AMS measurements of cosmogenic and supernova-ejected radionuclides in deep-sea sediment cores
J. Feige, A. Wallner, L.K. Fifield, G. Korschinek, S. Merchel, G. Rugel, P. Steier, S.R. Winkler, R. Golser
aa r X i v : . [ a s t r o - ph . E P ] N ov EPJ Web of Conferences will be set by the publisherDOI: will be set by the publisherc (cid:13)
Owned by the authors, published by EDP Sciences, 2018
AMS measurements of cosmogenic and supernova-ejected radionuclides indeep-sea sediment cores
J. Feige , a , A. Wallner , , L.K. Fifield , G. Korschinek , S. Merchel , G. Rugel , P. Steier , S.R. Winkler , and R.Golser University of Vienna, Faculty of Physics, VERA Laboratory, Währingerstrasse 17, 1090 Vienna, Austria Department of Nuclear Physics, The Australian National University, Canberra, ACT0200, Australia Physics Department, TU Munich, James-Franck-Straße, 85748 Garching, Germany Helmholtz-Zentrum Dresden-Rossendorf, Bautzner Landstraße 400, 01328 Dresden, Germany
Abstract.
Samples of two deep-sea sediment cores from the Indian Ocean are analyzed with accelerator massspectrometry (AMS) to search for traces of recent supernova activity ∼ Al, Mn and Fe, are extracted from the sediment samples. The cosmogenic isotope Be, which is mainly produced in theEarth’s atmosphere, is analyzed for dating purposes of the marine sediment cores. The first AMS measurementresults for Be and Al are presented, which represent for the first time a detailed study in the time periodof 1.7-3.1 Myr with high time resolution. Our first results do not support a significant extraterrestrial signal of Al above terrestrial background. However, there is evidence that, like Be, Al might be a valuable isotopefor dating of deep-sea sediment cores for the past few million years.
The search for evidence of recent supernova (SN) activ-ity close to the solar system has become a popular matterover the last years. It was first suggested by Ellis et al.(1996) [1] and Korschinek et al. (1996) [2] to search forgeological isotope anomalies. Since then not only terres-trial (e.g. [3], [4], [5]) but also lunar samples ([6], [7])have been successfully analyzed for traces of Fe, a ra-dionuclide ejected by SN explosions. Terrestrial archiveswith a homogeneous growth rate in an undisturbed envi-ronment, such as ice cores, deep-sea sediments, and ferro-manganese crusts and nodules are most suitable to searchfor SN-produced elements. In those archives material isideally accumulated at a constant rate and a possible SNsignal should be fixed to the time of its deposition. The en-hancement of Fe discovered by Knie et al. (2004) [5] ina ferromanganese crust was located in a layer correspond-ing to a time range of 1.7-2.6 Myr [8]. Due to the verylow growth rate of 2.37 mm Myr − the signal is containedmainly in a layer of 2 mm thickness and a more precisedating is not possible. The same is true for lunar samples,where no time information is available. The sedimentationrate is negligible and mixing of layers e.g. by micromete-oritic and dust impacts, the so called gardening e ff ect, maydisperse any signal [6]. Low sedimentation rate is an ad-vantage; the signal is deposited in a very thin layer and asmall sample volume covers a long time period. If a signa-ture is present, crust samples are better suited for obtain- a e-mail: [email protected] ing a first hint on the time of deposition; but there is onlya limited time resolution. The search in sediments and icecores with much higher time resolution, however, requiresmuch more sample material to be processed. In ice coresit is not possible to date back more than some 100 kyr [9].Therefore, these are not suitable archives for a SN trace2 Myr ago. Here, we extract radionuclides (namely Be, Al, Mn and Fe) from deep-sea sediments with accu-mulation rates higher than ferromanganese crusts, i.e. inthe order of mm kyr − , to obtain a good time resolution.The findings of an enhancement of SN-ejected ra-dionuclides in a deep-sea sediment would corroborate thetheory of a recent SN close to the solar vicinity. Not onlythe Fe signal identified by [5] is a valuable hint to suchevents in the local interstellar medium, but also the exis-tence of the Local Bubble, in which our solar system isembedded [10]. This interstellar cavity has been producedby ∼
20 SN explosions starting around 14 Myr ago in astellar moving group of stars belonging today to the theSco-Cen association. It has been shown, that enough Fematerial can be transported from these SN origin to Earth,to produce such a signal as observed in the ferromanganesecrust [11].
