60Fe deposition during the late Pleistocene and the Holocene echoes past supernova activity
Edited by Mark Thiemens, University of California San Diego, La Jolla, CA, and approved July 13, 2020 (received for review October 5, 2019)
Significance
Nearby supernova explosions shape the interstellar medium. Ejecta, containing fresh nucleosynthetic products, may traverse the solar system as a transient passage, or alternatively the solar system may traverse local clouds that may represent isolated remnants of supernova explosions. Such scenarios may modulate the galactic cosmic-ray flux intensity to which Earth is exposed. Varying conditions of the traversed interstellar medium could have impacts on climate and can be imprinted in the terrestrial geological record. Some radionuclides, such as 60Fe, are not produced on Earth or within the solar system in significant quantities. Their existence in deep-sea sediments demonstrates recent production in close-by supernova explosions with a continued influx of 60Fe until today.
Abstract
Nuclides synthesized in massive stars are ejected into space via stellar winds and supernova explosions. The solar system (SS) moves through the interstellar medium and collects these nucleosynthesis products. One such product is 60Fe, a radionuclide with a half-life of 2.6 My that is predominantly produced in massive stars and ejected in supernova explosions. Extraterrestrial 60Fe has been found on Earth, suggesting close-by supernova explosions ∼2 to 3 and ∼6 Ma. Here, we report on the detection of a continuous interstellar 60Fe influx on Earth over the past ∼33,000 y. This time period coincides with passage of our SS through such interstellar clouds, which have a significantly larger particle density compared to the local average interstellar medium embedding our SS for the past few million years. The interstellar 60Fe was extracted from five deep-sea sediment samples and accelerator mass spectrometry was used for single-atom counting. The low number of 19 detected atoms indicates a continued but low influx of interstellar 60Fe. The measured 60Fe time profile over the 33 ky, obtained with a time resolution of about ±9 ky, does not seem to reflect any large changes in the interstellar particle density during Earth’s passage through local interstellar clouds, which could be expected if the local cloud represented an isolated remnant of the most recent supernova ejecta that traversed the Earth ∼2 to 3 Ma. The identified 60Fe influx may signal a late echo of some million-year-old supernovae with the 60Fe-bearing dust particles still permeating the interstellar medium.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
The temperature and particle density of the interstellar medium (ISM) is shaped by winds from massive stars and stellar explosions. Near the solar system (SS), supernovae (SNe) have created a network of filaments, shells, and superbubbles (1–7). Superbubbles represent low-density cavities in space with typical densities of ∼0.005 hydrogen atoms per cm3 (10−26 g·cm−3) and exist for tens of millions of years. The SS has been located inside the Local Superbubble (LB) for at least the last 3 My and possibly more than 10 My (8). The LB extends for ∼60 to 100 pc from the sun in the galactic plane and forms an open structure perpendicular to the plane (galactic chimney) (3).
The LB does, however, contain some higher-density regions, and the SS is presently immersed in a small cluster of local interstellar clouds (CLIC). These are partially ionized individual clouds within the much-lower-density LB material, filling between 6 and 19% of the LB volume (4). Because the SS is moving at a considerable velocity sometimes exceeding 25 km·s−1 relative to these clouds, it has encountered them regularly, with the first encounter sometime between 44 and 150 ky BP (1, 6, 7). For a cloud length of ∼1 pc the cloud transit time becomes on the order of 40 ky. Presently, the SS is traversing the so-called local interstellar cloud (LIC) at a relative velocity of 25.7 km·s−1 (9). This small local cloud has a density of ∼0.2 hydrogen atoms per cm3 (about 40 times the density of the LB) and has a size of ∼5 pc (1). The SS entered the LIC sometime between 4 and 40 ky BP, but it will be leaving the LIC within the next few thousand years because it is passing very near the edge (1) (Fig. 1).
Fig. 1.
As a consequence, the SS has traversed different regions of ISM during the past several million years, which may have affected the heliosphere, the inner SS, and also the flux of interstellar dust and galactic cosmic rays at Earth (1, 10–13). Several questions arise: What is the origin of these interstellar clouds? Are they enriched in material from SNe, or are they simply representative of the average ISM? Will passage through such clouds modulate the galactic cosmic-ray flux into the inner SS?
Several formation scenarios have been proposed for the CLIC, including the LIC. The CLIC could be a result of interactions between the LB and its neighboring Loop I superbubble (14) or an array of ISM structures consisting of material that originated from the inside surface of the LB wall and these structures were generated by a distorted magnetic field (magnetic flux tube) (15), or it may represent a dense cloud that survived a shock wave from an expanding bubble (1, 4, 7).
To address these questions, we searched for a geological radioisotope record that may allow the mapping of signatures of such cloud transitions. We have measured concentrations of the radioisotope 60Fe in deep-sea sediments covering the last ∼33 ky. 60Fe has a half-life of (2.61 ± 0.04) My (16–18). Because it is not naturally produced on Earth, the presence of 60Fe is a sensitive indicator of supernova explosions within the last few million years. The 60Fe can reach Earth because it is trapped in interstellar dust grains that can penetrate into the SS (10–13, 19–24). It has been shown that there was significantly enhanced 60Fe deposition on Earth at 1.7 to 3.2 My (25–30) and ∼6 My BP (28) (Fig. 1), as well as a low present-day influx (31). 60Fe has also been detected in lunar samples (32), in the galaxy via gamma rays associated with its radioactive decay (33) and in galactic cosmic rays (34). In particular, these multiple influxes during the last 10 My, as observed in deep-sea ferromanganese crusts and sediments, suggest that Earth may have been exposed to waves of SN ejecta, or alternatively it may have traversed clouds of 60Fe-enriched dust. Passage through the CLIC could be a possible source for an enhanced 60Fe influx during the past few tens of thousands of years. Alternatively, 60Fe-bearing dust grains might permeate the interstellar space, including interstellar clouds unaffected by the ISM structure they pass. In such a scenario, the terrestrial 60Fe-influx pattern would not be correlated with the changing mass density faced by the SS while moving through the ISM.
