Surface Plasmon Effects On Organic Solar Cells

Review
Journal of
Nanoscience and Nanotechnology
Copyright 2014 American Scientic Publishers

All rights reserved
Printed in the United States of America
Vol. 14, 10991119, 2014

www.aspbs.com/jnn
Surface Plasmonic Effects on Organic Solar Cells

Ashraf Uddin and Xiaohan Yang
School of Photovoltaic and Renewable Energy Engineering,
The University of New South Wales, Sydney 2052, Australia
Most high-performance organic photovoltaic (OPV) devices reported in the literature have been fabricated using the bulk heterojunction (BHJ) concept. Typically, the optimum thickness of the active
layer for an OPV device is around 100 nm, or possibly less; such a thin layer can lead to low absorption of light. A thicker layer, however, inevitably increases the device resistance, due to the low carrier
mobilities and short exciton diffusion lengths in organic materials. This situation imposes a trade-off
between light absorption and charge transport efciencies in OPV devices, motivating the development of a variety of light-trapping techniques. Metallic nanoparticles (NPs) such as Ag, Au, etc.
and other metallic nanostructures are potential candidates for improving the light absorption due to
the localized surface plasmon resonance (LSPR). LSPR contributes to the signicant enhancement
of local electromagnetic elds and improves the optical properties of the nanostructure devices.
The excitation of LSPR is achieved when the frequency of the incident light matches its resonance
peak, resulting in unique optical properties; selective light extinction as well as local enhancement
of electromagnetic elds near the surface of metallic NPs. The resonance peak of LSPR depends
Delivered by Publishing Technology to: University of New South Wales
strongly on the size, shape, and the dielectric environment of the metallic NPs. In this review article,
IP: 149.171.67.164 On: Mon, 27 Jan 2014 22:47:18
progress on plasmonic enhanced OPV device performance is examined. The concepts of surface
Copyright: American Scientific Publishers
plasmonics for OPV devices, suitable plasmonic materials, location, optimum size and concentration
of NP materials within the device are explored.
Keywords: Organic Solar Cells, Surface Plasmonic Materials, Localized Surface Plasmon
Resonance.
CONTENTS
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Concept of Plasmonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1. Localized Surface Plasmons . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Scattering Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3. Effect of Nanoparticle Size . . . . . . . . . . . . . . . . . . . . . . . .
2.4. Effect of Material Properties . . . . . . . . . . . . . . . . . . . . . . .
2.5. Effect of Dielectric Medium . . . . . . . . . . . . . . . . . . . . . . .
3. Chemicals and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Device Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1. Device Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Location of Nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1. Particles on the Front . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2. Particles on the Rear . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3. Particles Embedded in the Active Layer . . . . . . . . . . . . . .
5.4. Near-Field Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Nanoparticle Size Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7. Nanoparticles Concentration Effects . . . . . . . . . . . . . . . . . . . . .
8. Conclusion and Outlooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References and Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Author to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2014, Vol. 14, No. 2
1. INTRODUCTION
1099
1101
1102
1102
1103
1103
1103
1104
1106
1107
1108
1108
1108
1109
1109
1111
1111
1114
1115
1117
1117
Organic semiconductors are currently being considered

for a wide range of commercial applications based
on organic light emitting diode (OLED), organic eldeffect transistor (OFET) and organic photovoltaic (OPV)
technologies. Organic photovoltaic technologies afford the
potential for low-cost photovoltaic via roll-to-roll processing, with recently demonstrated cell efciencies of
over 8%.1 Organic solar cells are lightweight and exible, and can be fabricated on glass and plastic substrates.2
The potential simplicity of OPV processing is far beyond
other current technologies.3 Furthermore, organic semiconductors have high absorption coefcients, which allow
very thin lms to be used.4 The reduced manufacturing
cost, coupled with reduction in material used, implies that
organic solar cells have the potential to make a signicant impact on the PV market. The highest reported power
conversion efciency (PCE) for an OPV device is 12%,
which is approaching what is deemed to be required for
commercialization.5 Considering the advantages and signicantly increased efciencies of OPV, it has resulted in
1533-4880/2014/14/1099/021
doi:10.1166/jnn.2014.9017
1099
Uddin and Yang
commercial production of primary OPV products and a

solar cells to extract the full potential of the cells. Plasfew public demonstration projects.6
monic is one such technique that involves trapping light
At present, OPV devices with the highest perforinside the cell by exciting surface plasmons on metalmance are based on the bulk heterojunction (BHJ) type
lic nanostructures.1115 Plasmonic materials are advantain which an electron-donor polymer material and an
geous due to the large electric eld enhancement near the
electron-acceptor fullerene material are blended to form
metal surface and strong dependence of the resonance frean active photovoltaic layer. The most commonly studquency on size, shape, material, dielectric environment and
ied polymer-fullerene system is based upon a blend of
location, all design parameters of metal nanoparticles.16 17
donor poly(3-hexylthiophene-2,5-diyl) (P3HT) and accepAdditionally, the optical properties of plasmonic materitor [6,6]-phenyl-C61 butyric acid methyl ester (PCBM),
als are electrically decoupled from the cell; hence this
in which the highest reported PCE values are in the range
light trapping technique does not affect the cell properties
of 4 to 6%.7 8 The theoretical PCE limits of this BHJ are
or require any major modication to device fabrication.16
9
over 23%. The difference between the theory and experiFor organic solar cells, the most commonly used plasmental PCE can be explained by ve reasons: namely the
monic metals are Ag and Au in spherical shape.1820 It has
energy level misalignment at the donor/acceptor interfaces,
been proposed that Al could be a good candidate since it
insufcient light trapping, low exciton diffusion lengths,
supports a good overlap between the plasmon resonance
nonradiative recombination and low charge carrier mobiland the natural absorption bands of OPV materials.21
ities. The OPV device structure needs a thin active layer
Recently, several studies have investigated the inuence of
due to the low exciton diffusion length ( 10 nm) for the
gold nanoparticle (Au NPs) concentration on plasmonic
high efciency device,10 but the thin active layer is insufenhancement of PCE for P3HT:PC60 BM solar cells.19 22
cient to absorb most of the incident light.
A maximum PCE enhancement of 23% was observed for
The light trapping structure is one of the potential ways
35 nm size Au NPs.22
to improve the PCE of BHJ by increasing the effective
Embedding the metal nanostructures near the active
optical path length within the active layer without the need
layer leads to enhanced electric eld near junction which
of altering its physical thickness. Due to the small physresults in more excitons being created within an exciton
ical dimensions of the photoactive layer, the ability to
diffusion length. One of the earliest reports for enhanceabsorb a large fraction
of the by
solar
spectrum is
a crucial to:
ment
of PCE of
of New
organic
solarWales
cells was using Au, Ag
Delivered
Publishing
Technology
University
South
parameter in determining the IP:
efciency
of
OPV
devices.
and
Cu
metal
clusters
at
the
ITO
copper phthalocya149.171.67.164 On: Mon, 27 Jan 2014 22:47:18
23
This highlights the need to incorporate
good lightAmerican
trapping Scientific
nine interface.
The purpose of this work was to use
Copyright:
Publishers
mechanisms that increase optical absorption in organic
the electric eld enhancement to increase the number
Ashraf Uddin obtained his Ph.D. degree in March 1991 in Semiconductor Physics from
Osaka University, Osaka, Japan. After his Ph.D., he worked at the R&D centre of Toshiba
Corporation, Japan as semiconductor materials and devices research scientist (19911997).
There he worked on opto-electronic devices for about three years and on poly-Si thin lm
transistor to develop a process technology on glass substrate for the fabrication of at panel
display. In 1997, he migrated to Australia and worked on IIIV semiconductor laser diodes
in ANU, Canberra. He worked at the School of Materials Science and Engineering, NTU,
Singapore as an Assistant Professor from 1 July 2001 to 30 June 2007. He then worked
in the Department of Electrical and Computer Engineering, King Abdulaziz University,
Jeddah, Saudi Arabia as an Associate Professor from August 2007 to December 2008.
Currently, he is an Associate Professor in the School of Photovoltaic and Renewable
Energy Engineering, UNSW as. He has published over 100 research papers in different international journals and in
conference proceedings. He has 8 Japanese and one US patents.
Xiaohan Yang received his B.S. in 2008 from Jiangnan University (China), M.Eng.
(2009), and Ph.D. (2013) in Photovoltaic Engineering from the University of New South
Wales. He also worked as a Research Assistant (20102013) and University Tutor (2010
2013) over the course of his Ph.D. study. His research interests include organic solar cells
and plasmonics for photovoltaic applications.
1100
J. Nanosci. Nanotechnol. 14, 10991119, 2014
Uddin and Yang
of photogenerated charge carriers which was successhave recently emerged. It then draws attention to important
areas of research for the future.
fully achieved with a maximum enhancement of a factor of 2.7 by using Cu clusters. After that, many works
related to plasmonic in organic solar cells have been
2. CONCEPT OF PLASMONICS
done in these years. In 2011, Wu et al. blended Au
Metal nanoparticles have optical properties which are difNPs into the PEDOT:PSS buffer layer to investigate the
ferent from bulk metals. Surface plasmons (SP) are the
24
effect of plasmonic in P3HT:PC60 BM devices. Lu et al.
coherent oscillations of conductive electrons in thin metal
also improved PCEs of organic solar cells by incorporatlms. Surface plasmons occur at the interfaces between
ing metal nanoparticles into the PEDOT:PSS layer. They
metals and dielectrics, where the real part of the dielecfound that the localized surface plasmon resonance (LSPR)
tric function (Re() of these layers have opposite signs.
induced light absorption enhancement region would be
This requirement is satised in the region from infrared
much boarder after combining Ag and Au NPs into the
to visible for metals, where metals have negative Re(.
25
PEDOT:PSS buffer layer. In 2012, Li et al. demonRe < 0 occurs where a material is strongly absorbing,
strated dual metallic nanostructures in single organic solar
this strong absorption is responsible for the high reectivity
cell by simultaneously incorporating Au NPs into the
of metals. Surface plasmons oscillations are accompanied
active layer and fabricating the Ag nanograting elecby an electromagnetic eld which has its maximum value at
26
trode by nanoimprinting method. A broadband absorpthe surface z = 0, and ceases at z = .28 This eld perpention enhancement and an improved PCE were achieved
dicular to the surface is said to be evanescent or near-eld
through this combination.
in nature, and decays exponentially with distance from the
Kim et al. deposited 13 nm, uniform-sized Ag nanoparinterface. A schematic representation of the surface plasticles below the photoactive layer of a P3HT:PC60 BM
mons is shown in Figure 1. Its properties depend on the
20
solar cell. The addition of this Ag layer lead to signilm thickness and the dielectric medium on both sides of
cantly improved efciency, due largely to increased short
the lm.
circuit current Jsc . External quantum efciency (EQE)
Surface plasmons polaritons (SPPs, propagating plasmeasurements showed that the addition of the plasmonic
mons)
have a larger momentum than the incident photons
layer enhanced photocurrent generation across the entire
with
a
given
frequency, which prevents power propagating
spectrum in which P3HT:PC60 BM solar cell is active.
Delivered by Publishing Technology to:
University
New South
Wales
away
from the of
surface,
or in other
words, the propagaYoon et al. investigated plasmon
enhanced
absorption
by
IP: 149.171.67.164 On: Mon,
27isJan
2014 22:47:18
tion
non-radiative.
Thus,
there
is
a
momentum
mismatch
depositing a self-assembled Ag nanoparticle
Copyright:monolayer
American Scientific
Publishers
between
the
photons
and
surface
plasmons,
hence
to gen27
from a colloidal solution. This layer had uniform partierate
SPPs
using
light,
the
momentum
mismatch
between
cles, with 4 nm diameter, and controllable particle spaclight and SPPs must be equated i.e., both the frequency
ing. This layer also improved the Jsc of P3HT:PC60 BM
and wave vector of the incident light must be xed. The
organic solar cells, which was attributed to enhanced optifrequency-dependent SP wave vector (also called the surcal density. Most investigations focus on P3HT as the
face plasmon dispersion relation) is given by:28
donor material, however, the spectral response of this poly
mer is severely limited due to its large band gap.
m d
k
=
k
(1)
spp
0
The plasmon enhanced performance of OPV devices for
m + d
different organic and plasmonic materials are examined
where the free space wave vector k0 = /c, d is the dielecwithin this review. The key purpose of this article is to
tric function of the adjacent medium and m is the dielectric
introduce the reader to the concept of surface plasmonic
function of the metal; where C is the speed of light.
for OPV devices and to explore the plasmonic materials which could produce high performance organic solar
Z
cells. The rst section introduces the reader to the concept of plasmonic for photovoltaic application is explored.
E
The main focus of this review is a detailed investigation of the plasmonic materials used for OPV devices.
Dielectric d > X
In Section 3, the organic materials, details the device fab+ + + + + + + + ++ + + + + + + + + Metal <
m
rication and operating characteristics. The characterisation
Surface
of OPV devices are explained in Section 4. The nal
charges
section looks into important results of achievement which
may help elucidate a deeper understanding of how plasFigure 1. Schematic representation of a propagating plasmon at the
mon enhanced the OPV performance, which is required
interface of a metal and dielectric with an electro-magnetic eld at its
for efcient device optimisation. The conclusion provides
maximum along the surface (x-axis, z = 0) and decaying exponentially
a future outlook for this technology. It also describes curalong x-axis. Reproduced with permission from [29], S. Pillai and M. A.
rent understanding of commercial aspects for this technolGreen, Plasmonics for photovoltaics, Comprehensive Renewable Energy,
edited by Ali, Elsevier, Oxford (2012), pp. 641656. 2012, Elsevier.
ogy, by highlighting initial existing investigations, which
1101
Uddin and Yang
2.1. Localized Surface Plasmons

