Oxygen Isotopic Compositions of San Carlos Olivine Standard NMNH 111312–42

Abstract

San Carlos olivine geostandard developed by Gene Jarosewich and co-workers at the Smithsonian Institution, named NMNH 111312–44, was the most widespread olivine geostandard used in olivine oxygen isotopic determination. However, the NMNH 111312–44 olivine grains have run out completely. Another set of San Carlos olivine, whose grains are larger, named NMNH 111312–42, was distributed by the Smithsonian Institution. Whether they have the same oxygen isotopic composition is still unclear. This study analyzed the oxygen isotopic compositions of NMNH 111312–42 and NMNH 111312–44, using a CAMECA IMS–1280 secondary ion mass spectrometer. The results show that NMNH 111312–42 (5.28 ± 0.17‰, 1SD, N = 35) and NMNH 111312–44 (5.22 ± 0.16‰, 1SD, N = 29) olivine grains have indistinguishable δ¹⁸O values within analytical error, suggesting that NMNH 111312–42 olivine grains can be taken as the same geostandard as NMNH 111312–44 olivine grains for high-precision oxygen isotope analysis.

1. Introduction

Natural forsterite-rich olivine grains have been used as elemental and isotopic geostandards during in situ analyses of olivine samples [1,2]. Among them, the San Carlos olivine standard developed by Gene Jarosewich and co-workers [3] at the Smithsonian Institution, named NMNH 111312–44, is the most widespread olivine geostandard.

The NMNH 111312–44 (mineral size: 0.250 mm–0.177 mm) was mined in the San Carlos Apache Reservation, which is one of Earth’s richest resources for olivine-rich rocks. It is located in San Mateo County, California, United States. Fournelle [4] suggested that the composition of the NMNH 111312–44 shows only slight major element variability (Fo89.6 to Fo90.5). Jarosewich et al. [3] calculated the ‘Boyd homogeneity indices’ for NMNH standards by measuring 100 spots on 10 grains of each standard. The range of heterogeneity in Si, Fe, and Mg was 0.81, 0.90, and 1.00, respectively, which were considered to be homogeneous (values < 3). The δ¹⁸O of NMNH 111312–44 was also considered to be homogeneous with the majority of values falling within the range of 5.2‰ to 5.3‰ (relative to VSMOW) [5,6,7].

Due to its homogeneity, a large number of labs have taken it as standard in analyses of geological samples. In particular, it was often used as reference material in olivine oxygen isotope determination. However, the NMNH 111312–44 olivine grains (referred to here as –44) have been completely consumed. Another San Carlos olivine, whose grains are larger (mineral size: 0.500 mm–0.250 mm), named NMNH 111312–42 (referred to as –42), was distributed by the Smithsonian Institution. Recent studies have regarded –44 and –42 as the same geostandard (e.g., Lambart. et al. [5]). Although they are chemically indistinguishable in terms of major and trace elements (Figure 1), whether they have the same oxygen isotopic composition is still unclear. In this study, the oxygen isotopic compositions of –42 and –44 were measured simultaneously and compared to determine whether –42 could replace –44 as the same geostandard for in situ oxygen isotope analysis.

2. Materials and Methods

Secondary ion mass spectrometer (SIMS) analysis is an in situ analytical technique for measuring trace element abundance and isotopic ratios of geological samples at μm size. Along with the development of this analytical method, in situ δ¹⁸O measurements have been extensively applied on silicate minerals and glasses (e.g., zircon, olivine, and quartz) [8]. During in situ analysis, standard samples were used to calibrate δ¹⁸O results of unknown samples. Therefore, standard samples are crucial in the application of in situ oxygen isotopic analysis.