Results
Trace concentrations of 60Fe were measured in five samples originating from two different piston cores, extracted during the Eltanin expeditions and provided by the Antarctic Marine Geology Research Facility at Florida State University. The cores, E45-21 (38°58.7′S, 103°43.1′E) and E50-02 (39°57.5′S, 104°55.7′E), were collected from sites ∼1,000 km southwest of Australia in the Indian Ocean at depths of ∼4,200 m (28, 35–39). E45-21 was one of the cores that showed enhanced 60Fe depositions between 1.7 and 3.2 My at depths of 400 to 700 cm (28). Here, the emphasis was on the top 13 cm of these cores that covers the past ∼33 ky. Three samples of 1-cm thickness were taken from core E45-21 and two from E50-02. These five samples encompass the depth range 0 to 13 cm with 3- to 4-cm intervals between them (Table 1).
Table 1.
| Sample ID | Core | Depth, cm | Time period, ka* | 60Fe detected | 60Fe-bgr expected† | 60Fe/Fe|meas, 10−15 atom/atom | 60Fe/Fe|corr, 10−15 atom/atom‡ | Fe conc., mg·g−1§ | 60Fe conc., 103 atom·g−1 | 60Fe inc. rates, atom·cm−2·y−1 |
|---|---|---|---|---|---|---|---|---|---|---|
| 85 | 45-21 | 0–1 | 0–2.8 | 6 | 0.9 | 4.10 | ||||
| 81 | 50-02 | 4–5 | 10–12.5 | 7 | 1.4 | 6.66 | ||||
| 83 | 45-21 | 6–7 | 16.3–19.1 | 3 | 0.45 | 2.77 | ||||
| 84 | 45-21 | 9–10 | 24.5–27.3 | 0 | 0.25 | 2.99 | ||||
| 82 | 50-02 | 12–13 | 30.0–32.5 | 3 | 0.7 | 4.75 | ||||
| All¶ | 0–13 | 0.0–32.5 | 19 | 3.6 | 4.87 | |||||
| Blank | Terrestrial | — | — | < 0.01 | ||||||
| Mean# | 47 samples | 1.7–3.2 Ma | 288 | 24.1 ± 1.8 |
All uncertainties (1-σ) are calculated using Feldman and Cousins statistics (40).
*
Sedimentation rate of 3.7 mm/ky and 4.0 mm/ky for cores 45-21 and 50-02, respectively (38); we assumed for the top layers the same sedimentation rates as measured for the deeper layers (1.7 to 3.2 Ma).
†
Background (bgr) events as expected from measured terrestrial blanks which are assumed to contain negligible amounts of 60Fe.
‡
Background-corrected.
§
Stable Fe concentrations (conc.) were measured by ICP-MS at HZDR with a typical uncertainty of 5%.
¶
Weighted with the total amount of analyzed material.
#
For comparison we list the mean value for the time period 1.7 to 3.2 Ma as given in ref. 28.
Iron and beryllium were leached from 3 g of these sediment samples at the Helmholtz-Zentrum Dresden-Rossendorf (HZDR) (39). A modified version of the method proposed by Bourlès et al. (41) and Merchel and Herpers (42) was applied (for more details see ref. 43). This gentle procedure extracts the authigenic fraction of the sample containing the soluble beryllium (10Be and 9Be) and iron (60Fe and stable iron). We assume negligible losses of extraterrestrial 60Fe relative to stable iron for that procedure (as tested by ref. 32). Purified beryllium oxide and iron oxide samples were mixed with Nb and Ag powder, respectively, for accelerator mass spectrometry (AMS) analysis (Materials and Methods) at HZDR (44) and the Australian National University (ANU) (17, 28). Since AMS measures the 60Fe/Fe isotope ratio, the concentrations of the stable Fe were determined via inductively coupled plasma mass spectrometry (ICP-MS) at HZDR so that 60Fe concentrations could then be derived.
A mean measurement background of 60Fe/Fe = (0.27 ± 0.08) × 10−16 was determined over the past 5 y of 60Fe AMS measurements on commercial iron oxide samples containing no 60Fe, equivalent to less than one identified detector event over approximately one full day of measurement (see also ref. 28).
For deeper layers sedimentation rates between 3.7 (Eltanin 45-21) and 4.0 mm/ky (Eltanin 50-02) were determined from magnetostratigraphy and terrestrial 10Be/9Be dating (28, 35, 38). The 10Be concentrations for the five surface sediment samples agree well with the decay-corrected data obtained for the deeper layers and confirm present-day ages of these samples (SI Appendix).
Assuming similar sedimentation rates for the first few centimeters of sediment as those measured in the deeper layers, the 1-cm-thick layers of the samples represent a time period of ∼2.5 ky per layer. Postdepositional particle redistribution by bioturbation and movement of the 60Fe in solution in the pore waters, however, does degrade this time resolution (45). We refer to independent measurements of Lee et al. (46) and assume a constant 7-cm mixing length over the entire time period (45, 46); this yields a total of ±9 ky time resolution per individual 1-cm sample.
The three samples from Eltanin 45-21 and two from Eltanin 50-02, therefore, provided a time record that extended from the present to 27.3 (10 cm) and 32.5 ky (13 cm), respectively. Taking mixing into account, the time range covers ∼40 ky.
The five individual sediment samples gave between zero and seven 60Fe events in the detector (Table 1). In total 19 60Fe events were detected compared to an expected background of 3.6 detector counts scaled from measurements of pure terrestrial iron samples (assumed to contain negligible 60Fe). The individual 60Fe events are listed together with the expected number of background events in Table 1. Correcting for background and applying counting statistics for a low number of detector events (40), we deduce an average isotope ratio of (68% confidence interval) and significant above background with >99.99% confidence. The measured 60Fe/Fe atom ratios for the individual samples are listed in Table 1. There is no significant difference within statistics between samples representing different time periods and hence no obvious trend with sample depth or age (Fig. 2).
