The Dyadic Radionuclide System 60 Fe / 53 Mn to Distinguish Interstellar from Interplanetary 60 Fe

. The discovery of live 60 Fe in a deep-sea crust with proposed interstellar origin followed by evidence for elevated interplanetary 3 He in the same crust raised the question on how to unambiguously identify the true production site of the identiﬁed 60 Fe. Here, we show the implementation of the dyadic radionuclide system 60 Fe / 53 Mn to serve as a tool for the identiﬁcation of surplus interstellar 60 Fe over interplanetary production. The recent updates in experimental 60 Fe and 53 Mn data from iron meteorites as well as in production rate models conﬁrm the validity and robustness of this dyadic system for future applications. we were able to establish a substantial meteorite database for both radionuclides 60 Fe and 53 Mn using accelerator mass spectrometry (AMS) [9]. AMS is the most sensitive experimental approach to directly determine concentrations of both radionuclides in terrestrial and extraterrestrial samples. Radiochemical neutron activation analysis


Introduction
Direct detection of freshly-synthesized supernova nuclides is one of the gold standards to study stars and their impact on the solar system. The radionuclide 60 Fe proved to be an excellent candidate because of its relatively long half-life of 2.6 Myr [1,2], allowing accumulation and detection over several million years.
The first indication of live 60 Fe in a deep-sea ferromanganese crust [3] followed by the clear discovery in a 2 Myr old layer [4] raised the question on its astrophysical origin. The proposed supernova (interstellar) origin [4] was subsequently challenged. Elevated levels of 3 He in the same crust were used as an argument to propose that the discovered 60 Fe was produced by cosmic-ray spallation in (micro-) meteorites and not by stars [5]. Further evidence for a global 60 Fe signal [6] was used as an indirect argument for the supernova hypothesis. 60 Fe was subsequently identified on timescales of years, thousands of years and millions of years in several archives, see [7] for a recent summary.
The clear identification of the astrophysical production site of any discovered 60 Fe requires the exclusion of anthropogenic 60 Fe, which was conclusively presented in [8]. Cosmogenic production is governed by cosmic-ray fluxes, target sizes and compositions, and production rates for radionuclides by primary and secondary cosmic rays outside Earth's atmosphere. The most viable candidates to establish cosmic-ray production rates of 60 Fe are meteorites, which could be used to distinguish between a rather constant cosmic dust influx plus single large object events such as meteoroids, comets or asteroids and solitary supernova 60 Fe influxes.
It is critical to deduce the absolute amount of interplanetary dust influx into a sample. This makes it particularly challenging, because a constant background flux of 60 Fe bearing cosmic dust modulated by single large events can mimic supernova influxes. Note, for deep-sea crusts we integrate over several hundred thousand to millions of years, which makes rare events like meteorite falls even over global scales probable. Therefore, a tool to unambiguously identify interstellar over cosmogenic 60 Fe is needed. Here, we show that by additionally measuring the long-lived radionuclide 53 Mn (t 1/2 = 3.7 Myr), the dyadic radionuclide system 60 Fe / 53 Mn could be used to overcome these problems and to disentangle interstellar from interplanetary 60 Fe influxes. It is worth mentioning that one can estimate the maximum cosmogenic 60 Fe concentration in a sample by measuring the stable target element Ni and assuming all of the Ni present is extraterrestrial Ni. This estimation is crude but practical for large 60 Fe influxes and low stable Ni abundances and fails for samples with low extraterrestrial but high terrestrial influxes. The hereby proposed system is applicable in any case.

