Environmental Engineering Reference
In-Depth Information
table 4.3
Scale Factors for Converting
202/198
Hg Fractionation Factors for Other Isotope
Ratios with
198
Hg as the Denominating Isotope
196/198
Hg
199/198
Hg
200/198
Hg
201/198
Hg
202/198
Hg
204/198
Hg
Reference
MD
equ
0.515
0.254
0.505
0.754
1.000
1.486
Young et al., 2002
MD
kin
0.252
0.502
0.752
1.000
1.492
Young et al., 2002
NV
0.466
0.107
0.497
0.700
1.000
1.654
Schauble, 2007
NV
0.080
0.471
0.674
1.000
1.499
Ghosh et al., 2008
NOTE
: Provided are factors following equilibrium and kinetic mass-dependent fractionation laws (MD
equ
and MD
kin
) and two solutions for
nuclear volume (NV) fractionation, based on modeled and experimentally derived nuclear radii.
Precision of the Analytical Measurement
Often, researchers also report the internal precision of an
isotope ratio measurement, usually expressed as two times
the standard error (2 SE) of repeated isotope ratio measure-
ments during a single analytical run. Although this value is
an indication for the precision of the individual measure-
ment, it does not take into account potential variability
introduced during sample preparation or by inherent sam-
ple inhomogeneities, which is often considerable for matri-
ces such as soils or sediments. At any rate, authors should
clearly state the level of uncertainty of their measurement
and how it was derived.
As for any other analytical measurement, it is important to
report a measure of uncertainty for each determination to
allow for the evaluation and comparison of data. The most
important parameter is the reproducibility of the overall
analytical method, including sampling, sample prepara-
tion, and measurement. For unknown samples, this can
be assessed only by replicate analysis—that is, measure-
ment of replicate sample preparations. This uncertainty is
often referred to as the external reproducibility (or some-
times called “between-run precision”) and characterizes
the distribution of independent measurements around the
population mean. The external reproducibility is expressed
in multiples of the standard deviation of the sample (in
analytical chemistry, commonly 1 SD) and requires the
measurement of at least three individual sample prepara-
tions. However, often there is not suffi cient sample material
(or mass of Hg) available to allow replicate Hg isotope ratio
analyses. In this situation, a conservative estimate of uncer-
tainty should be provided. A related sample that is similar or
close in matrix to the sample in question and that is avail-
able in suffi cient quantity could be measured repeatedly to
provide a measure of the combined uncertainty resulting
from sample preparation and measurement. Preferably, this
surrogate sample should be an actual environmental sam-
ple rather than a CRM to truly account for uncertainties
associated with in-house sample homogenization.
Another useful performance indicator is the accuracy of
the deviation measurement (accuracy, with which
Mercury Isotope Data of Natural Samples
At time of this writing, there have been only a dozen or
so published studies describing Hg isotope ratio variations
in natural samples. Although only 10 years ago many sci-
entists were skeptical that the Hg isotope system is subject
to natural fractionation in the fi rst place, some pioneer-
ing studies have now unequivocally shown that variations
exist and that Hg exhibits rich and often unexpected frac-
tionation patterns. Figure 4.1 illustrates the range of
202
Hg
that has been observed in nature so far; this is discussed in
more detail in the following section. Although initial deter-
minations concentrated on document variations in natural
Hg isotope ratios, more recent investigations have started
to focus on the (bio)geochemical processes causing the
observed differences. Newer studies have confi rmed that
analytical methods have matured to a point at which rou-
tine measurements have now become possible. Note that all
of the following
δ
202
Hg
can be determined). This can be evaluated by analyzing
a reference sample (CRM or in-house standard), which
has a different isotope composition than the bracketing
standard. Measured
δ
values conform to the notation described
in the previous section, report
δ
202
Hg, and were converted
from the original literature where necessary and possible.
δ
202
Hg for this secondary standard
should be consistent for each analytical session. The preci-
sion with which this deviation can be measured over an
extended period of time should also be the limiting value
when reporting the external reproducibility. Currently, the
reported precision for such comparisons is slightly better
than 0.10‰ (2 SD). Hence, apply ing today's technolog y and
instruments, measured Hg isotope ratio deviations (
δ
Sediments
H g i s of t of p e r a t i of s i in a s e r i e s of f s e d i m e in t s of b t a i in e d f r of m m a in y
different locations have been determined (Foucher and
Hintelmann, 2006). The
δ
202
Hg displayed a range of almost
5‰, ranging from
0.74‰. Large variations were
observed near a gold mine in New Brunswick, Canada,
where cyanide leaching mobilized large concentrations of
4.00 to
δ
202
Hg)
between two samples must be
0.10‰ to be signifi cant.
