Young asteroidal fluid activity revealed by absolute age from apatite in carbonaceous chondrite

Chondritic meteorites, consisting of the materials that have formed in the early solar system (ESS), have been affected by late thermal events and fluid activity to various degrees. Determining the timing of fluid activity in ESS is of fundamental importance for understanding the nature, formation, evolution and significance of fluid activity in ESS. Previous investigations have determined the relative ages of fluid activity with short-lived isotope systematics. Here we report an absolute 207Pb/206Pb isochron age (4,450±50 Ma) of apatite from Dar al Gani (DaG) 978, a type ∼3.5, ungrouped carbonaceous chondrite. The petrographic, mineralogical and geochemical features suggest that the apatite in DaG 978 should have formed during metamorphism in the presence of a fluid. Therefore, the apatite age represents an absolute age for fluid activity in an asteroidal setting. An impact event could have provided the heat to activate this young fluid activity in ESS.

C hondritic meteorites (chondrites) are composed of the solid minerals that have formed in the early solar system (ESS). They are the most precious samples to constrain primary processes that have been prevailed in the solar nebula 1,2 . However, most of the chondritic meteorites have been affected by later fluid activity to various extents [3][4][5] . Fluid activity has changed the mineralogy, chemistry and isotope compositions of chondrites through fluid-mineral interactions (that is, aqueous alteration, metasomatism and fluid-assisted metamorphism 6 ). Constraining when and where these fluidmineral interaction events took place is important to understand the thermal histories of chondrite parent bodies and the temporal-spatial evolution of fluids in the solar nebular and asteroidal settings 7 . The formation time of secondary minerals can further constrain the heating sources that have driven the fluid-mineral interaction events in ESS (for example, decay of 26 Al, heating during accretion of chondritic components to form asteroidal parent bodies and impact heating). Previous investigations have dated a few secondary minerals (for example, carbonate minerals, ferroan olivine, magnetite, nepheline and grossular) that have formed by low-temperature and metamorphic processes in the presence of fluids, based on short-lived isotopic systematics (for example, 53 Mn-53 Cr, 129 I-129 Xe and 26 Al-26 Mg systematics 7 ). Most of these studies revealed that fluid activity in ESS has started within 1-2 Myr after the formation of Allende Ca,Al-rich inclusions (CAIs) and lasted up to 15 Myr (refs 5,7). A few studies, however, found no resolvable excess of daughter nuclides 8 , indicating an extended period of fluid activity (425 Myr). However, no absolute ages have been reported for constraining fluid activities in ESS up to date. It also remains unknown how late fluid activities could be present in ESS and whether fluid activity is coupled with young metamorphic processes. This is partly due to the fact that dating using short-lived isotope systems is only applicable to events that occurred several to tens of millions after the solar system formation. In addition, to date, most of the secondary minerals (for example, phyllosilicate, carbonate and magnetite) in chondritic meteorites are not suitable for long-lived isotope geochronology.
Apatite is an important volatile-rich mineral Ca 5 (PO 4 ) 3 (F,Cl,OH) in chondrites. Apatite and merrillite [Ca 9 NaMg(PO 4 ) 7 ] in chondrites are thought to be secondary minerals, due to the large difference in condensation temperatures for Ca and P (refs 9,10). This is consistent with the absence of Ca-phosphate minerals in most primitive chondrites 1 . The formation of apatite in chondrites is usually attributed to fluid activity 5,9,11,12 , although their formation conditions may vary in different chondrites of various petrologic types (from type 1 to type 6). With the developments of uranium-lead (U-Pb) geochronology with secondary ion mass spectrometer (SIMS), apatite can be dated with detailed petrographic context in polished thin sections [13][14][15] . Therefore, combined with petrographic and geochemical observations, the age of apatite in chondrites can constrain the absolute ages of fluid activity in ESS and/or subsequent heating events (for example, refs 16,17).
DaG 978 is an ungrouped and low petrologic type (B3.5) carbonaceous chondrite with a low abundance of volatile elements 18 . It contains many relatively coarse chlorapatite grains (up to 280 mm in size 19 ). Here we report on the petrography, mineralogy, rare earth elements (REEs) and oxygen isotopic compositions, and U-Pb isotopic systematics of the apatite in DaG 978. Our data reveal that the apatite in DaG 978 could have formed during a metamorphic event in the presence of fluid. Its age represents the first absolute age of fluid activity in ESS.