Cosmogenic 60 Fe and 53 Mn from meteorites
In the last few years, we were able to establish a substantial meteorite database for both radionuclides 60 Fe and 53 Mn using accelerator mass spectrometry (AMS) [9]. AMS is the most sensitive experimental approach to directly determine concentrations of both radionuclides in terrestrial and extraterrestrial samples. Radiochemical neutron activation analysis and low-level counting have been used in the past to measure both radionuclides in meteorites but are not sensitive enough to search for supernova traces. For 60 Fe in particular, one needs to consider waiting times of decades after chemical purification for the in-growth of the daughter nucleus 60 Co to achieve secular equilibrium [10]. Note, large databases for the isotope 53 Mn exist without 60 Fe data [11]. AMS does not require any waiting time since live radionuclides are detected and not their decay. Currently, the only AMS facility in the world capable of measuring both radionuclides at natural levels is the Heavy Ion Accelerator Facility of ANU [2]. The Munich accelerator lab [12], which was pioneering the measurements of both radionuclides using a gas-filled magnet, has been closed in 2020. Clearly, there is the need for existing or new labs to establish new capabilities for 60 Fe and 53 Mn AMS.
The database with more than 30 single iron meteorites is complemented by an updated production rate model based on the INCL (Liège Intra Nuclear Cascade) code [9]. The predicted production rate ratio is 60 Fe . The common units of production rates are specific activities in disintegrations per minute per kg of target element or target mass, in this case stable Ni for 60 Fe and stable Fe for 53 Mn. The stable element content in terrestrial samples is dominated by terrestrial elements. The specific activity as a measure for the interplanetary influx is therefore not suitable, because the extraterrestrial target element concentration is not quantifiable. Hence, it is necessary to estimate reasonable values of extraterrestrial Ni / Fe in terrestrial samples. Importantly, only the ratio of stable Ni to Fe is needed and not the concentration of both elements. For terrestrial samples, the amount of cosmogenic radionuclides present is much smaller compared to extraterrestrial samples, where intrinsic production is dominant. This relaxes the requirement on the exact knowledge of Ni / Fe.
Using the half-lives of 2.6 Myr and 3.7 Myr, we back-calculate a half-life and target composition independent and directly measurable production rate ratio of 60 Fe / 53 Mn = (1.77 ± 0.07) × 10 −3 × Ni / Fe in units of at / at and wt.%. By factoring out Ni / Fe, the data becomes comparable between samples with different target element concentrations. This allows us to compare the model prediction with extraterrestrial samples (known extraterrestrial Ni / Fe) and with terrestrial archives (unknown extraterrestrial Ni / Fe). The possible values for extraterrestrial Ni / Fe in terrestrial samples are conservatively estimated to be less than 1. Solar system, chondritic and meteorite abundances show the clear trend of Ni Fe, most commonly Ni / Fe = 0.055 [13]. Iron meteorites, which are metal-rich and elevated in Ni, generally show a higher Ni / Fe of around 0.1. For comparison, Earth's crust has a low Ni to Fe ratio of 0.002, whereas the ratio in sea water is 0.28, but with low concentration of both metals [14]. Using the most common abundances of stable elements in the solar system, the cosmogenic 60 Fe to 53 Mn atomic ratio obtained in a measurement would be around 10 −4 .
In contrast to that, the stellar atomic nucleosynthesis ratio is estimated to be 10 0 -10 −2 [15]. The hypothetical interstellar 60 Fe would dominate over cosmogenic 60 Fe as an excess, whereas any supernova 53 Mn is only a fraction of the bulk cosmogenic 53 Mn. This makes any accompanied supernova 53 Mn challenging to identify over cosmogenic background with only one attempt so far by AMS [16]. Therefore, interstellar 60 Fe would appear as an enhanced 60 Fe to 53 Mn ratio over the cosmogenic baseline, which is dominated by cosmogenic 53 Mn.

