Condensation and Radial Transport of Filamentary Enstatite Crystals from Interplanetary Dust

Filamentary enstatite crystals, formed by gas-solid condensation in the solar nebula, are found in chondritic porous interplanetary dust particles of probable cometary origin. We measured the oxygen isotopic composition of four filamentary enstatite grains from the giant cluster interplanetary dust particle U2-20 GCP. These grains sample both the O-rich solar (∆O ≈ −30 h) and O-poor planetary (∆O ≈ 0 h) isotope reservoirs. Our measurements provide evidence for very early vaporization of dust-poor and dust-rich regions of the solar nebula, followed by condensation and outward transport of crystalline dust to the comet-forming region very far from the Sun. Similar processes are likely responsible for the crystalline silicates observed in the outer regions of protoplanetary disks elsewhere in the Galaxy.


2
Filamentary Enstatite Condensation Crystalline silicates are seen near and far from the central star in protoplanetary disks [1] and in comets, which formed in the outer Solar System [2,3]. Interstellar dust is almost entirely amorphous [4], so it is thought that crystalline silicates in the outer regions of protoplanetary disks were either transported there from regions close to the central star (where stellar outbursts can vaporize disk material that then condenses as crystalline dust) or formed in situ via an energetic process in the outer disk [e.g., 5].
Outward transport from the inner Solar System to the comet-formation region is thought to have been inhibited by a gap at Jupiter's orbit as early as 1 Myr after CAIs based on high-precision bulk isotope measurements of meteorites [6] and astronomical observations of protoplanetary disks [7]. A chondrule-like fragment 'Iris' from comet Wild 2 was measured to have no detectable radiogenic 26 Mg from the decay of 26 Al (t 1/2 = 0.72 Myr). Iris was estimated to have formed very late in the solar nebula's lifetime -more than 3 Myr after calcium-aluminum-rich inclusions (CAIs) in CV chondrites with the canonical 26 Al/ 27 Al initial ratio of 5.2 × 10 − 5. The mineralogy and O isotopic composition of Iris is similar to type II chondrules found in chondrites [8]. The existence of crystalline silicates in the outer regions of young protoplanetary disks, those that do not yet have gaps opened by forming planets, requires a mechanism distinct from those that formed chondrules like Iris.
Filamentary crystals were first predicted by Donn and Sears (1963) to form via vapor-to-solid condensation of nebular gas and be accreted into comets [9]. Roughly [11]. The likely astrophysical mechanism behind this solar-planetary difference is photochemical processing of the Solar System's parent molecular cloud by nearby massive stars [12]. Most CAIs and amoeboid olivine aggregates (AOAs), which likely condensed directly from gas very near the proto-Sun, have 16 O-rich isotopic compositions [13][14][15]. Chondrules, which likely formed in a dust-rich region of the disk [16], have roughly terrestrial or 'planetary' oxygen isotopic compositions [17]. Vaporizing a region of the nebula with an approximately solar dust-to-gas ratio (0.01) would produce a gas with roughly solar values of ∆ 17 O, whereas vaporizing a region with a greater-than-solar dust/gas ratio would produce a gas with higher ∆ 17 O values.
Prior to the Solar System's formation, the outer region of its parent molecular cloud was likely polluted by 26 Al-containing outflows from nearby Wolf-Rayet stars [18]. In the earliest stages of stellar evolution, before 26 Al had been homogenized throughout the disk, the mass accretion rate onto the proto-Sun was high and variable [19]. Fueled by this 'clumpy' mass accretion, the young proto-Sun likely emanated FU Orionis-like outbursts capable of vaporizing different regions of the disk [20] with variable dust/gas ratios [21,22]. Un-melted fine-grained CAIs and AOAs lacking 26 [13][14][15], indicating that they likely formed near the Sun in a region with a roughly solar dust/gas ratio. The acquired data are summarized in Table 1 and plotted on an oxygen three-isotope plot in Figure 2. Sample C is enriched in 16  and D are consistent with a planetary oxygen isotopic composition and plot within 2σ of one another and a previously studied ribbon [26]. Table 1 Oxygen isotopic compositions, expressed in delta notation as per mill ( ) deviation from terrestrial, of four filamentary enstatite samples from U2-20 GCP. Listed 2σ uncertainties were calculated by bootstrap resampling methods.
High-magnification bright-field TEM images of sample D (Fig. 1G) exhibit alternating high-and low-contrast bands that indicate stacking faults caused by the sporadic intergrowth of ortho-and clino-enstatite. These stacking faults are also indicated by the intense streaking parallel to theâ * -axis visible in the electron diffraction pattern. These crystallographic features suggest this ribbon condensed from a high-temperature vapor, forming proto-enstatite that converted to ortho-/clino-enstatite as it cooled [10].
Sample C is among the most 16 O-enriched Solar System objects ever measured: it plots within 2σ of the O isotopic composition of the Sun as inferred from Genesis measurements [11] and exceedingly rare 16 O-rich refractory inclusions in carbonaceous chondrites [13,15]. Sample C's extremely 16 O-rich Filamentary Enstatite Condensation composition is consistent with condensation from a gas produced by evaporation of a disk region with a roughly solar dust/gas ratio, likely near the central star or above the disk's mid-plane [20]. Samples A, B, and D and a previously studied ribbon from the same cluster [26] have Filamentary enstatite crystals that condensed from these different gas reservoirs were transported to the outer regions of the disk and subsequently accreted by the same (likely cometary) parent body.
The rate of outward transport in a protoplanetary disk is roughly proportional to the rate of mass accretion onto the central star [19,20]. Grains that condensed early and close to the Sun would be efficiently transported outwards.
Outer Solar System bodies such as comets likely accreted some of the oldest crystalline dust that formed close to the Sun. The migration of inner Solar System likely took place in the first ∼500 kyr, before Jupiter's core opened a Springer Nature 2021 L A T E X template gap in the disk [30]. A schematic of the formation and transport of filamentary enstatite to the comet forming region is shown in Figure 3.
Our results can explain the isotopic composition of refractory inclusions in outer Solar System bodies and the origin of crystalline dust in young disks elsewhere in the Galaxy. For example, the refractory particle 'Inti' from comet  We acquired 12×12-µm, 256×256-pixel ion raster images using the Wash U Cameca NanoSIMS 50. Each sample was pre-sputtered with a 78 pA Cs + beam for 300 s to remove any adsorbed water or insoluble organic matter. Measurements were collected using a 2 pA Cs + primary beam focused to approximately 100 nm. We simultaneously collected 16  Isotopic data were analyzed using a custom Matlab script. Un-processed data and the script used to analyse them (including ROIs used and all other dependent files) are available for public download [33]. We aligned the images from each cycle and defined regions of interest (ROIs), avoiding grain-and raster-edges to reduce topography-induced instrumental fractionation effects.
The ROIs were also drawn to avoid any adhering grains or potential contami-  [38,39]. Calculating delta-values in this way minimizes multiplicative confounding factors, such as instrumental mass fractionation (IMF), quasi-simultaneous arrival (QSA), and detector dead time. The reported uncertainties were calculated via bootstrap resampling (10 6 trials with replacement) of the pixels within each ROI. The calculated bootstrap uncertainties were found to be greater than or equal to statistical, though the deviation was found to be minimal in most cases. This method for calculating the uncertainty is more likely to yield a reliable estimate for a small number of counts and is therefore more appropriate for the small size of the studied samples (and resulting small number of total counts).