Fig. 2.
Interstellar dust particles ablate during passage through the atmosphere and the interstellar 60Fe atoms would therefore have been released from the dust particles when they arrived at the surface of the ocean (22). 60Fe is then transferred in the same way as terrestrial stable Fe to the sediment archive, via scavenging processes. The measured 60Fe/Fe isotope ratio can be converted into an incorporation rate for 60Fe (atoms per square centimeter per year) using the stable (terrestrial) Fe content as measured by ICP-MS (Table 1), the mean density of the archives (1.16 g·cm−3), and their sedimentation rates (3.7 and 4 mm·ky−1) (28, 38).
Here, deep-sea sediments are assumed to incorporate all 60Fe from the water column above due to the high particle reactivity which results in 100% transfer into the sediments. Hence, averaged over the entire time period we obtain a mean deposition rate of 60Fe atoms·cm−2·y−1, corresponding to an accumulated incorporation of atoms·cm−2 over 33 ky (Table 1). A previous preliminary value for a subset of this sediment gave a 2-σ upper limit of <3.6 60Fe atoms·cm−2·y−1 (28), consistent with our new value if the same time period is assumed for the sediment samples.*
We can compare this result with those from deep-sea ferromanganese crusts, another geological record where 60Fe has been successfully identified (26, 28). These crusts, however, have incorporation efficiencies for Fe that can be significantly lower than 100%. For example, “Crust-3,” recovered from the Pacific Ocean, has an incorporation efficiency of ∼17%. It has recently been studied with ∼1-mm layer depth resolution. Its top layer averages over the most recent 370 ky and gives an incorporation rate of 60Fe atoms·cm−2·y−1. Corrected for the crust incorporation efficiency this corresponds to a deposition rate of 60Fe atoms·cm−2·y−1, indicating a 60Fe influx averaged over this 10-times-longer period that is a factor of ∼2 to 3 lower than was found for the ∼33-ky average in the sediment. The top 1 mm layer of another crust (237KD) (26) suggests a similar result [assuming 10% incorporation (19, 28)] (Fig. 3). The detection of 60Fe in Antarctic snow (31) suggests a present-day influx (<20-y accumulation) of 60Fe atoms·cm−2·y−1 (Fig. 3), which is in line with the results from “Crust 3.” Note that if the sedimentation rate of present-day sediments would be different from the 1.7- to 3.2-My average the 60Fe deposition rate for the top layers would change accordingly.
Fig. 3.
The high-resolution data obtained here over a period of ∼33 ky would in principle allow a time-dependent 60Fe flux into the SS to be probed on time scales of ∼10 ky, since the residence time of Fe in the ocean is negligible [∼100 y (22)]. However, owing to the low 60Fe influx and the consequent detection of only a few 60Fe events per sample, the data obtained here allow us to generate a meaningful value for the 60Fe influx only for the average over 33 ky, but not for individual data points (Fig. 2).
Discussion
Our data demonstrate detection of an interstellar 60Fe influx over the recent past of ∼33 ky that appears roughly constant over this time. Note that the Antarctic snow data (31) also confirm the existence of a present-day influx (<20 y) at 2.4 σ significance (Fig. 3). Assuming that extraterrestrial 60Fe is homogeneously distributed over Earth’s surface (5.1 × 1018 cm2), we derive from our data that 5.9 × 1023 60Fe atoms (∼60 g) reached Earth during the past 33 ky.
ISM dust particles that incorporate 60Fe are the most probable means for 60Fe to enter the SS [neglecting the orders-of-magnitude lower 60Fe influx as highly energetic cosmic ray particles (34)] and to be deposited in terrestrial archives; dust overcomes the solar wind pressure and the SS magnetic field. Satellite-borne instruments on the Ulysses, Galileo, and Cassini space missions (11–13, 47–49) have detected ISM dust inside the SS, and the data suggest that a mass fraction of 3 to 6% of interstellar dust particles is presently able to penetrate deep into the SS and reach Earth. Taking the median value of 4.5%, the observed terrestrial deposition of 60Fe atoms·cm−2·y−1 can be converted to an ISM 60Fe concentration of (3.8 ± 1.7) × 10−12 60Fe atoms·cm−3 in dust for the local ISM. Note that it is necessary first to scale the deposition rate by a factor of 4 to take into account that deposition occurs over the surface of the globe, whereas the volume swept out by Earth’s passage through the ISM is proportional to its cross-sectional area. Here, we assumed a velocity of the SS relative to the dust particles in the LIC of 25.7 km·s−1 and an uncertainty in the dust penetration efficiency of 33% (Table 2). Interestingly, this derived 60Fe concentration, which is representative of the local ISM during the past ∼33 ky, is closely similar to the galaxy-averaged 60Fe concentration in the ISM of ∼4 × 10−12 atoms·cm−3 as deduced from the average SN rate in the galaxy and gamma-astronomy data (33).