This study analyzed the oxygen isotopic compositions of –42 and –44, using a CAMECA IMS–1280 secondary ion mass spectrometer at the Institute of Geology and Geophysics, Chinese Academy of Sciences, China. The CAMECA IMS-1280 SIMS is a large-geometry, double-focusing instrument, which has the same radius (585 mm) for both the electrostatic sector analyzer (ESA) and the sector magnet. It is equipped with two primary ion sources, the Duoplasmatron source, which uses oxygen gas, and the cesium ion source [9]. The Gaussian primary beam size could be adjusted to determine the δ¹⁸O value. Randomly selected –42 and –44 olivine grains, which were both provided by the Smithsonian Institution, were mounted in epoxy resin and polished. The polished mount is cleaned in high-purity ethanol using an ultrasonic bath and then vacuum-coated with high-purity gold prior to the SIMS analysis [10]. The oxygen isotopes of both –42 and –44 olivine grains were measured in two sessions (July and October 2020) to check their homogeneity and consistency. The instrument configuration and analytical protocol are similar to that of Tang et al. [11]. The Cs⁺ ions were used as the primary beam and worked in Gaussian mode to sputter olivine for oxygen isotope analysis. The primary beam size was ~15 μm in diameter, and the intensity was 3.8–6.5 nA. The primary beam worked in raster mode with an area of 10 × 10 μm² during analysis. The ¹⁶O and ¹⁸O ions were detected simultaneously by two Faraday cups, and the signals were amplified by 109 ohms and 1012 ohms resistors, respectively. A normal electron gun was used to compensate for the charging effect in the bombarding area. The width of the entrance slit was ~200 μm; the size of the field aperture was 5000 × 5000 μm²; the width of the energy slit was 40 eV. The magnification of the transfer system was configured as ~133. Each analysis consists of pre-sputtering, beam centering, and signal collecting processes. The collecting process consists of 12 cycles, and each cycle lasts 4 s. Typical ¹⁸O peak ion intensities of 1.1–1.8 × 10⁷ counts per second yield an internal 1σ-uncertainty of less than 0.1‰.

The measured ¹⁸O/¹⁶O ratios were normalized using Vienna Standard Mean Ocean Water compositions (VSMOW, ¹⁸O/¹⁶O = 0.00200520) [6]:

$${\left(\mathsf{\delta }{}_{}{}^{18}\mathrm{O}\right)}_{\mathrm{M}}=\left({\displaystyle \frac{{\left({}_{}{}^{18}\mathrm{O}/{}_{}{}^{16}\mathrm{O}\right)}_{\mathrm{M}}}{0.0020052}}-1\right)\times 1000‰$$

(1)

3. Results

Oxygen isotope measurements were carried out in two sessions to evaluate the homogeneity of both –42 and –44. The first session performed as individual measurements on eight –42 grains and six –42 grains, while the second session consisted of twenty-seven –42 grains and twenty-three –42 grains. These measurements aimed to assess the homogeneity and consistency of the δ¹⁸O values within and between the two sets of San Carlos olivine. The δ¹⁸O values of –42 grains ranged from 4.80‰ to 5.61‰, while that for –44 ranged from 4.89‰ to 5.53‰. The average δ¹⁸O values for –42 olivine grains in the two sessions were (5.32 ± 0.13)‰ (1SD, N = 8) and (5.26 ± 0.18)‰ (1SD, N = 27), respectively. The average δ¹⁸O values for –44 olivine grains were (5.17 ± 0.07)‰ (1SD, N = 6) and (5.24 ± 0.18)‰ (1SD, N = 23), respectively. Although the 1SD values in the second session showed a slightly wider spread than in the first session, they were still well within the acceptable limits for homogeneity evaluation. By combining all the analysis results, the average δ¹⁸O value for –42 olivine grains was calculated to be (5.28 ± 0.17)‰ (1SD, N = 35), while that for –44 olivine grains was (5.22 ± 0.16)‰ (1SD, N = 29) (

Table 1. The δ¹⁸O values of NMNH 111312–44 and NMNH 111312–42.