Therefore the resonance occurs at sp = p / 3 for a
metal sphere in vacuum where d = 1. The LSPR freIn contrast to SPPs, localized surface plasmons (LSPs) are
quency is dependent on the size, shape, material properoscillations of the electrons in conned geometries smaller
ties and the effect of the dielectric medium around the
than the wavelength of light, such as small metal partinanoparticles20 and gives metal nanoparticles its strong and
cles or voids in metallic structure. Movement of the conwell
dened colour. They determine the position and width
ductive electrons upon excitation with incident light leads
of the plasmon resonance. Due to the connement of the
to a build-up of polarization charges on the particle surSP to the metal nanoparticle, excitation of surface plasface as shown in Figure 2. This acts as a restoring force,
mons can result in selective photon absorption, scattering
allowing a resonance to occur at a particular frequency,
and a large enhancement of the local electric eld in the
which is described as the localized surface plasmon resclose vicinity of the metal nanoparticles. Hence, varying
onance (LSPR) frequency sp to distinguish it from bulk
these parameters offers the tunable resonance position to
resonances and resonance on metal surfaces. The LSPR is
engineer plasmonic structures to target weakly absorbing
predominantly dipolar in nature for small particles where
regimes of various types of solar cells. Therefore, the optir (where r is the radius of the particle and is the
cal properties of metal nanoparticles are clearly different
wavelength of the light). Furthermore, it has been shown
from SPPs. We will discuss these properties in more detail
that for silver nanoparticles (Ag NPs) and light with visiin the following sections.
ble wavelengths, higher order multipoles can be neglected
even for particle diameters as high as 100 nm.30
2.2. Scattering Mechanism
The surface roughness provided by small particles
The advantages of strong optical scattering associated
(r ) or voids which generates SPs locally can provide
with surface plasmon excitation, and local electric eld
the additional momentum for the coupling of incident light
enhancement, are utilized for light trapping and therefore
to the SP. The LSPs can therefore be excited by light with
increasing absorption in OPV cells. However for OPV
an appropriate frequency corresponding to the excitation
devices, the improved performance is mainly attributed to
31
light wave vector. In this case, they can also effectively
the scattering process of the plasmonic structures as the
decay with light emission/scattering. The LSPs with these
near eld is limited to only a few tens of nanometers.34
properties are currently being exploited in light-emitting
metal nanoparticles
are Wales
considered to be dipoles,
32 33
Delivered
by Publishing
Technology to: The
University
of New South
diode (LED) and PV
applications.
accepting
incoming
radiation
and
scattering it at large
IP:due
149.171.67.164
27 Jan 2014 22:47:18
The polarization of the sphere
to the presence On:
of anMon,
angles. InPublishers
the process, a fraction of the light is lost
Copyright: American Scientific
external eld can be calculated applying boundary condiinto the air and the rest is coupled into the photoactive
tions given by:
layer. The high refractive index of the absorbing layer
= 4 0 R3
m d
m + 2d
(2)
where m is the relative permittivity of the metal particle.

The real part Rem describes the polarizability whilst the
imaginary part Imm gives the energy dissipation in the
metal. From Eqs. (1)(2), very strong interaction occurs
with the incident eld when m = 2d . This occurs when
Rem = 2d and Imm = 0. The corresponding surface plasmon resonance frequency is:
Light
sp =
p
Electrical Field
(3)
1 + 2d
+
+
Figure 2. Schematic diagram of a localized surface plasmon with horizontal electric dipole excited by a normally incident light. Reproduced
with permission from [29], S. Pillai and M. A. Green, Plasmonics for
photovoltaics, Comprehensive Renewable Energy, edited by Ali, Elsevier,
Oxford (2012), pp. 641656. 2012, Elsevier.
1102
(for P3HT:PC60 BM (1:1)) the refractive index is 1.87

(500 nm) and 2.11 (600 nm)35 leads to a high density of available photonic modes. Hence, most of the scattered light is coupled into photoactive layer, i.e., fact fair ,
where fact is the fraction of light scattered into a photoactive layer and fair is the fraction of light scattered into
air. The scattering process is maximized at the resonance
position. The large angular distribution of the scattering
process ensures that most of the incident light is outside
the escape cone, facilitating total internal reection and
thereby is totally trapped and absorbed. The coupling is
most efcient when the dipole is as close to the photoactive layer as possible. A schematic representation of the
scattering process is shown in Figure 3. For particles on
the rear, the principle is the same, but the mode of excitation of the plasmons is different, leading to an enhanced
scattering cross-section.
For efcient light trapping using plasmons, Qscat (normalized scattering cross-section) and fact must be maximized, whilst Qabs (normalized absorption cross-section)
should be minimized. The size, shape, material, dielectric
environment and distance of the particles from the photoactive layer, all play important roles in the scattering
properties of the nanoparticles. Each of these design considerations will be discussed in the following section.
Uddin and Yang
Incident light
Spacer layer
Total internal
Reflection
Escape cone
Figure 3. Schematic diagram of the scattering process for particles on

the front of active layer in bulk heterojunction OPV device.
5 and 15 nm gold nanospheres (Au NSs) and doped them

into a PEDOT:PSS solution. The extinction spectra of Au
NSs with two sizes have similar peak positions, which
indicate that the resonance wavelengths of 5 and 15 nm
Au NSs are similar. However, the increase of PCE was
more pronounced for devices doped with 15 nm Au NSs.
The current density enhancement is more noticeable for
the device doped with 15 nm Au NSs. The additional photocurrent generated within 15 nm Au NSs device, which
is due to higher scattering efciency, compared to 5 nm
Au NSs. Similar results have been demonstrated by other
groups.26 39
2.4. Effect of Material Properties

The nanoparticle material also affects the plasmonic resonance. For organic solar cells, the most commonly used
plasmonic metals are silver, gold and copper. For gold and
copper, the absorption spectra lies in the visible region,
whereas for silver the corresponding transition is in the
ultraviolet (UV) region at 310 nm, leading to a highly
2
dispersive dielectric function at optical frequencies on
Im

(4)
Cabs =
excitation with light.40 This means that for a particle size

4
corresponding to the same mass thickness of gold and sil1 2
Cscat =
2
(5)
ver deposited, gold nanoparticles have a lower resonance
6
than silver nanoparticles, as seen in Figure 4, showing the
where is the polarizability of the particle, given by:
optical radiation efciencies with particle-size dependence
Delivered by
Publishing
Technology
to:
New
Walesradiation efciency
for University
Ag, Al, Auofand
Cu.South
The optical
1
(6) Mon,
= 3V IP: 149.171.67.164 On:
27 Jan
2014
22:47:18
(ratio
of
the
scattering
to
the
extinction)
is an important
+2
parameterPublishers
for plasmonic metals as it forms the basis for
where V is the volume of the particle and is the permitthe scattering effect. Recently it has been proposed that Al
tivity of the metal. The normalized scattering cross-section
could be a good candidate for organic solar cells since it
Qscat is given by Qscat = Cscat / r 2 where r 2 is the geosupports a good overlap between the plasmon resonance
metric cross section of the particle.
and the natural absorption bands of the OPV materials.21
Metal nanoparticles can exhibit both scattering and
Associated with metals are the parasitic losses which
absorption. The magnitude of each phenomenon is depenneed to be minimized, as scattering and absorption coexdent on the nanoparticle size. For smaller nanoparticles
ist. In the range of different plasmonic metals, it has been
(50 nm and less) absorption is the more dominant phefound that Ag is the metal with the lowest absorption
nomenon whereas for larger nanoparticles (> 50 nm)
and highest overall radiation efciency, which makes it
scattering is a more dominant process. As size increases,
the most popular choice for plasmonic applications. Alloythe scattering process increases signicantly due to its
ing the two metals in different ratios can yield resonances
6
dependence on r (where r is the radius of the nanoparin between the resonances of the pure gold and silver
3
ticle, as V r , which can be derived by substitutparticles.41 The resonance can also be tuned using nanoing Eq. (6) into Eq. (5). As the scattering cross-section
shells that have a dielectric as the core and a metal as the
depends strongly on the particle size, it is important to
shell.42 This reduces potential parasitic absorption in the
optimize this parameter. Increasing particle size enhances
metal.
the scattering process, which is advantageous for light
2.3. Effect of Nanoparticle Size
The extinction of the particles is dened as the sum of the
scattering and absorption and is a measure of the interaction of the incident radiation with the metal nanoparticles.
For small size particles, the absorption (Cabs ) and scattering (Cscat ) cross-sections are given by Bohren:36
trapping. However, beyond certain limits, increasing particle size results in increased retardation effects (due to
inhomogeneous polarization) and higher order multipole
excitation modes, which can decrease the efciency of the
scattering process.37
The size of the particles affects only the scattering efciency of the nanoparticles: the larger the size, the greater
the scattering and not the resonance position. This is consistent with the electrostatic theory and has been demonstrated by the work of Qiao et al.38 Qiao et al. synthesized
2.5. Effect of Dielectric Medium

