530
M. Vacha and S. Hirata
molecules (at 532 nm) is shown in Fig. 30.6d. The phosphorescence is emitted
uniformly throughout the film. On top of this uniform background, phosphorescence
with higher intensity is emitted close to the AgNW. The emission enhancement along
the nanowires is on the order of 1.1. In addition to the long AgNW structures, we
observe that the phosphorescence is enhanced much further (by a factor of 4–5) in
locations where two or more AgNW overlap. Such nanowire crossings (junctions)
form plasmonic hotspots with highly confined and enhanced local electric fields
which cause the strong enhancement of phosphorescence. Microscopic image of the
same sample area taken with the same excitation at 532 nm and detected in the region
of UC emission (below 492 nm) is shown in Fig. 30.6e. The UC and phosphorescence images are very similar. The enhancement of UC intensity on the AgNWs is
on the order of 1.3. In the junctions of the AgNWs, the enhancement increased up
to a factor of 15. In addition, there is an interesting difference between the UC and
phosphorescence patterns near the AgNW junctions. In the UC image, the enhanced
emission is clearly elongated along one of the nanowires extending from the junction.
This difference is evaluated by plotting 1D cross sections of the emission intensity.
In the cross sections along the direction of the AgNW (Fig. 30.6f), the UC (black
symbols) is significantly broader than the phosphorescence (red symbols). On the
other hand, there is no difference between the UC and phosphorescence in the cross
sections taken perpendicular to the AgNWs.
To evaluate and compare the extent of the spatial elongation (dispersion) of phosphorescence and UC emission, we fitted the experimental intensity cross sections
with a convolution of a Gaussian function with an exponential function. The Gaussian function represents a point spread function of the microscope, whereas the
exponential function describes diffusion of the triplet excitons as a possible cause
of the spatial dispersion. An example of such fitting is shown in Fig. 30.5f as solid
lines for both phosphorescence (red) and UC emission (black). The fits provide a
characteristic parameter of dispersion length. The dispersion lengths were evaluated
statistically by fitting the intensity profiles on several tens of AgNW junctions. The
average dispersion was 0.5 μm for phosphorescence and 1.2 μm for UC emission.
Thus, the UC emission is enhanced by the localized plasmon in AgNW junctions
in qualitatively different way than phosphorescence in the same locations, both in
terms of intensity enhancement and spatial dispersion.
To summarize the experimental observations, triplet states of the donor PtOEP are
generated by the 532 nm excitation within the whole sample and lead to phosphorescence of the donor and UC emission from acceptor. The junctions of individual
AgNW work as plasmonic hotspots and further enhance the generation of triplet
states of the donor, as evidenced by the increased intensity of phosphorescence from
these location (with enhancement factor of 4–5). The locations of the junctions have
thus higher density of donor triplets compared to the rest of the sample and work as
localized triplet sources. Here, we assume that the hotspots in the AgNW not only
generate the higher density of the donor triplets but also enhance the donor–acceptor
triplet energy transfer. Further, the plasmons also enhance the acceptor–acceptor
triplet–triplet energy transfer (triplet exciton diffusion) within the hotspots and along
the AgNW. The result of the enhancement of both types of energy transfer within the
M. Vacha and S. Hirata
molecules (at 532 nm) is shown in Fig. 30.6d. The phosphorescence is emitted
uniformly throughout the film. On top of this uniform background, phosphorescence
with higher intensity is emitted close to the AgNW. The emission enhancement along
the nanowires is on the order of 1.1. In addition to the long AgNW structures, we
observe that the phosphorescence is enhanced much further (by a factor of 4–5) in
locations where two or more AgNW overlap. Such nanowire crossings (junctions)
form plasmonic hotspots with highly confined and enhanced local electric fields
which cause the strong enhancement of phosphorescence. Microscopic image of the
same sample area taken with the same excitation at 532 nm and detected in the region
of UC emission (below 492 nm) is shown in Fig. 30.6e. The UC and phosphorescence images are very similar. The enhancement of UC intensity on the AgNWs is
on the order of 1.3. In the junctions of the AgNWs, the enhancement increased up
to a factor of 15. In addition, there is an interesting difference between the UC and
phosphorescence patterns near the AgNW junctions. In the UC image, the enhanced
emission is clearly elongated along one of the nanowires extending from the junction.
This difference is evaluated by plotting 1D cross sections of the emission intensity.
In the cross sections along the direction of the AgNW (Fig. 30.6f), the UC (black
symbols) is significantly broader than the phosphorescence (red symbols). On the
other hand, there is no difference between the UC and phosphorescence in the cross
sections taken perpendicular to the AgNWs.
To evaluate and compare the extent of the spatial elongation (dispersion) of phosphorescence and UC emission, we fitted the experimental intensity cross sections
with a convolution of a Gaussian function with an exponential function. The Gaussian function represents a point spread function of the microscope, whereas the
exponential function describes diffusion of the triplet excitons as a possible cause
of the spatial dispersion. An example of such fitting is shown in Fig. 30.5f as solid
lines for both phosphorescence (red) and UC emission (black). The fits provide a
characteristic parameter of dispersion length. The dispersion lengths were evaluated
statistically by fitting the intensity profiles on several tens of AgNW junctions. The
average dispersion was 0.5 μm for phosphorescence and 1.2 μm for UC emission.
Thus, the UC emission is enhanced by the localized plasmon in AgNW junctions
in qualitatively different way than phosphorescence in the same locations, both in
terms of intensity enhancement and spatial dispersion.
To summarize the experimental observations, triplet states of the donor PtOEP are
generated by the 532 nm excitation within the whole sample and lead to phosphorescence of the donor and UC emission from acceptor. The junctions of individual
AgNW work as plasmonic hotspots and further enhance the generation of triplet
states of the donor, as evidenced by the increased intensity of phosphorescence from
these location (with enhancement factor of 4–5). The locations of the junctions have
thus higher density of donor triplets compared to the rest of the sample and work as
localized triplet sources. Here, we assume that the hotspots in the AgNW not only
generate the higher density of the donor triplets but also enhance the donor–acceptor
triplet energy transfer. Further, the plasmons also enhance the acceptor–acceptor
triplet–triplet energy transfer (triplet exciton diffusion) within the hotspots and along
the AgNW. The result of the enhancement of both types of energy transfer within the