Cathodoluminescence Microscopy of Zircon in HP- and UHP-Metamorphic Rocks: A Fundamental Technique for Assessing the Problem of Inclusions versus Pseudo-Inclusions

This paper shows how a faulty approach to the study of mineral inclusions in zircon can lead to misleading interpretations of the geological context. We present and discuss two well-documented examples. Zircon grains separated from HP metamorphic jadeitite of the Rio San Juan Complex, Dominican Republic, and from UHP pyrope quartzite of the Dora Maira Massif, northern Italy, were studied using cathodoluminescence (CL) techniques, in combination with mineral inclusion and age data. In general, zircon from both localities shows inherited magmatic core domains with oscillatory zoning and metamorphic rims. The magmatic cores of zircon from the jadeitite yield ages of 115–117 Ma and host jadeite and omphacite which are of metamorphic origin and formed at about 78 Ma. Zircon from lawsonite blueschist, representing the country rock of the jadeitite, contains domains with oscillatory zoning that are nearly identical in age to the zircon cores from the adjacent jadeitite, and also contains younger metamorphic minerals such as lawsonite, albite, phengite (Si 3.68 ), chlorite, and omphacite. Similar observations were made on the magmatic cores of zircon from the pyrope quartzite. These are about 275 Ma in age and host pyrope, phengite (Si 3.55 ), talc, and kyanite, all of which formed during UHP metamorphism at about 35 Ma. Zircon from the biotite-phengite-gneiss country rock (metagranite) shows oscillatory zoning and yields ages that are identical to those of the magmatic cores of zircon from pyrope quartzite, which thus reflect granitic intrusion ages. The country-rock zircon also encloses metamorphic minerals with ages of about 35 Ma. Such minerals are, for example, garnet and phengite, as well as a polymineralic assemblage of clinopyroxene+garnet+phengite+quartz, that point to formation at UHP metamorphic conditions around 40 kbar/750 °C. Based on these examples we suggest an effective approach centered on key evidence from CL studies to show that magmatic domains of zircon may actually contain pseudo-inclusions which were not entrapped during an early stage of formation, but were instead introduced during later metamorphic or metasomatic events along microcracks representing pathways for fluid influx. Cathodoluminescence microscopy is thus an excellent tool for avoiding such pitfalls by allowing distinction between true inclusions and pseudo-inclusions in zircon.