Surface plasmons are sensitive to the refractive index of
the dielectric medium. The dielectric layer above and
below the particles also strongly inuences the coupling
of the radiation modes. This has important implications
in photovoltaic applications due to the different types of
dielectrics used such as passivation layers, hole conducting layers, etc. The presence of a substrate can cause
a red-shift of the plasma resonance wavelength of the
1103
Uddin and Yang
Figure 4. The optical radiation efciencies with particle-size dependence for Ag, Al, Au and Cu at wavelengths of (a) 400 nm and (b) 800 nm.
Reproduced with permission from [40], V. Dhumale, et al., Bull. Mater. Sci. 35, 143 (2012). 2012, Springer.
metal particles, and the peak can be further red-shifted by

completely surrounding or embedding the particle with a
dielectric medium.43 The electrons in this case sense the
presence of boundaries and modify their collective oscillation accordingly by depolarizing interaction with the adjacent medium, thereby red-shifting the resonance. Adding
metal nanoparticles to a dielectric medium can increase
its effective refractive index. Effective medium theories
can describe this effect if particles are small. Aspens has
given a general expression for the dielectric constant of
a mixed medium:44Delivered by Publishing Technology to: University of New South Wales
On: Mon, 27 Jan 2014 22:47:18
h
IP:
h 149.171.67.164
h
Figure 5. Publishers
Normalized scattering cross-section spectra for a 100 nm Ag
= fa a
+ Copyright:
fb b
(7) Scientific
American
+ ph
a + ph
b + ph
particle embedded in three different dielectrics (air, Si3 N4 and Si). Dipole
where h is the dielectric constant of the host medium,
a is that of one component of volume fraction fa while
b and fb are those for component b (Note: fa + fb = 1),
p is a polarization factor that can range from 0 to but
equals 1 for cylinders and 2 for spheres.
Maxwell-Garnett theory holds for dilute concentrations
say of component b, so h = a making the rst term on
the right zero and:
= a
a p 2fb + b 1 + pfb
a p + fb + b 1 fb
(8)
Since = n + ik2 where n and k are the real and

imaginary parts of the refractive index, it is easy to convert
between permittivities and indices if complex numbers can
be manipulated.
The dielectric layer or spacer layer is critical for solar
cells as it serves as a conducting or passivation layer in
both the front and rear of the cell. However, it has been
observed that the scattering cross section will also be modied by changing the thickness of the dielectric layer, due
to changes in the electric eld driving the resonance.45
The driving eld increases with spacer layer thickness for
particles located on the front, illuminated surface of the
cell whereas for particles on the rear there is a strong
asymmetry.
1104
(D) and quadrupole (Q) modes are indicated. Reproduced with permission from [17], H. A. Atwater, et al., Nat. Mater. 9, 205 (2010). 2010,
Nature Publishing Group.
Figure 5 shows the observed red-shift of the normalized

scattering cross-section Qscat (normalized to the geometric cross-section) for a 100 nm diameter silver nanoparticle totally embedded in dielectric mediums in increasing
order of refractive index 1 (air), 2 (Si3 N4 and 3.5 (Si).
The dipole wavelength can be shifted from around 400 nm
in air to beyond 1200 nm using Si, and this is an effective method of achieving spectral tunability using surface
plasmons. This effect has also been observed for organic
solar cells by adding PEDOT:PSS on top of Ag NPs.46
Kalfagiannis et al.46 investigated the optical properties of
Ag NPs by spectroscopic ellipsometry and found that SPR
peak of Ag NPs shifted from 430 nm to 477 nm after
PEDOT:PSS coating. This is expected by SPR condition
for surrounding environment with higher refractive index
than air, as that of PEDOT:PSS with a refractive index
of 1.46.
3. CHEMICALS AND MATERIALS

Normally, ve commercially available organic semiconductor materials PEDOT:PSS, P3HT, PCPDTBT, PC60 BM,
Uddin and Yang
(a)
(b)
OCH3
O
P3HT
PC60BM
(c)
(d)
OCH3
O
N
S
Figure 8.
solutions.
TEM images of 20 nm and 40 nm Ag nano-spheres in
As an example, Figure 8 shows the TEM images of

20 and 40 nm sizes Ag nano-spheres (NSs) in solutions.
PCPDTBT
PC70BM
Figure 9 displays optical absorbance spectra of 10, 20,
30 and 40 nm diameter Ag NSs. These absorbance specFigure 6. Molecular structure of organic materials: (a) P3HT
tra were measured for spin coated Ag NSs on the ITO
(b) PC60 BM (c) PCPDTBT and (d) PC70 BM.
substrate. A peak in the absorbance plots shows the plasmonic resonance position. As the particle size increases,
and PC70 BM, are used for the OPV devices. The molecuthe nanoparticle resonance red-shifts, this can be observed
lar structures of P3HT, PC60 BM, PCPDTBT and PC70 BM
in Figure 9. Additionally, absorbance curves are sharper
are shown in Figure 6. Silver nanospheres (0.02 mg/mL in
and intensities are higher for smaller size Ag NSs, they
aqueous citrate) with different diameters such as 10, 20,
tend to broaden and shrink for larger size Ag NSs. This
30, 40 nm, etc. can be purchased commercially. These Ag
is likely due to averaging over the range of sizes of the
NPs solutions can be used as received.
nanospheres and higher scattering efciency for larger Ag
The optical absorbance spectra of PCPDTBT, P3HT,
48
NSs.
For small
the scattering
University
of particles,
New South
Wales cross section is
PC60 BM and PC70 BM, are shown in Figure 7. The AM
given
in Eq.
(5). 22:47:18
Scattering depends on the size of the
IP: 149.171.67.164 On: Mon, 27 Jan
2014
1.5 G solar spectrum is also shown.Copyright:
This diagram
shows Scientific
American
particles. Publishers
As the size of the particles increases, scattering
a good correlation between the peak absorption of P3HT
increases, this is advantageous for light trapping. Different
( 500 nm) and the peak irradiance of the solar spectrum.
plasmonic materials used in OPV devices in the literatures
P3HT has a high absorption coefcient47 and can absorb up
are listed in Table I.
to 650 nm. PCPDTBT has a lower band gap and absorbs
Plasmonic technique permits considerable reduction in
from 300 nm to 850 nm. The extended spectral response of
the physical thickness of photoactive layers at the same
PCPDTBT indicates that using this polymer can enhance
time provides more efcient solar cells. Parasitic absorpthe spectral response of polymer:fullerene solar cells. The
tion loss in the metal nanoparticles is a limiting facPC70 BM material has better absorption prole overlaps
tor, more work needs to be done on loss reduction
with the solar spectrum, when compared to PC60 BM.
n
1.4
1.0
0.8
0.6
0.4
0.2
300
400
500
600
700
800
900
0.0
1000 1100 1200
Wavelength (nm)
Figure 7. Absorbance spectra of PCPDTBT (red solid), P3HT (red
dash), PC60 BM (black dash) and PC70 BM (black solid) overlapped with
the irradiance of the solar spectrum (blue).
10 nm Ag NSs (Peak: 408 nm)

Absorbance (a.u.)
1.2
Spectral Irradiance (Wm2nm1)
Absorption (a.u.)
PCPDTBT
PC70BM
P3HT
PC60BM
Solar Spectrum
350
400
450
500
550
600
650
700
750
800
Wavelength (nm)
Figure 9. Optical absorbance spectra of 10 nm (green circle), 20 nm
(red up-triangle), 30 nm (blue down-triangle) and 40 nm (brown square)
Ag NSs.
1105
Uddin and Yang
Table I. Plasmonic materials used and device performance data for different organic photovoltaic systems as reported in literature. The illuminations
used for IV tests were 1 sun.
Plasmon material
Active layer system
Jsc (mA/cm2 )
Voc (V)
FF
PCE (%)
PCE increase
Ref.
Ag NPs
Ag NPs
Ag NPs
Ag NPs
Ag NPs
Ag NPs
Ag NPs
Ag NSs
Ag NPs
Ag NTAs
Ag NW
Au NDs
Au NPs
Au NPs
Au NPs
Au NPs
Au NPs
Au NPs
Au NPs + grating
Au/Ag NPs
Au NPYs
Au NSs
Nanovoid
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PCBB
P3HT:PC60 BM
PCDTBT:PC70 BM
PCDTBT:PC60 BM
CuPc:C60
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
P3HT:PC60 BM
PFSDCN:PC60 BM
P3HT:PC60 BM
P3HT:C60
PBDTTT-C-T:PC70 BM
PTB7:PC70 BM
P3:PC60 BM
MEH-PPV:PC60 BM
P3HT:PC60 BM
106
1024
933
807
693
7
1229
1116
106
957
521
1113
1022
894
905
47
457
0602
1839
177
41
7439
053
0.62
0.61
0.58
0.58
0.59
0.42
0.64
0.60
0.86
0.89
0.50
0.58
0.59
0.63
0.62
1.10
0.55
0.27
0.76
0.71
0.55
0.78
0.42
0.55
0.51
0.53
0.59
0.35
0.55
0.54
0.55
0.53
0.50
0.57
0.70
0.62
0.52
0.50
0.61
0.55
0.63
0.69
0.42
0.45
0.44
3.69
3.20
2.82
2.75
2.23
1.20
4.30
3.60
4.60
4.52
1.32
3.65
4.24
3.51
2.91
2.17
1.53
0.09
8.79
8.67
1.10
2.36
0.20
21% (3.05)
8.5% (2.95)
17% (2.41)
65% (1.67)
70% (1.31)
45% (2.20)
25% (3.43)
29% (2.80)
11% (5.10)
6.6% (4.24)
38% (0.96)
20% (3.04)
19% (3.57)
13% (3.10)
23% (2.37)
32% (1.64)
30% (1.18)
58% (0.057)
16% (7.59)
20% (7.25)
206% (0.36)
19% (1.99)
285% (0.052)
[20]
[49]
[46]
[51]
[52]
[27]
[53]
[54]
[50]
[55]
[56]
[57]
[24]
[58]
[59]
[60]
[61]
[62]
[26]
[27]
[42]
[38]
[63]
through proper design. Moreover, controlling of the pardiagram of organic solar cell device structure on glass
ticle size, shape and distribution is another crucial task,
substrate. The device is built on a glass substrate which
which depends on nanoparticle fabrication methods. Nanocan also be transparent exible. The substrates are usuimprinting methodDelivered
enables control
over above parameters;
by Publishing
Technology to:
of New South
Wales (PET). The anode
allyUniversity
glass or polyethylene
terephthalate
27 Jan
22:47:18 oxide layer, usually indium
however, the feature sizes in IP:
this 149.171.67.164
case are limited byOn:
theMon,
consists
of2014
a semitransparent
Copyright:
American
template which is produced by other
lithography
tech- Scientific
tin oxide Publishers
(ITO). Its role is to allow light to pass through,
niques. Therefore, it is necessary to nd a cost-effective
and to collect holes from the device. A layer of the
method to produce desirable nanoparticles. In OPV area,
conductive polymer poly(3-4-ethylenedioxy-thiophene)
the most commonly used plasmonic metals are Ag and Au
poly(styrenesulfonate) (PEDOT:PSS) is applied between
in spherical shape. From the commercial point of view, Ag
the anode and the photoactive layer. This thin layer is
NPs is the suitable material for a OPV devices. The potenspin coated on top of the ITO surface. The PEDOT:PSS
tial of using other materials (i.e., Al and Cu) may contribute
layer serves as a hole conducting and electron blocker.
to more cost-effective and efcient organic solar cells.
It smooths the ITO surface, seals the active layer from
oxygen, and prevents the anode material from diffusing
into the active layer, which can lead to trap sites.64 The
4. DEVICE STRUCTURE
light absorbing photoactive layer containing the donor
The device structure of organic solar cells is different from
and acceptor material is sandwiched between two electraditional silicon wafer based solar cells. The photoactrodes. For laboratory devices, this layer is spin coated
tive layer is a blend of conjugated polymers as electron
from a common solution which contains the polymer donor
donors and fullerene derivatives as electron acceptors. This
and fullerene acceptor suspended in an appropriate solphotoactive layer is sandwiched between two electrodes
vent or mixture of solvents. The cathode is usually aluwith proper work functions. Figure 10 shows a schematic
minium, although calcium or silver is sometimes used. The
function of the cathode is to collect electrons from the
Active
Al or Ag
Layer
device. This cathode layer is usually deposited by thermal
evaporation. The energy band diagram of device structure
PEDOT:PSS
ITO/PEDOT:PSS/P3HT:PC60 BM/Al is shown in Figure 11.
A schematic diagram of an The OPV device with light
trapping
by nano-structure materials consist of an ITO
ITO
coated glass substrate, nanospheres (NSs) buffer layer,
Glass
a hole conducting layer PEDOT:PSS, PCPDTBT:PC70 BM
blend photoactive layer and an Ag cathode is shown in
Figure 10. Schematic diagram of organic solar cell device structure.
The ITO and Al/Ag layers are used as anode and cathode, respectively.
Figure 12.
1106
Uddin and Yang

(a)
Vacuum Level
Energy Level (eV)
3.2
P3HT
8 mm
4.2
4.8
Al
2 mm
4.28
ITO
PEDOT:PSS
5.1
12 mm
2 mm
PCBM
5.2
12 mm
6
6.0
(b)
Figure 11. Energy band diagram of the OPV device structure. Values
for energy levels were taken from Refs. [65, 66].
4.1. Device Fabrication