Interstellar 60 Fe identification
The very first utilization of this principle happened alongside the first indication of supernova 60 Fe in a deep-sea crust [3]. There, it was already estimated, that the atomic ratio of cosmogenic 60 Fe to 53 Mn should be in the order of 10 −4 . This strong experimental argument was not adopted in the following studies, possibly due to limited data on extraterrestrial production rates and meteorite samples. Seventeen years later in 2016, this dyadic radionuclide system yielded evidence for supernova 60 Fe on the surface of the Moon [17]. Lunar regolith was analyzed for 60 Fe and 53 Mn and compared to four different meteorites before the database was significantly extended. The specific activity ratio was calculated, since the lunar surface itself is the target for the cosmic ray interactions. The meteorite activity ratio was determined to be 60 Fe / 53 Mn = (2.68 ± 0.35) × 10 −3 in units of dpm kg −1 [Ni] / dpm kg −1 [Fe]. This agrees perfectly with the current model value of (2.5 ± 0.1) × 10 −3 and the meteorite database value of 2.6 × 10 −3 . The lunar surface ratio was determined to be (10 − 50) × 10 −3 , much higher than for meteorites. The measured atomic 60 Fe / 53 Mn of lunar material would be misleading because of the low lunar Ni / Fe of 0.001 -0.01, which is similar to the ratio on Earth and significantly lower than for meteorites. The specific activity reveals the clear enhancement, which shows that for lunar material the knowledge of Ni / Fe is crucial.
More recently, 60 Fe was discovered in Antarctic surface snow [8]. Besides ruling out anthropogenic production, the identification of its astrophysical origin was important since the Antarctic snow covered only the most recent 20 years of terrestrial history and could therefore reveal an influx right at this moment. Here, the specific activity ratio approach was not possible, because the stable elements in the sample are of terrestrial origin. The same meteorite data from [17] was converted to atomic ratios with a value of 60 Fe / 53 Mn = (1.9 ± 0.2) × 10 −3 × Ni / Fe in units of at / at and wt.%. Similarly, this again agrees perfectly with the outlined model value of (1.77 ± 0.07) × 10 −3 × Ni / Fe and the meteorite database value of 1.8 × 10 −3 × Ni / Fe. The measured 60 Fe to 53 Mn ratio in Antarctic snow was around 17 × 10 −3 . The extraterrestrial Ni / Fe was conservatively estimated to be less than 1. This yields a more than two orders of magnitude enhancement over cosmogenic background for chondritic composition Ni / Fe = 0.055. There is still a clear enhancement even for the extreme case of Ni / Fe = 1. The interstellar origin of the discovered 60 Fe becomes now even clearer with the updated model and the significantly extended meteorite database (Figure 1).  Table 1: Summary of 60 Fe / 53 Mn ratios from the theoretical model, the meteorite database and the samples lunar regolith as well as Antarctic snow. Depending on the origin of the analyzed sample, one has to compare any measured data with the first or the second 60 Fe / 53 Mn ratio for extraterrestrial or terrestrial samples, respectively. The unit dpm / kg X refers to the specific activity ratio with X being the respective target element for the production of 60 Fe or 53 Mn. In particular for lunar regolith, the measured atomic ratio is misleading because of the low intrinsic Ni / Fe. 60 Fe / 53 Mn ratios around 10 −3 indicate cosmogenic production, whereas higher ratios indicate a surplus of interstellar 60 Fe.

Sample
Ni  [17] Lunar regolith 0.001 -0.01 (10 − 50) × 10 −3 10 −5 − 10 −4 Koll et al. [8] Antarctic snow . . . . . . 17 × 10 −3 Figure 1: Antarctic snow data [8] in green with the updated, independent meteorite model in red and available meteorite data in black [9]. The enhancement of 60 Fe over the expected cosmogenic ratio is clearly visible. The datapoint is exemplarily displayed for the most likely stable element abundance Ni / Fe = 0.055. By establishing a significant meteorite database for both radionuclides and having an independent model, the dyadic radionuclide system 60 Fe / 53 Mn is now robust enough to be widely applied for the search of interstellar 60 Fe in any terrestrial or extraterrestrial archive. The model values for the production rate ratios are in very good agreement with now 31 single meteorite measurements, see Table 1. This system allows to unambiguously identify supernova produced 60 Fe over cosmogenic background.