OCCURRENCE OF PRIMARY MAGMATIC COPPER IN PLAGIOCLASE MEGACRYSTS OF BASALTIC LAVAS, OREGON

Mafic igneous rocks are commonly associated with various metals and there are many examples of basalts that host small amounts of native copper. Such copper is more common among the matrix phases of basalts and could represent a final stage of a melt or a secondary alteration. Worldwide, there are only a few occurrences of copper within early-crystallizing phenocrysts of plagioclase, and this fact makes the ‘sunstones’ of Oregon as interesting as they are spectacularly beautiful. As we reported earlier (Badur et al., 2020; GSA Abstracts with Programs) native Cu exists in the 'sunstones' as thin platelets with crystallographically regulated orientations (‘copper schiller’) which tend to form through exsolution and are predominantly located in the cores of the highest quality gemstones. These labradorite megacrysts (~An₆₇) have strikingly homogeneous major and trace element distributions and internally homogeneous ⁸⁷Sr/⁸⁶Sr ratios, similar to those observed in plagioclase phenocrysts of the Columbia River Basalt Group's Steens Basalt. The homogeneous nature of all these data suggests that after crystallization, the crystals have not experienced the significant chemical change (diffusive mass transport, alteration, weathering). Recent EMPA measurements at Auburn University show that more finely grained crystals of matrix plagioclase in a typical tholeiitic texture show an expected enrichment of potassium from their cores to their rims, with finely grained phases in the mesostasis of the basalts that are very rich in K and Si (they appear to be sanidine and quartz). This petrography is consistent with the growth of a late generation of plagioclase and matrix phases during fractional crystallization and eruption of the basaltic lavas. Encouraged by these observations, we are preparing samples of phenocrysts and matrix for separate laser ⁴⁰Ar/³⁹Ar age studies in the ANIMAL facility and hope to present these results later this year, with the prediction that the matrix phases will give a more faithful record of the timing of basalt eruption. As to the overall model of sunstone development, we propose that throughout their initial crystallization, copper was dispersed homogeneously, and subsequent cooling was adequate at a rate that allowed Cu exsolution but was too rapid for major and trace elements to diffuse.