Trace element incorporation into quartz: A combined study by ICP-MS, electron spin resonance, cathodoluminescence, capillary ion analysis, and gas chromatography

Abstract

Pegmatite quartz from different occurrences in Norway and Namibia was investigated by a combination of ICP-MS, Electron Spin Resonance (ESR), Capillary Ion Analysis (CIA) and Gas Chromatography (GC) to quantify trace elements in very low concentrations and to determine their position in the quartz structure. The studied quartz samples show similar geochemical characteristics with low contents of most trace elements. Remarkable are the elevated concentrations of Al (36–636 ppm), Ti (1.6–25.2 ppm), Ge (1.0–7.1 ppm), Na (5.2 to >50 ppm), K (1.6 to >100 ppm) and Li (2.1–165.6 ppm). These elements are preferentially incorporated into the quartz lattice on substitutional (Al, Ti, Ge) and interstitial (Li, Na, K) positions. Li + was found to be the main charge compensating ion for Al, Ge and Ti, whereas some ppm of Na and K may also be hosted by fluid inclusions. Ti may be incorporated as substitutional ion for Si or bound on mineral microinclusions (rutile). The results of the ESR measurements show that there may be a redistribution of alkali ions during irradiation. The diamagnetic [AlO 4/M +] 0 center transforms into the paramagnetic [AlO 4] 0 center, whilst the compensating ions diffuse away and may be captured by the diamagnetic precursor centers of [GeO 4] 0 and [TiO 4] 0 to form paramagnetic centers ([TiO 4/Li +] 0, [GeO 4/Li +] 0). In general, fluid inclusions in pegmatite quartz can be classified as H 2O-CO 2-NaCl type inclusions with water as the predominant volatile. Among the main elements hosted by fluid inclusions in quartz are Na, K, NH 4, Ca, Mg and the anionic complexes Cl −, NO 3 −, HCO 3 − and SO 4 2−. Gas analysis of trapped fluids shows volatile components in the following order of abundance: H 2O > CO 2 > N 2(+) ≥ CH 4 > COS > C 2 and C 3 hydrocarbons. Additionally, traces of Co, Ni, Zn, Pb, and Cu were detected by CIA in fluid inclusions of some samples. There are indications that the REE and Rb are also bound in fluid inclusions, however, the concentrations of these elements are too low to be measured by CIA. Assuming that the REE preferentially occur in fluid inclusions, the typical chondrite normalized REE distribution patterns with tetrad effects and a distinct negative Eu anomaly would reflect the composition of the mineralizing fluid. For a number of elements, especially those with extremely low concentrations, the “type” of incorporation in quartz could not directly be determined. We conclude that these ions either are too large to substitute for the small Si 4+ ion or they are not soluble in the mineralizing fluids to be hosted by fluid inclusions. Some of these elements, which are concentrated in the specific mineralization of certain pegmatites, are not present in elevated concentrations in the paragenetic pegmatite quartz itself. This was observed, for instance, for Be, Cs and Rb in the Li (Be-Cs-Rb) pegmatites of Rubicon or for Nb and Ta for Nb-Ta bearing pegmatites from Norway. It may be concluded that the concentrations of these trace elements in quartz do not reflect the mineralization and that these elements thus, cannot be used as petrogenetic indicator.