Filamentary Enstatite Condensation
After NanoSIMS measurements, we prepared the remaining portions of samples C and D for analysis via transmission electron microscopy (TEM).
Using the Omniprobe micro-manipulator, we transferred the ribbon remnants from the Au-foil substrate to a liftout grid to which they were secured with a thin Pt weld. The samples were subsequently thinned to electron transparency (∼200 nm thick) and briefly polished with a 5 kV, 48 pA Ga + beam (∼30 s per side). Throughout the process, the Ga + ion beam was used only sparingly to minimize potential amorphization effects.
Electron diffraction measurements were performed with the FEI Titan3  Uncertainties are 2σ. Also plotted: a ribbon (U2-20, E) from the same cluster [26], Genesis solar measurements [11], amoeboid olivine aggregates (AOAs) from Efremovka [13], chondrules (LL3) [17], a CAI-like object ('Inti') found in comet Wild 2 [27], mean comet Wild 2 fines [28], CH CAIs [14,23], and CM CAIs [15]. (B) Zoom-in of the dashed region in (A). TFL, terrestrial fractionation line; PCM, primitive chondrule mineral line. Fig. 3 Illustration of the proposed formation and transport of filamentary enstatite grains with dichotomous oxygen isotopic compositions. The grains were transported from different regions of the inner disk to the outer disk quickly after they formed, and before a forming Jupiter opened a gap in the disk. The cometary and asteroidal parent bodies (NC, noncarbonaceous; CC, carbonaceous chondrites) formed from material that was local to different regions of the solar nebula at the time they formed [6].