral parameters that are considered to most closely relate to the geology of a project.
The INTERP-11 method is more effective when some information relating to project geology can be provided, as this information is used to guide the system in what to consider. The data is delivered in a spreadsheet format, making thematic mapping of parameters in 2D or 3D very simple, once the sample spectra are linked with their spatial information.
ALS offers a combined data acquisition and interpretation package (HYP-PKG), however if only the raw spectral data is needed, then TRSPEC-20 is available. Conversely, if only interpretation is needed and the raw spectra file can be provided, then INTERP-11 is available.
The HYP-PKG package code includes the TRSPEC-20 and INTERP-11 procedures together, to provide a cost-effective solution for the acquisition and interpretation of VIS-SWIR spectral data.
An economical package combining TerraSpec® 4 HR scanning and aiSIRISTM expert spectral interpretation.
The value of hyperspectral mineralogy in exploration and geometallurgy increases substantially with larger sample volumes. Discounts are available for large submittals covering entire drilling campaigns.
|Raw spectral files in ASD or ASCII format, and spreadsheet with mineral assemblages and spectral parameters related to the project geology.|
|INTERP-11||Rapid and accurate interpretation of hyperspectral scans by the aiSIRISTM expert software.||Spreadsheet with mineral assemblages and spectral parameters related to the project geology.|
|TRSPEC-20||Spectral scan using the TerraSpec® 4 HR spectrometer. Crushed reject or RC chips are recommended as the optimal sample type. *For pulverised samples request TRSPEC-21.||Raw spectral files in ASD or ASCII format.|
Many minerals that form during hydrothermal alteration events can be identified using hyperspectral reflected light, particularly in the short-wave infra-red (SWIR). Several relevant minerals associated with other deposit types can also be identified, and the technique can also be applied in deeply weathered regolith environments, where it can be used to interpret paleo water tables, redox fronts, and transported-residual regolith boundaries. Objective mineral identification greatly assists in mapping mineral distribution. Mineral maps can be used to guide exploration activities, or to inform geometallurgical models to optimise mine performance.
However, not all minerals have diagnostic features in the visible and short-wave infra-red (VIS- SWIR) wavelength ranges, and so they cannot be identified. Common examples of minerals that lack diagnostic features in the VIS-SWIR wavelength range include feldspars, quartz, phosphates, spinels and sulphides.
It is also important to emphasise that this type of spectral scanning provides a qualitative output (i.e. whether a mineral is present or not). In some cases, the strength of a spectral feature may correlate with the modal abundance in a sample, however this is not uniform across all minerals. Consequently, the concept of “spectral abundance” is sometimes used to highlight mineral phases that are returning a relatively strong signal.
Yes, however these sample forms are not considered optimal. With core, there is often subjectivity about where to sample as there can be a lot of variation in 1m of core. Pulp material, while certainly homogeneous and representative for most other analytical methods, tends to suffer from light scattering and a relatively high albedo. This means that the signal to noise ratio tends to be lower than for drill chips or crushed material, and spectral features become more difficult to identify and extract. Drill chips/crushed material often represent the best compromise.
Interpreted outputs following spectral data collection include spectral mineral matches and
calculated indices. The spectral mineral match is basically where the spectra of a sample is
compared to known materials in a database, and the best or most likely mineral in the sample is
estimated. Often, the two closest mineral matches can be provided for both the VIS and SWIR
regions. Of course, this only applies to the minerals having diagnostic features in the VIS or
Spectral indices are typically numerical outputs that relate to things such as the wavelength of a spectral feature (e.g. at peak maxima), the width of a feature, the depth of a feature or the ratio of one feature to another. These indices can be used to infer the composition of some minerals (e.g. Fe-rich, intermediate or Mg-rich chlorites), relative spectral abundance, and even the amount of water incorporated into the crystal structure. In turn, this data may sometimes be used to infer the likely character of the hydrothermal fluid from which they formed, including relative temperature, pH and redox potential.