4.1.1. Solution Preparation
4.1.1.1. P3HT:PC60 BM. Normally, regioregular P3HT
and PC60 BM are carefully weighed by an analytical balance. After that, P3HT blend with PC60 BM and then
dissolve in chlorobenzene to produce a P3HT:PC60 BM
(15:15 mg/mL) solution. This solution is then stirred with
a magnetic stirrer on a hotplate for several hours at 40
C.
After stirring, the solution needs to be ltered using a
0.45 m polytetrauoroethylene (PTFE) syringe lter.
Delivered
by Publishing Technology to: University of New South Wales
4.1.1.2. PCPDTBT:PC
70 BM. PCPDTBT and PC70 BM
Figure 13. (a) Schematic drawing shows physical dimensions of patIP: 149.171.67.164
27 Jan 2014 22:47:18
need to be weighed by an analytical
balance and On:
then Mon,
terned ITO glass and (b) photograph of the ITO pattern design on the
Publishers
dissolved in 1,2-dichlorobenzene Copyright:
with a ratioAmerican
of 2:7 Scientific
glass for device
fabrication.
(8.9 mg/mL: 31.1 mg/mL). This solution is then stirred
overnight with a magnetic stirrer on a hotplate in a glove
Subsequently, a NSs plasmonic buffer layer needs to be
box at 40
C. After stirring, the solution needs to be lspin coated on the ITO substrate at 2000 rpm for plastered using a 0.45 m PTFE syringe lter.
monic studies. Different sizes of NSs such as Ag, Au, Al,
The patterned ITO glass substrate (e.g., 12 mm
etc. plasmon-enhanced layer can be used with the aim of
12 mm 1.2 mm, 5 /square) is needed for the fabricaoptimizing the performance of devices. Then, a layer of
tion of OPV devices as shown in Figure 12. The design
the hole conductive polymer PEDOT:PSS (Sigma-Aldrich
of an ITO substrate is shown in Figure 13. As the thick483095 or Clevios P VPAI 4083 ltered with a 0.45 m
ness of metal contacts is usually 100 nm, without patterned
PTFE syringe lter) is spin coated on the top of ITO or Ag
ITO, the probes may penetrate the electrodes and shunt the
NSs lm to form a thin lm of about 50 nm thickness. This
organic solar cells.
substrate is then dried at 140
C in an oven for 10 min to
The ITO glass substrate is then ultrasonicated in deionremove the water. Next, a 100 nm thick P3HT:PC60 BM
ized (DI) water; acetone and isopropyl alcohol, each for
or PCPDTBT:PC70 BM blend is deposited by spin coat10 min to remove all probable contaminate materials.
ing. Spin coating parameters may be changed for different
blends subject to solution density and viscosity.67
Figure 14 shows a P3HT:PC70 BM lm thickness as
Active
Al
a function of spin speed, the spinning time was kept
Layer
at 30 sec. For P3HT:PC70 BM (1:1) dissolved in 1,2PEDOT:PSS
dichlorobenzene (20 mg/ml), a spin speed of 1000 rpm can
be chosen in order to achieve 100 nm lm thickness. Then,
P3HT:PC60 BM or PCPDTBT:PC70 BM blend lms need to
ITO
be placed in a petri dish for 30 minutes solvent annealing.
Ag NSs
Glass
Device fabrication will be completed by the thermal evaporation of a 100 nm thick Al or NSs layer at a background
Figure 12. Schematic diagram of organic solar cell device structure
pressure of 105 mbar as shown in Figure 12. Finally, silwith nano-spheres buffer layer used for light trapping. The ITO and Al
ver paste can be deposited for better charge collection.
layers were used as anode and cathode, respectively.
1107
Uddin and Yang

250
Incident light
P3HT:PC70BM in 1,2-dichlorobenzene (20 mg/ml)
Measurement points
Film Thickness (nm)
200
Fitting curve
Front
150
100
Embedded
50
Rear
400
600
800 1000 1200 1400 1600 1800 2000 2200
Figure 15. Schematic diagram of the light trapping schemes for solar
cells using suface plasmons: front, embedded and rear case.
Spin Speed (rpm)

Figure 14. Spin coated lm thickness as a function of spin speed for
P3HT:PC70 BM in 1,2-dichlorobenzene solution (20 mg/ml).
Then, the device need to be thermally annealed at 130 C

for 10 min in nitrogen atmosphere in a glove box.
the resonance position and scattering properties of the

plasmons. Additionally, the location of the metal nanoparticles is another important parameter when designing a
solar cell. In this section, the effect of placing the nanoparticle on the front, rear or embedded in the active layer as
shown in Figure 15, is discussed. The effect of a spacer
layer thickness, the location of the metal nanoparticles
from the absorbing layer, is also critical in determining the
fraction of the coupled radiation into the active layer (fact ).
4.2. Characterization
All the transmission electron microscopy, uv-vis
absorbance spectra, external quantum efciency (EQE) and
currentvoltage (IV ) measurements can be performed
in air on non-encapsulated devices at room temperature.
TEM images can be
taken by by
a JEOL
1010 transmission
Delivered
Publishing
Technology to:
of New
South Wales
5.1.University
Particles on
the Front
electron microscope. Absorbance
can be measured using
IP: 149.171.67.164
On: Mon,
27
Jan
2014
22:47:18
The particles are located on the front side of device
American
a Perkin Elmer uv-vis spectrometer Copyright:
(Lambda 1050).
EQE Scientific Publishers
facing the incident light as shown in Figure 15. The
data can be measured with a QEX10 quantum efciency
scattering mechanism for this geometry is described in
measurement system from PV Measurements, Inc. Current
Section 2.2. The scattering of light by excitation of surdensityvoltage characteristics can be measured using a
face plasmons can improve the efciency of solar cells,
2
Keithley 2400 source meter under a 100 mW/cm , AM
which is achieved by trapping the light within the active
1.5 G spectrum from a solar simulator. All measurelayer to enhance the absorption. Most reported organic
ments should be performed within one/two days of device
solar cells with absorption enhancements are for nanoparfabrication. The samples should be stored in a nitrogen
ticles on the front (nanoparticles on top of ITO).51 52 55 57
desiccator to minimize the effects of oxidation. Physical
Literature suggests that the increased efciencies of the
lm properties are usually analyzed using optical absorpplasmonic OPV devices with the Ag NPs on top of ITO
tion, AFM and TEM, whilst electronic performances are
are driven by improved Jsc and the low values of device
examined by EQE and IV measurements. Absorption
series resistance Rs .46 The enhanced photocurrents indican be calculated by 1-transmission-reection. AFM
cate that LSPR induces more photogenerated charge carriimages are normally captured by a Bruker dimension icon
ers by stronger absorption within an active layer, resulting
scanning probe microscopy (SPM). Tapping mode can
from the stronger enhancement in electric eld near the
be used to scan an area of 5 m 5 m on the sample
metal nanoparticles.20 Kalfagiannis et al. found that the
surface. TEM images are usually taken by a JEOL 1010
saturation current of the device with the Ag embedded in
transmission electron microscope for Ag NSs in solution.
the PEDOT:PSS layer is comparatively higher, indicating
Solar simulator illumination intensity should be calibrated
lower Rs values.46 This is because individual Ag NPs may
using a monocrystalline silicon reference cell (with KG2
act as hole conductors, transporting holes more efciently.
window) calibrated by Newport Corporation. The detailed
However, recent studies have led to a better understandexperimental methods may vary for different studies.
ing of the scattering mechanisms, and identied the disadvantages of this geometry.34 52 68 Metal nanoparticles
5. LOCATION OF NANOPARTICLES
generally absorb more light in the shorter wavelength (UV
regime) regions. For this geometry, any potential absorpAs previously mentioned, the size and shape of the
tion in the metal cannot be avoided and can compromise
nanoparticles, together with the metal properties and the
dielectric medium around the nanoparticles, determines
absorption in the cells. Morfa et al. observed that Ag
1108
Uddin and Yang
NPs exhibit a decrease in quantum efciency at 450 nm

for plasmon enhanced organic solar cells.52 This feature
coincides with the dip observed in the transmission spectra owing to absorption of the Ag NPs. Secondly, this
geometry leads to a suppression of light absorption in
the regions below the resonance wavelength.45 For wavelengths below plasmon resonance, Fano resonance effect
can cause destructive interference between the transmitted
and scattered light, and thus can cause reection rather
than enhanced coupling.17 In this region the metal nanoparticles would be highly reective, thereby compensating for
the scattered light in the sub-resonance wavelength regime.
5.2. Particles on the Rear
The metal nanoparticles are placed on the side opposite the
illuminated side of the solar cell as shown in Figure 15.
The excitation of the nanoparticles in this geometry is by
Figure 16. Current densityvoltage characteristics of three devices: the
the light that is transmitted through the cell, which would
reference sample (red squares), samples with the Ag NPs on top of ITO
(blue dots) and samples with the Ag NPs on top of active layer (magenta
otherwise be lost from the cell. Hence, there are no loss
triangles). The inset gure uses semi-log to show J V characteristics
mechanisms for this case. By placing nanoparticles on the
under dark (inset-left) and linear t for a sample with the NPs embedrear, any potential absorption in the metal can be ruled
ded in PEDOT:PSS layer (inset-right). Reproduced with permission from
out, as shorter wavelength light is absorbed in the cell.
[46], N. Kalfagiannis, et al., Sol. Energy Mater. Sol. Cells 104, 165
Additionally, the suppression of absorption due to interfer(2012). 2012, Elsevier Ltd.
ence effects is avoided. This allows tunability of the resonance position over the entire region of interest. However,
exhibit good processing compatibility and hence can be
very few reports investigated this geometry in organic solar
easily incorporated into the absorber matrix. For organic
cells as the evaporated metal cathode on the rear acts as
University
of New
Souththat
Wales
solar
cells, it has
been found
adding metal nanopara back reector which alreadyIP:
provides
a
good
light
trap149.171.67.164 On: Mon,
27 Jan
2014
22:47:18
ticles
into
the
active
layer
not
only enhances local
46
ping scheme. Kalfagiannis et al. have
comparedAmerican
the per- Scientific Publishers
Copyright:
60 7072
absorption,
it
also
leads
to
enhance
carrier mobilformances of P3HT:PC60 BM bulk-heterojunction organic
ity,
since
metal
nanoparticles
are
more
conductive
than
solar cells when Ag NSs are incorporated in two different
organic materials.60 70 On the other hand, the incorporated
places as shown in Figure 16. Incorporation of Ag NSs on
metal nanoparticles will modify the nanoscale morpholtop of ITO resulted in a better overall performance with
ogy of the polymer/fullerene blend, especially with high
an increased efciency of 17%. Alternatively, adding Ag
NPs concentration, this makes it difcult to extract carriNSs on top of the active layer achieved 25% more phoers before they undergo recombination, and consequently
tocurrent generation compared with reference cells. The
reduce cell performances.60 70 Therefore, it may be necPCE was improved in comparison to the control device,
46
essary to isolate direct contact of the metal nanoparticles
but still lag behind the values of previous geometry. It
with the photoactive layer using a dielectric layer to preis expected that the LSPR of Ag NSs increases the probvent exciton quenching and charge trapping.73 74
ability of exciton generation and dissociation while high
Wang et al. have investigated the effects of Au
percentage of PC60 BM on the top of active layer effecNPs incorporated into the active layer of organic solar
tively transfers electrons to the cathode.69 Therefore, placcells with a newly synthesized donor polymer poly
ing the plasmonic NPs in direct contact with the active
[2,7-(9,9-dioctyluorene)-alt-2-((4-(diphenylamino)phenyl)
layer plays a key role in the harvesting of the SPR effect.
thiophen-2-yl) malononitrile] (PFSDCN).60 It was found
As a result, an increase in the value of short circuit current
that increased Au NPs concentration leads to improve
Jsc compared with the Ag NPs located on top of ITO was
observed. However, series resistance Rs and overall perthe optical absorption enhancement, which is due to
formance of solar cells with Ag NPs on top of active layer
LSPR effect of Au NPs as discussed in previous sections.
are worse than those of samples with Ag NPs on top of
However, this enhancement is not consistent with the
ITO, which imply that the metal nanoparticles inside the
overall observed performances of devices. To understand
photoactive layer could act as carrier traps and decrease
the reasons for the drop in PCE after the addition of
cell performance.70
> 1 wt% Au NPs, Wang et al. fabricated hole-only and
electron-only devices for determining the hole and elec5.3. Particles Embedded in the Active Layer
tron mobilities as shown in Figure 17. In low Au NPs
concentration condition, the increases of carrier mobility
Metal nanoparticles are embedded in a photoactive layer
can be explained by the introduction of hopping sites,
as shown in Figure 15. Organic materials and metals
1109
Uddin and Yang
Figure 17. Effects of Au NPs concentration on the hole and electron

