Iron and nickel atoms in cometary atmospheres even far from the Sun

In comets, iron and nickel are found in refractory dust particles or in metallic and sulfide grains1. So far, no iron- or nickel-bearing molecules have been observed in the gaseous coma of comets2. Iron and a few other heavy atoms, such as copper and cobalt, have been observed only in two exceptional objects: the Great Comet of 18823 and, almost a century later, C/1965 S1 (Ikeya–Seki)4–9. These sungrazing comets approached the Sun so closely that refractory materials sublimated, and their relative abundance of nickel to iron was similar to that of the Sun and meteorites7. More recently, the presence of iron vapour was inferred from the properties of a faint tail in comet C/2006 P1 (McNaught) at perihelion10, but neither iron nor nickel was reported in the gaseous coma of comet 67P/Churyumov–Gerasimenko by the in situ Rosetta mission11. Here we report that neutral Fe i and Ni i emission lines are ubiquitous in cometary atmospheres, even far from the Sun, as revealed by high-resolution ultraviolet–optical spectra of a large sample of comets of various compositions and dynamical origins. The abundances of both species appear to be of the same order of magnitude, contrasting the typical Solar System abundance ratio. High-resolution ultraviolet and optical spectra of a large sample of comets show that Fe i and Ni i lines are ubiquitous, even when the comets are far from the Sun.

In comets, iron and nickel are found in refractory dust particles or in metallic and sulfide grains 1 . So far, no iron-or nickel-bearing molecules have been observed in the gaseous coma of comets 2 . Iron and a few other heavy atoms, such as copper and cobalt, have been observed only in two exceptional objects: the Great Comet of 1882 3 and, almost a century later, C/1965 S1 (Ikeya-Seki) [4][5][6][7][8][9] . These sungrazing comets approached the Sun so closely that refractory materials sublimated, and their relative abundance of nickel to iron was similar to that of the Sun and meteorites 7 . More recently, the presence of iron vapour was inferred from the properties of a faint tail in comet C/2006 P1 (McNaught) at perihelion 10 , but neither iron nor nickel was reported in the gaseous coma of comet 67P/Churyumov-Gerasimenko by the in situ Rosetta mission 11 . Here we report that neutral Fe i and Ni i emission lines are ubiquitous in cometary atmospheres, even far from the Sun, as revealed by high-resolution ultraviolet-optical spectra of a large sample of comets of various compositions and dynamical origins. The abundances of both species appear to be of the same order of magnitude, contrasting the typical Solar System abundance ratio.
The spectra of about 20 different comets have been collected since 2002 over a large range of heliocentric distances (0.68 to 3.25 au) with the UVES spectrograph mounted on the 8-m UT2 telescope of the ESO Very Large Telescope (VLT; Extended Data Table 1). The use of a dichroic beam splitter and a narrow 0.4″ slit provided a resolving power (λ/∆λ) of about 80,000 over the wavelength range 304-1,040 nm. The slit length of about 10″ typically covers around 7,500 km of the coma at a distance of 1.0 au and, except for a few cases, the slit was centred on the comet nucleus. A uniform procedure 12 was used for the acquisition and reduction of all the data.
Close examination of these spectra revealed the omnipresence of emission lines of neutral atoms of iron (Fe i) and nickel (Ni i) in comets as far as 3.25 au from the Sun, with up to about 40 Fe i lines and 25 Ni i lines for some of them (Extended Data Fig. 1). These lines are weak and located in the blue part of the spectrum (<450 nm), where there are plenty of bright molecular emissions, making blends unavoidable, explaining in part why they have been missed until now. We searched for the lines of the other metals that were identified in comet Ikeya-Seki 7,8 , in particular neutral chromium, the most abundant after nickel, but we did not find any of them. Ni i lines were recently identified in the interstellar comet 2I/Borisov 13 .
The metallic lines, in contrast to the molecular lines, have only a short spatial extent, as already noted for Ikeya-Seki 7 , despite very different conditions of temperature and irradiation. The few spectra not centred on the nucleus do not show these lines, or show them only faintly. In the spectra obtained during the close encounter of comet 103P/Hartley 2 with Earth ( Fig. 1), these lines show a radiance that is inversely proportional to the projected distance to the nucleus, p. Such a profile corresponds to ejection from the surface of the nucleus or a short-lived parent 5 and a constant expansion velocity, resulting in a density distribution proportional to p −2 . Spectra taken at various position angles indicate that the distribution of metals in the inner coma is nearly isotropic, which suggests collisional dragging and an initial velocity high enough to hide radiative pressure effects.