Two marine sediment cores from the Indian Ocean, Eltaninpiston cores ELT49-53 and ELT45-21, are available for
PJ Web of Conferences analyzation. These were collected from the South Aus-tralian Basin, 38 ◦ ◦ ◦ ◦ − for ELT49-53and 3.7 mm kyr − for ELT45-21 by magnetostratigraphicanalysis. However, changes in the velocity of bottomwater currents triggered by climate change may alter thegrain size distribution of the sediment, prevent depositionor even erode deposited material. Due to these environ-mental e ff ects the sediment accumulation rate varies withtime [12]. Sections have been provided from depths of120-697 m below the sea floor, spanning a time intervalfrom 1.7-3.1 Myr. Altogether, there are 26 samples of coreELT45-21 and 45 samples of core ELT49-53 available foranalysis. Each sample covers a length of 1 cm, represent-ing an average time period of ∼ Al(t / = Mn (t / = Fe (t / = Fe concentration was observed in a ferromanganesecrust [5].Extraterrestrial (i.e. stellar) Al is mainly producedin three di ff erent environments in massive stars via the Mg(p, γ ) Al reaction: the central core burning in mainsequence stars, the C and Ne burning shell in the laterstages of the star, and during explosive Ne burning [15].A natural terrestrial background of Al is constantly gen-erated by spallation reactions from cosmic rays of Ar inthe Earth’s atmosphere amounting to a global mean of1.3 × atoms cm − yr − [16].The dominant stellar production of Mn is within ex-plosive silicon and oxygen burning inside of 2 M ⊙ of amassive star [17]. A background contribution arises fromextraterrestrial input of dust, meteorites and micromete-orites onto Earth. A target element for the production of Mn by nuclear reactions of cosmic ray protons and sec-ondary neutrons is iron, there is also a smaller contributionfrom nickel [18]. The annual flux of Mn has been esti-mated to be approximately 200 atoms cm − yr − [19].In contrast to Al and Mn, the influx of Fe comingfrom meteorites and micrometeorites should be very low.Here, the target element is nickel only; the production rateof Fe from cosmic rays is around three orders of magni-tude lower than the production rate of Mn [18]. Stellar Fe is produced in the late burning stages of massive starsvia neutron capture on Fe. The half-life of Fe is only44.5 days, hence the production of Fe competes with thebeta-decay of Fe. Hot temperatures above 4 × K arerequired to produce a su ffi cient amount of neutrons limit-ing the synthesis of Fe to the shell burning of He and Cand to explosive Ne burning [15].Estimations of the fluence of these isotopes in the ma-rine sediment cores and the probability of detecting a po-tential signal with AMS have been presented earlier [8]. Inthe case of Al and Mn we would expect an exponential decay profile with increasing depth in a core. A positivevariance of the data in the decay curve might indicate aSN signature. The height and broadness of such a signaldepends on various e ff ects such as the amount of materialejected in the SN explosion, the dimension of the incom-ing shock wave containing the material, the time it takesfor the radionuclides to be deposited, etc.Additionally to these three radionuclides chosen forSN detection, Be (t / = Al it is produced in the Earth’s atmosphere. It orig-inates from spallation induced by cosmic rays of mainlynitrogen and oxygen leading to an average total flux of6.5 × atoms cm − yr − [22]. Chemical sample preparation was adapted from the leach-ing procedure described by Bourlès et al. [23] and Fitoussiet al. [24]. This technique is designed to extract the au-thigenic fraction, which includes the isotopes of interest,from the detrital phase. Dissolving the whole sample ma-terial leads to enhanced extraction of the stable elementsfrom the detrital component, which dilute the signal (seesection 2.3). Further chemical separation of Al, Be, Fe,and Mn from the leachate is an updated version of Merchel& Herpers [25], to adjust for higher concentration of ele-ments like Ca, Na, and K originating from a total startingmass of 3 g of sediment.The samples were leached for 1 hour in 60 mlof 0.04 M NH OH · HCl in 25% (V / V) acetic acid at(90 ± ◦ C, then for 7 hours at (95 ± ◦ C [23]. After tak-ing an aliquot for stable Be-determination by inductivelycoupled plasma mass spectrometry (ICP-MS) about 600 µ g of stable Be-carrier was added