mobilities in the active layer. Reproduced with permission from [60],
C. C. D. Wang, et al., J. Mater. Chem. 22, 1206 (2012). 2012, The
Royal Society of Chemisry.
as the average hopping distance becomes smaller when

the NPs concentration in the active layer increases. With
high Au NPs concentration, the NPs-induced morphology
change dominates the charge transport process and thus
both the hole and electron mobility are expected to be
reduced. It is noted that the PCE start to decrease when Au
Figure 18. AFM phase images of the surface of the active layer of
NPs concentration > 1 wt%, although the carrier mobility
P3HT:PCBM BHJ solar cells (a) without Ag NPs, (b) with 16:1 P3HT:Ag
keeps on maintaining increment until the Au NPs conNPs, (c) with 8:1 P3HT:Ag NPs, and (d) with 4:1 P3HT:Ag NPs, respeccentration reaching 2 wt%. This indicates that besides the
tively. Reproduced with permission from [70], M. Xue, et al., Appl. Phys.
carrier mobility, there should be other operation processes
Lett. 98, 253302 (2011). 2011, American Institute of Physics.
of devices affectedDelivered
by Au NPs.by Publishing Technology to: University of New South Wales
149.171.67.164
On: Mon,
27 Jancontinuous
2014 22:47:18
To further investigate this IP:
geometry,
Xue et al. examdisplays
enhancement after 630 nm, as a result
ined the properties of hybrid P3HT:PC60 BM:Ag NSs solar
of changing the surrounding medium of Ag NSs, which
cells using AFM and photoinduced charge extraction with
causes the red-shift of LSPR.43 It is observed that plas70
a linearly increasing voltage (Photo-CELIV). Figure 18
monic EQE enhancement above 700 nm is higher than
shows AFM phase images of the active layers with varydevice using structure in Figure 12. It is expected that
ing concentrations of Ag NPs. Signicant phase separaplacing Ag NSs in direct contact with the active layer has
tion of the Ag NSs from P3HT:PC60 BM active layer was
played a key role in the harvesting of the LSPR effect.70 As
observed as the concentration of Ag NSs is increased from
a result, an increase in the plasmon enhancement compared
16:1 to 4:1 P3HT:Ag. The phase-segregated Ag NPs form
with the Ag NSs located on top of ITO was observed.
their own sub-network, with both the extent of the netHowever, the EQE values in short wavelength region (400
work and its degree of interconnectivity increasing with
600 nm) are signicantly lower than control device, which
increasing loading of the NPs. The formation of the Ag
imply that the metal nanoparticles located in the front surNSs sub-network can also cause the reduced number of
face of photoactive layer could act as carrier traps and
charges extracted from the device. Carriers trapped on the
decrease cell performance at the blue region by parasitic
NP sub-network signicantly enhance recombination, as
absorption and recombination.75
suggested by the photo-CELIV measurements, which limThe placement of the Ag NSs buffer layer is crucial in
its the ability for excitons to dissociate. To optimize plasdetermining its impact on the device. There exists a trademon enhanced BHJ organic solar cells, it is necessary to
off related to the distance between the buffer layer and
nd a method to prevent the carriers from localizing on
the photoactive layer. Placing the buffer layer in contact
the metal NPs so that NPs would only enhance device
with the photoactive layer can enhance the LSPR effect
performance.
of Ag NSs. However, using this geometry would signiTo further investigate the effects of Ag NSs locations,
cantly increase the surface recombination.76 There should
a comparison on device performance using different cell
be some physical separation between the two. The elecstructures has been made. Figure 20 shows EQE spectric eld caused by the excited surface plasmons decays
tra of P3HT:PC70 BM devices with structure as shown in
exponentially with distance from the buffer layer.77 This
Figure 12 for 20 and 40 nm Ag NSs/PEDOT:PSS. The
implies that there is some non-obvious physical separation
result of P3HT:PC70 BM device is consistent with other
required, between the Ag NSs buffer layer and the photoacresults. The device structure PEDOT:PSS/Ag NSs EQE
tive layer of the device, which will allow for the maximum
1110
Uddin and Yang
improvement performance. Further optimization of the

plasmon enhanced organic bulk-heterojunction devices can
be achieved by changing both the diameter of the Ag NSs
and the thickness of the PEDOT:PSS buffer layer.
P3HT:PCBM
P3HT:PCBM on 20 nm Ag NSs
P3HT:PCBM on 40 nm Ag NSs
60
Absorption (%)
5.4. Near-Field Effect

Surface plasmons can be utilized for photovoltaic applications by three different mechanisms. These are
(a) the scattering from the metal nanoparticles that also
act as dipoles (far-eld effect),
(b) the near-eld enhancement and
(c) direct generation of charge carriers in the active layer.
70
50
40
30
20
10
0
300
400
500
600
700
800
900
Most reported photocurrent enhancement results for

Wavelength (nm)
inorganic devices are explained by the rst mechanism of
Figure 19. Optical absorbance spectra of P3HT:PC60 BM blend lms
scattering. The organic devices can be enhanced by both
with Ag NSs 20 nm (red solid), 40 nm (blue solid) and control blend
scattering and near-eld effect.
(black dash). Reproduced with permission from [11], X. Yang, et al.,
The electric eld near an emitting dipole has nonPhys. Status Solidi RRL 1 (2012). 2012, John Wiley and Sons, Inc.
radiative components with its eld strength many times
of absorbance compared to PC60 BM, in the range of 450 to
stronger than the eld radiated from the dipole, local600 nm. The presence of a vibronic structure in regioreguized within a few tens of nanometers around the metal
lar P3HT is observed at 600 nm. This vibronic structure
nanoparticle.78 If a photoactive layer is close to the eld,
is due to the ordered state achieved by the regioregular
the absorption is enhanced as a result of the high density of
P3HT.82 Incorporation of the Ag NSs buffer layers slightly
states of the photons. This effect can be further enhanced
enhances
the absorbance spectra. The total optical absorpby decreasing the spacing with the active layer and size
tion of P3HT:PC60 BM blend lms in the spectral range
of the particles. Following this effect, increased numbers
of 350650 nm are increased 4 and 6% as a result of
of excitons were generated
near
DA interface
within
Delivered
bythe
Publishing
Technology
to: University of New South Wales
incorporation
of the
20 and 40 nm Ag NSs, respectively.
a photoactive layer, increasing
the
absorption
capability
IP: 149.171.67.164 On: Mon,
27 Jan 2014
22:47:18
79
This
is
attributed
to
light
they used 5American
nm size Scientific Publishers scattering of the Ag NSs and the
before they recombine. The fact that
Copyright:
increased electric eld in the photoactive layer by excited
particles is a clear indication that these particles would
localized surface plasmons around the Ag NSs.38
not support strong scattering. Also, the surface plasmon
The stronger scattering by the larger Ag NSs can effecresonance exists far from the regime where photocurrent
tively
improve the PCE; the total scattering was indeed
enhancement was observed.
enhanced, but the back scattering power dominated the
In the near-eld, metal nanoparticles can directly excite
forward fraction with an increase in the size of the Ag
excitons when placed close to the photoactive layer. This
NSs. It should also be noted that estimated IQE actually
can play an important role in having the nanoparticles
80
provides the minimum IQE of the OPVs because the estiinside the active layer for organic solar cells, rather than
mated absorption consisted of the absorption by not only
on the front or rear of the cells as is the case for solid state
the active layer but also the NPs and the electrodes. The
devices.81 However, many other factors like interface and
IQE values were not greatly affected by the Ag NSs at all
surface effects on the near-eld still need to be understood
sizes, except for Ag NS-94, in which the IQE value near
and also call for the improved cell design? At this stage,
400, 450 nm showed a slight decrease, conceivably due to
applying the particles on the front or rear looks like a betthe absorption by the MNPs themselves. The unaffected
ter option as the optical properties of the metal nanoparIQE result strongly supports the fact that the enhanced
ticles are totally isolated from the electrical properties of
Jsc resulted from the absorption enhancement from the
the solar cell, allowing easy and independent optimization
plasmonic
scattering by the Ag NSs. In fact, not only
of the plasmon properties.
the scattering power but also the scattering direction must
be considered, as the backward scattered power does not
6. NANOPARTICLE SIZE EFFECTS
contribute to the absorption enhancement of the proposed
Figure 19 shows the uv-vis absorbance spectrum of
OPVs.
P3HT:PC60 BM blend lms with and without Ag NSs
Figure 20 displays the EQE spectra of P3HT:PC60 BM
buffer layer. For the control blend, with no buffer layer
devices with and without Ag NSs buffer layer. The sec(black dash), two main absorption peaks at 350 and
ondary axis displays the enhancements in EQE, due to the
510 nm are observed, which correspond to PC60 BM and
addition of an Ag NS buffer layer. This is a unitless multiP3HT, respectively. Absorption in the blend lm is conplication factor. Although OPV devices have not been optimized, signicant changes in electronic performance as a
tributed to primarily by P3HT. It has a higher magnitude
1111
Uddin and Yang
leads to an improvement in both of these characteristics,

as it is believed that the crystallinity of the P3HT phase
35
1.5
has improved. After the addition of Ag NSs below the
30
PEDOT:PSS layer, the values of Voc decreased slightly,
1.4
suggesting that the interface between PEDOT:PSS and
25
active layer was modied by the Ag NSs. However, the
1.3
20
addition of Ag NSs led to an improvement of photocurrent
and FF. For the devices prepared with Ag NSs, the Jsc
1.2
15
(FF) values were 9.11 (0.49) and 9.62 (0.49) mA/cm2
10
for 20 nm and 40 nm Ag NSs, respectively. As a result, the
1.1
PCEs were increased to 2.08% and 2.16%, respectively.
5
1.0
This represents an increase of 24% and 29% for 20 nm
0
and 40 nm Ag NSs, with reference to the thermally treated
300
400
500
600
700
800
control device. The larger-sized Ag NSs led to a larger
Wavelength (nm)
enhancement of Jsc and PCE, which is thought to result
Figure 20. EQE spectra of P3HT:PC60 BM devices with Ag NSs 20 nm
from enhanced scattering interactions. With the incorpo(red solid), 40 nm (blue solid) and control device (black dash) (primary
ration of Ag NSs, an additional photocurrent is created
axis). Enhancements of EQE for the devices with 20 (red circle) and
due to the enhancement of the photo-generation of exci40 nm (blue square) size Ag NSs buffer layer (secondary axis). Reprotons associated with enhanced electric eld intensity from
duced with permission from [11], X. Yang, et al., Phys. Status Solidi RRL
1 (2012). 2012, John Wiley and Sons, Inc.
the localized surface plasmon resonance. This plasmonic
coupling of light is a major contributor to the improved
electronic transport through the blend lms.
result of localized surface plasmons resonance have been
The placement of the Ag NSs buffer layer is crucial in
observed. The EQE prole of the control device (black
determining its impact on the device. There exists a tradedash) follows primarily the prole of optical absorbance of
off related to the distance between the buffer layer and
the P3HT:PC60 BM blend lm, and has a maximum EQE
the photoactive layer. Placing the buffer layer in contact
value 29.9% at 510 nm. With the presence of an Ag
with the photoactive layer can signicantly increase the
NSs buffer layer, the
maximum
EQE
values
are
increased
Delivered by Publishing Technology to: University of New27South Wales
surface
recombination,
and thus, there should be some
to 35.2 and 36.6% at 520 IP:
nm149.171.67.164
for 20 and 40 nmOn:
AgMon,
27 Jan
2014 22:47:18
physical
separation
between
the two. However, the electric
NSs, respectively. The calculated maximum
enhancements
eld caused by the excited surface plasmons decays expoin the spectral range 300650 nm are 1.19 and 1.24 for 20
nentially with distance from the layer.52 This implies that
and 40 nm Ag NSs, respectively.
there
is some non-obvious physical separation required,
Figure 21 displays the current densityvoltage (J V )
between
the Ag NS buffer layer and the photoactive layer
characteristics of P3HT:PC60 BM devices with and withof the device, which allows for maximum improvement to
out Ag NSs. The as-cast device illustrates poor device
be obtained. Further optimization of the plasmon enhanced
characteristics with Jsc = 511 mA/cm2 and FF = 0.33, as
organic bulk-heterojunction devices can be achieved by
evidenced by the almost linear prole of the J V curve.
changing both the diameter of the Ag NSs and thickness