Table 2.
| 0–0.02 ka | 0–33 ka | 0–10 Ma average | 1.7–3.2 Ma | |
|---|---|---|---|---|
| Snow (31) | ELT | ELT and crust | ELT peak | |
| Deposition rate, atom·cm−2·y−1 | 5.1 ± 0.5 | 24 ± 2 | ||
| 60Fe-SS flux at 1 AU, atom·cm−2·y−1 | 14 ± 4 | 20 ± 2 | 98 ± 8 | |
| 60Fe ISM flux, 103 atom·cm−2·y−1 | 0.11 ± 0.06 | 0.31 ± 0.14 | 0.45 ± 0.15 | 2.1 ± 0.8 |
| Velocity, km·s−1* | 25.7 | 25.7 | ∼15 ± 5 | ∼15 ± 5 |
| Distance traversed per My, pc·My−1 | 26 | 26 | 15 ± 5 | 15 ± 5 |
| Distance traversed, pc | 0.0005 | 0.9 | 153 ± 51 | 22 ± 7 |
| 60Fe ISM particle conc., 10−12 60Fe·atom·cm−3 | 1.3 ± 0.7 | 3.8 ± 1.7 | 9.6 ± 4.3 | 47 ± 22 |
| ISM particle density, H-atom·cm−3 | 0.2 | 0.2 | 0.005 | — |
The 60Fe deposition rate for the 10-My case is a combined value deduced from the Eltanin sediment (ELT), crust (28), and Antarctic snow data (31). The 60Fe ISM flux is calculated assuming a dust particle penetration probability from the ISM into the SS of (4.5 ± 1.5)%. For comparison of different influx scenarios, we assumed here the present density of the LB to be simply representative for the past 10 My (150 pc, with 1 pc = 3.086 × 1018 cm). This low value will not reflect the real average density over the past 10 My but will be representative for a large fraction of the time during the past 10 My. The dust penetration probability was assumed to be identical for the 0 to 33 ky and 0 to 10 My time periods.
The 60Fe concentration of (3.8 ± 1.7) × 10−12 atoms·cm−3 in the LIC measured in the present work is an order of magnitude lower than the value of ∼47 × 10−12 atoms·cm−3 averaged over the 1.7-to-3.2-My peak. This elevated 60Fe influx between 1.7 and 3.2 My (and between 5.5 and 7 My) is most readily interpreted in terms of transient SN ejecta passing Earth (28). Given the much lower influx during the last 33 ky, during which the SS spent at least some of the time within the LIC, it seems unlikely that the LIC (and possibly the CLIC as well) represents a residual structure of a broken-up SN ejecta shell from this younger event. Note that the 60Fe influx over the past 33 ky seems comparable to the (not-decay-corrected) influx measured for the older, ∼6- to 7-My event of ∼1.5 60Fe atoms·cm−2·y−1 in its peak (28). The intensity of the present-day influx would fit to a Super-AGB star production having typically much lower nucleosynthesis yields compared to SNe (50) but there are no stellar candidates within a few parsecs, and additionally the stable isotope signature in the local ISM, such as He and Ne, would not fit this scenario (51, 52).
Conclusions
Our data clearly show that there is 60Fe in the local ISM, confirming present-day Antarctic snow data (31). Koll et al. (31) noted that a sharp increase in the flux of 60Fe would be expected around the time when the SS entered the LIC, if the LIC is the origin of the detected 60Fe. Our data, although based on low statistics, do not indicate such an increase during the past 33 ky, even though it is likely that the SS entered the LIC during this time (1, 4), and hence that at least the oldest sediment sample should represent a time period before entering the LIC.
Possible scenarios for the presence of 60Fe in the local ISM might be as follows:
1)
Gradual fading away of the transient passage of the supernova debris that produced the 1.7-to-3.1-My peak.
2)
3)
The LIC may be an independent ISM structure that survived the passage of the most recent SN ejecta (1.7 to 3.2 Ma) (54). If the 60Fe observed here is coming from the LIC, it could represent the residue from an older event (note e.g., the enhanced 60Fe influx ∼6 Ma, where radioactive decay has reduced an originally higher 60Fe concentration).
To discriminate between these possibilities, it will be necessary to extend the sediment data further into the past, and in particular to fill in the gap between the ∼40 ky covered by the present work and ∼1 Ma. If the 60Fe influx is found to increase steadily toward the peak at 1.7 to 3.1 Ma, then the first explanation would be favored. If, on the other hand, the influx is found to be lower at >40 ky than at present, this would favor the LIC as the source of the 60Fe and the third possibility would be the more likely. Existing crust data already cover this region in principle, but the time resolution is insufficient to distinguish between the above scenarios.
Materials and Methods
Sample Measurement.
In general, individual samples contained between some 103 and ∼104 60Fe atoms per g sediment, equivalent to one 60Fe-decay every 400 to 4,000 y in a 1-g sediment sample. Thus, we have applied the most sensitive technique, AMS (55–58), to directly count the minute amounts of ISM 60Fe, via its isotope ratio 60Fe/Fe. The corresponding 60Fe concentrations were derived from the concentrations of the stable Fe determined via ICP-MS. The 60Fe measurements were performed with the 14UD accelerator at the ANU (17, 28, 59, 60).
Details on the Determination of 60Fe.
The AMS measurements determine the atom ratio, for example 60Fe/AFe, where AFe (A = 54 or 56) is a stable isotope measured as an ion current in a Faraday cup in front of the particle detection system. We expect 60Fe/56Fe ratios between 10−17 and a few 10−16 from all samples. We have developed the capability to detect trace amounts of 60Fe in terrestrial archives by AMS with an overall efficiency (atoms detected/atoms in the sample) of 0.5‰ (28), that is, one 60Fe event in the detector would represent ∼2,000 60Fe atoms in the analyzed sample. The ANU setup has been optimized for high measurement efficiency and selectivity.
