Biomedical Engineering Reference
In-Depth Information
and we assume that a similar ratio also holds for H
2
P. This means that at very high
H
2
P concentrations and long waiting times the TOPO ligand shell becomes nearly
completely replaced by H
2
P and the related
K
M
is much larger than
K
L
.
Thus, the influence of the ligand exchange dynamics on the QD PL efficiency
(quenching) may be outlined as follows. Sample preparation is accompanied by
diluting the original QD-toluene solution followed by a new equilibrium of TOPO
coverage in case that there is a relatively low concentration of TOPO in the
sample. This implies an overall reduction of the QD surface coverage by TOPO
ligands resulting in a decrease of the PL quantum yield due to the formation of
more unsaturated surface states [
30
,
56
]. Such surface states have even without
additional quenchers different PL quantum yields as compared to the near band
edge or excitonic states [
56
]. Additionally, this will allow for further attachment of
quenchers, either those which are already inherently present or are added during a
titration procedure. The time scale of such processes is in the present experiments
typically between 200 and 2000 s (see Figs.
4.23
and
4.24
) and depends critically on
the experimental conditions such as the type of QD, TOPO, (unknown) impurities
and/or dye concentration. However, besides processes on a time scale below 2000 s,
a further reorganization of the surfactant shell (accompanied by increased quenching
of QD PL) occurs on much longer time scales which can be still observed. It might
be related to the formation of new TOPO structures such as surfactant islands on
the QD surface or TOPO micelles in the solution. In principle, the configuration
of even parts of the ligands on the surface of semiconductor nanocrystals plays a
critical role for their stability [
161
]. Another plausible explanation might be related
to an oxidative degradation of the QD surface on these long time scales, especially
in the case of amine ligands and PDI molecules [
30
,
31
,
74
,
162
]. We would like
to emphasize that also at these long time scales a still increasing attachment of
(m-Pyr)
4
-H
2
P as identified by further increasing FRET has been observed (see
Fig.
4.23
).
From all these facts and discussions it follows that that non-FRET quenching
in “QD-Dye” nanoassemblies might be related to depletion of ligands by the
respective dye molecules. Following the argument that non-coordinated Zn
2
+
atoms (and the corresponding surface, intra-band states) decrease the PL intensity
according to their absolute number [
30
] replacement of ligands by dye molecules
will increase the number of uncoordinated Zn
2
+
since pyridyl anchors of the given
dye molecules can at most saturate two bonds. Since dye molecules replace several
ligands due to their large volume the PL quenching will increase. However, the
related number of removed ligands necessary to explain the observed non-FRET
efficiency was much too large with respect to the volume of the dye molecule.
Here, we suggest that this “unusual quenching mechanism” could be related to the
crystal structure of the QD which results with respect to the nature of ligand sites
in different binding strengths and/or different luminescent properties and/or charge
localization features of intra-band states. In case that there are various attachment
sites with varying binding strengths it is reasonable to assume that these sites are in a
different way sensitive to dye attachment either due to a formation of (dye specific)
trap states, which quench the QD PL effectively, or are more easily exchanged by