Thermal treatment with the optimal temperature of 130 C
of Ag NS buffer layer.
Figure 22 shows the UV-vis absorbance spectra of
P3HT:PCBM ref (as-cast)
PEDOT:PSS and PCPDTBT:PC70 BM blend lms with and
10
P3HT:PCBM ref (130 C)
without Ag NSs buffer layer using cell structure as shown
P3HT:PCBM + 20 nm Ag NSs
in Figure 12. The secondary axis displays the enhance8
ments in absorbance, due to the addition of an Ag NSs
buffer layer. For the control blend, with no Ag NSs buffer
6
layer (black dashed line), two main absorption peaks at
400 nm and 760 nm are observed. Both PCPDTBT
4
and PC70 BM contribute absorption in the region of the
400 nm peak, whilst absorption in the region of
the 760 nm peak is attributed only to PCPDTBT.83
2
Absorption in the blend lm is contributed to primarily by PCPDTBT, especially in long wavelength region.
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
PC70 BM has an absorption in the spectral range of 300 to
Voltage (V)
650 nm. The concentration ratio of 2:7 for PCPDTBT and
PC70 BM was selected, which is due to a balance between
Figure 21. J V characteristics of P3HT:PC60 BM devices with Ag NSs
light absorption and change transport within the active
20 nm (red solid), 40 nm (blue solid) and control devices (as-cast: black
layer.84
solid, annealed: black dash), recorded under illumination at 100 mW/cm2 .
1.6
40
Enhancement (X)
Current Density (mA/cm2)
EQE (%)
P3HT:PCBM
EQE enhancement (20 nm)
EQE enhancement (40 nm)
1112
Uddin and Yang

90
40
20
0
300
400
500
600
700
800
900
30
60
45
20
30
10
15
0
1000
300
Wavelength (nm)
Figure 22. Optical absorbance spectra of PEDOT:PSS and
PCPDTBT:PC70 BM active layer with Ag NSs 10 nm (green solid),
20 nm (red solid), 30 nm (blue solid), 40 nm (brown solid) and control
blend (black dash) (primary axis). Enhancements of absorbance for
the lms with 10 nm (green circle), 20 nm (red up-triangle), 30 nm
(blue down-triangle) and 40 nm (brown square) Ag NSs buffer layer
(secondary axis).
75
EQE enhancement (%)
60
Reference Cell
10 nm Ag NSs
20 nm Ag NSs
30 nm Ag NSs
40 nm Ag NSs
40
EQE (%)
80
Absorbance Enhancement (%)
Absorbance (a.u.)
PEDOT:PSS + active layer

with 10 nm Ag NSs
with 20 nm Ag NSs
with 30 nm Ag NSs
with 40 nm Ag NSs
0
400
500
600
700
800
900
1000
Wavelength (nm)
Figure 23. EQE spectra of PCPDTBT:PC70 BM devices with Ag NSs
10 nm (green solid), 20 nm (red solid), 30 nm (blue solid), 40 nm (brown
solid) and control device (black dash) (primary axis). Enhancements of
EQE for the devices with 10 nm (green circle), 20 nm (red up-triangle),
30 nm (blue down-triangle) and 40 nm (brown square) Ag NSs buffer
layer (secondary axis).
Current Density (mA/cm2)
at 400 nm. The calculated maximum EQE enhancements

in the spectral range 300900 nm are 14%, 18%, 25% and
Incorporation of the Ag NSs buffer layers slightly
30% for 10, 20, 30 and 40 nm Ag NSs, respectively. The
enhances the absorbance spectra. The optical absorption
enhancements of lms with Ag NSs buffer layers have
larger-sized Ag NSs led to a larger enhancement of EQE
peaks around 400 nm, which correspond to absorptions of
in the red wavelength region.
Ag NSs. And the absorbance displays continuous enhanceFigure 24 shows the J V characteristics of PCPDTBT:
ment after 550 nm, which benets from plasmonic resPC70 BM devices with and without Ag NSs. The control
onance of Ag NSs. The PEDOT:PSS buffer layer, as a
device exhibited a Voc of 551 mV, a Jsc of 10.16 mA/cm2 ,
IP: 149.171.67.164 On: Mon,
27 Jan 2014 22:47:18
low refractive index media, causesCopyright:
the red-shift
of the Scientific
and a FF Publishers
of 0.40, resulting in a PCE of 2.24%. After the
American
85 86
plasmonic resonance position.
The overall absorptions
addition of Ag NSs below the PEDOT:PSS layer, the valof PEDOT:PSS and PCPDTBT:PC70 BM blend lms are
ues of Voc and FF remained unchanged, suggesting that
increased as a result of incorporation of Ag NSs buffer
the interface between PEDOT:PSS and active layer was
layers. This is attributed to light scattering of the Ag NSs
not affected by the Ag NSs. Some portion of the Ag NSs
and the increased electric eld in the photoactive layer by
surface was probably modied by PEDOT:PSS. The phoexcited localized surface plasmons around the Ag NSs.87
tocurrents, however, were improved as the particle size
Figure 23 displays the EQE spectra of PCPDTBT:
was increased. For the devices prepared with Ag NSs, the
PC70 BM devices with and without Ag NSs using cell
Jsc values were 10.77, 10.97, 11.32 and 11.54 mA/cm2
structure in Figure 12. The secondary axis displays the
12
enhancements in EQE, due to the addition of a Ag NSs
Reference Cell
buffer layer. For all compared devices, the EQE at wave10 nm Ag NSs
10
lengths shorter than 335 nm is identical. This indicates
20 nm Ag NSs
30 nm Ag NSs
that the measurements were reliable; the condition of the
40 nm Ag NSs
8
tested devices was stable during the entire study. The
EQE prole of control device (black dashed line) pri6
marily follows the prole of optical absorbance of the
PCPDTBT:PC70 BM blend lm, and has a maximum EQE
4
value of 40% at 420 nm. With the presence of a Ag
NSs buffer layer, the EQE values increase after 560 nm for
2
different size Ag NSs, this correlates with the plasmonic
resonance position in the absorbance spectra. The reason
0
for the EQE enhancements around 340 nm is not fully
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Voltage (V)
understood. Similar results have been observed in poly-Si
thin lm solar cells. It is suggested that the enhancement is
Figure 24. J V characteristics of PCPDTBT:PC70 BM devices with Ag
related to a higher order plasmonic resonance (quadrupole
NSs 10 nm (green solid), 20 nm (red solid), 30 nm (blue solid), 40 nm
resonance).88 89 Moreover, a reduction in EQE due to para(brown solid) and control device (black dash), recorded under illumination at 100 mW/cm2 .
sitic absorption and a red-shift of EQE peaks are observed
1113
Uddin and Yang
for 10, 20, 30 and 40 nm size Ag NSs, respectively. As

a result, the PCEs increased to 2.35%, 2.38%, 2.45% and
2.56%, respectively. This represents an increase of 5%,
6%, 9% and 14% for 10, 20, 30 and 40 nm size Ag
NSs, respectively with reference to the control device. The
larger-sized Ag NSs led to a larger enhancement of Jsc and
PCE, which is thought to result from enhanced scattering
interactions. With the incorporation of Ag NSs, an additional photocurrent is created due to the enhancement of
the photogeneration of excitons associated with enhanced
electric eld intensity from the localized surface plasmonic
resonance. The enhancement in Jsc reects the improved
photocurrent generation for long wavelengths observed in
Figure 23. This plasmonic coupling of light is a major contributor to the improved electronic transport through the
blend lms.
7. NANOPARTICLES
CONCENTRATION EFFECTS
The Ag NSs concentration effects are evaluated in
P3HT:PCBM active layer as shown in Figure 25. The
optimum concentration of Ag NSs is needed to be
estimated for the optimization of OPV device performance.

Figure 25(a) shows the performances enhancement of the
OPVs with Ag NSs embedded in PEDOT:PSS at various doping concentrations. The active layer absorbance is
enhanced with concentration in the wavelength range from
350 to 500 nm due to LSP effect.90 The surface of Ag
NSs is stimulated after surface plasmon around the particle. Incident light are coupled in very broad spectrum
range and stimulate LSP effect. According to Mie theory,91
when the radius of nanoparticle meets
03
(9)
2
incident light can be Rayleigh scattered by the particle,
and the scattering light have the same wavelength as incident light. According to Rayleigh scattering condition,
incident light can be trapped and then scattered into any
direction as the Ag NSs are far smaller than the incident light wavelength. Asymmetrically scattered light into
the active layer with the larger permittivity which effectively increases the optical path length.92 The optical path
of light is enhanced signicantly, and the absorbance of
active layer is improved.
r

IP: 149.171.67.164 On: Mon, 27 Jan 2014 22:47:18
Figure 25. Devices characteristics and spectral responses of plasmonic OPVs at various concentrations of Ag NSs. (a) The PCE (black circles), Jsc
(blue triangles (up)), Voc (red squares), and FF (green triangles (down)) enhancements of the OPVs with Ag NS embedded at various concentrations,
(b) the EQE enhancement at various concentrations, (c) the absorption enhancement at various concentrations, and (d) the IQE enhancement at various
concentrations; low (12 108 /cm2 ; medium (13 109 /cm2 ); optimum (45 109 /cm2 ); and high (90 109 /cm2 ). The EQEs were enhanced primarily
by the absorption enhancement, which was concentrated near the LSPR region (400500 nm). Reproduced with permission from [95], S. W. Baek,
et al., Sci. Reports 3 (2013), article:1726; DOI: 10.1038/Srep01726. 2013, Nature Publishing Group.
1114
Uddin and Yang
According to the Mie theory, the interaction between

incident photons and the plasmonics follows the Maxwells
equations of nanoparticles internal absorption and external scattering electromagnetic eld. From its primary
solution, the absorption/scattering cross section is
obtained9394 as

2
9NV3/2
s
=
(10)
c
1 + 22 2 + 22
the concentration of the Ag NSs was higher than the optimum concentration, the EQE was degraded throughout
nearly all of the wavelength range, especially at a longer
wavelength, while the absorption enhancement was virtually identical to that at the optimal concentration, as shown
in Figures 25(b) and (c).
Large clusters of Ag NSs were formed during the
deposition of PEDOT:PSS at a high concentration of Ag
NSs. Figure 4(c) reveals that the peak position was not
shifted at a high concentration, thereby showing that
where N is the bulk density of nanoparticles, = 2 /
the inter-particle distance was not reduced further by the
is the frequency of incident electromagnetic eld, s is the
increased density of the Ag NSs from the optimum
dielectric constant of surrounding media, V is the volume
concentration.96 Figure 25(a) reveals that the Ag NSs were
of nanoparticle, c is the speed of light in vacuum, and
well dispersed at the optimal concentration. However, as
1 and 2 are the real part and imaginary part of metal
dielectric constant, also known as external scattering and
the concentration increased further, partially aggregated
internal absorption of metal nanoparticles.
Ag NSs were indeed observed. At a high doping concentration of 90 109 cm2 , the actual total number den(11)
1 + 22 2 + 22 = 0