For such experiments where only a few counts are expected, it is crucial to suppress completely any interfering backgrounds due to stable elements of the same mass (e.g., 60Ni in the case of 60Fe), molecules of the same mass, and tails of the hugely more abundant stable isotopes (e.g., 54,56,57,58Fe). This is achieved using the high energies of ∼170 MeV available from the 14UD tandem accelerator operating at ∼14 MV in combination with a gas-filled magnet detection system. At these ultralow levels, the elimination of interference from the stable isobar 60Ni becomes extremely challenging. Although AMS completely destroys and removes molecular ions of mass 60 in the beam with the use of the accelerator and subsequent mass filters, the stable nuclide 60Ni behaves in exactly the same way through the accelerator and subsequent analyzers and hence is still present at levels up to 10 orders of magnitude above the rare 60Fe. Spatial separation of the 60Fe and 60Ni isobars was achieved in a gas-filled magnet, allowing the great majority of the 60Ni to be blocked from entering the final particle detector (17, 28, 60). A multielectrode ionization chamber, which determines not only the energy but also the rate of energy loss of each ion and its position, then allows the few 60Fe events to be clearly separated from the residual 60Ni. The gas-filled magnet reduces the 60Ni intensity (typically 105 to 106 counts per s) by a factor of ∼104 to a rate that the ionization detector can handle comfortably (61, 62). This setup provides a powerful means to reject the 60Ni isobaric background but accepts essentially all of the 60Fe events.
Due to the low number of expected 60Fe events, the measurement background was carefully monitored with samples of iron oxides that contained negligible 60Fe. The measurement background achieved for the system was 60Fe/Fe = (2.7 ± 0.8) × 10−17. The unknown samples were measured relative to standard-type samples (isotope ratios known with ∼10% accuracy) which were based on material from the Technical University of Munich and on material from recent meteorite studies (63). They had been previously cross-calibrated against a material from the Paul Scherrer Institute with well-known isotope ratios (64). These samples confirmed the validity of scaling the system settings between the different masses, with the data showing a reproducibility of ∼5%.
The measurement procedure was a slow sequence of 10-s measurements of the 54Fe10+ ion current, alternating with counting periods of 60Fe11+ ions that were typically 20 to 30 min for the real samples and blanks but were reduced to 3 to 5 min for the standard-type materials. All samples were measured repeatedly.
The five individual samples resulted in 6 to 20 mg of Fe2O3 which were subdivided into three to seven individual sputter samples for AMS. Sputtering times to fully consume the material were between 2 and 12 h per sample; this corresponds to a total counting time of ∼30 h and a similar time for blank samples. On average the crust samples gave ∼0.6 60Fe detector events per hour compared to <0.1 events per hour for blanks. Overall, distributed over three AMS beam times, 19 detector events were registered for the sediment samples compared to two detector events for blanks.
Arguments against a 60Fe Influx from Meteorites or Micrometeorites in the Past 40 ky.
Although we do not expect any significant 60Fe production on Earth (28, 31), some 60Fe will be deposited continuously through interplanetary material that bombards Earth. In space, highly energetic cosmic rays (predominantly galactic cosmic protons) will—via secondary neutrons and protons—produce 60Fe through spallation reactions on Ni target atoms in interplanetary objects. Some earlier studies suggested that micrometeorites (MM) captured by Earth could also be responsible for the 60Fe increase observed in crust 237KD in the period ∼2 to 3 My BP (65, 66).
We can estimate an interplanetary contribution to the total 60Fe influx from the total amount of interplanetary material accreting on Earth: About (30,000 ± 20,000) tons of cosmic dust reaches Earth (67) per year, predominantly through MM and interplanetary dust particles. Objects that are more massive contribute less than 1% of the total mass flux. With measured concentrations of 60Fe in micrometeorites or interplanetary dust of 0.51 dpm (decays per minute)·kg−1 Ni (1 × 1012 60Fe atoms·kg−1 Ni) (63, 68, 69) and a Ni content in cosmic dust (interplanetary particles) of ∼1% (CI chondrites) we calculate a flux of 0.06 60Fe atoms·cm−2·y−1, evenly spread over Earth’s surface (Fig. 3). We measured a mean 60Fe flux of 3.5 ± 1.0 60Fe atoms·cm−2·y−1 for the past 33 ky, that is, the observed 60Fe influx is 58 ± 17 times higher than expected from a constant influx of interplanetary particles.
We also measured the Ni content in the leachates of the five individual sediment samples by means of ICP-MS. The sediments contained on average 40 µg leachable Ni per g sediment, compared to 4,900 µg of leached Fe per g sediment (Table 1) (28). Thus, we obtain an element concentration ratio [Ni]/[Fe] of 0.008 atom/atom (see also refs. 28 and 39). With 1 × 1012 60Fe atoms per kg Ni, we calculate for the sediment samples a 60Fe concentration of 4 × 104 atom·g−1, which is five times higher than the measured ratio (Table 1). Thus, if 20% of the total Ni in the archives would be of (micro)meteoritic origin, we could account for the measured 60Fe. The leachable Ni concentration was found to vary by less than 10% in all five sediment samples. For comparison, the 30,000 tons per year of cosmic dust influx (see also refs. 70–73 indicating an even lower cosmic dust influx), assuming 10% is Fe (1% is Ni) and distributed equally across Earth’s surface, corresponds to a stable ISM Fe (Ni) concentration in the sediments of 1.3 μg/g sediment (0.13 μg/g for Ni). The extraterrestrial stable Fe influx would be less than a per mille contribution to the total Fe in the sample (4,900 µg/g) and as a consequence the leached fraction of the sediment is completely dominated by stable Fe of terrestrial origin. Hence, we must assume a similar scenario for Ni. In summary, under the assumptions made above of 60Fe production in (micro)meteorites, an interplanetary source for the elevated 60Fe influx is not supported (see also the discussion in the supplement to ref. 28).
Arguments against a Significant 60Fe Production from Solar Events in the past 40 ky.
Independent from interstellar sources, extreme solar proton events can also lead to a significantly increased atmospheric production of cosmogenic radionuclides which eventually can become incorporated in terrestrial archives. Recent detections of excursions in the 14C record at AD 774/775 and 993/994 as well as in 14C, 10Be, and 36Cl at ∼660 BC have, for example, been found in tree rings and ice cores, respectively, which points to an extraterrestrial origin (refs. 74–77 and references therein). Such short-term solar events require time resolutions in the archive of the order of years. The total 60Fe production for such events will be very low because of these short time scales and because of the scarcity of the required Ni target nuclei in the atmosphere needed for 60Fe production. Therefore, such events would not be detectable in our samples.
Data Availability Statement.
All relevant data are included in the paper and SI Appendix.
Acknowledgments
We thank the Antarctic Marine Geology Research Facility, Florida State University (C. Sjunneskog) for providing the sediment cores. This work was funded by Austrian Science Fund project AI00428, through the European Science Foundation Collaborative Research Project CoDustMas; Australian Research Council projects DP140100136, DP180100495, and DP180100496; German Academic Exchange Service project 56266169; and The Group of Eight Australia–Germany Joint Research Cooperation Scheme. J.F. acknowledges a stipend from the University of Vienna and R.G. support from the European Cooperation in Science and Technology “ChETEC” Action (CA16117). We also acknowledge financial support from the Australian Government for the Heavy Ion Accelerator Facility at ANU through the National Collaborative Research Infrastructure Strategy. ICP-MS measurements were performed by S. Beutner (HZDR).
Supporting Information
Appendix (PDF)
- Download
- 308.83 KB
References
1
P. Frisch, V. V. Dwarkadas, “Effect of supernovae on the local interstellar material” in Handbook of Supernovae, A.W. Alsabti, P. Murdin, Eds. (Springer International Publishing AG, 2016).
2
D. P. Cox, P. R. Anderson, Extended adiabatic blast waves and a model of the soft X-ray background. Astrophys. J. 253, 268 (1982).
3
R. K. Smith, D. P. Cox, Multiple supernova remnant models of the local bubble and the soft X-ray background. Astrophys. J. Suppl. Ser. 134, 283 (2001).
4
P. C. Frisch et al., The galactic environment of the Sun: Interstellar material inside and outside of the heliosphere. Space Sci. Rev. 146, 235–273 (2009).
5
D. Breitschwerdt et al., The locations of recent supernovae near the Sun from modelling (60)Fe transport. Nature 532, 73–76 (2016).
6
S. Redfield, J. L. Linsky, The structure of the local interstellar medium. IV. Dynamics, morphology, physical properties, and implications of cloud-cloud interactions. Astrophys. J. 673, 283 (2008).
7
P. Frisch, D. G. York, The Galaxy and the Solar System, (University of Arizona Press, Tucson, 1986), pp. 83–100.
8
B. Fuchs, D. Breitschwerdt, M. A. de Avillez, C. Dettbarn, C. Flynn, The search for the origin of the Local Bubble redivivus. Mon. Not. R. Astron. Soc. 373, 993 (2006).
9
R. Lallement, R. Ferlet, A. M. Lagrange, M. Lemoine, A. Vidal-Madjar, Local cloud structure from HST-GHRS. Astron. Astrophys. 304, 461 (1995).
10
E. Gruen et al., Discovery of Jovian dust streams and interstellar grains by the Ulysses spacecraft. Nature 362, 428 (1993).
11
A. J. Westphal et al.; 30714 Stardust@home dusters, Interstellar dust. Evidence for interstellar origin of seven dust particles collected by the Stardust spacecraft. Science 345, 786–791 (2014).
12
I. Mann, Interstellar dust in the solar system. Annu. Rev. Astron. Astrophys. 48, 173–203 (2010).
13
N. Altobelli et al., Interstellar dust flux measurements by the Galileo dust instrument between the orbits of Venus and Mars. J. Geophys. 110, A07102 (2005).
14
D. Breitschwerdt, M. J. Freyberg, R. Egger, Origin of H I clouds in the local bubble. I. A hydromagnetic Rayleigh-Taylor instability caused by the interaction between the Loop I and the local bubble. Astron. Astrophys. 361, 303 (2000).
15
P. D. Cox, L. Helenius, Flux-tube dynamics and model for the origin of the local fluff. Astrophys. J. 583, 205 (2003).
16
G. Rugel et al., New measurement of the 60Fe half-life. Phys. Rev. Lett. 103, 072502 (2009).
17
A. Wallner et al., Settling the half-life of 60Fe: Fundamental for a versatile astrophysical chronometer. Phys. Rev. Lett. 114, 041101 (2015).
18
K. M. Ostiek et al., Activity measurement of 60Fe through the decay of 60mCo and confirmation of its half-life. Phys. Rev. C 95, 055809 (2017).
19
A. Wallner et al., Abundance of live 244Pu in deep-sea reservoirs on Earth points to rarity of actinide nucleosynthesis. Nat. Commun. 6, 5956 (2015).
20
E. Dwek, The evolution of the elemental abundances in the gas and dust phases of the galaxy. Astrophys. J. 501, 643–665 (1998).
21
J. Ellis, B. D. Fields, D. N. Schramm, Geological isotope anomalies as signatures of nearby supernovae. Astrophys. J. 470, 1227 (1996).
22
B. J. Fry, B. D. Fields, J. R. Ellis, Radioactive iron rain: Transporting 60Fe in SN dust to the ocean floor. Astrophys. J. 827, 48 (2016).
23
G. Korschinek, T. Faestermann, K. Knie, C. Schmidt, 60Fe, a promising AMS isotope for many applications. Radiocarbon 38, 68 (1996).
24
T. Athanassiadou, B. D. Fields, Penetration of nearby supernova dust in the inner solar system. New Astron. 16, 229–241 (2011).
25
K. Knie et al., Indication for supernova produced 60Fe activity on earth. Phys. Rev. Lett. 83, 18–21 (1999).
26
K. Knie et al., 60Fe anomaly in a deep-sea manganese crust and implications for a nearby supernova source. Phys. Rev. Lett. 93, 171103 (2004).
27
C. Fitoussi et al., Search for supernova-produced 60Fe in a marine sediment. Phys. Rev. Lett. 101, 121101 (2008).
28
A. Wallner et al., Recent near-Earth supernovae probed by global deposition of interstellar radioactive 60Fe. Nature 532, 69–72 (2016).
