Biomedical Engineering Reference
In-Depth Information
order to improve counting statistics and reduce (personnel) costs. However,
automated analysis requires high-quality spectra or micrographs that are
rich in contrast. The automation algorithms must be carefully tested with
appropriate reference samples, including variant particle morphologies. Only
both reliable algorithms and larger datasets will help improve the reliability
of data obtained from individual analysis.
2.4.2 Ensemble-Averaging Characterization
Techniques for Nanoparticles (EAA)
Ensemble-averaging techniques determine nanoparticle property distribu-
tions by integrating the collective response of a particle ensemble to a stimu-
lus. The main advantage of ensemble-averaging techniques is the size of the
ensemble that is analyzed with high acquisition speed owing to collective
signal generation. The interpretation of data from ensemble-based analysis
generally requires model assumptions for the individual particle response to
the stimulus in order to derive particle property distributions. For example, in
DLS for stimulus, a laser beam is focused on the sample, scattered by individ-
ual particles and results in collective interference that shows fluctuations due
to Brownian particle motion. Theoretical models are then used to describe
individual light scattering and particle motion. Intensity fluctuation interpre-
tation requires complicated data deconvolution in order to obtain the particle
size. The validity of the underlying model assumptions can sometimes not be
easily tested by the nonexpert user. The interpretation of measurement data
of ensemble-averaging analysis techniques therefore requires awareness for
possible experimental error sources. Especially, unexpected particle features
such as exotic shape, spontaneous agglomeration, or surface charging effects
may result in systematically misinterpreted data. Therefore, the applied model
assumptions generally require verification by microscopic analysis of a repre-
sentative set of individual particles.
2.4.3 Sample Preparation Issues
Sample preparation for toxicological studies and material characterization is
a very demanding task and a crucial basis for reliable results. Artifacts and
severe systematic errors can arise from incorrect sample preparation proce-
dures. Insufficient stability of nanoparticle dispersions and accuracy of mate-
rial dosing are closely related, omnipresent problems. Aging phenomena of
nanoparticles, such as accelerated surface oxidation, imply additional com-
plications for reliable and reproducible studies. Incomplete dispersion and
uncontrolled reagglomeration makes primary particle size determination
from nonmicroscopic methods invalid. Controlled suspension concentration
by means of dilution will not be successful in case of unexpected sedimenta-
tion or dish wall adherence. The practical determination of particle concentra-
tion in liquid fractions taken from dispersions is not easily achieved with high
