Geoscience Reference
In-Depth Information
oxides, or glass fibers. Sieve filters have much more consistent pore sizes and are typically
made from plastic films (e.g., polycarbonate or polysulfone capsule filters).
Filters can be a significant source of organic matter contamination and must be cleaned
appropriately. Glass fiber filters such as Whatman GF/F are one of the most popular choices
of filters owing to their rapid flow rate, high loading capacity and ease of cleaning (combust
at 450°C > 4 hours). All polycarbonate or polysulfone capsule filters should be cleaned as
recommended by the manufacturer and flushed with copious amounts of laboratory-grade
water. Silver filters have also been employed successfully with respect to DOM absorbance
and fluorescence analyses (Lapworth et al.,
2009
). It is recommended for all filter types
that a suitable volume of sample water (dependent on the loading capacity of the filter) is
rinsed through before collection of any sample water and that all surfaces that come into
contact with the sample should be triple rinsed with filtrate (e.g., sample bottles). Filter
holders must be precleaned as described in
Section 4.2.3
and the filtration system thor-
oughly tested for any contamination (see
Section 4.2.2
). Vacuum, pressure, and gravity
filtration are all routinely applied by researchers in their filtration units. The filtration pres-
sure should be maintained at the lowest necessary to pass the filter in a reasonable period
of time with respect to any potential microbial action and filtration should be conducted out
of direct light. High pressures have been shown to result in lysis of cells during filtration
(Rosenstock and Simon,
1993
) and subsequent “contamination” of the filtrate. Clogging
of filters reduces the flow rate and also may reduce the nominal pore size of the filter and
over time lead to the lysis of cells into the filtrate and so should be avoided. In-line filtra-
tion typically of a large prescreen pore size filter (e.g., 10, 1.2 µm) before a small pore size
filter (e.g., 0.2 µm) is recommended in many freshwater systems to avoid clogging and
associated issues if desiring to filter to small pore sizes (e.g., Ahad et al.,
2006
; Saraceno
et al.,
2009
).
4.3.2 Effects of Filtration on Fluorescence
The impact of filtration has been examined with respect to DOM fluorescence both in
laboratory and
in situ
-based studies. Fluorescence of organic matter was examined in six
contrasting U.K. freshwaters in unfiltered, 1.2 µm filtered, and 0.2 µm filtered waters to
compare the effects of filtered to unfiltered water on organic matter fluorescence and to
examine the effects of filtration at different pore sizes on DOM fluorescence in a study by
Baker et al. (2007). The six chosen sites ranged from those with very good through those
with bad water quality and ranged from free-flowing rivers to slow moving and regulated
canal waters and also lake water. The study by Baker et al. (2007) focused on two fluoro-
phores in detail: tryptophan-like fluorescence at an excitation of 225-230 nm and an emis-
sion range of 335-350 nm and humic-like fluorescence at an excitation of 230-245 nm
and an emission range of 395-430 nm. All of the samples in the study showed for the
tryptophan-like fluorescence a significant decrease between the unfiltered water and the
1.2 µm filtered water (5-71% decrease, mean = 35 %;
Table 4.1
) and the unfiltered water
and the 0.2 µm filtered water (32-86% decrease, mean = 58%;
Table 4.1
). The results