sity of approx. 55 109 cm2 , including large clusters,
was slightly higher than the optimized concentration of
94
is the Mie scattering resonance condition. When meets
approx. 45 109 cm2 due to aggregation despite the fact
Eq. (11), its value is called as Mie resonance frequency,
that the doping concentration had doubled. Approximately
and the cross-section area of the interaction between inci5% of the Ag NSs were large clusters that had formed
dent light and nanoparticles is far greater than the geomefrom smaller Ag NSs. The large Ag NS cluster may have
try area of the nanoparticles. Thus, a bigger plasma eld is
reduced the contact area between the PEDOT:PSS and the
stimulated to make the incident light bound into medium
active layer, P3HT:PCBM, considerably reducing the FF
material, and light absorbance of the medium is enhanced.
value. The Voc value was not affected in the low to optiThen Eq. (10) can be simplied as
mum
concentration
range,
but itWales
showed a slight decrease
Delivered by Publishing Technology to: University
of New
South
at
a
higher
concentration.
It
is
likely
that the Voc drop was
N
(12)
IP: 149.171.67.164 On: Mon, 27 Jan 2014 22:47:18
caused byPublishers
the exposure of Ag clusters on the PEDOT:PSS
The higher the Ag NPs density is, the greater the
surface,97 increasing the number of short paths for charge
absorption-scattering cross-section area is, and the more
carriers.
greatly the electromagnetic eld is enhanced which results
in capturing more photons.
8. CONCLUSION AND OUTLOOKS
The Jsc and PCE values were enhanced in proportion to
High efciency plasmonic OPV cells are demonstrated by
the particle concentration and were optimized at a concenAg NSs into the anodic PEDOT:PSS layer. The PCE was
tration of 45 109 cm2 as shown in Figure 25(a). Howenhanced to 7.6% by incorporating Ag NSs in a PCDTBTever, when the Ag NP concentration was higher than the
based OPV, mainly because Jsc was signicantly improved
optimal value, both the Jsc and FF values were degraded.
by approximately 13%. In addition, a PCE of 8.6% was
The Voc value decreased slightly as well, as shown in
achieved in PTB7-based OPVs after embedding the Ag
Figure 25(a). In particular, Jsc shows a noticeable decrease
NSs. The EQE was remarkably enhanced near the LSPR
with an increase in concentration of the Ag NSs, essenwavelength of the Ag NSs, and the improvement was
tially contributing to the reduced PCE. To understand
primarily caused by the enhancement of the absorption
the concentration dependence of the PCE enhancement,
capability. Unchanged conductivity of the PEDOT:PSS
the EQE and total absorption of the plasmonic OPVs
lms by embedding the Ag NSs proves also that the Jsc
at various concentrations are shown in Figures 25(b)
was enhanced primarily by an optical effect. The near
4(d). Figure 25(b) shows that the EQE was proportionally
eld optical microscopy scattering signals strongly supenhanced near the LSPR wavelength until the concentraport that the scattering efciency tuned by the size of the
tion reached the optimum value. In addition, the absorpAg NSs inuenced the device performance. This study
tion enhancement follows the EQE tendency, as shown in
demonstrates the possibility of manipulating the plasmonic
Figure 25(c). The IQE enhancement is nearly constant at
optical properties of metal NSs as a powerful optical engiall wavelengths, except for the high concentration, as
neering tool in OPVs. These improvements are due to the
shown in Figure 25(d), providing further verication that
light scattering of Ag NSs and the electric eld generthe EQE is enhanced by the optical absorption enhanceated by the excited surface plasmons. Moreover, a trade-off
ment. Note that with the increase in the density, the peak
related to the distance between the Ag NSs buffer layer
position is blue-shifted due to the increased dipolar coupling as indicated in Figures 25(b) and (c). However, when
and the photoactive layer has been identied. The direct
1115
Uddin and Yang
efciency. Although the optical properties and the subsecorrelation of the photovoltaic characteristics with the optical, electrical and morphological properties of plasmonic
quent improvement in light-harvesting have mostly been
devices, aiming at the quantication of the contribution
the focus of studies to date, the electrical properties of
of different NP-mediated mechanisms, will be a vital
NP-based OPVs are rather complex and revealing their
advancement. Therefore, low-cost and large area nanofabexact role on solar cell performance remains a challenge.
rication techniques need to be developed and are highly
In this respect, careful consideration must be given for the
desirable. Implementation of such emerging techniques for
type of NP used to avoid misalignment at the electronic
photon management may bring the vision of OPV comband level which can introduce charge trapping states and
mercialization closer to reality.
therefore a reduction in the Jsc and degradation in the
In-depth understanding of the OPV performance
Voc which is directly derived from the HOMOLUMO
enhancement mechanisms with respect to inter-relationlevels of the donor and acceptor materials. Research on
ships between the properties of the NPs (nature, size,
the optical properties of plasmonic OPVs should always
shape, distribution) with the performance parameters is
be performed in conjunction with an investigation of the
essential. This understanding should be supported by
respective electrical properties.
extensive theoretical modelling, providing guidelines for
the different NP parameters. Additional challenges will
ABBREVIATIONS
involve reaching the required parameters determined from
OPV Organic photovoltaic
simulation through NP synthesis and functionalization.
OLED
Organic light emitting diode
Parasitic absorption loss in metal nanoparticles is a limOFET
Organic eld-effect transistor
iting factor; more work needs to be done on loss reducBHJ Bulk-heterojunction
tion through proper design. Controlling of the particle size,
PCE Power conversion efciency
shape and distribution is another crucial task. In OPV area,
PC
60 BM [6,6]-phenyl-C61 butyric acid methyl ester
the most commonly used plasmonic metals are Ag and Au
PC70 BM [6,6]-phenyl-C71 butyric acid methyl ester
in spherical shape. The potential of using other materiPCPDTBT poly[2,1,3-benzothiadiazole-4, 7-diyl [4, 4als (i.e., Al and Cu) may contribute to more cost-effective
bis(2-ethylhexyl)-4H-cyclopenta[2,1and efcient organic solar cells. Future optimization of
b:3,4-b ]dithiophene-2,6-diyl]]
the plasmon assisted organic devices can be achieved by
Delivered by Publishing Technology to: University
of
New South Wales
PTB7 Poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2changing the size, shape, dielectric
medium and surface
IP: 149.171.67.164
On: Mon, 27 Jan 2014b:4,5-b
22:47:18

coverage of metal nanoparticles. Emphasis
should
also be Scientific Publishers ]dithiophene-2,6-diyl][3-uoroCopyright:
American
2-[(2-ethylhexyl)
carbonyl]thieno[3,4-b]
given to controlled methodologies for incorporating these
thiophenediyl]]
NPs into the cell (coating as a separate layer, mixing with
PEDOT:PSS Poly(3,4-ethylenedioxy-thiophene)
the donoracceptor solution, polarity, solubility, etc.). Furpoly(styrenesulfonate)
thermore, the use of NPs of various sizes and shapes and
NPs Nanoparticles
with different LSPR bands can be hierarchically placed in
Ag NPs Silver nanoparticles
the form of single or multiple layers into OPV devices,
Ag NSs Silver nanosphers
with the aim of optimizing the match between the device
Au NPs Gold nanoparticles
absorption and the solar spectrum. To further advance the
Au NSs Gold nanosphers
OPV technology, it is important to nd the most approCu NPs Cupper nanoparticles
priate combination of donor and acceptor materials used
Al NPs Aluminium nanoparticles
in organic solar cells. This requires careful consideration
LSPR Localized surface plasmon resonance
of the individual material band gaps, which affects the
LSP Localized surface plasmon
amount of absorption and thus the maximum obtainable
Jsc Short circuit current density
Jsc , and the diagonal band gap of the heterojunction, which
IV Currentvoltage
governs the maximum obtainable Voc . Additionally, the
Voc Open circuit voltage
ground state and excited state energy offsets, which deterfact Fraction of light scattered into a photoacmine the ability for excitons to dissociate, need to be contive layer
sidered. Narrowing the band gap can improve Jsc , but Voc
fair Fraction of light scattered into air
may correspondingly be reduced due to the higher HOMO
Cscat Scattering cross-section
level that results. Moreover, other parameters such as carQabs Normalized absorption cross-section
rier mobility, crystallinity and molecule chain packing also
Qscat Normalized scattering cross-section
affect Jsc .
FF Fill factor
Besides this, the use of multi-junction or tandem devices
EQE External quantum efciency
with multiple bandgaps, in which NPs with different
IQE Internal quantum efciency
LSPR bands are placed into different polymer active
ITO Indium tin oxide
layers and/or the interfaces, provides simultaneous benets and has the potential to further enhance the conversion
Rs Series resistance
1116
Uddin and Yang
SP
SPM
SPPs
TEM
AFM
UV
HOMO
LUMO
Surface plasmons
Scanning probe microscopy
Surface plasmons polaritons
Transmission electron microscopy
Atomic force microscopy
Ultraviolet
Highest occupied molecular orbital
Lowest unoccupied molecular orbital.
15. E. A. Parlak, T. A. Tumay, N. Tore, S. Sarioglan, P. Kavak, and