to each leachate. Pre-cipitation with NH and removal of hydroxides of Al,Be and Fe from the solution resulted in slow oxidation ofMn + to Mn + and thus, delayed precipitation and separa-tion of MnO(OH) after a few hours. Purification to de-crease the isobar ( Cr) content is achieved by reprecipita-tion, i.e. dissolution in HNO and H O and precipitationwith KClO in the heat. Repeated washing and drying re-sult in targets of MnO , suitable for AMS measurements.Iron was separated from Al and Be by dissolving in HCland applying the solution to an anion exchange column[25]. Nickel containing the isobar of Fe, i.e. Ni, whichwas already mainly reduced in the first precipitation step,is further decreased during this anion exchange. Al and Bewere separated from each other by cation exchange. In thisstep, the boron-fraction containing B is further reduced.All three elements, Al, Be and Fe, were precipitated withNH as hydroxides, washed, dried, and for Be and Alignited to oxides at 900 ◦ C. The concentrations of Be and Al Mn and Fe inthe marine sediment samples are measured as isotopic ra-tios, i.e. the ratio of the radionuclide relative to the sta-ble isotope, using accelerator mass spectrometry (AMS). eavy Ion Accelerator Symposium 2013
A beam of negative ions is produced in a Cs sputter source,preaccelerated and analyzed through a combination of anelectrostatic analyzer and an injector magnet. Most im-portantly, isobaric molecular interferences are excludedin AMS by injecting the negative particles into a tandemaccelerator, where molecules break up by stripping pro-cesses in a foil or gas stripper. The now positively chargedhigh-energy ions pass another mass spectrometer, wherebreak-up products of molecules consisting now of di ff erentmasses are filtered from the beam (e.g. [26]). Atomic sta-ble isobars, which are much more abundant in nature thanthe corresponding isobaric radionuclide, will pass throughthe AMS setup in the same manner, and will be suppressedby specific methods for di ff erent elements. Therefore,AMS is highly sensitive and capable of quantifying verysmall isotopic ratios as low as 10 − , which makes it verysuitable for a wide range of applications [27], in particularalso astrophysical applications [28].Several di ff erent laboratories are involved in the AMSmeasurements. Al is analyzed at VERA (Vienna Envi-ronmental Research Accelerator), Austria, a 3 MV tandemaccelerator [29]. For AMS Al − ions are selected. Dedi-cated isobaric suppression is not necessary, as the stableisobar Mg does not form negative ions.The DREAMS (Dresden AMS) facility [30], a6 MV tandem accelerator at Helmholtz-Zentrum Dresden-Rossendorf, Germany, is used for collection of Be / Bedata in the marine sediment targets. At this laboratory, Bsuppression is achieved by a 1 µ m thick silicon nitride ab-sorber foil placed between the high-energy 90 ◦ analyzingmagnet and a 35 ◦ electrostatic analyzer [30]. Comparativemeasurements will carried out at VERA.Su ffi cient suppression of the interfering stable isobars Cr and Ni of Mn and Fe, respectively, is onlypossible by acceleration to very high energies in the or-der of 100 MeV in combination with a gas-filled mag-net (e.g. [31]). Separation is achieved by interaction withthe gas atoms leading to di ff erent average charge states ofthe radionuclide and its stable isobar depending on theiratomic numbers. Therefore, isobars are deflected di ff er-ently and take separate trajectories in the magnetic field.AMS machines with terminal voltages larger than 10 MVare needed to achieve the energies required for separation.Facilities capable of detecting Mn and Fe at the ex-pected low levels of our study are the 14 MV MP tandemaccelerator at the Maier-Leibnitz-Laboratory in Garching,Germany [31] and the HIAF (Heavy Ion Accelerator Fa-cility), a 15 MV pelletron accelerator at the ANU in Can-berra, Australia [32]. Mn and Fe will be measured atthe latter laboratory.
The distribution of Be / Be in a time range of 1.7-3.1 Myrmeasured with the DREAMS facility is presented in Fig. 1.Here, the in-house standard SMD-Be-12 with a Be / Bevalue of (1.70 ± × − [30] has been used to nor-malize the data. The error bars originate from a compo-sition of statistical uncertainty from the AMS measure- æ æ æ æ æ æ æ æææ æææææ æ æ æ ææ æ æ æ æ æ æà à à àà à à ELT 49 - - æà Age @ Myr D B e (cid:144) B e @ - D Figure 1.
