Sender: |
|
Subject: |
|
From: |
|
Date: |
Mon, 12 May 1997 16:44:19 +0100 |
Reply-To: |
|
Parts/Attachments: |
|
|
Jean-Pierre,
I did not x-ray the material. My conclusions are based on the following
observations.
Ni holders exhibited tarnish that has the appearance of Ni oxide after
running garnet samples yielding about 0.2 to 0.1 micromoles of O2. The
tarnish is visible after fluorinating two or three samples of this size
using IR laser heating. Delta 18O values for these standard samples were
consistently low by approximately 2 to 3 per mil (gases analyzed as O2).
Au coating of the entire holder eliminated the fractionation (delta values
were correct) and the Ni exhibited no tarnish beneath the gold coating.
Where the gold coating was too thin, as evidenced by formation of visible
tarnish, delta values were again too low. Correlation between formation of
visible tarnish on the Ni holders and isotopic fractionation indicates
that oxygen is being lost to the Ni.
In addition, we found a correlation with grain size. Larger quartz grains
(McMaster OQQ quartz standard provided by Yuch-Ning Shieh) did not exhibit
the isotope fractionation nor did the Ni holder show evidence of oxidation
for these grains (larger than 200 microns). Smaller grains (NCSU quartz)
showed large negative fractionations and significant oxidation of Ni.
Experiments with the CO2 laser indicate that the difference is the extent
to which the laser strikes the Ni holder. Smaller grains (or those that
are more difficult to heat for any reason) require more rastering and the
laser therefore strikes the holder directly throughout much of the
analysis. Larger grains take up most of the laser energy as they melt. In
the former case the Ni holder is heated to 40C or higher while in the
latter case little heating of Ni occurs. Oxidation is accelerated by the
temperature of the metal substrate. Thus an apparent grain size effect.
-Ed Young
Earth Sciences
University of Oxford
|
|
|