INWARD RADIAL MIXING OF INTERSTELLAR WATER ICES IN THE SOLAR PROTOPLANETARY DISK

Primitive meteorites (chondrites) were formed 4.56 Ga years ago by the accretion of different components produced in the accretion disk: refractory inclusions, chondrules, Fe–Ni metal, and volatile-rich matrix (Marrocchi & Libourel 2013; Scott & Krot 2014). Submillimeter water ice grains also represented a significant proportion of the building blocks of chondrites, leading to episodes of hydrothermal alteration that have strongly modified their petrography (Brearley 2006; Marrocchi et al. 2014). Two principle sources are generally considered regarding the origin of water ices in the chondrite-accretion regions, with local water being directly condensed from the gas and interstellar water coming from the coldest part of the solar system (Lunine 2006). However, the respective proportions of local and interstellar water ices incorporated along with rock during chondrite accretion are still a matter of debate (Alexander et al. 2012).

The hydrogen isotopic composition (expressed as D/H ratio) is a powerful tool for understanding the origin of water accreted by primitive meteorites (Duprat et al. 2010; Alexander et al. 2012; Piani et al. 2012, 2015; Jacquet & Robert 2013). Bulk carbonaceous chondrites are enriched by a factor of 5–10 in deuterium (120–230 × 10⁻⁶; Jacquet & Robert 2013) relative to the initial isotopic composition of the solar system (i.e., 25 × 10⁻⁶; Geiss & Gloeckler 2003). The D/H of comets is generally higher by a factor 2–3 relative to chondritic values with D/H values ranging from (161 ± 24) to (530 ± 70) × 10⁻⁶ (Altwegg et al. 2015; Marty et al. 2016). The distribution of the D/H of whole rock carbonaceous meteorites is asymmetric with a mean isotopic composition of (149 ± 3) × 10⁻⁶ and a significant tail toward high D/H ratios (Robert 2006; Jacquet & Robert 2013). Considering only hydrated CM chondrites, bulk D/H ratios present a significant D enrichment (Figure 1; mean D/H = (145 ± 3) × 10⁻⁶) relative to the putative local CM water (D/H = (83 ± 1.5) × 10⁻⁶) determined from the correlation observed in CM chondrites between D/H and C/H ratios (Alexander et al. 2012). In addition, hydrated interplanetary dust particles show (i) mineralogical similarities with CM chondrites and (ii) D/H isotopic compositions close to CMs, but with a significant tail extending toward cometary values (Bradley et al. 2007). Taken together, such features suggest that carbonaceous chondrites could have accreted variable contributions of water ices formed at different heliocentric distances (Jacquet & Robert 2013).

Zoom In Zoom Out Reset image size

Oxygen isotopic compositions of aqueously formed minerals in hydrated chondrites can be also used to better estimate the contribution of interstellar water ices in CM chondrites (Sakamoto et al. 2007; Fujiya et al. 2015). Oxygen isotopic compositions are expressed in delta units (‰) relative to the Standard Mean Ocean Water (SMOW; $\delta {}^{\mathrm{17,18}}{\rm{O}}$ =[(^17,18O/¹⁶O)_sample/(^17,18O/¹⁶O)_SMOW − 1] × 1000) and mass-independent variations of oxygen isotopes are described by the parameter ${\rm{\Delta }}{}^{17}{\rm{O}}$, defined as ${\rm{\Delta }}{}^{17}{\rm{O}}=\delta {}^{17}{\rm{O}}-0.52\times \delta {}^{18}{\rm{O}}$ (Marrocchi & Chaussidon 2015). Outer disk water ices are expected to present significant ^17,18O-rich enrichments $({\rm{\Delta }}{}^{17}{\rm{O}}\gg 0)$ relative to the terrestrial values due to the self-shielding of ¹⁶O-rich nebular CO gas by UV light (Clayton 2002; Lyons & Young 2005; Yurimoto et al. 2008). Meteoritic carbonates represent direct snapshots of the isotopic compositions of alteration fluids and can in theory be used to decipher the origin of water in the chondrite-forming region. However, water-rich chondrites are characterized by a high degree of alteration and the dissolution of primary carbonates and subsequent precipitation of secondary carbonates have removed any record of their water ice origin (Lee et al. 2014). The recent discovery of Paris—a new CM chondrite characterized by a very low degree of alteration (Hewins et al. 2014; Marrocchi et al. 2014; Pignatelli et al. 2016)—offers a unique opportunity to (i) determine the origin of water ice grains accreted in Paris and (ii) quantitatively estimate the respective contributions of D-poor and ¹⁶O-rich local inner solar system water (${\rm{\Delta }}{}^{17}{\rm{O}}\approx 0$; hereafter W-ISS, Alexander et al. 2012) and D-rich and ^17,18O-rich interstellar water (${\rm{\Delta }}{}^{17}{\rm{O}}=20$ to 90‰; hereafter W-ISM; Sakamoto et al. 2007). In this paper, we thus report the bulk D/H ratio of the Paris chondrite and the oxygen isotopic analysis of Paris Ca-carbonates. We used the data to quantitatively estimate the proportion of interstellar ices in the inner solar system during the accretion of hydrated chondrites.