29
P. Ludwig et al., Time-resolved 2-million-year-old supernova activity discovered in Earth’s microfossil record. Proc. Natl. Acad. Sci. U.S.A. 113, 9232–9237 (2016).
30
G. Korschinek et al., Supernova-produced 53Mn on Earth. Phys. Rev. Lett. 125, 031101 (2020).
31
D. Koll et al., Interstellar 60Fe in Antarctica. Phys. Rev. Lett. 123, 072701 (2019).
32
L. Fimiani et al., Interstellar 60Fe on the surface of the moon. Phys. Rev. Lett. 116, 151104 (2016).
33
R. Diehl, Nuclear astrophysics lessons from INTEGRAL. Rep. Prog. Phys. 76, 026301 (2013).
34
W. R. Binns et al., Observation of the 60Fe nucleosynthesis-clock isotope in galactic cosmic rays. Science 352, 677–680 (2016).
35
E. Allison, M. T. Ledbetter, Timing of bottom-water scour recorded by sedimentological parameters in the south Australian basin. Mar. Geol. 46, 131–147 (1982).
36
L. A. Frakes, USNS Eltanin sediment descriptions, cruises 32-45 (Contribution 33, Sedimentary Research Laboratory, Florida State University, Tallahassee, 1971).
37
L. A. Frakes, USNS Eltanin sediment descriptions, cruises 4-54 (Contribution 37, Sedimentary Research Laboratory, Florida State University, Tallahassee, 1973).
38
J. Feige et al., Limits on supernova-associated 60Fe/26Al nucleosynthesis ratios from AMS measurements of deep-sea sediments. Phys. Rev. Lett. 121, 221103 (2018).
39
J. Feige et al., The search for supernova-produced radionuclides in terrestrial deep-sea archives. Publ. Astron. Soc. Aust. 29, 109–114 (2012).
40
G. J. Feldman, R. D. Cousins, Unified approach to the classical statistical analysis of small signals. Phys. Rev. D Part. Fields 57, 3873–3889 (1998).
41
D. Bourlès, G. M. Raisbeck, F. Yiou, 10Be and 9Be in marine sediments and their potential for dating. Geochim. Cosmochim. Acta 53, 443–452 (1989).
42
S. Merchel, U. Herpers, An update on radiochemical separation techniques for the determination of long-lived radionuclides via accelerator mass spectrometry. Radiochim. Acta 84, 215–219 (1999).
43
J. Feige et al., “AMS measurements of cosmogenic and supernova-ejected radionuclides in deep-sea sediment cores” in EPJ Web of Conferences, (2013), Vol. 63,.
44
S. Akhmadaliev, R. Heller, D. Hanf, G. Rugel, S. Merchel, The new 6MV AMS-facility DREAMS at Dresden. Nucl. Instr. Method. B 294, 5 (2013).
45
L. R. Teal, M. T. Bulling, E. R. Parker, M. Solan, Global patterns of bioturbation intensity and mixed depth of marine soft sediments. Aquat. Biol. 2, 207–218 (2008).
46
H. Lee et al., Distribution and inventories of 90Sr, 137Cs, 241Am and Pu isotopes in sediments of the Northwest Pacific Ocean. Mar. Geol. 216, 249 (2005).
47
N. Altobelli et al., Cassini between venus and earth: Detection of interstellar dust. J. Geophys. Res. 108, 8032 (2003).
48
A. Li, “Interstellar grains - The 75th anniversary” in Light, Dust, and Chemical Evolution, R. Saija, C. Cecchi-Pestellini, Eds. (Journal of Physics: Conference Series, IOP Publishing, Bristol, UK, 2005), Vol. 6, pp. 229–248.
49
B. J. Fry, B. D. Fields, J. R. Ellis, No escape from the supernova! Magnetic imprisonment of dusty pinballs by a supernova remnant. Astrophys. J. 894, 109 (2020).
50
C. L. Doherty, P. Gil-Pons, H. H. B. Lau, J. C. Lattanzio, L. Siess, Super and massive AGB stars - II. Nucleosynthesis and yields - Z = 0.02, 0.008 and 0.004. Mon. Not. R. Astron. Soc. 437, 195 (2014).
51
E. Salerno et al., Measurement of 3He/4He in the local interstellar medium: The collisa experiment on Mir. Astrophys. J. 585, 840–849 (2003).
52
G. Gloeckler, L. Fisk, Composition of Matter, (Springer, 2007), Vol. 27.
53
M. M. Schulreich, D. Breitschwerdt, J. Feige, C. Dettbarn, Numerical studies on the link between radioisotopic signatures on Earth and the formation of the Local Bubble. I. 60Fe transport to the solar system by turbulent mixing of ejecta from nearby supernovae into a locally homogeneous ISM. Astron. Astrophys. 604, A81 (2017).
54
J. D. Slavin, The origins and physical properties of the complex of local interstellar clouds. Space Sci. Rev. 143, 311 (2009).
55
H.-A. Synal, “Developments in accelerator mass spectrometry” in 100 Years of Mass Spectrometry, K. Blaum, Y. Litvinov, Eds. (Int. J. Mass Spectrom, 2013), Vol. 349–350, pp. 192–202.
56
W. Kutschera, “Applications of accelerator mass spectrometry” in 100 Years of Mass Spectrometry, K. Blaum, Y. Litvinov, Eds. (Int. J. Mass Spectrom, 2013), Vol. 349-350, pp. 203–218.
57
P. Steier et al., Analysis and application of heavy isotopes in the environment. Nucl. Instrum. Methods Phys. Res. B 268, 1045 (2010).
58
A. Wallner et al., Novel method to study neutron capture of 235U and 238U simultaneously at keV energies. Phys. Rev. Lett. 112, 192501 (2014).
