4.5 Article

Accurate determination of ferric iron in garnets by bulk Mossbauer spectroscopy and synchrotron micro-XANES

Journal

AMERICAN MINERALOGIST
Volume 97, Issue 10, Pages 1726-1740

Publisher

MINERALOGICAL SOC AMER
DOI: 10.2138/am.2012.4107

Keywords

Garnet; almandine; andradite; grossular; Mossbauer spectroscopy; XANES; ferric iron

Funding

  1. NSF [EAR-0439161, EAR-0809459, EAR-0809253, EAR-08088899]
  2. Department of Energy (DOE)-Geosciences [DE-FG02-92ER14244]
  3. DOE [DE-AC02-98CH10886]
  4. Directorate For Geosciences
  5. Division Of Earth Sciences [1219761] Funding Source: National Science Foundation

Ask authors/readers for more resources

Fe3+/Sigma Fe Measurements of Fe3+/Sigma Fe in geological materials have been intractable because of lack of access to appropriate facilities, the time-consuming nature of most analyses, and the lack of precision and reproducibility in most techniques. Accurate use of bulk Mossbauer spectroscopy is limited by largely unconstrained recoilless fraction (f), which is used to convert spectral peak area ratios into valid estimates of species concentrations and is unique to different mineral groups and compositions. Use of petrographic-scale synchrotron micro-XANES has been handicapped by the lack of a consistent model to relate spectral features to Fe3+/Sigma Fe. This paper addresses these two deficiencies, focusing specifically on a set of garnet group minerals. Variable-temperature Mossbauer spectra of the Fe2+-bearing almandine and Fe3+-bearing andradite end-members are used to characterize f in garnets, allowing Fe3+/Sigma Fe to be measured accurately. Mossbauer spectra of 19 garnets with varying composition were acquired and fit, producing a set of garnet-specific standards for Fe3+ analyses. High-resolution XANES data were then acquired from these and 15 additional previously studied samples to create a calibration suite representing a broad range of Fe3+ and garnet composition. Several previously proposed techniques for using simple linear regression methods to predict Fe3+/Sigma Fe were evaluated, along with the multivariate analysis technique of partial least-squares regression (PLS). Results show that PLS analysis of the entire XANES spectral region yields the most accurate predictions of Fe3+ in garnets with both robustness and generalizability. Together, these two techniques present reliable choices for bulk and microanalysis of garnet group minerals.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.5
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available