Search

Poster; International audience; The rates and chronologies of valley incision are closely modulated by the tectonic uplift of active mountain ranges and were controlled by repeated climate changes during the Quaternary. The continental collision between the Iberian and Eurasian plates induced a double vergence orogen, the Pyrenees, which has been considered as a mature mountain range in spite of significant seismicity (e.g. Chevrot et al., 2011) and evidence of neotectonics (e.g. Goula et al., 1999). Nevertheless, recent studies indicate that the range may have never reached a steady state (Ford et al., in press). One option for resolving this controversy is to quantify the incision rates since the Miocene by reconstructing the vertical movement of geometric markers such as fluvial terraces. However, the few available ages from the Pyrenean terrace systems do not exceed the middle Pleistocene. Thus, to enlarge the time span of this dataset, we studied alluvium-filled horizontal epiphreatic passages in limestone karstic networks. Such landforms are used as substitutes of fluvial terraces because they represent former valley floors (e.g. Palmer, 2007; Audra et al., 2013). They record the transient position of former local base levels during the process of valley deepening. The Têt river valley (southern Pyrenees) was studied near the Villefranche-de-Conflent limestone gorge where 8 cave levels have been recognized over a vertical height of 600 meters. Given that 26Al/10Be cosmogenic burial dating in this setting was limited to the last ∼5 Ma (Calvet et al., 2015), here we used the cosmogenic 10Be/21Ne method in order to restore a more complete chronology of valley incision (e.g. Balco & Shuster, 2009; McPhilipps et al., 2016). Burial age results for alluvial deposits from 12 caves document incision rates since the Langhian (∼14 Ma). Preliminary results indicate a history of valley deepening in successive stages. The data show a regular incision rate of 70–80 mm/ka from the Langhian to the Messinian, then a decrease to a mean incision rate of ∼15 mm/ka. Together with these burial ages, pre-burial catchment denudation rates were estimated and do not exceed 21 mm/ka. Moreover, no acceleration during the Quaternary is recorded. This method also allows to infer that the Pla des Horts plateau is of Burdigalian age.

The DREAMS (DREsden AMS) facility consists of a sophisticated 6 MV accelerator system [1], but also provides two chemistry laboratories for external users. One lab is used for preparation of 10Be, 26Al, 41Ca, 53Mn and 60Fe targets. The other one is dedicated to halide targets (36Cl, 129I), thus, any use of Cl- or S-compounds such as HCl or H2SO4 is strictly prohibited. Separation protocols are applied to calcite- and quartz-rich samples for in-situ projects [2-4]. Atmospheric 10Be has been leached from marine sediments and Mn-nodules for dating purposes [5] and chemistry refined for bigger samples and heavier nuclides (26Al, 53Mn, 60Fe) [6]. After adapting standard protocols [7] introducing simple Mn-separation by delayed hydroxide separation, higher isobar concentrations (53Cr) have been found asking for further cleaning by ion exchange. With intent to speed-up and simplify the separation procedures for ice and meteorite samples [8], difficulties have been arisen e.g. carryover of Ag+-ions into MnO2, and shortly after overcome. Another chemistry challenge probably mastered is the dissolution of meteoritic troilite (FeS) without losing natCl-carrier before equilibrium with 36Cl, plus suppression of massive isobar amounts. Remaining issues also influencing the quality of AMS-data, such as incorrectly measured stable isotope concentrations (9Be, 27Al), are usually underestimated and harder to tackle. [1] Akhmadaliev et al., NIMB 294 (2013) 5. [2] Merchel et al., Quat. Geo. 22 (2014) 33. [3] Zech et al., Paleo3 369 (2013) 253. [4] Yildirim et al., Tectonics 32 (2013) 1107. [5] Feige et al., Ludwig et al. & Rodrigues et al., AMS-13. [6] Feige et al., EPJ Web Conf. 63 (2013) 03003. [7] Merchel & Herpers, RCA 84 (1999) 215. [8] Smith et al., AMS-13.

