Fractionation of Oxygen Isotopes by Thermal Ionization Mass Spectrometry Inferred from Simultaneous Measurement of (17)O/(16)O and (18)O/(16)O Ratios and Implications for the (182)Hf-(182)W Systematics.

Accurate (182)Hf-(182)W chronology of early planetary differentiation relies on highly precise and accurate tungsten isotope measurements. WO3(-) analysis by negative thermal ionization mass spectrometry requires W(17)O(16)O2(-), W(17)O2(16)O(-), W(18)O(16)O2(-), W(17)O3(-), W(17)O(18)O(16)O(-), and W(18)O2(16)O(-) isotopologue interference corrections on W(16)O3(-) species ( Harper et al. Geochim. Cosmochim. Acta 1996 , 60 , 1131 ; Quitté et al. Geostandard. Newslett. 2002 , 26 , 149 ; Trinquier et al. Anal. Chem. 2016 , 88 , 1542 ; Touboul et al. Nature 2015 , 520 , 530 ; Touboul et al. Int. J. Mass Spectrom. 2012 , 309 , 109 ). In addition, low ion beam intensity counting statistics combined with Faraday cup detection noise limit the precision on the determination of (18)O/(16)O and (17)O/(16)O relative abundances. Mass dependent variability of (18)O/(16)O over the course of an analysis and between different analyses calls for oxide interference correction on a per integration basis, based on the in-run monitoring of the (18)O/(16)O ratio ( Harper et al. Geochim. Cosmochim. Acta 1996 , 60 , 1131 ; Quitté et al. Geostandard. Newslett. 2002 , 26 , 149 ; Trinquier et al. Anal. Chem. 2016 , 88 , 1542 ). Yet, the (17)O/(16)O variation is normally not being monitored and, instead, inferred from the measured (18)O/(16)O variation, assuming a δ(17)O-δ(18)O Terrestrial Fractionation Line ( Trinquier et al. Anal. Chem. 2016 , 88 , 1542 ). The purpose of the present study is to verify the validity of this assumption. Using high resistivity amplifiers, (238)U(17)O2 and (238)U(18)O2 ion beams down to 1.6 fA have been monitored simultaneously with (235,238)U(16)O2 species in a uranium certified reference material. This leads to a characterization of O isotope fractionation by thermal ionization mass spectrometry in variable loading and running conditions (additive-to-sample ratio, PO2 pressure, presence of ionized metal and oxide species). Proper determination of O isotope composition based on the simultaneous analysis of the (18)O/(16)O and (17)O/(16)O ratios could prevent tens of ppm bias or more on the (182)W/(184)W and (183)W/(184)W ratios.