Results
General petrography and mineralogy of DaG 978.  Fig. 1). Lath-shaped olivine grains replacing low-Ca pyroxene occur in some ferromagnesian chondrules and chondrule fragments. Nepheline grains are present as inclusions in some of these lath-shaped olivine grains and as lamellae in anorthite grains. Apatite is common in DaG 978 and has a wide occurrence (described below).
Petrography and mineralogy of apatite and merrillite. Coarse-grained apatite (420 mm in size) has three major textural occurrences in DaG 978 ( Fig. 1). First, the majority of apatite grains in DaG 978 are closely associated with altered CAIs that can contain secondary minerals such as nepheline, Na-rich plagioclase and troilite ( Fig. 1a; Supplementary Fig. 1). Most of the apatite grains associated with altered CAIs are subhedral-euhedral in shape with well-developed crystal forms (Fig. 1a). Second, a few apatite grains occur at the margins of large FeNi metal grains within or outside of chondrules (Fig. 1b). Apatite is rarely included in FeNi metal, which usually contains diopside, olivine, chromite and merrillite as small inclusions. Third, many of the anhedral to euhedral apatite grains occur as isolated crystals in fine-grained matrix of DaG 978. The spatial distribution of some apatite grains in the matrix resembles a chain of beads (Fig. 1c). Regardless of the different textural occurrences, many of the apatite grains in DaG 978 are closely associated with anhedral to euhedral grains of high-Ca pyroxene and/or olivine (Fig. 1d). Some of the high-Ca pyroxene and olivine grains are included in coarse-grained apatite. Merrillite is common in FeNi metal, mainly, as inclusions. Most merrillite grains in DaG 978 are o20 mm in size, only a few grains as large as 50 mm.

Discussion
The mineralogical features in DaG 978 indicate that both metamorphic process and metasomatic process have affected this chondrite 19 . The main evidence of metamorphism in DaG 978 is the chemical changes of some primary minerals (that is, olivine, chromite and FeNi metal) in chondrules and refractory inclusions. The chemical changes of primary minerals require thermally driven volume diffusion of some elements (that is, Mg, Fe, Cr and Ni). The olivine-chromite thermometer suggests a maximum equilibrium temperature of B580-680°C, indicating mild thermal metamorphism 19 . The real metamorphic temperature could be lower than this temperature, because the olivine-chromite pairs in a type 3.5 chondrite might not have reached chemical equilibrium 21 . On the other hand, lathshaped olivine, nepheline, Ca-phosphates and troilite in CAIs could not have formed with the metamorphic process alone. Instead, they could have formed during metamorphism in the presence of fluid (that is, replacement of primary minerals by fluids and/or direct precipitation from fluids 19,22 ). For example, the formation of apatite that is closely associated with CAIs requires that P and Cl migrated from a source outside the CAIs. Without the presence of fluid, it is difficult to interpret the migration of these elements in DaG 978 (ref. 11). Lath-shaped olivine is widely interpreted as a product of fluid-assisted metamorphism 6,22 . The lack of phyllosilicate minerals in DaG 978 indicates that the fluid-mineral interaction should have taken place at temperatures of 4200°C (refs 4,6). This requirement is generally consistent with the temperature deduced from the olivine-chromite geothermometry and those (B400°C or higher) for other chondrites of around type 3.5 (refs 6,23). Therefore, the thermal event drove the diffusion of some elements in primary minerals. It might also result in the onset of fluid activity in DaG 978. During or intermediately subsequent to the thermal event, the fluids interacted with the minerals in DaG 978 and formed the secondary minerals. In summary, the apatite in DaG 978 should be a product of fluid-assisted metamorphism on the parent body 11 . Close textural association between apatite, and both FeNi metal and altered CAIs indicates that FeNi metal and CAIs could be the main sources of P and Ca for the formation of apatite in DaG 978, respectively. Chondrule mesostasis could be another source of P because some mesostasis in primitive chondrites contain P 2 O 5 up to 3.5 wt% (ref. 24). The source of Cl-rich fluid cannot be constrained based on the current observation. However, the partition coefficients of F and OH between apatite and aqueous fluids are much higher than that of Cl (ref. 25). Therefore, the low contents of F and calculated H 2 O in apatite from DaG 978 indicate that the fluids from which apatite formed must be highly depleted in F and H 2 O and highly enriched in Cl.
