188
Both methods provide the chemical composition based on
heating the sample and analyzing the decomposition products (Ivleva et al. 2016). Pyrograms or ion chromatograms
are obtained that can be compared to references, equivalent
to spectra of spectroscopic techniques (Löder and Gerdts
2015; Dümichen et al. 2017; Shim et al. 2017).
Most studies use these methods to analyze preselected
particles. Recently Fischer and Scholz-Böttcher (2017) presented also for Pyr-GC-MS an approach independent of a
prior visual sorting by analysing whole filters on which previously purified samples had been concentrated. That also
TED-GC-MS can be used to analyze subsamples of environmental samples without pre-selection to identify MPs has
been shown by Dümichen et  al. (2017). An advantage of
TED-GC-MS presented by Dümichen et al. (2017) is that a
relatively high sample amount of up to 100 mg can be processed, which, depending on the condition of the environmental sample, obviates the need for sample purification.
Chemical imaging approaches developed for μFTIR and
Raman spectroscopy eliminate the need for a visual preselection. Therefore, the purified samples are concentrated
on filters that are directly scanned. The filter chosen for the
analysis has to be compatible to the method by not interfering with the sample analysis (Käppler et  al. 2015; Löder
et al. 2015). For μFTIR the use of Focal plane array (FPA)
detectors have substantially improved the time needed for
the analysis of whole filter areas (Löder et  al. 2015; Tagg
et al. 2015; Käppler et al. 2016; Mintenig et al. 2017; Primpke
et al. 2017b). Although the imaging using FPA is independent of a prior visual selection of potential MPs, the approach
presented by Löder et al. (2015) still involves an operatorbased selection of MPs based on their spectral signature.
Therefore, advances are automated approaches independent
of human bias like it has been recently presented by Primpke
et al. (2017b).
Shim et al. (2017) recently reviewed the advantages and
disadvantages of currently used methods for identification of
MPs. Furthermore, Elert et al. (2017) added to the comparison a classification of the different techniques in terms of
restrictions, requirements and the analytical information
received.
The major advantage of thermal analysis is the simultaneous analysis of polymer and containing additives, while the
major disadvantage is the destruction of the sample by combustion. While thermal analyses provide mass-related results
only, spectroscopic analyses are normally non-destructive
and provide particle-related results (Shim et al. 2017). The
lower size limit for μFTIR is at 10 μm due to the diffraction
limit (Löder and Gerdts 2015; Shim et al. 2017), whereas for
Raman spectroscopy particles down to 1 μm size can be analyzed (Ivleva et  al. 2016). Residual water hampers FTIR
analysis while for Raman spectroscopy fluorescence of residues of the environmental matrix is a problem as well as the
interference from pigments (Imhof et al. 2016; Käppler et al.
2016; Shim et al. 2017). Käppler et al. (2016) showed that
Raman imaging provides a better identification of
MPs < 20 μm when compared to using FPA-μFTIR in transmission mode but with the major drawback for Raman imaging that the measurement time was more than 100-times
higher than the μFTIR analysis. Currently, μFTIR imaging of
large filter areas is considerably faster than Raman imaging,
even when reducing the resolution for Raman imaging,
resulting in a comparable quality to FTIR imaging (Käppler
et al. 2016).
All above mentioned methods share the commonality that
to avoid misinterpretation of spectra and programs alike as
well as identifying dyed MPs, efficient sample purification is
of utmost importance (Löder and Gerdts 2015; Crichton et al.
2017; Fischer and Scholz-Böttcher 2017; Maes et al. 2017b).
When choosing the most appropriate method: time demand,
size range, and sample preparation have to be considered.
Furthermore, thermal analysis should be used when a fast
assessment of mass-related data is required, while spectroscopic analysis provides particle-related data but might take
considerably longer. A holistic approach would involve FTIRanalysis of MPs down to 10  μm, Raman- analysis for MPs
below 10 μm and a subsequent thermal analysis.
Biological Effects of Microplastics on Biota
Although research on MPs in aquatic systems regarding
monitoring and abundance in animals has dramatically
increased in the last years, profound knowledge about the
effects of MPs on biota is still scarce (Ribeiro et al. 2017).
Here, we give a short overview about investigated consequences of MP exposure, methods, and their effects on
organisms.
Images circulating the media of sea turtles, dolphins or
seals entangled in plastic bags and other macroplastics are
well known, but what about the plastic we do not see?
Microplastics can pose a danger to organisms, when they are
ingested (Avio et al. 2017). Reasons for ingestion in the first
place are either MPs being mistaken for food or prey due to
similarities in size, shape or color (Wright et  al. 2013) or
because the organism is not selective with food particles,
which is, for example, the case for most filter and deposit
feeders (Van Cauwenberghe et  al. 2015). Although filter
feeders often possess some mechanisms to avoid particles
that are too big or inedible, MPs are very similar to actually
nutritious food and thus sorting mechanisms might not work
(Ward and Shumway 2004).
T. Hamm et al.
Précédent

- 197/259

Suivant