2.1. Water Content and Bulk D/H Isotopic Composition

2.2. In Situ Oxygen-isotope Measurements

The three aliquots of Paris show reproducible water content ([H₂O⁺] = 4.78 wt% ± 0.03) and hydrogen isotopic composition (Figure 1; D/H = (167 ± 0.2) × 10⁻⁶). These results are consistent with previous measurements reported for CM chondrites (Robert & Epstein 1982; Kerridge 1985; Robert 2006; Alexander et al. 2012).

The two sections of Paris present heterogeneous alteration degrees characterized by significant differences in abundance of Fe–Ni metal beads (Figure 2(a)). Thirty-five Ca-carbonate grains were selected in the fresh areas of Paris sections (Figures 2(b) and (c)) for oxygen isotopic analysis. The O-isotopic compositions of Paris carbonates vary widely, from 24.2‰ to 40.8‰ in $\delta {}^{18}{\rm{O}}$ and 11.6‰ to 23.8‰ in $\delta {}^{17}{\rm{O}}$ and reveal the existence of two distinct populations of carbonates (Figure 3(a); Table 1). When combined with the literature data from other CM chondrites (Benedix et al. 2003; Tyra et al. 2012, 2016; Lee et al. 2013; Horstmann et al. 2014), the carbonate O-isotopic compositions define two statistically different trends (Figure 3(b)). The ISS trend is characterized by large isotopic variations, with ${\rm{\Delta }}{}^{17}{\rm{O}}\lt 0$ and a slope of 0.65 in a $\delta {}^{17}{\rm{O}}\mbox{--}\delta {}^{18}{\rm{O}}$ diagram (Figure 3(b)). A secondary trend shows less isotopic variation, but has ${\rm{\Delta }}{}^{17}{\rm{O}}\gt 0$ and a steeper slope >1 (Figure 3(b)). The inflection point between these two trends is estimated to have $\delta {}^{18}{\rm{O}}=39.0\unicode{x2030}$ and $\delta {}^{17}{\rm{O}}=18.5\unicode{x2030}$ (Figure 3(b)). The trends observed in the $\delta {}^{17}{\rm{O}}\mbox{--}\delta {}^{18}{\rm{O}}$ diagram demonstrate the occurrence of two types of carbonates that recorded aqueous fluids with different origins in the solar protoplanetary disk.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 1.  Oxygen Isotopic Compositions of Paris Ca-carbonates

