Paper No. 5
Presentation Time: 2:05 PM

IN-SITU 4HE/3HE ANALYSIS OF VERTICALLY MOUNTED ZIRCONS


VERMEESCH, Pieter1, SCHWANETHAL, James1, CARTER, Andrew2 and JEFFRIES, Teresa E.3, (1)Earth Sciences, University College London, Gower Street, London, WC1E 6BT, United Kingdom, (2)Department of Earth and Planetary Sciences, Birkbeck, University of London, Malet Street, London, WC1E 75x, United Kingdom, (3)Core Research Laboratories, The Natural History Museum, Cromwell Road, London, SW7 5BD, United Kingdom, j.schwanethal@ucl.ac.uk

In-situ U-Th-He dating of zircon offers significant advantages over conventional ‘whole grain degassing and dissolution’:

1. Order of magnitude higher sample throughput.

2. U-Th-He/U-Pb double-dating by default.

3. The ability to generate depth profiles and thus reconstruct the 3-dimensional distribution of U,Th and He which contains valuable thermal history information.

In-situ dating also introduces new challenges:

1. The need to accurately measure the absolute concentration of U, Th and He by micro-analysis.

2. Measuring the radioactive parent and daughter in different volumes, which reduces the accuracy in compositionally zoned grains.

We have developed a novel method to address both of these challenges:

1. We avoid the need for absolute concentration measurements by normalising all the raw mass spectrometer measurements to a standard of known U-Th-He age. The need to measure the volume of the laser ablation pits is removed by normalising to stoichiometric 29Si for U and Th, and to proton-induced 3He for He.

2. The effect of (concentric) compositional zoning is minimised by mounting the zircons vertically between two teflon wafers, and ‘drilling’ parallel to the c-axis.

We have tested the new method on 58 grains of Fish Canyon Zircon (FCZ, ~27.8 Ma), using a 470 Ma old Sri Lanka zircon as an age standard, an Agilent 4500 LA-ICP-MS with 193 nm laser for 238U/29Si and 232Th/29Si measurements, and a Noblesse sector field mass spectrometer with 213 nm laser for 4He/3He measurements. The central age of the data falls within two standard errors of FCZ (29.8 +/- 3.6 Ma), but the single-grain age estimates are overdispersed with respect to the formal analytical uncertainty by 47%. We suspect this excess scatter is due to:

1. Possible gradients within the sample vials during proton irradiation.

2. The effect of non-concentric compositional zoning.

Both problems can be tested and solved with further experiments. The proton-gradients can be removed by frequently changing the orientation of the sample vials during irradiation. The compositional zoning problem can be mitigated by reducing the depth of the He ablation pit. To increase sensitivity to this lower signal, we are able to reduce the volume of out extraction line.