Sample	δ18O	2SE	Average δ18O
Sancarlos42@1 Sancarlos42@2 Sancarlos42@3	5.31	0.08	5.28 ± 0.17
	5.36	0.13
	5.06	0.07
Sancarlos42@4 Sancarlos42@5	5.36	0.11
	5.32	0.12
Sancarlos42@6 Sancarlos42@7 Sancarlos42@8 A5755-San42@1	5.38	0.11
	5.50	0.10
	5.27	0.11
	5.49	0.04
A5755-San42@2 A5755-San42@3	5.32	0.05
	5.55	0.07
A5755-San42@4	5.25	0.09
A5755-San42@5	5.47	0.03
A5755-San42@6	5.36	0.08
A5755-San42@7	5.49	0.08
A5755-San42@8	5.61	0.08
A5755-San42@9	5.50	0.11
A5755-San42@10	5.33	0.09
A5755-San42@11	5.18	0.08
A5755-San42@12	5.36	0.07
A5755-San42@13	5.11	0.08
A5755-San42@14	5.21	0.08
A5755-San42@15	5.25	0.09
A5755-San42@16	5.36	0.12
A5755-San42@18	5.16	0.10
A5755-San42@19	5.24	0.08
A5755-San42@20	5.22	0.08
A5755-San42@21	5.27	0.05
A5755-San42@22	5.04	0.10
A5755-San42@23	5.05	0.10
A5755-San42@24	4.80	0.06
A5755-San42@25	5.17	0.07
A5755-San42@26	5.10	0.09
A5755-San42@27	5.04	0.12
Sancarlos44@1	5.05	0.14	5.22 ± 0.16
Sancarlos44@2	5.17	0.12
Sancarlos44@3	5.18	0.13
Sancarlos44@4	5.11	0.14
Sancarlos44@5	5.24	0.13
Sancarlos44@6	5.24	0.07
A5755-San44@1	5.07	0.05
A5755-San44@2	5.53	0.08
A5755-San44@3	5.35	0.06
A5755-San44@4	5.53	0.08
A5755-San44@5	5.38	0.06
A5755-San44@6	5.41	0.05
A5755-San44@7	5.25	0.08
A5755-San44@8	5.35	0.08
A5755-San44@9	5.06	0.06
A5755-San44@10	5.03	0.10
A5755-San44@11	5.01	0.08
A5755-San44@12	4.89	0.09
A5755-San44@13	5.05	0.09
A5755-San44@14	5.22	0.10
A5755-San44@15	5.27	0.07
A5755-San44@16	5.18	0.09
A5755-San44@17	5.32	0.11
A5755-San44@18	5.47	0.05
A5755-San44@19	5.28	0.10
A5755-San44@20	5.13	0.11
A5755-San44@21	5.40	0.07
A5755-San44@22	5.22	0.07
A5755-San44@23	5.04	0.08

).

4. Discussion

4.1. Homogeneity of San Carlos Olivine Grains

The homogeneity of geostandard material in oxygen isotope compositions is crucial in determining whether it can be used as a geostandard material for SIMS oxygen isotope analysis [12]. It can be assessed by evaluating the standard deviation (1SD) of the measured δ¹⁸O values. The 1SD of δ¹⁸O for different –42 olivine grains in the two sessions were 0.13‰ (N = 8) and 0.18‰ (N = 27), respectively. These relatively small standard deviations indicate that the variations in oxygen isotope compositions among the different –42 grains are minimal and well within acceptable limits. Such low variability demonstrates the homogeneity of the –42 grains, suggesting that they are isotopically consistent and suitable for use as a geostandard material in SIMS analysis. Despite the small difference observed between the average δ¹⁸O values of the –42 and –44 olivine grains, they are indistinguishable within analytical error (Figure 2). The 1SD values for the –44 olivine grains (0.07‰ for N = 6 in the first session and 0.18‰ for N = 23 in the second session) also suggest that these grains exhibit similar levels of isotopic homogeneity as the –42 grains. Therefore, the –42 olivine grains could be taken as the same geostandard material for the SIMS oxygen isotope analysis as the –44 olivine grains.

4.2. The Accuracy and Precision for Oxygen Isotope Analysis

The oxygen isotopic compositions of olivine are critical for understanding geological and planetary processes, providing insights into research such as, but not limited to, mantle dynamics and crustal evolution [13]. An accurate determination of olivine’s oxygen isotope values is essential for utilizing these data in tracing geological processes. Calibration during analysis relies on standard samples like the San Carlos olivine geostandard. Therefore, the uniformity of San Carlos olivine and potential differences between –42 and –44 olivine grains can significantly influence calibration accuracy.

