Comparing Different Approaches to Determining the Bulk Composition and Phase Proportions of Exsolved Oxides
Bibliographic record
Abstract
Igneous magnetite and ilmenite from plutonic rocks commonly show microtextures that reflect the progress of magmatic evolution and subsolidus re-equilibration. To investigate the trend of magmatic oxide crystallization, a bulk composition prior to subsolidus modification needs to be established. The scale of these microtextures, typically consisting of various generations of exsolution, can be smaller than the resolution of electron microprobe analysis. Element zoning, presence of alteration phases, and sample surface imperfections (e.g., voids, relief) can further contribute to obtaining inaccurate results. Broad or defocused beam analysis of such heterogeneous phases to acquire “homogenized” compositions is still commonly applied despite the analytical errors known to be associated with this practice [1, 2]. Combining single spot analyses with modal abundances is an alternative approach to calculating a bulk composition. Quantitative X-ray mapping where each pixel is fully quantified [3] has the advantage of preserving spatial context and allows filtering of mapping artifacts [4, 5]. In this work, we applied different analytical approaches to estimating the bulk compositions and phase proportions in exsolved Fe-Ti oxides from the Skaergaard intrusion, East Greenland, to compare the robustness and advantages and disadvantages of each approach. We acquired X-ray mapping data and spot analyses of magnetite on a JEOL iHP200F field emission electron microprobe, equipped with five WDS spectrometers and two EDS spectrometers (Bruker XFlash 6130) at the University of British Columbia. Analytical conditions used were an accelerating voltage of 12kV to improve spatial resolution and a beam current of 100 nA with a focused beam for X-ray mapping. We analyzed for Fe kα, Ti kα, Al kα, V kα and O kα, using mean atomic number backgrounds [6]. All X-ray data were quantified using the PROZA matrix correction algorithm [7], and interference corrections were applied for Ti kα overlap on O kα and V kα (Figure 1). In addition, we acquired defocused beam analyses using the same analytical conditions on all mapping locations with beam sizes defocused to 25 microns and 50 microns for comparison. Where possible, we also acquired focused spot analyses of the oxide phases. Different clustering approaches were used to estimate modal abundances within exsolved magnetite including JEOL Phase Map Maker and Phase Analysis Program, CalcImage (Hartigan-Wong k-means clustering, Probe Software, Inc.), and Fiji’s Xlib plugin for unsupervised clustering [8, 9]. We also used AMICS, Bruker’s automated mineral analysis and characterization software, to create high-resolution phase maps. AMICS utilizes advanced machine vision technology to segment backscatter-electron images and acquire EDS spectra for each segment. EDS spectra can be matched automatically to a database of reference phases until all phases are identified (Figure 2). The most robust results were determined using quantitative X-ray mapping with subsequent filtering to remove artifacts. This approach adequately handles compositional zoning and has the added benefit of preserving the spatial context of compositional variation in exsolved Fe-Ti oxides. Example of fully quantified element maps of magnetite with ilmenite oxyexsolution lamellae. Each pixel corresponds to a complete analysis as shown by element totals close to 100 wt% across the map. Phase characterization in AMICS segmentation mode. BSE image is acquired (left) and segmented based on gray level variation, detecting very subtle variations in the image (middle). EDS spectra are acquired for each segment and used to construct a highly detailed phase map (right) (green: magnetite; purple: ilmenite; gray: pleonaste; white: voids).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".