Timescales of quartz crystallization estimated from glass inclusion faceting using 3D propagation phase-contrast x-ray tomography: examples from the Bishop (California, USA) and Oruanui (Taupo Volcanic Zone, New Zealand) Tuffs V31C-2801 Ayla Pamukcu ([email protected])1, Guilherme A. R. Gualda ([email protected])1, Alfred T. Anderson, Jr. ([email protected])2 1Department of Earth and Environmental Sciences, Vanderbilt University; 2Department of Geophysical Sciences, The University of Chicago BACKGROUND Glass Inclusion Faceting & Timescales RESULTS Geology/Samples Faceting & Residence Times 1.55 (b) • • Oruanui Tuff The extent to which a glass inclusion is faceted can be used as a proxy for residence time of a crystal. Quartz is ideal for studies of glass inclusion faceting because Si is the only diffusing element. Taupo Volcanic Zone, New Zealand • Eastern California, USA • Erupted ~26.5 ka, 10 phases • Erupted ~760 ka, many phases classified broadly into early- and late-erupted deposits (Wilson & Hildreth, 1997) 500 ~530 km3 of ash fall, pyroclastic density currents, and intracaldera material • • Large inclusions facet more slowly than smaller inclusions - greater volume of material to diffuse Inclusions near center of crystal will be more faceted than those at edge - included for a longer time Samples: • P1577 (ash fall) • ORN-016 (ash fall) • ORN-067 (ash fall) Propagation Phase-Contrast X-ray Tomography Propagation phase-contrast tomography provides a new way to study glass inclusions in situ and in 3D. Edgeenhancement permits quantification of glass inclusion geometry, with a slight decrease in image resolution. Images obtained here have 2.77 μm/voxel resolution, such that inclusions >10 voxels (~30 µm) in one direction can be quantitatively resolved, though smaller inclusions can be qualitatively resolved. a. b. Glass Inclusion Shapes 50 45 40 Abundance 35 d. • Samples (all early-erupted): • BB08-21b (ash fall) • BC17-Ia15 (ignimbrite) • F8-15 (ash fall) 25 P1577 n = 35 ORN-016 n = 1 ORN-067 n = 72 BB08-21b n = 69 BC17-Ia15 n = 71 F8-15 n = 109 P1577 n = 35 ORN-016 n = 1 ORN-067 n = 75 BB08-21b n = 70 BC17-Ia15 n = 72 F8-15 n = 110 20 30 25 20 15 10 0 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0+ Blue: volume representing the intersection of inclusion and ellipsoid; Red: excess volume of inclusion; Green: excess volume of ellipsoid 0 17 Aspect Ratio (Max/Min Axis) • 500 μm f. 5 5 Edge-detected absorptioncontrast tomographic image e. 3D rendering of inclusion circled in b & d. Inclusion (solid) is overlain with a fitted ellipsoid (dotted). 1.05 SNSVR = 1 104 105 106 • • Increasing aspect ratio indicates more elongation of inclusions. ORN-067 inclusions are more elongated than those in Bishop and other Oruanui samples. 20 25 30 35 40 45 50+ Equivalent Radius (μm) • • F8-15 inclusions are dominantly more spherical (peak between 1.0-1.2) and are generally more spherical than inclusions in • other Bishop and Oruanui samples. 1.15 1.05 ENSVR = 1 0.95 103 107 P1577 ORN-016 ORN-067 BB08-21b BC17-Ia15 F8-15 104 105 106 107 Inclusion Volume (μm3) Partially Faceted • Large SNSVR values indicate faceted or elongate inclusions. Large ENSVR values indicate faceted inclusions. • Generally, smaller inclusions have larger SNSVR/ENSVR values smaller inclusions are more faceted than larger inclusions. • Scatter in SNSVR values reflects more elongated shape of some inclusions (see Glass Inclusion Shapes). • All inclusions are faceted to some extent (ENSVR, SNSVR >1). • Most small inclusions are fully faceted (lie on 1:1 line), but larger inclusions are not fully faceted (lie above 1:1 line). • Residence times must be greater than those recorded by fully faceted inclusions but less than those recorded by partially faceted inclusions. 400 300 200 P1577 ORN-016 ORN-067 BB08-21b BC17-Ia15 F8-15 100 0 0 100 200 300 400 500 600 CONCLUSIONS 1. Propagation phase-contrast x-ray tomography can be used successfully to image glass inclusions in quartz crystals. Image resolution of 2.77 μm/voxel allows inclusions with a volume greater than 1000 voxels (~21,000 μm3) to be quantitatively resolved and measured. Inclusions smaller than this size can be qualitatively resolved. 2. Glass inclusions show a wide array of aspect ratio in both Oruanui and Bishop Tuff samples. Inclusions in Oruanui sample ORN-067 are more elongate in shape than in other samples from both systems. Inclusions in F8-15 are dominantly more spherical in shape and are more spherical than inclusions in other Oruanui and Bishop samples. 15 d. Edge-detected propagation phasecontrast tomographic image f. High-silica rhyolite Glass Inclusion Sizes 10 e. • RESULTS a. Standard absorption-contrast tomographic image c. 1.15 Time To Fully Facet (a) b. Standard propagation phasecontrast tomographic image c. ~1000 km3 of ash fall, ignimbrite, and intracaldera material 99% high-silica rhyolite, 1% more mafic material • 1.25 Bishop Tuff • Both size and position of a glass inclusion within a crystal influence the extent to which an inclusion will facet in a given time interval: 1.35 Inclusion Volume (μm3) 600 • 1.45 0.95 103 Above: Long Valley Caldera and earlyand late-erupted Bishop Tuff outflow deposits (from Manville & Wilson, 2004) Faceting of glass inclusions occurs over time at magmatic temperatures through dissolution and reprecipitation. This volume transported by this process is ΔV. • • Right: Oruanui ash fall deposits Time To Observed Shape (a) • Left: Oruanui ash flow deposits Abundance − ∆ = ∆ 8 Ω Ellipsoid-normalized surface area –tovolume ratio (ENSVR) Sphere-normalized surface area –tovolume ratio (SNSVR) (a) 1.25 Equivalent radius = radius of a sphere of the same volume as a given inclusion. Minimum resolvable inclusion has equivalent radius of ~17 μm. BC17-Ia15 and F8-15 have a greater abundance of inclusions than the BB08 and Oruanui samples. Sample F815 has a noticeably greater abundance of large inclusions than other samples. Size distribution and abundances of ORN-067 inclusions are similar to that of BC17-Ia15 and F8-15 samples. P1577 and BB08-21b samples have noticeably fewer inclusions. 3. Bishop Tuff samples have a greater abundance of inclusions, particularly sample F8-15, which has a strikingly large number of large inclusions. 4. SNSVR and ENSVR values for both Oruanui and Bishop samples suggest smaller inclusions are more faceted than larger inclusions and that all inclusions are faceted to some extent. 5. Many small inclusions in both the Oruanui and Bishop Tuffs are fully faceted, but larger inclusions are not yet entirely faceted. The residence times calculated from the excess volumes of the inclusions and the best fit ellipsoid (volume transported during faceting) suggest residence times of 101-103 a (including 2σ error) in both Oruanui and Bishop samples. ACKNOWLEDGEMENTS This work was funded by an NSF EAPSI to A. Pamukcu and NSF EAR-1151337 to G. Gualda. Use of the Advanced Photon Source, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science by Argonne National Laboratory, was supported by the U.S. DOE under Contract No. DE-AC02-06CH11357. Additional thanks go to M. Rivers for assistance at APS.