The Be / Be data from DREAMS in the two marinesediment cores ELT49-53 (black circles) and ELT45-21 (graysquares) with increasing age, not corrected for half-life. The datafits well to the values expected from exponential decay (solidline). æ æ æ æ æ æ æ æææ æææææ æ æ æ ææ æ æ æ æ æ æà à à àà à à
ELT 49 - - æà Age @ Myr D A l (cid:144) A l @ - D Figure 2.
The Al / Al data vs age of the two marine sedimentcores. Exponential decay is indicated by the solid line. ments (which are usually between 1-2 %) and of the uncer-tainty from stable isotope ICP-MS measurements. No pre-cise information on the uncertainties of ICP-MS data areavailable yet; commonly they are in the range of 3-5 %.First estimations from repeated measurements of severalsamples lead to an average uncertainty of 4 %. The pre-liminary chronology of the sediment cores as indicated inFig. 1 and 2 has been extracted from magnetostratigraphicdata of [12]. The Be / Be data of the two cores have beenfitted with an exponential decay function using the half-life of Be. Extrapolating this curve to an initial value of Be / Be at the surface results in (1.07 ± × − . Thisratio falls within the range of values presented by [23] forthe Indian Ocean.A variability of Be / Be within both cores is indicatedin the data shown in Fig. 1. Variations are commonly as-sociated with a change in the production of cosmogenic Be in the Earth’s atmosphere, which is induced by thevariability of the geomagnetic field (see [33]). It is possi-ble, that melting ice after glacial periods releases Be into
PJ Web of Conferences the ocean, which would lead to an increase of the Be / Beratio ([34], [35]). The Be concentration is influenced byclimate changes, which might lead to changes in the bot-tom water circulation and an increase or decrease of thesedimentation rates [12]. It has to be further investigated,which e ff ects play a significant role in our samples.The Al / Al data for the two sediment cores has beenmeasured with the VERA facility, Vienna. We obtain verylow isotopic ratios of ∼ − , requiring a long measure-ment time. Each sample was sputtered for several hoursuntil complete exhaustion to obtain a statistical uncertaintyof better than 10 %. There are several reasons for the smallisotopic ratios of Al / Al compared to Be / Be. The at-mospheric production rate of Al is approximately 10 times lower than of Be [16] and the age of the samplescorrespond to three half-lifes of Al, which means a lotof Al has already decayed. In addition stable aluminiumis a major component in deep-sea clay sediments. This ispartially suppressed by leaching rather than dissolving thewhole material during the chemical separation.The data shown in Fig. 2 were normalized to the stan-dard material AW-V-2 and AW-V-3 with Al / Al val-ues of (2.71 ± × − and (3.65 ± × − , respec-tively [36]. The combined Al / Al distribution through-out the marine sediment cores is fitted with one exponen-tial decay curve based on the half-life of Al. A surfacevalue of (2.6 ± × − is deduced. The Al / Al ratiosin the two cores seem to agree with each other very well,an indication that Al might be a suitable isotope for dat-ing of marine sediment samples. A significant signal ofa recent close-by SN explosion can not be extracted fromthe presented data.
We have analyzed two deep-sea sediment cores from theIndian Ocean for a potential SN signal. They span a timeperiod between 1.7 and 3.1 Myr. We studied samples of3 kyr integration time for every 20-30 kyr. Radionuclidesanalyzed so far are Al and Be, where the latter is mea-sured for an independent dating of the individual samples.Our data represents for the first time a detailed study inthis time period with high time resolution. Mn and Fewill be measured in the same sediment samples.The samples have been chemically processed with aseparation technique following [23] and [25] to extractthe elements Al, Be, Mn, and Fe. To date, most sam-ples have been analyzed by AMS. Measurements of Aland Be have been performed at the VERA and theDREAMS facilities, respectively. The data are compati-ble with ratios expected from atmospheric production bycosmic rays. Although more challenging to measure, dueto low Al / Al ratios, these adapt to the exponential fitvery well. Like Be / Be, the Al / Al data seem to agreewith each other in the two sediment cores, making Al apotential radionuclide for dating purposes. A significantSN signal can not be identified yet. Longer measurementsof the Al / Al targets will be carried out to reduce thestatistical uncertainty. The Be / Be data presented here was collected withDREAMS. Comparative measurements with VERA are inprogress.New sample material of the marine sediments surfacewill be chemically prepared and analyzed to compare theresults with extrapolations from our current measurementsand with Al / Al and Be / Be ratios expected in the In-dian Ocean. A potential SN signal is best identified bymeasurements of radionuclides with a low background,such as Fe. These measurements are performed at theANU, Canberra, and will be compared with results from Al and Mn. First measurements of Mn are plannedto take place later in 2013, also at ANU, Canberra.