Once freed in a collisionless environment, the atoms are bathed in solar radiation, and would conceivably tend towards an excitation temperature of the order of the colour temperature of the Sun, about 5,800 K. To analyse their emission spectra, we first considered a simple three-level atomic model, as done previously for the analysis of Ikeya-Seki data 4,14 . We then built a more realistic multilevel atomic model, taking into account the high-resolution structure of the solar spectrum (Supplementary Information). This allowed us to compute the production rates of Ni i and Fe i for each comet. In the case of an isotropic expansion at a constant velocity v, the column densities C(p) and the production rates Q are linked by the relation C(p) = Q(4vΔp) −1 , where Δ is the geocentric distance. The observed average density over the slit area A is Owing to the atmospheric blurring, we used instead the convolution of the 1/p profile with a 1″ Gaussian. We adopted the frequently assumed value of v = 0.85 r −1/2 km s −1 for the expansion velocity 15 . Extended Data Table 2 gives the Fe i and Ni i column densities and the corresponding production rates for each spectrum. The quantities found are very small. For the Jupiter-family comet 103P, they correspond to only about 1 g of iron ejected every second, compared to about 100 kg of water, making these elements minor constituents of the coma.
To compare these abundances to those of other species usually observed in comets, we derived from our spectra the production rates of the radicals OH (a daughter product of water), CN and CO 2 + (Extended Data Table 3), and collected CO and H 2 O infrared and submillimetre measurements from the literature. Extended Data Fig. 2 shows that the abundances of iron and nickel atoms are well correlated with the other species and the comets' activity level. Of particular interest is the Nature | Vol 593 | 20 May 2021 | 373 high metallic abundance in the distant and chemically peculiar comet C/2016 R2 relative to its other elements, except CO and CO 2 + . This suggests a possible link between Fe, Ni and the carbon oxides. This water-poor comet had a high activity, driven by a large CO production rate of about 10 29 molecules per second, at about 3 au (refs. 16,17 ).
The average Ni i/Fe i abundance ratio is about unity (log(Ni/Fe) = −0.06 ± 0.31), that is, an order of magnitude higher than the solar value 18 (−1.25 ± 0.04) or the ratio measured in comet Ikeya-Seki (−1.11 ± 0.09), and it does not depend on the heliocentric distance or the comet origin. In our sample, comet 103P has the lowest value (−0.64 ± 0.07) and the carbon-chain-depleted comet 73P the highest value (0.60 ± 0.23), both at heliocentric distances of r ≈ 1 au (Fig. 2).
The comet blackbody equilibrium temperature of its surface is expected to be around T ≈ 280r −1/2 K with r expressed in astronomical units, that is, about 340 K for the comet observed at the closest distance to the Sun (0.68 au) and about 150 K for the most distant one (3.25 au). These temperatures are much lower than those needed to vaporize refractory dust grains, as well as iron and nickel in metallic form or in sulfides 10 . We therefore explore several possibilities to explain how Fe and Ni atoms are released at such low temperatures and why the Ni/Fe ratio is enhanced.
The β parameter that characterizes the ratio between the radiation pressure and the gravity, which is about 6 for iron 10 , is too small to alter substantially the velocity field in the vicinity of the nucleus and to decrease the column density of iron relative to nickel. Moreover, comet Ikeya-Seki, which should show the largest effects, has a normal (solar) abundance ratio. The high Ni/Fe ratio observed must therefore be representative of the sublimating material or the sublimation process.
Iron in meteorites is known to be distributed between silicates, sulfides and metallic iron, with silicates and metallic iron requiring higher temperatures (about 1,200 K) to sublimate than sulfides (about 600 K), whereas nickel is found only in sulfides and the metal phase 19,20 . We may thus expect a higher Ni/Fe ratio if sublimation occurs at temperatures lower than 1,000 K. This is supported by the fact that FeNi alloys and sulfides formed in the low-temperature range are Ni-rich, such as kamacite and pentlandite 21 . Although some comets may actually be Ni-rich, partial sublimation of such species could explain the high Ni/Fe ratios that we measure in comets far enough from the Sun. Fe and Ni, as well as Ni-rich sulfides such as pentlandite, have been found in cometary material and interplanetary dust particles, often in the form of nanometre-sized particles [22][23][24] . Assuming equal numbers of Fe and Ni atoms in such compounds would explain the relative abundance close to 1 in average, but not the large over-or under-abundance    correspond to ecliptic comets with short periods (<200 years), external comets (in blue) have a semi-major axis of a < 10,000 au, and new comets (in black) come directly from the Oort cloud (a > 10,000 au).