	$\delta {}^{18}{\rm{O}}\,(\unicode{x2030} )$	2σ	$\delta {}^{17}{\rm{O}}\,(\unicode{x2030} )$	2σ	${\rm{\Delta }}{}^{17}{\rm{O}}\,(\unicode{x2030} )$	2σ
#1	24.18	0.23	12.09	0.47	−0.48	0.38
#2	25.01	0.27	11.63	0.78	−1.37	0.68
#3	25.45	0.27	12.14	0.73	−1.09	0.63
#4	27.37	0.25	13.40	0.56	−0.83	0.46
#5	28.54	0.63	14.65	0.59	−0.19	0.46
#6	29.41	0.21	15.25	0.38	−0.04	0.36
#7	29.59	0.17	13.72	0.50	−1.66	0.44
#8	29.60	0.26	15.01	0.49	−0.38	0.31
#9	29.95	0.24	14.59	0.55	−0.99	0.46
#10	30.46	0.29	14.23	0.78	−1.61	0.67
#11	31.71	0.32	16.44	0.79	−0.04	0.32
#12	31.78	0.61	16.71	0.58	0.19	0.45
#13	31.81	0.67	16.99	0.48	0.45	0.38
#14	32.15	0.17	15.88	0.51	−0.84	0.44
#15	32.74	0.20	16.79	0.43	−0.23	0.48
#16	33.01	0.19	15.60	0.63	−1.56	0.56
#17	33.16	0.16	15.98	0.75	−1.26	0.69
#18	33.58	0.75	16.95	0.51	−0.52	0.40
#19	33.74	0.64	16.71	0.81	−0.83	0.64
#20	34.33	0.58	16.70	0.57	−1.15	0.44
#21	36.43	0.20	16.91	0.55	−2.03	0.47
#22	36.79	0.13	19.80	0.36	0.67	0.46
#23	37.31	0.13	19.61	0.51	0.21	0.38
#24	37.57	0.54	20.40	0.52	0.87	0.40
#25	37.60	0.23	20.57	0.50	1.01	0.68
#26	37.85	0.19	20.16	0.37	0.48	0.46
#27	38.25	0.18	19.73	0.53	−0.16	0.40
#28	38.69	0.13	18.98	0.39	−1.14	0.34
#29	38.83	0.64	23.80	0.59	3.61	0.46
#30	39.41	0.73	23.15	0.57	2.65	0.39
#31	39.45	0.60	20.29	0.45	−0.22	0.35
#32	39.51	0.71	23.00	0.55	2.46	0.38
#33	39.71	0.16	20.78	0.43	0.14	0.41
#34	40.66	0.54	20.70	0.46	−0.45	0.36
#35	40.76	0.14	20.71	0.67	−0.49	0.61

Note. ${\rm{\Delta }}{}^{17}{\rm{O}}$ errors were calculated using the errors of the ¹⁷O/¹⁸O ratio.

Download table as:  ASCIITypeset image

The main carbonate ISS population is indistinguishable within error from the trend defined by the matrix phyllosilicates and bulk O-isotopic compositions of CM chondrites (slope 0.65; Figure 3(b)). This continuous trend does not follow a mass-dependent fractionation trend (Figure 3(b)), implying that the carbonate O-isotopic compositions were not affected by terrestrial alteration (Tyra et al. 2012) and did not result from fluid circulation along a temperature gradient, which would have produced a trend with a slope of 0.52. Instead, it implies that carbonate O-isotopic compositions are a direct proxy for the degree of O-isotopic equilibration between ^17,18O-rich fluids and ¹⁶O-rich anhydrous minerals. In addition, the fluids from which W-ISS carbonates precipitated had near terrestrial ${\rm{\Delta }}{}^{17}{\rm{O}}$ values, demonstrating that most water ices accreted by CM chondrites had a dominantly local origin from the inner solar system (i.e., ${\rm{\Delta }}{}^{17}{\rm{O}}\approx 0$; Alexander et al. 2012). However, the significant departure of certain carbonates from the W-ISS trend (Figures 3(a) and (b)) implies the presence of interstellar water with ${\rm{\Delta }}{}^{17}{\rm{O}}\gg 0$ within CM chondrites (Figure 3(b)).

The contribution of interstellar water ices can be quantitatively estimated from isotopic mass balance calculations. The O-isotopic compositions of fluids from which a carbonate located at the inflection point precipitated (Figure 4(a)) were calculated at different temperatures (10°C–150°C) using the equation $1000\mathrm{ln}{\rm{\Delta }}$ ${}^{18}{\rm{O}}=(2.78\times {10}^{6}\,{{\rm{T}}}^{-2})$ $-\,3.39$ $=\ \delta {}^{18}{{\rm{O}}}_{\mathrm{carb}}$ $-\ \delta {}^{18}{{\rm{O}}}_{\mathrm{water}}$ (Kim & O'Neil 1997). In the absence of a contribution from ^17,18O-rich interstellar water, the fluid compositions excepted for the inflection point define a mass-dependent trend that is located slightly below the TFL (defined by the colored points; Figure 4). Hence, for each temperature, we calculated the O-isotopic composition of a fluid $(\delta {}^{\mathrm{17,18}}{{\rm{O}}}_{\mathrm{fluid}})$ resulting from mixing of inner solar system water $(\delta {}^{\mathrm{17,18}}{{\rm{O}}}_{{\rm{W-ISS}}})$ with varying proportion of ^17,18O-rich interstellar water (Sakamoto et al. 2007) $(\delta {}^{\mathrm{17,18}}{{\rm{O}}}_{{\rm{W-ISM}}}=+180\unicode{x2030} )$, according to the following isotopic mass balance:

$$\begin{eqnarray}&&\delta {}^{\mathrm{17,18}}{{\rm{O}}}_{\mathrm{fluid}}=f\times \delta {}^{\mathrm{17,18}}{{\rm{O}}}_{{\rm{W}} \mbox{-} \mathrm{ISS}}+(1-f)\times \delta {}^{\mathrm{17,18}}{{\rm{O}}}_{{\rm{W}} \mbox{-} \mathrm{ISM}},\end{eqnarray} \tag{ 1 }$$

where f is the concentration of inner solar system water ices relative to the concentration of interstellar water ices. The results demonstrate that W-ISM carbonates precipitated from alteration fluids that accommodated up to 8% of ^17,18O-rich interstellar water (Figure 4(a)).

Zoom In Zoom Out Reset image size

Regardless of the temperature of carbonate precipitation, there is a significant difference in the slopes of the trends defined by the W-ISM carbonates and the slopes of the W-ISS/W-ISM mixing lines (Figure 4(a)). This could be due to variations in the O-isotopic compositions of interstellar water. A recent report confirms the extreme $\delta {}^{17}{\rm{O}}-\delta {}^{18}{\rm{O}}$ values measured in the Acfer 094 chondrite (Sakamoto et al. 2007; Nittler et al. 2015), but also points out the existence of less-^17,18O-rich interstellar water in another chondrite (Nittler et al. 2015) (i.e., MIL 07687; $\delta {}^{\mathrm{17,18}}{\rm{O}}=34\unicode{x2030} ;{\rm{\Delta }}{}^{17}{\rm{O}}=+16\unicode{x2030} )$. For the latter O-isotopic composition, the isotopic mass balance calculations reproduce the carbonate W-ISM trend (Figure 4(b)) and demonstrate the contribution of ^17,18O-rich interstellar water ices at a level of up to 35% (Figure 4(b)).

Our results confirm the unique nature of the pristine CM chondrite Paris that kept the record of the presence of interstellar water due to its limited degree of alteration (Hewins et al. 2014; Marrocchi et al. 2014). At the exception of Maribo and LON 94091 (Lee et al. 2013; Horstmann et al. 2014), the O-isotopic compositions of Ca-carbonates (Figure 3(a)) of other CM chondrites suggest that they did not accrete significant amounts of interstellar water. In addition, Paris's carbonates show O-isotopic compositions that fall on both trends in the oxygen three-isotope plot (W-ISS and W-ISM; Figure 3(a) and Table 1), suggesting that the accretion of interstellar water ice grains were heterogeneous and contributed to different extents to the alteration fluids. Due to the low permeabilites of CM chondrites (Bland et al. 2009), the alteration processes are expected to be isochemical with minimal fluid flow and mainly controlled by the existence of microenvironments (<100 um; Bland et al. 2009). Considering Paris, these microenvironments were thus characterized by varying abundances of interstellar water relative to local water, leading to the precipitation of W-ISS and W-ISM carbonates (Figure 3(a)).

Our oxygen isotopic data reveal that interstellar water ices contributed significantly to the total Paris water budget. This is confirmed by the bulk D/H ratio of Paris that shows significant D enrichment (i.e., D/H = (167 ± 0.2) × 10⁻⁶; Figure 1) relative to (i) the mean (D/H) of CM chondrites (139 × 10⁻⁶) and (ii) the putative D/H of local CM water (i.e., D/H = (82 ± 1.5) × 10⁻⁶; Alexander et al. 2012). Such a result is in line with the asymmetric distribution of bulk D/H ratios toward cometary values observed in carbonaceous chondrites, which suggests significant contribution of water ice grains coming from the outer part of the solar system (Robert 2006). In addition, even though the D/H ratio of only one other W-ISM carbonate-bearing CM is available (Alexander et al. 2012) (i.e., LON 94091), we note that this CM shows also a significant enrichment in deuterium (Alexander et al. 2012) (i.e., D/H = 152 × 10⁻⁶). Given the initial D/H inferred for inner solar system water (D/H = (82 ± 1.5) × 10⁻⁶; (Alexander et al. 2012)) and the recent determination of the D/H of the Comet 67P (i.e., D/H = (530 ± 70) × 10⁻⁶; Altwegg et al. 2015), the D enrichment observed in LON 94091 and Paris can be explained by a contribution of 15.5% and 19% of interstellar water ice grains, supporting our mass balance calculation based on the O-isotopic composition of carbonates. Thus, we propose that the abundance of interstellar water varies among CM chondrites, affecting their isotopic D/H and O-isotopic compositions at both the bulk and mineral scales.