The precision required for oxygen isotope analysis varies depending on the geological application. For instance, oxygen isotope fractionation between olivine and basaltic melts at mantle temperatures is typically limited to ≤1‰, while fractionations between olivine and coexisting minerals such as orthopyroxene and spinel are smaller than 1.5‰ at magmatic temperatures [14]. These small fractionations highlight the importance of high-precision measurements in resolving subtle isotopic variations. Oxygen isotopic compositions of olivine have revealed compositional heterogeneity in Earth’s deep mantle, attributed to the subduction of oceanic slabs over 3.3 billion years ago [15]. In that study, two groups of olivine with distinct oxygen isotope compositions (4.9–5.4‰ and 3.6–4.7‰) were identified. Accuracy requirements for such analyses typically range from ±0.10‰ to ±0.60‰, with ±0.20‰ sufficing in most cases. Even considering a slight δ¹⁸O discrepancy (0.06‰) between –44 and –42 olivine grains, this is well within the acceptable precision for igneous rock analysis (±0.2‰ to ±0.3‰ [16]).

4.3. Comparison with Other Olivine Reference Materials

Compared to –44, the larger size of –42 olivine grains offers practical advantages for in situ analysis by accommodating more analytical points per grain, reducing variability introduced by inter-grain differences, and minimizing standard sample waste. These larger grains also reduce errors introduced by inter-grain variability, further enhancing analytical reliability.

In recent years, some new olivine standard samples for oxygen isotope analysis have been developed (

Table 2. The information of several new olivine standard samples.

Sample	Locality	Average δ18O (1SD)
06JY06OL	Jingyu, Northeast China	5.20 ± 0.03‰
06JY29OL		5.30 ± 0.13‰
06JY31OL		5.27 ± 0.15‰
06JY34OL		5.25 ± 0.07‰
09XDTC1-24OL	Eastern Tianshan, NW China	3.91 ± 0.25‰
JAY03-3	Jilin province, Northeast China	16.37 ± 0.11‰
JAY02-4		18.29 ± 0.14‰

The data were obtained from Tang et al. [17] and Peng et al. [18].

). Tang et al. [17] evaluated the potential as SIMS reference materials of five olivines from mantle peridotite xenoliths and dunite. The obtained results revealed that the oxygen isotopic values of these samples are homogeneous and suitable to be used as reference materials for in situ oxygen isotope microanalysis. The recommended δ¹⁸O value for these five samples ranged from (3.91 ± 0.25)‰ to (5.30 ± 0.13)‰. Peng et al. [18] developed two new olivine geostandard samples from the Jian forsterite jade deposit and examined the homogeneity of their oxygen isotope. The recommended δ¹⁸O values are (16.37 ± 0.11)‰ and (18.29 ± 0.14)‰, respectively. Notably, these two types of olivine samples have extremely high magnesium contents (Fo > 99), making them potential reference materials only for SIMS oxygen isotope analysis of high-magnesium olivine.

When considered alongside the recent advancements in olivine geostandards, the –42 olivine grains represent a valuable addition to the suite of available geostandards. These materials collectively provide researchers with a diverse and reliable set of tools for precise and accurate oxygen isotope analysis, advancing our understanding of geochemical processes across a wide range of geological environments.

5. Conclusions

The results of this study show that the oxygen isotope composition of NMNH 111312–42 is homogeneous. Given the requirement for accuracy in olivine oxygen isotopic analysis, the olivine geostandard NMNH 111312–42 can be taken as the same geostandard as NMNH 111312–44 for SIMS oxygen isotope analysis. Given the larger size, NMNH 111312–42 could be even more suitable as the geostandard than NMNH 111312–44 in in situ oxygen isotope microanalysis. NMNH 111312–42 could be a valuable tool for advancing the accuracy and reliability of oxygen isotope studies in olivine, supporting a wide range of geological and geochemical research applications.