Article
of Ni i observed in, for example, comets Garradd and 103P, or in the carbon-chain-depleted comets 21P and 73P. This interpretation requires temperatures above approximately 600 K, still higher than expected at heliocentric distances larger than 0.4 au. However, small grains can be heated at temperatures higher than blackbody equilibrium temperatures, for example, superheating of submicrometre-sized fluffy aggregates 25 . Even higher temperatures can be reached for smaller grains, such as metallic nanoparticles 26 . Collisions of high-velocity nanoparticles with cometary dust grains could break the matrix in which Fe and Ni are embedded and produce impact vapour with a temperature of the order of 1,000 K (ref. 27 ). Several mechanisms can thus potentially provide the necessary heating, especially if a considerable amount of iron and nickel is in the form of nanoparticles. The refractory elements Na, K, Si and Ca have been found by the Rosetta spacecraft in the gaseous coma of 67P at a large distance from the Sun (3 au) and attributed to ion-induced sputtering of the nucleus surface material by the solar wind, but Fe and Ni have not been reported 28 . We do not observe the light refractory elements in our spectra, and nucleus sputtering would not be active for comets closer to the Sun owing to the much denser coma and would not produce the correlations between Fe, Ni and the other volatile species that we observed (Extended Data Fig. 2).
Organometallic complexes such as [Fe(PAH)] + , carbonyls such as Fe(CO) 5 and Ni(CO) 4 , and even iron pseudocarbynes have been proposed as possible constituents of cometary or interstellar material [29][30][31][32] . The strong correlation of the production rates of iron, nickel and carbon oxides for all the comets in our sample led us to evaluate the possibility of the carbonyl hypothesis. We estimated the sublimation temperatures and sublimation rates of both Fe and Ni carbonyls (Extended Data  Fig. 3). These temperatures are only slightly higher than that of CO 2 and indicate that, if present in comets, these carbonyls can sublimate at low temperatures and at large distances from the Sun, in contrast to silicates and sulfides. This could explain why carbonyls have not been found in interplanetary dust particles, whereas they have been recently identified in the Lewis Cliff 85311 meteorite 33 . Furthermore, the higher rate of sublimation of Ni(CO) 4 compared to Fe(CO) 5 (Extended Data Fig. 3), about a factor of 10 at temperatures around 300 K (a temperature typical of the diurnal temperature of the nucleus 34 ), might provide a simple explanation to the Ni/Fe overabundance, although this scenario depends on the efficiency of the photo-dissociation of the carbonyls. Interestingly, similar computations for Cr, the next most abundant metal in the Sun after Ni, show that the sublimation rate of Cr(CO) 6 is lower by a factor of about 100 with respect to Fe(CO) 5 , which means that CrI would be a factor of around 10,000 less abundant than Fe i, explaining the non-detection of the CrI lines. A detailed photo-chemical model analysis, which is beyond the scope of this paper, would be needed to verify whether this scenario can actually reproduce the measured abundances. However, the discovery of free iron and nickel atoms in comets indicates that important constituents of the nucleus or processes in the coma are still missing, possibly bringing new important constraints on the composition of comets and the formation of the Solar System.

Online content
Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41586-021-03435-0. We consider relative abundances f x of 10 −3 -10 −5 × f x (H 2 O) for both Fe(CO) 5 and Ni(CO) 4 , and we adopt n = 10 13 cm −3 as in ref. 54 . The resulting sublimation temperatures of the iron and nickel carbonyls (97-108 K and 74-82 K, respectively, depending on f x ) are between the sublimation temperatures of H 2 O and CO 2 (152 K and 72 K), whereas CO sublimates at 25 K (ref. 54 ). The sublimation rate (in molecules cm −2 s −1 ) from the surface of a pure ice into vacuum can be expressed as 57 : Z x (T) = P v,x (T)(2πm x kT) −1/2 , where T is the ice temperature and m x the mass of species x.