Biomedical Engineering Reference
In-Depth Information
1.2 s [
74
], which is considerably longer than observed for uncapped CdSe QDs
[
30
]. This difference might be due the addition of a ZnS shell which exhibits a
stronger ligand binding [
106
] as is also imposed by our findings by comparing the
quenching efficiencies for CdSe and CdSe/ZnS QDs. However, for TOPO ligands,
this time may be considerably longer since the binding strength of TOPO is much
higher than the one of amines [
106
]. In any case, the reduction of coverage of the
QD surface by removing ligands [
30
] or upon dilution [
34
] results in PL quenching.
This quenching is either due to the creation of unsaturated bonds causing a lower
PL quantum yield or due to an increased attachment of impurities and/or titrated
quenchers, respectively. In case that a fast ligand exchange is also present in our
experiments, a titration step will be followed by two processes, that is an increased
ligand desorption due to dilution and the replacement of ligands (L) by quenchers
(M), in our case H
2
P.
A fast ligand exchange rate would be consistent with our titration experiments,
since PL quenching is observed within less than 60 s after a single titration step.
However, in case that ligand exchange is much slower, the immediate PL response
after titration points towards the presence of a few “free” attachment sites, which
correspond to the volume of about 2-3 TOPO ligands per H
2
P. Depending on the
binding strength and the number of anchoring functional groups the equilibrium
constant
K
k
det
will be different for ligands (
K
L
) and quencher molecules
(
K
M
). This has been clearly proven in recent experiments [
62
,
74
] which revealed
that at a given molar ratio the quenching efficiency depends drastically on the
number and kind of (
ortho
-,
meta
-,
para
-) pyridyl substituents of H
2
P(seeFig.
4.6
).
The complexation of (m-Pyr)
4
H
2
P with CdSe/ZnS QD is estimated by the constant
K
M
=
k
att
/
10
5
-10
7
M
−
1
[
101
], which is by at least three orders of magnitude larger than
≈
K
L
50 M
−
1
observed for octylamine ligands [
30
]. Due to the high value of
K
M
,
most of H
2
P molecules will in the course of time be attached to the QD surface, at
least at reasonable low concentrations. In fact, upon titration at molar ratios
x
=
1,
the spectral shifts of the H
2
P absorption and fluorescence spectra (see Fig.
4.8
)as
well as changes of the H
2
P emission decay times (see Fig.
4.5
b, d) are in accordance
with the above considerations.
Since
K
M
<
K
L
, we would expect that upon the presence of a dye molecule at
the QD surface the ligand shell is replaced at least proportional to the effective
volume of the molecule, at least if the QD surface is assumed to be homogenously
covered by ligands. In case of a completely closed ligand shell, for any QD with
d
CdSe
=
3.5 nm, the total average number of TOPO ligands per QD will be close to
120 [
94
]. Taking into account H
2
P and TOPO known volumes (see Fig.
4.7
)[
62
]
we expect that H
2
P will replace about three TOPO molecules, which would result
in a maximum number of about 40 H
2
P on the respective QD surface. However,
as can be seen from Fig.
4.20
a, PL quenching is still increasing up to
x
20,
although the spectral shifts have saturated and the fluorescent decay times stay
constant [
62
,
123
] indicating a local replacement of TOPO by H
2
P as follows from
theoretical considerations [
48
]. Notably, it is known from a comparison of ensemble
and single-molecule experiments on PDI molecules [
74
,
94
] that the number of
attached PDI molecules to a QD is much less than that given by the molar ratio
x
,
≈