Although both the apatite in DaG 978 and those in metamorphosed ordinary chondrites (OCs) could have formed though fluid-assisted metamorphism, their REE geochemistry and petrographic textures may reflect various contributions from metamorphic processes and metasomatic processes. Most of apatite grains in type 4-6 OCs exhibit negative Eu anomalies 11,26 . The negative Eu anomaly of the apatite from type 4-6 OCs could be a result of Eu equilibrium between apatite and recrystallized plagioclase during thermal metamorphism, with most Eu incorporated into plagioclase 9,11 . However, the apatite in DaG 978 has a positive Eu anomaly, similar to those observed in the Allende meteorite (type 3) and some of the apatite grains in type 3.9 and type 4 OCs 9,11,14 . Such positive Eu anomalies indicate the lack of chemical equilibrium between apatite and plagioclase. Instead, they may reflect the chemical features of fluids from which apatite formed 9 . The petrographic textures of apatite in DaG 978 do support that fluid-mineral interaction should have played a key role for its formation. For example, fluid activity can well interpret the euhedral apatite crystals in matrix and those associated with CAIs. On the contrast, the apatite grains in highly metamorphosed OCs are mainly irregular in shape 11 . In addition, a pathway of fluid migration can best interpret the chain-like distribution of some apatite grains in matrix (Fig. 1c). In summary, the geochemical and petrographic features of  apatite in DaG 978 also reflect the onset of mild metamorphic process in the presence of an aqueous fluid in a type 3 chondrite parent body. Thus, the age of apatite records the timing of fluid activity in DaG 978. The 207 Pb/ 206 Pb isochron age (4,450 ± 50 Myr) of the apatite obtained in this study has two important implications. First, in the literature, all ages (short-lived isotope chronologies, Sr isotope composition and Rb-Sr isotope chronology) about fluid activity in ESS are relative ages or model ages 7,27,28 . The apatite age obtained in this study is the first absolute age of fluid activity in ESS up to date. Second, the age of apatite in DaG 978 is B117 Myr younger than the oldest CAI 29 . This indicates that fluid activity may extend for a long time interval in ESS, from a few Myr to B117 Myr after CAI formation. It has been suggested that the solar nebula has a lifetime of B10 Myr (ref. 30). Therefore, the young age of the apatite in DaG 978 demonstrates that this fluid activity must have taken place on the parent body 7 .
Decay of short-lived radionuclides and collisional heating has been proposed as significant heat sources to thermal metamorphism of OCs 21 . For example, the main heating source for fluid activity within B15 Myr after CAI formation could be the decay of short-lived radionuclides (for example, 26 Al and 60 Fe (refs 7,31)). However, short-lived radionuclides cannot be a viable heat source for a thermal event that occurred 117 Myr after CAI formation. Instead, collisional heating in late ESS may provide enough heat to activate the metamorphic event in the presence of fluid on the parent body of DaG 978. It has been suggested that there are frequent impact events at 4.47±0.03 Ga in the main belt after the Moon-forming impact event, based on Ar-Ar ages of ordinary chondrites 32 and phosphate U-Pb ages of ordinary chondrites 16,17 , and dynamic modelling 33 . On the basis of this excellent consistency between the apatite 207 Pb/ 206 Pb isochron age and the timing of impact events proposed in the literature, it is most likely that an impact event may have provided the heat to activate the metamorphic event at 4,450 Myr on the parent body of DaG 978.
In addition, recent studies suggested that some large asteroids (that is, Ceres) have water evaporation from localized regions, indicating subsurface fluid activity 34 . Some of these water evaporation events could be related to impact events. Some of the evaporation events could be due to cryo-volcanism 34 . Thus, we cannot totally exclude the possibility that DaG 978 might have derived from the subsurface of a large asteroid, where fluid activity might have lasted up to 117 Myr after CAI formation.
In summary, the apatite in DaG 978 has an origin by fluid-assisted metamorphism and was formed at 4,450 Myr, B117 Myr later than CAI formation. Its age is the first absolute age about fluid activity in ESS. The late formation of apatite in DaG 978 represents a young metamorphic event in the presence of fluid in the parent body setting. An impact event at B4,450 Myr may have provided the heat to activate this fluid activity in late ESS. Alternatively, this meteorite might be ejected from a large (possibly Ceres like) asteroid, where subsurface fluid activities could last up to B117 Myr after CAI formation.