Since the acceptance tests of the French 5 MV accelerator mass spectrometry facility ASTER in 2007, routine measurement conditions for the long-lived radionuclides 10Be and 26Al have been established. Yearly sample throughput as high as over 3300 unknowns has been reached for 10Be in 2010. Unacceptable cross-contamination for volatile elements has been largely solved by an ion source upgrade allowing 36Cl measurements at ASTER. However, recent long-term tests using 35Cl/37Cl samples with strongly varying ratios have shown that identical targets lead to different 35Cl/37Cl results at the 2-4% level when being measured after a time gap of 24 hours while the source is running other samples. Besides time dependent mass fractionation, another very likely reason for this effect might be source memory, thus, asking for sophisticated measurement strategies and improved data evaluation and eventually further ion source improvement. Finally, after establishing quality assurance by cross-calibration of secondary in-house 26Al and 41Ca standards and taking part in round-robin exercises of 10Be and 36Cl, a two-step cross-calibration of secondary in-house 129I standards has been performed. The NIST 3231 containing 129I/127I at 0.981 x 10-6 has been used for step-wise dilution with NaI to produce gram-quantities of lower-level standards for every-day use. The resulting material SM-I-9 (129I/127I: ~1 x 10-9) has been measured vs. AgI produced from the two NIST ampoules with (0.982+0.012) x 10-8 solution using minimum chemistry. In a second stage, SM-I-10 and SM-I-11 with ratios of ~1 x 10-10 and ~1 x 10-11, respectively, have been cross-calibrated against SM-I-9. Individual uncertainties of the traceable secondary standards are 1.3-1.4 % (2σ), mainly originating from the given uncertainty of the primary NIST 3231 at 10-8. The cross-contamination for iodine is in the range of 0.4-0.6% within the first 20 hours of running the source.

The HVE 5 MV ASTER AMS national facility at CEREGE was accepted in 2007. Since then we have continued to optimize performance for 36Cl. Cl-36 analyses use AgCl, a Cs negative ion sputter source, Ar stripping to +5 in the terminal of the Tandetron™ accelerator at 5 MV and a Si3N4 postacceleration stripper foil to enhance suppression of 36S relative to 36Cl. The major challenges to obtaining the desired performance for Earth science applications are ion source memory, mass fractionation effects, 36S isobar suppression and sensitivity. Redesign of the SO110 ion source by HVE to change the size of the aperture and the shape of cathode significantly reduced ion source memory to less than ~0.1%, a level that can be compensated for by matching standards to samples. We observe small systematic drifts in 35Cl/37Cl ratios over time, the source of which is not yet determined. Measurement of standards indicates that this effect limits the precision of 35Cl/37Cl ratio determination to about 2%. 36S is suppressed in several ways. First, the sample chemistry has been designed to reduce S to very low levels. Second, cathodes are constructed of low-S nickel, enabling direct target loading without the use of AgBr pre-packing. Third, a post-acceleration Si3N4 stripper foil differentially absorbs energy from 36Cl and 36S. A ubsequent electrostatic deflector is then able to suppress 36S by a factor of ~5x10-3. Differential energy loss in the detector further suppresses 36S by about 10-4, for an overall suppression of 5x10-7. 36S count rates are typically equivalent to a background 36Cl/Cl of -15. At typical 35Cl currents of 20 μA samples with 36Cl/35Cl of 5x10-14 can be measured to ± 6% statistical uncertainty with one hour of analysis time. Typical machine blanks have 36Cl/Cl ~2x10-15.

We explore a new method for documenting the long-term retreat rate of seacliffs based on measurements and modeling of Be-10 concentration transects across present-day shore platforms. The proposed forward numerical model relies on a scenario of sea-level rise since the last deglaciation, and predicts the shape of Be-10 concentration transects as a function of prescribed cliff recession and vertical coastal platform downwearing rates. Two independent transect features allow fitting the long-term recession rate model to field observations: a sharp Be-10 concentration drop predicted at the former stationary location of the cliff during the last glacial period similar to 100 ka ago, and a characteristic dome shape whose magnitude is directly related to the recession rate of the cliff. A retreating chalk cliff site from the English Channel coast of France, at Mesnil-Val, where the 7 m tidal range broadly exposes the shore platform, was selected to test the proposed method. Although retreat rates were too high to pinpoint the predicted Be-10 concentration drop at the last glacial cliff position, the Be-10 concentration of the flints sampled across the shore platform is consistent with the expected dome shape. When modeled using the proper tidal range and proposed Holocene sea-level rise history, the Be-10 data yield a cliff retreat rate since the mid-Holocene of 11-13 cm/yr. This is consistent with a 30-year-long measurement record, strongly supporting the utility of the Be-10 method.

The first international round-robin exercise for the measurement of the long-lived radionuclide 10Be has been conducted. Ten participating accelerator mass spectrometry (AMS) facilities have each measured three samples at the 10-12-10-14 10Be/9Be level. All results have been made traceable to the NIST SRM 4325 standard to avoid additional discrepancies that arise when different facilities use different calibration materials. Hence, the data concentrates on pure measurement distinctions. Multivariate statistical investigations have been performed to reveal a bias between facilities, i.e. two distinguished groups could be identified. Maximum discrepancies between two single facilities are in the range of 6-31% depending on the absolute 10Be/9Be value. These findings should be considered when comparing 10Be data produced at one AMS facility with that produced at another facility, which is e.g. often the case for in-situ 10Be dating studies. Round-robin exercises are a very helpful tool as part of an overall quality assurance scheme to improve the accuracy, and not only the precision, of AMS data.

On a sub-millennial time scale the spatial distribution of erosion is controlled to first order by tectonics, relief, and possibly precipitation, and secondly by vegetation, lithology, temperature and human activity. The Himalayas form a very distinct orographic barrier with a pronounced rainfall gradient from the South to the North and have a very rugged terrain, causing highly dynamic surface processes and fast erosion rates. Thus, the Himalayas provide an ideal site of investigation to study erosion and constrain its controlling factors. In this contribution we present an integrated comparison of mean catchment erosion rates, calculated from in-situ produced 10Be cosmogenic isotope concentration in river sands (representative for millennial time scales) and suspended sediment measurements (integrating the annual to decadal time spans). We discuss erosion rates and patterns in the context of precipitation-landscape features of the studied catchments. The samples cover all major rivers, and several minor tributaries of the Narayani watershed (30,000 km2) in central Nepal. They represent all lithologies, topographic units and climate regimes across the Himalayan range. The erosion rates, both from cosmogenic nuclide analysis and suspended sediment measurements, range from 0.1 to 4 mm/yr. These agree well between the two methods and also with already published data for the major outlet stations at the Himalayan front. However, in the Middle and High Himalayas the cosmogenic erosion rates are significantly higher than those from suspended sediment measurements. While on the short term (intra-annual) a clear relation between precipitation and erosion can be observed, the cosmogenic erosion rates show no clear dependency with the basin wide precipitation pattern. Furthermore, no relation could be observed with the dominant lithological units and the degree of glaciation. Our observations confirm the overall established relationship between erosion rates, relief and slope, showing clearly that topography exerts a predominant control on spatial erosion rates on the millennial timescale. However, we observe a different relationship between main stream basins (> 250 km2) and small tributary basins (< 250 km2). Small basins show in general lower erosion rates than larger basins for respectively the same topographic characteristics.

Medium-energy AMS facilities such as the British 5 MV-NEC machine at “SUERC”, the French 5 MV-HVEE-machine “ASTER” and the two German 6 MV-HVEE-machines “DREAMS” at Dresden [1] and “Cologne AMS” have been recently installed in Central Europe. Of course, these machines need physicists to get them running but also scientists to establish AMS chemistry on-site. As it is not advisable to change simultaneously two ”things”, i.e. machine and chemistry, a close cooperation with the teams of ASTER and VERA helped to check the new sample preparation laboratories of DREAMS. Generally, a ”good” AMS sample has two features: high stable isotope current and low isobar concentration. High chemical yields and low concentrations of other elements originating from the matrix or chemical products used are less important, but may play a role if e.g. a matrix contains Ti being introduced into BeO-targets as shown by µ-XRF [2] and recent µ-PIXE analyses of final AMS-targets at HZDR. A processing blank with low radionuclide/stable nuclide ratio is, however, essential for projects near the detection limit. For high sample throughput and reasonable costs a fast, easy and cheap chemical separation is also favourable. The results from first samples processed at DREAMS are excellent: For 10Be-AMS-targets isolated from quartz-rich river sediments (Himalaya, erosion rate study) 9Be-currents had been as high as ASTER standards and machine blanks. Corresponding processing blanks were in the same order as the machine blank (1x10-15), thus, more than one order of magnitude lower than the lowest sample ratio. Chlorine-36 targets originating from calcite-rich boulders from a medieval rockfall had been well-measurable with blank-corrections lower than 8 %. Finally, 36Cl prepared from river terraces (Anatolian Plateau, Turkey, uplift rate study) showed also perfect performance at VERA. References: [1] Akhmadaliev et al., this meeting. [2] S. Merchel et al., NIMB 266 (2008) 4921. Acknowledgments: Thanks to T. Schildgen and C. Yildirim (Potsdam) performing 36Cl chemistry tests at DREAMS.

Cosmonuclide concentrations in surface samples are function of both duration of exposure to cosmic radiation and surface stability. From the measurement of a single cosmonuclide concentration in a single surface sample, it is thus impossible to quantify simultaneously the exposure duration and the denudation rate affecting the studied object. In the case of a simple exposure history, measurements of two cosmonuclide concentrations within the same surface sample may however theoretically be used to estimate both unknowns. A more constraining approach to accurately quantify both exposure time and denudation rate is to take advantage of the fact that the effective production attenuation length of neutrons is significantly shorter than that of muons. The neutron-induced cosmonuclide concentrations thus reach steady-state with respect to denudational loss much more rapidly than the muon induced ones. Consequently, the near-surface cosmonuclide concentrations mainly resulting from interactions with neutrons might be used to estimate the denudation rates while the several meters depth concentrations mainly resulting from interactions with muons might be used to estimate the exposure duration. A unique well constrained depth profile thus permits determination of both the exposure time and the denudation rate. In the case of abandoned material, inheritance due to previous exposition to cosmic rays can be revealed. Multi-nuclide depth profiles are also excellent tools for better deducing the physical parameters of the particles involved in the production of these cosmogenic radionuclides. This approach has been applied twice: 1.) 10Be and 26Al along a 25 meters pure quartz core from Galicia, Spain, and 2.) 10Be, 26Al and 36Cl along a ~11 meters carbonate and quartzose conglomerates core from La Ciotat, S.E. France. Both projects confirm that the density of the material is one of the most sensitive parameters. The latter study also allows to precise the still debatable spallation production rate of 36Cl from Ca.

Since the acceptance tests of ASTER in March 2007, routine measurement conditions for the long-lived radionuclides 10Be and 26Al have been established. Sample throughput as high as over 3300 unknowns has been reached for 10Be in 2010. Steady numbers for 26Al within the last three years settle down ~300 real samples. Unacceptable cross-contamination for volatile elements has been largely solved by an ion source upgrade [1]. Thus, the second frequent nuclide measured at ASTER in 2010 is 36Cl with ~480 sample targets. The enhancement with respect to immediate short-term sample to sample by the new ion source is impressive allowing minimal measurement time losses such as for initial burn-in periods of 5 min for virgin targets and waiting periods between data acquisition of two sample runs of 2 min. However, recent long-term tests using 35Cl/37Cl samples with strongly varying ratios have shown that identical targets lead to different 35Cl/37Cl results at the 2-4% level when being measured after a time gap of 24 hours while the source is running other samples. Reasons for this such as source memory, time dependent mass fractionation, drift of the Faraday-cup measurement system or something else are not yet clear. Finally, after establishing quality assurance at ASTER by cross-calibration of secondary in-house 26Al and 41Ca standards [1] and taking part in round-robin exercises of 10Be and 36Cl, we performed a two-step cross-calibration of secondary in-house 129I standards. The two ampoules of NIST 3231 containing 129I/127I at 0.981x10-6 have been used for step-wise dilution with NaI (MERCK, suprapur, 99.99 %) to get gram-quantities of lower-level standards for every-day use. The material SM-I-9 (~1x10-9) has been measured vs. AgI produced from the two NIST ampoules with (0.982+0.012)x10-8 solution using minimum chemistry. In a second stage, SM-I-10 and SM-I-11 with ratios of ~1x10-10 and ~1x10-11, respectively, have been cross-calibrated against SM-I-9. Individual uncertainties of the traceable secondary standards are around 1.5 %, mainly originating from the given uncertainty of the primary NIST 3231 at 10-8. References: [1] M. Arnold et al., NIMB 268 (2010) 1954.

As a first step towards using world-wide a common 36Cl standard-type material for the normalization of 36Cl AMS results, we have initiated an interlaboratory comparison, but with differences in the execution of the project compared to other round-robin exercises: a) participants have given up anonymity to better optimize discussion and conclusions, b) preliminary results have been shown at various scientific meetings and workshops to discuss first trends and to increase the number of participants, c) participants had the possibility to repeat their AMS measurements and/or change measurement setups or data evaluation to improve individual data [e.g. 1]. We are aware and acknowledge that these later changes by some of the participants, which led to different overall data compared with the initial results, are not fully consistent with the idea of a normal proficiency test layout. However, these data changes were accepted to better meet the main objectives of the intercomparison: to identify and eliminate differences between AMS laboratories, thus, improving the 36Cl data quality world-wide as soon as possible. The evaluation of the final results of the eight participating AMS laboratories for three synthetic AgCl samples with 36Cl/Cl ratios at the 10-11 (SM-Cl-11), 10-12 (SM-Cl-12), and 10-13 (SM-Cl-13) level shows no difference in the sense of simple statistical significance [2]. However, more detailed statistical analyses indicate a certain interlaboratory bias and an under-estimation of uncertainties by some of the laboratories. The round-robin data demonstrate that 36Cl/Cl results from two individual AMS laboratories can differ by up to 17% (Fig. 1). Thus, the necessity for further work on harmonizing the 36Cl business on a world-wide scale and on improving the measurements is obvious. Fig. 1: Regression of normalized results for the three investigated samples in the order of ascending average. [1] V. Alfimov et al., Ion Beam Physics, ETH Zurich, Annual Report (2009) 13. [2] S. Merchel et al., Anal. Bioanal. Chem., submitted.

O podoru v Velikem vrhu (Košuta, Karavanke) ni zanesljivih zgodovinskih zapisov, vendar pa le ti obstajajo o podoru na Dobraču (25.1.1348), ki je od Velikega vrha oddaljen 46 km. Podor je povzročil potres in naša hipoteza je bila, da je tudi podor v Velikem vrhu posledica istega dogodka. Tako smo s pomočjo metode datiranja površinske izpostavljenosti analizirali vzorce matične kamnine v steni Velikega vrha ter vzorce s površine podornih blokov. Ugotavljali smo vsebnost 36Cl, ki se je začel tvoriti po podoru. Na podlagi poznavanja števila atomov 36Cl na gram Ca na leto izpostavljenosti, čas dogodka (podor) izračunamo iz koncentracij 36Cl izmerjenih s pomočjo pospeševalnika (AMS). Prvi rezultati kažejo, da sta se podora na Dobraču in Velikem vrhu posledica potresa v Furlaniji Julijski krajini. Starost podora v Velikem vrhu je (740 +- 71) a. Za bolj zanesljive rezultate bo potrebno izboljšati sam model in upoštevati več dejavnikov - predvsem nadmorsko višino in klimatske razmere. There are no reliable historical data about the Veliki vrh rockfall (Košuta, Karavanke Mountains) but the records exist about the Dobrach rockfall (25th of January 1348) 46 km to the West of Veliki vrh. The rockfall was triggered by the earthquake and our hypothesis was that Veliki vrh rockfall was induced by the same event. Therefore we used the surface exposure dating method to analyze the bedrock samples in the Veliki vrh rock face as well as the samples of the boulder surface. The content of 36Cl, that started to produce after the earthquake, was measured. On the basis of the known number of 36Cl per gram Ca per year of exposure, the time of the rockfall can be calculated from the concentration of 36Cl measured by accelerator mass spectrometry (AMS). Preliminary results show the correlation of the two rockfalls (Dobrach and Veliki vrh) as the consequence of the earthquake in Friuli Venezia Julia. The age of Veliki vrh rockfall is (740 +- 71) a. For more reliable results further research should be conducted using also the data on altitude and climate.

In contrast to siliceous environments, there is a severe lack of cosmogenic nuclides, which can be used for in-situ dating of calcareous environments. Thus, we have investigated other nuclides than 36Cl as possible dating tools by cross-calibration. Cosmogenic 10Be is highly contaminated with atmospheric 10Be and cannot be removed quantitatively, even by a very sophisticated chemical cleaning procedure. Only working on clay-free calcite provides correct 10Be data, giving a 2.7 times higher production rate of 10Be from CaCO3 than from SiO2. Though, the production rate of 26Al is only ~4.6% (CaCO3 relative to SiO2), 26Al can be easily determined in calcite, as the low intrinsic 27Al concentration yields to nearly as high 26Al/27Al as within corresponding quartz. The measurement of 41Ca, mainly produced via thermal-neutron-capture, is hindered by very low 41Ca/Ca: < 2.5x10-15.

Driven by the progress in AMS and its spreading application within geosciences, measurements of increasing numbers of samples with low isotopic ratios will be required in the future. Therefore, we have examined the linearity of 10Be/9Be as a function of isotope ratio by distributing 3 secondary standards (dilutions of NIST4325: 10-12-10-14) to 9 AMS labs. The problem of low ratio samples is even more crucial for 36Cl mainly due to the high volatility of chlorine. Thus, we have prepared large quantities of 3 36Cl/Cl solutions from a certified 36Cl activity (NIST4943) by dilution with NaCl. AgCl precipitated from these solutions (10-11-10-13) has been distributed to 9 AMS labs. Some measurements are still ongoing. First results from 6 labs for each nuclide show that these interlaboratory exercises are very valuable.

The Monturaqui impact crater is the only meteorite impact related structure yet found in Chile. It is localized at 3,015 m altitude in the precordillera near the southern end of Salar de Atacama. It is a simple crater of ~400 m diameter and ~34 m of depth [1], first referred as an impact crater by [2]. The age of the crater was estimated as older than 0.1 Ma by [3] by thermoluminiscence analysis. We are reporting the first absolute ages of the Monturaqui impact crater following two approaches: a) the terrestrial age of the impactor by measuring the residual activities of 10Be, 26Al, 36Cl, 41Ca, 59Ni, 60Fe, and 53Mn in selected iron shale samples, which corresponds to the remaining altered fragments of the impactor, inferred to be an iron meteorite, and b) in-situ ages obtained through the use of long-lived terrestrial cosmogenic radionuclides 10Be and 26Al in the granite outcrops exposed to cosmic radiation starting after the impact. Chemical preparation of targets suitable for accelerator mass spectrometry (AMS) have been performed after [4] (for the iron meteorite sample) and a combination of slight modifications of [4] and [5] (for the granite samples). AMS measurements of 10Be and 26Al have been performed at the French 5 MV-AMS facility ASTER, 36Cl at CAMS, LLNL, USA, and 53Mn at the Maier-Leibnitz-Laboratory (MLL). We can compare our measured radioactivities with depth-depending production rates from sophisticated theoretical Monte-Carlo calculations [priv.com. I. Leya]. As these production rates are a function of the chemical composition (of the impactor in space), remaining fragments are highly altered and precise chemical analyses could not yet be achieved, certain further assumptions are influencing the following discussion of our, thus preliminary, data. The longest-lived radionuclide 53Mn (t1/2=3.7 Ma), normalized to a fully corroded Fe2O3-sample, is the least sensitive nuclide to a varying terrestrial age, thus, providing us with a shielding depth of 62-71 cm. The best fit of the measured shortest-lived radionuclide 36Cl (t1/2=0.3 Ma) with theoretical production rate at that depths is for a terrestrial age of 500-600 ka. The 26Al-activity goes along with that age. Though, the measured 10Be-concentration is far too high in comparison to the theoretical production rate, which are based on an average carbon-content of 0.1% (as Canyon Diablo). As earlier studies [6,7] demonstrated the great influence of inhomogeneous distributed trace elements like C, S, and P on the production rates of lighter cosmogenic radionuclides in iron meteorite samples. Finally, under the contrary assumption of no corrosion of the impactor, the whole discussion changes only slightly: Deeper shielding position (66-80 cm), but as production rates of 53Mn and 36Cl are influenced the same way, the terrestrial age will not change. Our second approach using terrestrial cosmogenic radionuclides leads to concordant results for 10Be only: The minimum in-situ exposure age of two samples from the crater wall could be calculated to 200-250 ka. A larger age of excavation is very likely due to the subsequent erosion of the crater walls. We are looking forward to measurements of the most sensible 41Ca (t1/2=0.1 ka) that might improve the accuracy of this age. References: [1] H. Ugalde et al., MAPS 42 (2007) 2153. [2] J. Sanchez, W. Cassidy, J. Geophys. Res. 71 (1966) 4891. [3] V.F. Buchwald, Handbook of iron meteorites, Univ. of California Press, Berkeley. Vol. 1 (1975) 262. [4] S. Merchel, U. Herpers, RCA 84 (1999) 215. [5] E. T. Brown et al., GCA 55 (1991) 2269. [6] I. Leya, Michel R., Lunar Planet. Sci. 29 (1998) #1172. [7] I. Leya et al., MAPS 32 (1997) A78.

