Carbonate and silicate intercomparison materials for cosmogenic 36Cl measurements

Silke Mechernich, Tibor Dunai, Steven Binnie, Tomasz Goral, Stefan Heinze, Alfred Dewald, Irene Schimmelpfennig, Karim Keddadouche, Georges Aumaître, Didier L. Bourlès, Shasta Marrero, Klaus Wilcken, Krista Simon, David Fink, Fred. M. Phillips, Marc Caffee, Laura Gregory, Richard Phillips, Stewart Freeman, Richard ShanksM. Akif Sarikaya, Stefan Pavetich, Georg Rugel, Silke Merchel, Naki Akçar, Serdar Yesilyurt, Susan Ivy-Ochs, Christof Vockenhuber

Research output: Contribution to journalArticlepeer-review


Two natural mineral separates, labeled CoCal-N and CoFsp-N, have been prepared to serve as intercomparison material (ICM) for in situ-produced cosmogenic 36Cl and natural chlorine (Clnat) analysis. The sample CoCal-N is derived from calcite crystals in a Namibian lag deposit, while the sample CoFsp-N is derived from a single crystal of alkali-feldspar from a Namibian pegmatite. The sample preparation took place at the University of Cologne and a rotating splitter was used to obtain homogeneous splits of both ICMs. Forty-five measurements of CoCal-N (between 1 and 16 per facility) and forty-four measurements of CoFsp-N (between 2 and 20 per facility) have been undertaken by ten target preparation laboratories measured by seven different AMS facilities. The internal laboratory scatter of the 36Cl concentrations indicates no overdispersion for half of the laboratories and 3.9 to 7.3% (1σ) overdispersion for the others. We show that the CoCal-N and CoFsp-N splits are homogeneous regarding their 36Cl and Clnat concentrations. The grand average (average calculated from the average of each laboratory) yields initial consensus 36Cl concentrations of (3.74 ± 0.10) × 106 at 36Cl/g (CoCal-N) and (2.93 ± 0.07) × 106 at 36Cl/g (CoFsp-N) at 95% confidence intervals. The coefficient of variation is 5.1% and 4.2% for CoCal-N and CoFsp-N, respectively. The Clnat concentration corresponds to the lower and intermediate range of typical rock samples with (0.73 ± 0.18) µg/g in CoCal-N and (73.9 ± 6.8) µg/g in CoFsp-N. We discuss the most relevant points of the sample preparation and measurement and the chlorine concentration calculation to further approach inter-laboratory comparability. We propose to use continuous measurements of the ICMs to provide a valuable quality control for future determination of 36Cl and Clnat concentrations.
Original languageEnglish
JournalNuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms
Early online date30 Jan 2019
Publication statusE-pub ahead of print - 30 Jan 2019


  • accelerator mass spectrometry
  • terrestrial cosmogenic nuclides
  • round robin
  • intercomparison material
  • consensus values


Dive into the research topics of 'Carbonate and silicate intercomparison materials for cosmogenic 36Cl measurements'. Together they form a unique fingerprint.

Cite this