Methods
Petrography and major element analysis. Petrographic textures of apatite in DaG 978 were mainly observed with JEOL 7000F field-emission scanning electron microscope (SEM) at Hokkaido University and JEOL 6490 SEM at Nanjing University. Both SEM instruments are operated at an accelerating voltage of 15 kV. Mineral chemistry of apatite was determined using a JEOL 8100 electron probe micro-analyzer (EPMA) at Nanjing University. A defocused beam (10 mm in diameter) at a beam current of 10 nA and an accelerating voltage of 15 kV was used. A few natural and synthetic materials were used standards. All EPMA data are reduced with the ZAF (atomic number-absorption-fluorescence) correction procedure. Before EPMA analysis, all Ca-phosphate minerals were qualitatively measured by energy dispersive spectrometers (EDS) installed on the SEM instruments. Since SEM-EDS results show high chlorine and a very low intensity of fluorine, Cl and F are the first and second elements to be measured, respectively.
Oxygen isotope systematics. Oxygen isotope compositions of apatite and associated silicate minerals were determined with the Cameca IMS-1270 instrument at Hokkaido University. Before measurements, the polished sections of DaG 978 were carbon-coated. The primary ion beam was mass filtered positive Cs þ ions of 20 keV and the beam spot size was 8-10 mm in diameter. A primary beam current of 100 pA was used to obtain a count rate of negative 16 O ions of B2 Â 10 7 c.p.s. A normal-incident electron gun was used for charge compensation of the sputtered area. A mass resolving power of B5,500 was used to separate 17 O from 16 OH. Negative 16 O ions were detected with an axial Faraday cup, while negative 17 O and 18 O ions were detected with an axial electron multiplier detector, in magnetic peak-jumping mode. The instrumental mass fractionation effect was corrected using San Carlos olivine. The reported uncertainties of the individual analyses are expressed in 2 s.d., which were estimated by considering both the internal error of each measurement and the reproducibility of the standard measurements. After the measurements, all spots were evaluated using filed-emission SEM.
REE geochemistry. The REE compositions of apatite and merrillite were determined with the Cameca IMS-6f instrument at Hokkaido University. Before measurements, the polished sections of DaG 978 were carbon-coated. The procedure was similar to that described in refs 35,36. The primary beam was mass filtered 16 O À of -14.5 keV and irradiated on the sample surface to a diameter of B25 mm. The primary beam current is 10-15 nA. Kinetic energy filtering was used to reduce interferences from molecular ions by offsetting the sample acceleration voltage (-100 eV). The energy bandwidth was 20 eV. The exit slit was set to a mass resolving power of B500. Positive secondary ion intensities were counted for 10 s with an electron multiplier detector in magnetic peak-jumping mode. Each analysis included 15 cycles. Relative sensitivity factors between secondary ion intensity and concentration for each REE (relative to Ca) were determined using the Takashima augite, for which REE contents have been well determined by instrumental neutron activation analysis 37 .
Uranium-lead isotopic systematics. The U-Pb isotopic system of apatite was determined using the Cameca IMS-1280HR instrument at the Institute of Geology and Geophysics, Chinese Academy of Sciences, China, following the analytical procedure of refs 15,38. Before measurements, the polished sections of DaG 978 were carbon-coated. The O 2 À primary beam was accelerated at À 13 kV, with an intensity of B10 nA. The ellipsoidal spot is B20 Â 30 mm in size. The secondary ions were extracted at an initial energy of 10 keV. The entrance slit and field aperture were set to a mass resolution power of B8,000. The energy slit was closed to a bandwidth of 60 eV to reduce the energy dispersion. Detection of secondary ions took the advantage of both mono-collector mode and multi-collector mode. For each analysis in this study, secondary ion beam intensities of 40 Ca 2 31 P 16 O 4 þ , 238 U þ and 238 U 16 O 2 þ were detected in L1-EM by peak-jumping sequences. In the second sequence, 204 Pb, 206 Pb and 207 Pb were detected with multi-collector electron multipliers L2, L1 and C, respectively. In the third sequence, 238 U, 232 Th 16 O and 238 U 16 O were measured with L1, H1 and H2, respectively. The Pb/U ratios were calibrated with an empirical correlation between Pb þ /U þ and UO 2 þ /U þ ratios, normalized to the 1,160 Myr apatite standard from the Prairie Lake alkaline-carbonatite complex in Ontario, Canada 13,15 . Each analysis included 10-20 cycles. The U/Pb and Pb/Pb isotope ratios were calculated by averaging the ratios of measurement counts. Uncertainties for individual analyses are reported as 2 s.d. The 207 Pb/ 206 Pb isochron and total Pb/U isochron ages were calculated at the 95% confidence level using the Isoplot/Ex 3.0 software 39 . Data availability. The data shown and discussed in this paper are presented in the Supplementary Information.