In-situ produced cosmogenic nuclides have proved to be valuable tools for quantifying Earth's surface processes. Here, the work-horses are 10Be and 26Al in quartz-rich minerals, and 36Cl in Ca- or K-rich minerals. Several attempts to find new matrix-product-pairs have been yet performed, especially with respect to broaden the time-scale to both more ancient [1] and more recent (historic) times. Thus, we have investigated other nuclides than 36Cl as possible dating tools by cross-calibration in accompanied calcite- and quartz-rich samples from Antarctica (DV3 & Joh) and Southern France (Ciot). AMS measurements of 10Be and 26Al have been performed at the French 5 MV-AMS facility ASTER, 36Cl at CAMS, LLNL, USA, and 41Ca at the Maier-Leibnitz-Laboratory. As we could only perform a single run for 41Ca measurements, all results can be regarded as preliminary. Ratios between different nuclides from the same matrix (CaCO3) and ratios of 10Be or 26Al from CaCO3 and SiO2 can be compared with pure physical model calculations [2] giving us experimental terrestrial production rates for 10Be, 26Al and 41Ca from Ca and CaCO3. As shown earlier [3], cosmogenic 10Be is highly contaminated with atmospheric 10Be and cannot be removed quantitatively from calcite samples, even by an improved chemical cleaning procedure [4]. Only working on clay-free calcite provides correct 10Be data, giving a 2.7 times higher production rate of 10Be from CaCO3 than from SiO2. Though, the production rate of 26Al is only ~4.6% (CaCO3 relative to SiO2), 26Al can be easily determined in calcite, as the low intrinsic 26Al concentration yields to nearly as high 26Al/27Al as within corresponding quartz. The measurement of 41Ca, mainly produced via thermal-neutron-capture, is hindered by very low 41Ca/Ca: -15. Of course, 41Ca counting statistics are poor: Measured 41Ca/Ca values are based on total counts of 1-5. Nevertheless, the reproducibility (Joh & Joh-WC) is excellent. All our data are lower than the already published one from rock samples, i.e. lower than the six surface samples (3-63 x 10-15) of Henning et al. [5] and Kutschera et al. [6] and lower than the surface and strongly shielded sample at 11 m depth (3.4-7.6 x 10-15) of Middleton et al. [7]. The low 41Ca/Ca ratios make it very unlikely that 41Ca could be generally used for in-situ dating of calcareous environments, especially as there is little hope that background level for CaF2-targets will improve, thus asking for very sophisticated and time-consuming CaH2-target preparation and handling [9]. Acknowledgments We appreciate the help of D. Lal (UCSD), M. Arnold and G. Aumaître (CEREGE), J. Lachner and I. Dillmann (TU Munich), S. Nardon (ENI S.p.A., Milan) and J. Borgomano (U Marseille). This work was partially funded within the framework of CRONUS-EU (Marie-Curie Action 6th FP; #511927). References [1] J.M. Schaefer et al., EPSL 251 (2006) 334. [2] J. Masarik et al., NIMB 259 (2007) 642. [3] S. Merchel et al., Quat. Geochronol. 3 (2008) 299. [4] S. Merchel et al., NIMB, in review. [5] W. Henning et al., Science 236 (1987) 725. [6] W. Kutschera et al., Radiocarbon 31 (1989) 311. [7] R. Middleton et al., Radiocarbon 30 (1989) 305. [8] G. Korschinek and W. Kutschera, NIM144 (1977) 343.

cited By 3; International audience; no abstract