BGR Bundesanstalt für Geowissenschaften und Rohstoffe

Go to:

AFP Method

Multi-disciplinary research in the frame of the AFP project on a wide range of ore concentrate samples has revealed the following key parameters which can be used to verify the documented origin of an ore concentrate sample in question:

  • Geochemical composition (major and trace elements) of ore minerals
  • Geological age of ore minerals
  • Mineralogical composition and variability of ore concentrates.

In the LA-ICP-MS laboratoryIn the LA-ICP-MS laboratory Source: BGR

A streamlined analytical protocol has been established to obtain the analytical data necessary for AFP:

  • Preparation of polished sections from mineral concentrates
  • Quantification of mineral proportions in ore concentrates using scanning electron microscopy (SEM) combined with automated mineralogy.
  • Determination of the chemical composition and of uranium-lead isotopes of about 50 individual ore mineral grains representative for a given concentrate sample using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS).

Analytical data obtained from “reference samples” of known origin (mine) are stored in a database. The claimed origin of a “control sample” taken along a supply chain (from mine to exporter) may be verified by cross-referencing its compositional data with the information stored in the reference database using statistical methods. Further details of the AFP method are given in the AFP manual.


Sampling

AFP sampling, Kibuba, BurundiAFP sampling, Kibuba, Burundi Source: BGR

Ore pre-concentrates directly sampled on-site are used as reference samples for AFP. Sampling is done by accredited samplers according to specific standards. The precise position of the sample location has to be determined by GPS and detailed site information needs to be documented. Pre-concentrates provide a sufficient number of ore mineral grains for analysis, still reflecting the characteristic features of a deposit including deposit-specific gangue or accessory minerals. A single AFP sample ideally consists of 50-200g pre-concentrate material depending on the grain size. One or more samples need to be taken at each individual production site.







Sample Preparation

Preparation of a polished sectionPreparation of a polished section Source: BGR

An aliquot of the AFP sample is mounted in epoxy resin. After the resin is cured, a polished section is prepared by grinding and polishing with gradually smaller grain sizes of the abrasives until a uniformly flat, scratch- and relief-free, highly reflective surface is obtained. This polished section is subsequently used for scanning electron microscopy (SEM) and laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) analysis.






Analysis

Scanning electron microscope (SEM)Scanning electron microscope (SEM) Source: BGR

Scanning electron microscopy (SEM) combined with automated mineralogy is used to obtain modal mineralogy as well as the mineralogy of each individual grain on the surface of the polished section. The latter is used to define about 50 ore mineral grains which are analyzed by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) to obtain their chemical compositions and an age estimate of the sample.
The analytical results (geochemical, geochronological, and mineralogical data) generated by SEM and LA-ICP-MS are stored in a database managed by BGR and the AFP Management Unit in Bujumbura.




Reference:
Gäbler, H.-E., Melcher, F., Graupner, T., Bahr, A., Sitnikova, M.A., Henjes-Kunst, F., Oberthür, T., Brätz, H., Gerdes, A., 2011. Speeding Up the Analytical Workflow for Coltan Fingerprinting by an Integrated Mineral Liberation Analysis/LA-ICP-MS Approach. Geostandards and Geoanalytical Research 35, 431-448.

Data evaluation

The application of AFP is to check the documented origin of a mineral shipment in case of doubt. A sample is taken from this shipment in doubt, the sample is analyzed and the analytical results are evaluated by comparison with data from a reference sample database where mine specific information on ore minerals is stored. The final result is a statement whether the documented origin of the shipment in doubt is credible or not.
The evaluation of the data is done by statistical methods which take into account typical properties of ore concentrates like:

  • huge variability, e.g. in trace element compositions,
  • not normal or lognormal distribution of the data,
  • the fact that ore mineral concentrates are not representative aliquots of an ore body in a statistical sense, and
  • the possible occurrence of several geochemically different ore bodies at a single mine site.

In order to meet these requirements, methods are used which are for example based on the comparison of statistical distributions of sample properties (e.g. trace element concentrations). In collaboration with experts from the fields of forensic sciences, informatics, statistics, and geology these methods are continuously improved and cross-checked.

References:
Gäbler, H.-E., Rehder, S., Bahr, A., Melcher, F., Goldmann, S., 2013. Cassiterite fingerprinting by LA-ICP-MS. Journal of Analytical Atomic Spectrometry 28, 1247-1255.

Gäbler, H.-E., Schink, W., Goldmann, S., Bahr, A., Gawronski, T., 2017. Analytical fingerprint of wolframite ore concentrates. Journal of Forensic Sciences, 62/4, 881-888.

Martyna, A., Gäbler, H.-E., Bahr, A., Zadora, G., 2018. Geochemical wolframite fingerprinting – the likelihood ratio approach for laser ablation ICP-MS data. Analytical and Bioanalytical Chemistry 410, 3073-3091. - URL: https://link.springer.com/article/10.1007/s00216-018-1007-9

Gäbler, H.-E., Schink, W., Gawronski, T., 2020. Data evaluation for cassiterite and coltan fingerprinting. Minerals, 10, 926. – URL: https://www.mdpi.com/2075-163X/10/10/926

Imprint

This Page: