tmpA4E7 TMP
tmpA4E7 TMP
tmpA4E7 TMP
de
FULL PAPER
www.MaterialsViews.com
DOI: 10.1002/adom.201300354
56
wileyonlinelibrary.com
1. Introduction
www.advopticalmat.de
www.MaterialsViews.com
FULL PAPER
wileyonlinelibrary.com
57
www.advopticalmat.de
FULL PAPER
www.MaterialsViews.com
820
2nl cos 2
(1)
Figure 2. Shaping the QD emission profile by FabryProt modes in the AAO templates. a,b) Schematic system showing the photoluminescence measurements of QDs on an AAO template (a), with the collected photoluminescence spectra of the QDs deposited on the AAO templates (dark blue) and
on quartz substrate (red) shown in (b). The film thicknesses, , of the AAO templates are 1.7 m and 5.2 m, respectively. The transmission spectra
of the pristine AAO templates are also included for reference (green). The photoluminescence spectra have been normalized for the purpose of clarity.
c,d) Schematic model used for calculating the radiation power of a dipole source located 10 nm above the AAO template (c), with the calculating results
shown in (d). The dashed blue lines in (c) draws a unit cell used in the simulations. e) Comparison of the photoluminescence decay dynamics of the
QDs on a quartz substrate and on AAO templates of various thicknesses. The collection wavelength is 790 nm. f) Calculated relative total decay rate of
the dipole on the AAO templates with different thicknesses. The dashed line indicates the wavelength where the experimental measurements were taken.
58
wileyonlinelibrary.com
www.advopticalmat.de
www.MaterialsViews.com
(=
2T 2
: D ( r 0 , T)
3g0
(2)
where
T is the transition frequency, D (r 0 , T) is the PLDOS,
and : is the transition dipole moment. In our study, the AAO
template serves equivalently as a photonic cavity with multiple
resonances, manifested by a series of transmission maxima. At
these resonances, the electromagnetic environment and therefore the PLDOS will be modified strongly. As a result, once coupled with the AAO template, the emission profile of the QDs is
modulated according to the transmission maxima that are in
resonance with the radiative electronic transitions of the QDs.
This is clearly manifested by the coincidence between the photoluminescence spectra (Figure 2b) and the calculated PLDOS
FULL PAPER
the spectra above 750 nm were given. In order to have a systematic study, we prepared five templates of various film thicknesses ranging from 1.7 m to 6.9 m with a fixed pore size.
The emission spectra collected from QDs on two typical AAO
templates and on a quartz substrate are presented in Figure 2b,
along with the corresponding transmission spectra of the two
AAO templates. One can see that the emission spectrum of the
QDs on the quartz substrate shows a conventional Lorentzianshaped profile with an emission peak locating at 790 nm and
full width at half maximum (FWHM) of 92 nm. The relatively
broad emission spectrum stems from the inhomogeneous
broadening of the exciton line of the QDs due to their natural
size variation, and is strongly modified by the presence of the
AAO template. As shown in Figure 2b, the relatively broad
emission evolves into various photoluminescence peaks with
much smaller FWHM locating in the vicinity of the transmission maxima of the AAO templates. Most interestingly, with
increasing the template thickness, the number of transmission
maxima of the AAO template and therefore the narrowing of
the photoluminescence peaks of the QDs are increased (Supporting Information Figure S5). Such behaviors clearly indicate that the AAO templates coerce the QD emission into their
transmission profiles, and thus shape the corresponding emission spectra.
The spectral shaping effect was further analyzed via theoretical calculations, where the experimental system was modeled by placing a classical point dipole above the AAO template
(Figure 2c). The emission spectrum of the dipole is characterized by recording its radiation power, which is closely related
to the PLDOS at the dipole's location. Figure 2d shows the calculated emission spectra for the dipole placed on two templates
with film thicknesses of 1.7 m and 5.2 m, respectively, the
spectral profile of which are in excellent agreement with the
measured photoluminescence spectra of the QDs. Such spectral shaping is attributed to the resonant coupling of the QD
emission with the photonic cavity modes supported by the
AAO templates, which can accordingly modify the spontaneous
decay rate of the QDs. The QDs used in our study have a nominal quantum yield of 30%.[54] Due to the negligible absorption
of the AAO templates in the investigated spectrum range, in
the following analysis the change of the spontaneous decay rate
is approximated to be that of the total decay rate of the QDs
themselves. In quantumtheory, the spontaneous decay rate of
an emitter positioned at r 0 is expressed as[55,56]
distribution at the back side of the template (Supporting Information Figure S6), verifying that modification of the PLDOS
by the AAO template is responsible for the emission spectrum
reshaping of the semiconductor QDs.
The total decay rate (including nonradiative and radiative),
, of an emitter is the inverse of its emission lifetime. Thus,
the above observations from time-averaged photoluminescence
measurements can be further corroborated by time-resolved
photoluminescence measurements. Figure 2e presents the
transient photoluminescence measurements of the QD emission on different substrates. All of the fluorescence decay traces
were measured at 790 nm. The decay of the QDs on silica follows a single exponential function with a lifetime of 84.47 ns.
For QDs on the AAO templates, their photoluminescence decay
traces follow a two-component exponential behavior with the
form
IPL (t) = Af e t/tf + As e t/ts
(3)
where Af and As are the weight factors of the fast and slow processes, respectively. tf and ts are the corresponding lifetimes.
The lifetimes of the slow processes extracted from the measured intensity decay curves are the same as that of QDs deposited onto the quartz substrate, while the fast processes exhibit
much shorter lifetimes, as shown in Figure 2e. These findings
indicate that the spontaneous decay of the QDs is accelerated
by the AAO template due to the increase of the PLDOS. This
can be further unveiled by calculating the relative total decay
rate via normalizing the total radiation power of the dipole
source on the templates with that of the same dipole in free
space (Figure 2f). The calculated results show that the relative
total decay rates exhibit apparent oscillation behaviors synchronizing with the calculated emission spectra of the dipole. A
quantitative comparison can be made between the ratio of the
measured lifetimes for QDs on the three AAO templates and
that of the calculated relative total decay rates, where the measured 1/tf,1.7 m:1/tf,5.2 m:1/tf,6.9 m = 1:1.31:1.26 is very close to
the calculated relative total decay rate tot,1.7 m:tot,5.2 m:tot,6.9
m = 1:1.54:1.49 if the first term is normalized to 1 for comparison purpose. Most interestingly, both of the experimental
measurements and theoretical results show that the radiative
acceleration is most pronounced for QDs deposited on the
5.2 m-template. This is because, in such cases, the transmission maximum of the template at 790 nm coincides with the
intrinsic emission peak of the QDs, giving rise to the strongest
coupling between the template and QDs.
It is known that the photoluminescence bands of the QDs
are strongly dependent on their size. We want to point out that
the above emission spectral profile shaping effect is a general
behavior regardless of the size of the QDs we used. The spectral
shaping is due to the resonant coupling of the QDs emission
with the FabryProt interference supported by the AAO templates. The coupling strength depends on the spectral overlap
between the photoluminescence of the QDs and the Fabry
Prot interference of the AAO template. Therefore, once the
emission bands of the QDs lie within the wavelength range
where the FabryProt interference takes place, the spectral
shaping effect will still occur. We have measured the photoluminescence spectra of QDs with two different sizes deposited
wileyonlinelibrary.com
59
www.advopticalmat.de
FULL PAPER
www.MaterialsViews.com
Figure 3. Photoluminescence of QDs deposited onto metal nanowire-loaded AAO templates. a) Schematic system of an AAO template loaded with
metal nanowires. b) Transmission electron microscopy (TEM) image of typical Au nanowires prepared using the AAO templates. c) Extinction spectra
of the pristine AAO template (green) and hybrid structures with a Ag nanowire array (dark blue) and a Au nanowire array (red) inside the pores of
the AAO templates. The film thicknesses of the AAO templates are 5.2 m. d) Photoluminescence spectra of the QDs deposited onto pristine AAO
template (dark blue), AuAAO (green), and AgAAO hybrid templates (red). e) Comparison of the electric field intensity enhancement distribution
(cross-sectional view) between an AAO template loaded with Au nanowires (left panel) and a pristine AAO template (right panel). The calculation was
carried out at an excitation wavelength of 700 nm with an incidence angle of 26 according to the photoluminescence measurement setup. The hybrid
template is modeled as Au nanowire arrays immersed into a uniform matrix with refractive index of 1.61. The length and diameter of the nanowires
were taken as those from the TEM measurements.
60
wileyonlinelibrary.com
www.advopticalmat.de
www.MaterialsViews.com
FULL PAPER
Figure 4. a) Normalized extinction spectra of the hybrid templates with Ag nanowire arrays inside the pores of the AAO templates. The hybrid
plasmon resonances of the templates locate at 740 nm (dark blue), 780 nm (green), and 805 nm (red). The film thicknesses of the AAO templates are
5.2 m. b) Photoluminescence spectra of the QDs deposited onto the three hybrid templates in (a). The spectra collected for QDs deposited onto
quartz substrate (yellow) and pristine AAO template (purple) are also included for comparison.
Enhancement factor at
800 nm
Integrated enhancement
factora)
QDs-Ag-740
2.1
1.7
QDs-Ag-780
4.2
3.9
QDs-Ag-805
4.9
4.0
Pristine AAO
1.3
a)
wileyonlinelibrary.com
61
www.advopticalmat.de
www.MaterialsViews.com
FULL PAPER
3. Conclusion
We have prepared pristine AAO templates with strong Fabry
Prot resonances by a facile and low-cost electrochemical
method. The FabryProt resonances are utilized to shape
the emission spectral profile of the semiconductor QDs. The
underlying mechanism was investigated via transient photoluminescence characterizations and theoretical calculations,
which indicate that tuning of the spectral shape is governed by
the modified PLDOS. The AAO templates have been further
employed to direct the growth of metal nanowire arrays. After
loaded with Au or Ag nanowires, the hybrid templates can further enhance the photoluminescence intensity of the QDs due
to the plasmon-induced electric field intensity enhancements.
We believe that these results can greatly expand the functionalities of the AAO templates and provide a new avenue for
controlling the emission properties of QDs beneficial to the
development of future photonic and optoelectronic applications.
4. Experimental Section
Preparation of the AAO Templates: The AAO templates were prepared
using a two-step anodization process. First, the aluminum sheets
(purity 99.999%) were degreased in acetone and electropolished under
a constant voltage of 16 V for 4 min in a mixture of HClO4 (volume
percentage of 72%) and C2H5OH (volume ratio of 1:3) at 0 C to smooth
the surface morphology. In the first and second anodization processes,
treated aluminum sheets were exposed to a H2SO4 solution (0.3 M) under
constant voltage of 19 V in an electrochemical cell at a temperature of
4 C. The alumina layer produced by the first anodization process was
then removed by wet chemical etching in a mixture of phosphoric acid
(0.15 M) and chromic acid (0.60 M) at 60 C for 1 h. The barrier layer of
AAO templates was thinned by stepwise potential reduced anodization
steps. The final reduced potential was 6 V. The film thicknesses of the
AAO templates were controlled by anodization time. In the final step the
underlying aluminum substrate was dissolved using aqueous CuCl2
(1.5 M) solution.
Growth of the Metal Nanowire Arrays Using AAO Templates: The
metal nanowire arrays were grown in the pores of the AAO templates
by alternative current electrodeposition in an electrolyte. For growth of
the Ag nanowire array the reactants were composed of AgNO3 (0.03 M)
and H2SO4 (0.03 M), while for the Au nanowire array, HAuCl4 (0.01 M)
and H2SO4 (0.1 M) were used. During the electrodeposition a Pt foil
was utilized as counter electrode. After the metal nanowire arrays with
desired lengths were obtained the underlying aluminum substrate was
dissolved using aqueous CuCl2 solution (1.5 M).
Instrumentation: The extinction spectra were measured on a
PerkinElmer Lamda 950 UV-visible-NIR spectrophotometer. TEM
imaging was performed on a JEOL 2010HT microscope at 100 kV. For
preparation of the TEM samples, the AAO matrixes were removed by
dissolving the templates containing metal nanowires in aqueous NaOH
solution (1.25 M). SEM imaging was conducted on a Zeiss Auriga-3934
microscope operating at 20.0 kV.
62
wileyonlinelibrary.com
D (r , T) =
2T
Im Tr G (r , r , T)
Bc 2
(4)
where Im stands for the imaginary part and Tr denotes the trace
of the Green tensor matrix in brackets. The Green tensor can be
obtained as
E r , T = iT: 0 G r , r , T j T
(5)
www.advopticalmat.de
www.MaterialsViews.com
Supporting Information
Supporting Information including SEM images of the AAO templates
with various film thicknesses and the corresponding transmission
spectra, SEM image of QDs dispersed on Si substrate and the backside
of the AAO template, photoluminescence spectra of the QDs deposited
on the AAO templates with various thicknesses, spectral shaping effects
of the QDs with different sizes, and calculated PLDOS at the backside
of the AAO templates is available from the Wiley Online Library or from
the author.
Acknowledgements
Z.-K. Zhou, D. Y. Lei, and J. M. Liu contributed equally to this work.
This work was financially supported by the National Basic Research
Program of China (Grant No. 2010CB923200), the National Natural
Science Foundation of China (Grant Nos. U0934002, 20828001,
11204385, 51202300), the Ministry of Education of China (Grant No.
V200801), the Doctoral Fund of Ministry of Education of China (Grant
No. 20120171120012), the Natural Science Foundation of Guangdong
Province (Grant No. S2012040007814), the Fundamental Research Funds
for the Central Universities (Grant No. 12lgpy45), and Fund of Education
Department of Guangdong Province (Grant No. 2012LYM_0011). D. Y. L.
and X. L. acknowledge the grants 1-ZVAW and 1-ZVAL administrated by
the Hong Kong Polytechnic University. Zhang-Kai Zhou, Dang Yuan Lei,
and Jiaming Liu contributed equally to this work.
Received: August 20, 2013
Revised: October 13, 2013
Published online: November 13, 2013
FULL PAPER
index was 1.61. The diameter of the air hole was 22 nm while the lattice
constant was 45 nm. An electrical point dipole was set 10 nm above the
center of the arrays. A block with the size of 0.7 0.6 5.4 m3 was
employed to separate the array and the PMLs. The PMLs were set as
1.4 1.3 6.4 m3 with general type. To get a good mesh, the dipole
was surrounded with a 4-nm-diameter sphere. Furthermore, a point
and another sphere with radius of 4 nm was set 10 nm below the upper
surface of the template at the center of the array. Fined meshes were
used inside the block while extra-fine meshes were defined for the PMLs
to get good absorption. The scattering boundary was placed outside of
the PMLs.
wileyonlinelibrary.com
63
www.advopticalmat.de
FULL PAPER
www.MaterialsViews.com
[41] N. Livneh, A. Strauss, I. Schwarz, I. Rosenberg, A. Zimran,
S. Yochelis, G. Chen, U. Banin, Y. Paltiel, R. Rapaport, Nano Lett.
2011, 11, 1630.
[42] Y. C. Jun, R. Pala, M. L. Brongersma, J. Phys. Chem. C 2010, 114, 7269.
[43] A. G. Gurto, G. Volpe, T. H. Taminiau, M. P. Kreuzer, R. Quidant,
N. F. van Hulst, Science 2010, 329, 930.
[44] M. Ringler, A. Schwemer, M. Wunderlich, A. Nichtl, K. Krzinger,
T. A. Klar, J. Feldmann, Phys. Rev. Lett. 2008, 100, 203002.
[45] G. Vecchi, V. Giannini, J. Gmez Rivas, Phys. Rev. Lett. 2009, 102,
146807.
[46] L. Zhao, T. Ming, H. J. Chen, Y. Liang, J. F. Wang, Nanoscale 2011, 3,
3849.
[47] T. Ming, L. Zhao, Z. Yang, H. J. Chen, L. D. Sun, J. F. Wang,
C. H. Yan, Nano Lett. 2009, 9, 3896.
[48] E. C. Le Ru, P. G. Etchegoin, J. Grand, N. Flidj, J. Aubard, G. Lvi, J.
Phys. Chem. C 2007, 111, 16076.
[49] K. Tanaka, E. Plum, J. Y. Ou, T. Uchino, N. I. Zheludev, Phys. Rev.
Lett. 2010, 105, 227403.
[50] J.-B. Li, Y. Yu, X.-N. Peng, Z.-J. Yang, Z.-K. Zhou, L. Zhou, J. Appl.
Phys. 2012, 111, 123110.
[51] Z.-K. Zhou, X.-N. Peng, Z.-J. Yang, Z.-S. Zhang, M. Li, X.-R. Su,
Q. Zhang, X. Y. Shan, Q.-Q. Wang, Z. Y. Zhang, Nano Lett. 2011,
11, 49.
64
wileyonlinelibrary.com