59
L. K. Fifield, S. G. Tims, T. Fujioka, W. T. Hoo, S. E. Everett, Accelerator mass spectrometry with the 14UD accelerator at the Australian National University. Nucl. Instrum. Methods Phys. Res. B 268, 858–862 (2010).
60
M. Martschini et al., New and upgraded ionization chambers for AMS at the Australian National University. Nucl. Instrum. Meth. B 438, 141 (2019).
61
M. Paul et al., Heavy ion separation with a gas-filled magnetic spectrograph. Nucl. Instrum. Methods Phys. Res. A 277, 418 (1989).
62
K. Knie et al., High-sensitivity AMS for heavy nuclides at the Munich tandem accelerator. Nucl. Instrum. Meth. B 172, 717 (2000).
63
I. Leya et al., 53Mn and 60Fe in iron meteorites - new data and model calculations. Meteorit. Planet. Sci. (2020) in press.
64
D. Schumann, N. Kivel, R. Dressler, Production and characterization of 60Fe standards for accelerator mass spectrometry. PLoS One 14, e0219039 (2019).
65
S. Basu, F. M. Stuart, C. Schnabel, V. Klemm, Galactic-cosmic-ray-produced 3He in a ferromanganese crust: Any supernova 60Fe excess on Earth? Phys. Rev. Lett. 98, 141103 (2007).
66
F. M. Stuart, M. R. Lee, Micrometeorites and extraterrestrial He in a ferromanganese crust from the Pacific Ocean. Chem. Geol. 322–323, 209 (2012).
67
S. G. Love, D. E. Brownlee, Heating and thermal transformation of micrometeoroids entering the Earth’s atmosphere. Icarus 89, 26–43 (1991).
68
R. Trappitsch, I. Leya, Cosmogenic production rates and recoil loss effects in micrometeorites and interplanetary dust particles. Meteorit. Planet. Sci. 48, 195–210 (2013).
69
K. Knie et al., Accelerator mass spectrometry measurements and model calculations of iron-60 production rates in meteorites. Meteorit. Planet. Sci. 34, 729–734 (1999).
70
B. Peucker-Ehrenbrink, Accretion of extraterrestrial matter during the last 80 million years and its effect on the marine osmium isotope record. Geochim. Cosmochim. Acta 60, 3187–3196 (1996).
71
P. Gabrielli et al., Meteoric smoke fallout over the Holocene epoch revealed by iridium and platinum in Greenland ice. Nature 432, 1011–1014 (2004).
72
J. M. C. Plane, Cosmic dust in the earth’s atmosphere. Chem. Soc. Rev. 41, 6507–6518 (2012).
73
G. Cremonese, P. Borin, E. Martellato, F. Marzari, M. Bruno, New calibration of the micrometeoroid flux on earth. Astrophys. J. 749, L40 (2012).
74
F. Miyake, K. Nagaya, K. Masuda, T. Nakamura, A signature of cosmic-ray increase in AD 774-775 from tree rings in Japan. Nature 486, 240–242 (2012).
75
F. Miyake, K. Masuda, T. Nakamura, Another rapid event in the carbon-14 content of tree rings. Nat. Commun. 4, 1748 (2013).
76
F. Mekhaldi et al., Multiradionuclide evidence for the solar origin of the cosmic-ray events of ad 774/5 and 993/4. Nat. Commun. 6, 8611 (2015).
77
P. O’Hare et al.; ASTER Team, Multiradionuclide evidence for an extreme solar proton event around 2,610 B.P. (∼660 BC). Proc. Natl. Acad. Sci. U.S.A. 116, 5961–5966 (2019).
Information & Authors
Information
Published in
Cover image: Pictured is a southern rockhopper penguin from Pebble Island in the Falklands/Malvinas archipelago. Juliana A. Vianna et al. sequenced genomes from 18 penguin species to reconstruct the evolutionary history of penguins and found that the family likely originated in New Zealand or Australia during the Miocene epoch. Intensification of the Antarctic Circumpolar Current 11.6 million years ago drove penguin range expansion and diversification, with changes in thermal niches associated with changes in genes governing thermal regulation, oxygen metabolism, and diving capacity. See the article by Vianna et al. on pages 22303–22310. Image credit: Keith N. Barnes (Pitta Environmental Consulting, Baltimore, MD).
Classifications
Copyright
© 2020. Published under the PNAS license.
Data Availability Statement.
All relevant data are included in the paper and SI Appendix.
Submission history
Published online: August 24, 2020
Published in issue: September 8, 2020
Keywords
Acknowledgments
We thank the Antarctic Marine Geology Research Facility, Florida State University (C. Sjunneskog) for providing the sediment cores. This work was funded by Austrian Science Fund project AI00428, through the European Science Foundation Collaborative Research Project CoDustMas; Australian Research Council projects DP140100136, DP180100495, and DP180100496; German Academic Exchange Service project 56266169; and The Group of Eight Australia–Germany Joint Research Cooperation Scheme. J.F. acknowledges a stipend from the University of Vienna and R.G. support from the European Cooperation in Science and Technology “ChETEC” Action (CA16117). We also acknowledge financial support from the Australian Government for the Heavy Ion Accelerator Facility at ANU through the National Collaborative Research Infrastructure Strategy. ICP-MS measurements were performed by S. Beutner (HZDR).
Notes
This article is a PNAS Direct Submission.
*Note that in ref. 28 the time-averaged deposition rate of the top layers was originally calculated by assuming a much longer time period of <300 ky covered by these layers; that is, based on two detector events, this assumption resulted in <0.2 60Fe atoms·cm−2·y−1 (1-σ), which, however, changes to <1.8 atoms·cm−2·y−1 (1-σ) for the updated time period of 33 ky. The isotope ratio and 60Fe concentration and total 60Fe deposition are, however, independent of the time period.
Authors
Competing Interests
The authors declare no competing interest.
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFGet Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.