Acknowledgments
This work is funded by the Austrian Science Fund (FWF),project P20434 and I428 (EUROCORES project EuroGE-NESIS, subproject CoDustMas, funded via the EuropeanScience Foundation). This research used samples and dataprovided by the Antarctic Marine Geology Research Fa-cility (AMGRF) at Florida State University. The AMGRFis sponsored by the U.S. National Science Foundation.We would like to thank the accelerator sta ff at Dresden-Rossendorf for their support. Furthermore, we thank AlineRitter (HZDR) for stable isotope measurements and theCEREGE-team, especially Didier Bourlès, for sharing de-tailed information on leaching procedures. References [1] Ellis, J. et al., Astrophys. J., , 1227-1236 (1996)[2] Korschinek, G. et al., Abstracts of the 7th InternationalConference on Accelerator Mass Spectrometry, Radio-carbon, , 68-69 (1996)[3] Knie, K. et al., Phys. Rev. Lett., , 18-21 (1999)[4] Fitoussi, C. et al., Phys. Rev. Lett., , id.121101(2008)[5] Knie, K. et al., Phys. Rev. Lett., , id.171103 (2004)[6] Cook, D. L. et al., Lunar and Planetary Institute Sci-ence Conference Abstracts, , id.1129 (2009)[7] Fimiani, L. et al., Lunar and Planetary Institute Sci-ence Conference Abstracts, , id.1279 (2012)[8] Feige, J. et al., Publ. Astron. Soc. Aust., , 109-114(2012)[9] EPICA community members, Nature, , 623-628(2004)[10] Fuchs, B. et al., Mon. Not. R. Astron. Soc., , 993-1003 (2006)[11] Breitschwerdt, D. et al., Astron. Nachr., , 486-496, (2012)[12] Allison, E. & Ledbetter M. T., Mar. Geol., , 131-147 (1982)[13] Holden, N. E., Pure & Appl, Chem., , 941-958,(1990)[14] Rugel, G. et al. 2009, Phys. Rev. Lett., ,id.072502 (2009) eavy Ion Accelerator Symposium 2013 [15] Limongi, M. & Chie ffi , A., Astrophys. J., , 483-500 (2006)[16] Auer, R. M. et al., Earth Planet. Sc. Lett., , 453-462 (2009)[17] Meyer, B. S., Chondrites and the ProtoplanetaryDisk, ASP Conference Series, , 515-526, (2005)[18] Merchel, S. et al., Nucl. Instrum. Meth. B, , 806-811 (2000)[19] Auer, R. M., Applications of Al-26 in AtmosphericResearch (PhD thesis, University of Vienna, 2008) 120[20] Korschinek, G. et al., Nucl. Instrum. Meth. B, ,187-191 (2010)[21] Chmele ff , J. et al., Nucl. Instrum. Meth. B, , 192-199 (2010)[22] Masarik, J. & Beer, J., J. Geophys. Res, , D11103(2009)[23] Bourlès, D. et al., Geochim. Cosmochim. Ac., ,443-452 (1989)[24] Fitoussi, C. & Raisbeck, G.M., Nucl. Instrum. Meth.B, , 351-358 (2007)[25] Merchel, S. & Herpers U., Radiochim. Acta, , 215-219 (1999) [26] Kutschera, W. et al., Nucl. Instrum. Meth. B, ,47-50 (1997)[27] Kutschera, W., Int. J. Mass Spectrom.,http: // dx.doi.org / / j.ijms.2013.05.023 (2013)[28] Wallner, A., Nucl. Instrum. Meth. B, , 1277-1282(2010)[29] Steier, P. et al, Nucl. Instrum. Meth. B, , 445-451(2005)[30] Akhmadaliev, S. et al., Nucl. Instrum. Meth. B, ,5-10 (2013)[31] Knie, K. et al., Nucl. Instrum. Meth. B, , 717-720(2000)[32] Fifield, L. K. et al., Nucl. Instrum. Meth. B, , 858-862 (2010)[33] Ménabréaz, L. et al., J. Geophys. Res., , B11101(2012)[34] Aldahan, A. A., Mar. Geol., , 147-162 (1997)[35] Wang, L. et al., Geochim. Cosmochim. Ac., , 109-119 (1996)[36] Wallner, A. et al., Nucl. Instrum. Meth. B,172