The presence of interstellar water ices from the outer part of the solar system in CM chondrites requires inward radial transport in the protoplanetary disk, which could be achieved either by gas drag (Lunine 2006) or turbulent diffusion (Jacquet & Robert 2013). An alternative scenario could involve impacts between asteroids formed in the main belt and ice-rich bodies coming from the outer part of the solar system due to planetesimal migration induced by the early core formation of Jupiter and Saturn (Grazier et al. 2014). Interestingly, the latter scenario would have generated collisions at velocities low enough to have allowed fragments to be re-accreted and this could therefore account for the highly brecciated nature of CM chondrites (Brearley 2006; Briani et al. 2012). In addition, this model predicts that a large fraction of water and volatiles in the main belt originates from the outer solar system, in line with (i) the finding of this study and the recent detection of ammoniated phyllosilicates at the surface of (1) Ceres (De Sanctis et al. 2015) and (ii) disk ionization modeling that suggests that terrestrial oceans should contain 7 to 30%–50% of interstellar ices (Cleeves et al. 2014).

Taken together, these results reveal the presence of interstellar water ices in the accretion zone of carbonaceous chondrites at a level of 8%–35%. This is consistent with the contribution of interstellar ices at a level of 9% in ordinary chondrites determined from the D/H ratios of organic matter/phyllosilicates intergrowths (Piani et al. 2015). As the accretion of carbonaceous chondrites may have predated the accretion of ordinary chondrites (Jacquet et al. 2012), this would suggest that the inward transport of interstellar water ices took place early in the history of the Solar protoplanetary disk. Thus, isotopic compositions of chondrites and hydrodynamical modeling both show that outer disk materials contributed significantly to the delivery of water and volatile elements in the chondrite- and Earth-accretion regions (Cleeves et al. 2014; Marty et al. 2016).

The in situ analysis of the oxygen isotopic compositions of Ca-carbonates in the pristine CM chondrite Paris revealed the presence of two distinct trends in a $\delta {}^{17}{\rm{O}}\mbox{--}\delta {}^{18}{\rm{O}}$ diagram. They are respectively characterized by (i) large isotopic variations, with ${\rm{\Delta }}{}^{17}{\rm{O}}\lt 0$ and a slope of 0.65 and (ii) less isotopic variation, with ${\rm{\Delta }}{}^{17}{\rm{O}}\gt 0$ and a steeper slope >1. While the former is coherent with an inner solar system origin of water, the latter highlights the presence of interstellar water in Paris. Based on mass balance calculations, we estimate that Paris contains 8 to 35% of interstellar water. The bulk D/H of Paris present significant D enrichment (D/H = (167 ± 0.2) × 10⁻⁶) relative to the mean D/H of CM chondrites ((139 ± 3) × 10⁻⁶) and the putative D/H of local CM water ((82 ± 1.5) × 10⁻⁶). This confirmed the presence of interstellar water in the Paris chondrite. Our results are coherent with previous studies that suggest efficient radial mixing of interstellar ices from the outer zone of the Solar protoplanetary disk.

We are grateful to Guillaume Avice for his helpful discussions and to Nordine Bouden for his assistance with oxygen isotopic measurements. We thank Thomas Rigaudier and Christian France-Lanord for the determination of the D/H isotopic composition of the Paris chondrite. This work was funded by l'Agence Nationale de la Recherche through grant ANR-14-CE33-0002-01 SAPINS (PI Yves Marrocchi). This is CRPG contribution #2440 and SAPINS contribution #07.