High Sensitivity Mapping of Melt Inclusions in Miocene Zircons of Central Anatolian Volcanic Province (CAVP), Cappadocia, Turkey

Melt inclusions originate from small juvenile melt droplets trapped at magmatic pressures and temperatures during crystallization of their host mineral. Thus, melt inclusions retained by their host crystals can uniquely preserve evidence for the thermochemical conditions in the magma during crystal growth. Zircon is a resistant mineral even under magmatic conditions, and it is common in many different rock types (igneous, metamorphic, and sedimentary). Moreover, zircon crystallization significantly affects trace element concentrations of the melt during processes such as fractionation, melt separation, and/or retention of accessory phases in the residual melt. Its potential as a host mineral for melt inclusions, however, has not been fully realized, mainly because of the small size of zircon and its inclusions.

Here, we developed a new technique for ion imaging of elemental distributions in melt inclusions in zircon, and applied it to melt-inclusion bearing zircon crystals from selected Miocene ignimbrites of the Central Anatolian Volcanic Province, Turkey. The high-sensitivity ion imaging of zircon provides information about the 2-D distribution of critical elements in the crystal and its inclusions, and element distributions can be directly compared to cathodoluminescence (CL) patterns of the host. High-sensitivity element maps were obtained using a CAMECA 1280-HR IMS at Heidelberg University for areas of 25×25 µm at 2-3 µm lateral resolution. Ion images for each element containing 128×128 pixels raw intensity values were initially processed using instrumental software (WINImage©) to accumulate data from measurement cycles into a single image data. Each element map was then recorded as a grayscale image with intensities encoded in each pixel. The raster images for each element was further processed using ImageJ© and ARCGIS© programs, where each element map was converted to a color scale expressing the appropriate value ranges and the data obtained on the same trace element for each zircon in different units were reduced to the same legend values. The color ion images obtained from the grayscale images of each map were overlaid onto CL images to correlate trace element abundances with growth regions visible in CL images.

Imaging has the important advantage compared to spot analyses of melt inclusions that contamination from the wall of the host mineral can be mitigated. For this, Zr ion images were used as controls for selecting ROIs (Regions of Interest) in order to eliminate pixels with mixed signals at the interface between zircon and the inclusion due to the finite width of the ion beam. High resolution imaging of melt inclusions and zircon allowed re-evaluating zircon-melt partitioning behavior of important trace elements for natural melt compositions. Partitioning values for elements with comparatively low abundances in the melt relative to zircon (Y, Th, U and Dy) are slightly lower than spot analyses and previously published results but they all follow a similar trend with predicted partitioning coefficients. 

This research was financed by The Scientific and Technological Research Council of Turkey within the research program number of 2214/A.