F. Turksoy, Sol. Energy Mater. Sol. Cells 110, 58 (2013).
16. M. Kang, T. Xu, H. J. Park, X. Luo, and L. J. Guo, Efciency
enhancement of organic solar cells using transparent plasmonic Ag
nanowire electrodes. Adv. Mater. 22, 4378 (2010).
17. H. A. Atwater and A. Polman, Plasmonics for improved photovoltaic
devices. Nat. Mater. 9, 205 (2010).
18. F. X. Xie, W. C. H. Choy, C. C. D. Wang, W. E. I. Sha, and D. D.
S. Fung, Improving the efciency of polymer solar cells by incorporating gold nanoparticles into all polymer layers. Appl. Phys. Lett.
99, 153304 (2011).
19. D. D. S. Fung, L. Qiao W. C. H. Choy, C. Wang, W. E. I. Sha, F. Xie,
Acknowledgment: The authors would like to thank the
and S. He, Optical and electrical properties of efciency enhanced
Australian Centre for Advanced Photovoltaics, UNSW
polymer solar cells with Au nanoparticles in a PEDOTPSS layer.
staff and technicians for their support. This program has
J. Mater. Chem. 21, 16349 (2011).
been supported by the Australian Government through the
20. S.-S. Kim, S.-I. Na, J. Jo, D.-Y. Kim, and Y.-C. Nah, Plasmon
Australian Renewable Energy Agency (ARENA). We are
enhanced performance of organic solar cells using electrodeposited
Ag nanoparticles. Appl. Phys. Lett. 93, 73307 (2008).
also grateful to all of our OPV group members for useful
21. V. Kochergin, L. Neely, C.-Y. Jao, and H. D. Robinson, Aluminum
discussions and support during this work.
plasmonic nanostructures for improved absorption in organic photovoltaic devices. Appl. Phys. Lett. 98, 133305 (2011).
22. X. Chen, L. Zuo, W. Fu, Q. Yan, C. Fan, and H. Chen, Insight
References and Notes
into the efciency enhancement of polymer solar cells by incor1. Z. He, C. Zhong, X. Huang, W. Y. Wong, H. Wu, L. Chen, S. Su, and
porating gold nanoparticles. Sol. Energy Mater. Sol. Cells 111, 1
Y. Cao, Simultaneous enhancement of open-circuit voltage, short(2013).
circuit current density, and ll factor in polymer solar cells. Adv.
23. O. Stenzel, A. Stendal, K. Voigtsberger, and C. von Borczyskowski,
Mater. 23, 4636 (2011).
Enhancement of the photovoltaic conversion efciency of copper
2. M. C. Quiles, T. Ferenczi, T. Agostinelli, P. G. Etchegoin, Y. Kim,
phthalocyanine thin lm devices by incorporation of metal clusters.
T. D. Anthopoulos, P. N. Stavrinou, D. D. C. Bradley, and J. Nelson,
Sol. Energy Mater. Sol. Cells 37, 337 (1995).
Morphology evolution via self-organization and lateral and vertical
24.
J.-L.
Wu, F.-C. Chen, Y.-S. Hsiao, F.-C. Chien, P. Chen, C.-H. Kuo,
diffusion in polymer: Fullerene solar cell blends. Nat. Mater. 7, 158
M. H. Huang, and C.-S. Hsu, Surface plasmonic effects of metallic
(2008).
nanoparticles on
performance
of polymer bulk heterojunction
Delivered
by Publishing
Technology
to: University
of the
New
South Wales
3. A. Facchetti, -conjugated
polymers
for organic electronics
and
solar
cells.
ACS
Nano
5,
959
(2011).
photovoltaic cell applications. Chem.
Mater.
23,
733
(2010).
IP: 149.171.67.164 On: Mon, 27 Jan 2014 22:47:18
25. L. Lu, Z. Luo, T. Xu, and L. Yu, Cooperative plasmonic effect of
4. U. Zhokhavets, T. Erb, G. Gobsch,
M. Al-Ibrahim,
and Scientific
Copyright:
American
Publishers
ag and au nanoparticles on enhancing performance of polymer solar
O. Ambacher, Relation between absorption and crystallinity of
cells. Nano Lett. 12, 59 (2012).
poly(3-hexylthiophene)/fullerene lms for plastic solar cells. Chem.
26. X. Li, W. C. H. Choy, L. Huo, F. Xie, W. E. I. Sha, B. Ding, X. Guo,
Phys. Lett. 418, 347 (2006).
Y. Li, J. Hou, J. You, and Y. Yang, Dual plasmonic nanostructures
5. L. Sguillon, Heliatek achieves 12% organic solar cell efciency,
for high performance inverted organic solar cells. Adv. Mater. 24,
http://optics.org/news/4/1/36.
3046 (2012).
6. F. C. Krebs, J. Fyenbo, D. M. Tanenbaum, S. A. Gevorgyan,
27. W.-J. Yoon, K.-Y. Jung, J. Liu, T. Duraisamy, R. Revur, F. L.
R. Andriessen, B. van Remoortere, Y. Galagan, and M. Jorgensen,
Teixeira, S. Sengupta, and P. R. Berger, Plasmon-enhanced optical
The OE-A OPV demonstrator anno domini 2011. Energy Environ.
absorption and photocurrent in organic bulk heterojunction photoSci. 4, 4116 (2011).
voltaic devices using self-assembled layer of silver nanoparticles.
7. W. Ma, C. Yang, X. Gong, K. Lee, and A. J. Heeger, Thermally
Sol. Energy Mater. Sol. Cells 94, 128 (2010).
stable, efcient polymer solar cells with nanoscale control of the
28. H. Raether, Surface Plasmons on Smooth and Rough Surfaces and
interpenetrating network morphology. Adv. Funct. Mater. 15, 1617
on Gratings, Springer-Verlag, Berlin (1988).
(2005).
29. S. Pillai and M. A. Green, Plasmonics for photovoltaics, Compre8. M. S. Ryu, H. J. Cha, and J. Jang, Effects of thermal annealing of
hensive Renewable Energy, edited by Ali, Elsevier, Oxford (2012),
polymer: Fullerene photovoltaic solar cells for high efciency. Curr.
pp. 641656.
Appl. Phys. 10, S206 (2010).
30. Y. C. Nah, S. S. Kim, J. H. Park, H. J. Park, J. Jo, and D. Y. Kim,
9. T. Kirchartz, K. Taretto, and U. Rau, Efciency limits of organic
Enhanced electrochromic absorption in Ag nanoparticle embedded
bulk heterojunction solar cells. J. Phys. Chem. C 113, 17958 (2009).
conjugated polymer composite lms. Electrochem. Commun. 9, 1542
10. A. C. Mayer, S. R. Scully, B. E. Hardin, M. W. Rowell, and M. D.
(2007).
McGehee, Polymer-based solar cells. Mater. Today 10, 28 (2007).
31. V. Z. Anatoly and I. S. Igor, Near-eld photonics: Surface plasmon
11. X. Yang, A. Uddin, and M. Wright, Plasmon enhanced light absorppolaritons and localized surface plasmons. J. Optics A: Pure and
tion in bulk heterojunction organic solar cells. Phys. Status Solidi
Appl. Optics 5, S16 (2003).
RRL 1 (2012).
32. S. Pillai, K. R. Catchpole, T. Trupke, G. Zhang, J. Zhao, and M. A.
12. E. Stratakis and E. Kymakis, Nanoparticle-based plasmonic organic
Green, Enhanced emission from Si-based light-emitting diodes using
photovoltaic devices. Mater. Today 16, 133 (2013).
surface plasmons. Appl. Phys. Lett. 88, 161102 (2006).
13. J. You, X. Li, F. Xie, W. E. I. Sha, J. H. W. Kwong, G. Li, W. C.
33. W. E. I. Sha, W. C. H. Choy, Y. G. Liu, and W. C. Chew, Near-eld
H. Choy, and Y. Yang, Surface plasmon and scattering enhanced
multiple scattering effects of plasmonic nanospheres embedded into
low-bandgap polymer solar cell by a metal grating back electrode.
thin-lm organic solar cells. Appl. Phys. Lett. 99, 113304 (2011).
Adv. Energy Mater. 2, 1203 (2012).
34. J. Zhu, M. Xue, H. Shen, Z. Wu, S. Kim, J.-J. Ho, A. Hassani14. Q. Gan, F. J. Bartoli, and Z. H. Kafa, Plasmonic enhanced organic
Afshar, B. Zeng, and K. L. Wang, Plasmonic effects for light conphotovoltaics: Breaking the 10% efciency barrier. Adv. Mater. 25,
2385 (2013).
centration in organic photovoltaic thin lms induced by hexagonal
1117
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
Uddin and Yang
periodic metallic nanospheres. Appl. Phys. Lett. 98, 151110

polymer photovoltaics made by generating silver nanoparticles in
polymer:fullerene bulk-heterojunction structures. Sol. Energy Mater.
(2011).
Sol. Cells 92, 397 (2008).
A. J. Moule, J. B. Bonekamp, and K. Meerholz, The effect of
54. C. Ma, W. Qin, X. Xu, M. Li, X. Han, L. Yang, S. Yin, J. Wei,
active layer thickness and composition on the performance of bulkand F. Zhang, Plasmon-enhanced organic solar cells with solutionheterojunction solar cells. J. Appl. Phys. 100, 94503 (2006).
processed three-dimensional Ag nanosheets. Sol. Energy Mater. Sol.
C. F. Bohren and D. R. Huffman, Absorption and Scattering of Light
Cells 109, 227 (2013).
by Small Particles, Wiley-VCH, New York (2008).
55. B. Wu, T. Z. Oo, X. Li, X. Liu, X. Wu, E. K. L. Yeow, H. J.
S. Pillai and M. A. Green, Plasmonics for photovoltaic applications.
Fan, N. Mathews, and T. C. Sum, Efciency enhancement in bulkSol. Energy Mater. Sol. Cells 94, 1481 (2010).
heterojunction solar cells integrated with large-area Ag nanotriangle
L. Qiao, D. Wang, L. Zuo, Y. Ye, J. Qian, H. Chen, and S. He,
arrays. J. Phys. Chem. C 116, 14820 (2012).
Localized surface plasmon resonance enhanced organic solar cell
56. M.-G. Kang, T. Xu, H. J. Park, X. Luo, and L. J. Guo, Efciency
with gold nanospheres. Appl. Energy 88, 848 (2011).
enhancement of organic solar cells using transparent plasmonic ag
K. Inho, L. T. Seong, J. D. Seok, L. W. Seong, and L. Kyeongnanowire electrodes. Adv. Mater. 22, 4378 (2010).
Seok, Size effects of metal nanoparticles embedded in a buffer
57. J. H. Lee, J. H. Park, J. S. Kim, D. Y. Lee, and K. Cho, High
layer of organic photovoltaics on plasmonic absorption enhancement.
efciency polymer solar cells with wet deposited plasmonic gold
J. Phys. D: Appl. Phys. 45, 65101 (2012).
nanodots. Org. Electron. 10, 416 (2009).
V. Dhumale, P. Shah, R. Sharma, and K. Tanabe, Effects of parti58. D. D. S. Fung, L. Qiao, W. C. H. Choy, C. Wang, W. E. I. Sha,
cle size and surrounding media on optical radiation efciencies of
F. Xie, and S. He, Optical and electrical properties of efciency
spherical plasmonic metal nanoparticles. Bull. Mater. Sci. 35, 143
enhanced polymer solar cells with Au nanoparticles in a PEDOT:PSS
(2012).
layer. J. Mater. Chem. 21, 16349 (2011).
W. Wu, P. N. Njoki, H. Han, H. Zhao, E. A. Schiff, P. S.
59. X. Chen, L. Zuo, W. Fu, Q. Yan, C. Fan, and H. Chen, Insight
Lutz, L. Solomon, S. Matthews, and M. M. Maye, Processing
into the efciency enhancement of polymer solar cells by incorcore/alloy/shell nanoparticles: Tunable optical properties and eviporating gold nanoparticles. Sol. Energy Mater. Sol. Cells 111, 1
dence for self-limiting alloy growth. J. Phys. Chem. C 115, 9933
(2013).
(2011).
60. C. C. D. Wang, W. C. H. Choy, C. Duan, D. D. S. Fung, W. E. I. Sha,
A. Kirkeminde, M. Retsch, Q. Wang, G. Xu, R. Hui, J. Wu, and
F.-X. Xie, F. Huang, and Y. Cao, Optical and electrical effects of
S. Ren, Surface-passivated plasmonic nano-pyramids for bulk hetgold nanoparticles in the active layer of polymer solar cells. J. Mater.
erojunction solar cell photocurrent enhancement. Nanoscale 4, 4421
Chem. 22, 1206 (2012).
(2012).
61. S. Shahin, P. Gangopadhyay, and R. A. Norwood, Ultrathin organic
F. J. Beck, A. Polman, and K. R. Catchpole, Tunable light trapping
bulk heterojunction solar cells: Plasmon enhanced performance using
for solar cells using localized surface plasmons. J. Appl. Phys. 105,
Au nanoparticles. Appl. Phys. Lett. 101, 53109 (2012).
114310 (2009). Delivered by Publishing Technology to: University of New South Wales
62. C.-H. Poh, L. Rosa, S. Juodkazis, and P. Dastoor, FDTD modeling
D. E. Aspnes, Plasmonics and effective-medium
theories. ThinOn:
SolidMon, 27 Jan 2014 22:47:18
IP: 149.171.67.164
to enhance the performance of an organic solar cell embedded with
Films 519, 2571 (2011).
Publishers
gold nanoparticles.
Optical Mater. Exp. 1, 1326 (2011).
S. Pillai, F. J. Beck, K. R. Catchpole, Z. Ouyang, and M. A.
63. N. N. Lal, B. F. Soares, J. K. Sinha, F. Huang, S. Mahajan, P. N.
Green, The effect of dielectric spacer thickness on surface plasmon
Bartlett, N. C. Greenham, and J. J. Baumberg, Enhancing solar
enhanced solar cells for front and rear side depositions. J. Appl.
cells with localized plasmons in nanovoids. Optic. Exp. 19, 11256
Phys. 109, 73105 (2011).
(2011).
N. Kalfagiannis, P. G. Karagiannidis, C. Pitsalidis, N. T.
64. B. R. Saunders, Hybrid polymer/nanoparticle solar cells: PreparaPanagiotopoulos, C. Gravalidis, S. Kassavetis, P. Patsalas, and
tion, principles and challenges. J. Colloid Interface Sci. 369, 1
S. Logothetidis, Plasmonic silver nanoparticles for improved organic
(2012).
solar cells. Sol. Energy Mater. Sol. Cells 104, 165 (2012).
65. Y. Zhou, M. Eck, and M. Kruger, Bulk-heterojunction hybrid solar
C. J. Brabec, M. Heeney, I. McCulloch, and J. Nelson, Inuence
cells based on colloidal nanocrystals and conjugated polymers.
of blend microstructure on bulk heterojunction organic photovoltaic
Energy Environ. Sci. 3, 1851 (2010).
performance. Chem. Soc. Rev. 40, 1185 ( 2011).
66. C.-Y. Liu, Z. C. Holman, and U. R. Kortshagen, Hybrid solar cells
S. Pillai, K. R. Catchpole, T. Trupke, and M. A. Green, Surface
from p3ht and silicon nanocrystals. Nano Lett. 9, 449 (2008).
plasmon enhanced silicon solar cells. J. Appl. Phys. 101, 93105
67. B. T. Chen, Investigation of the solvent-evaporation effect on spin
(2007).
coating of thin lms. Polymer Eng. Sci. 23, 399 (1983).
S. Woo, J. H. Jeong, H. K. Lyu, Y. S. Han, and Y. Kim, In situ68. Z. Ouyang, S. Pillai, F. Beck, O. Kunz, S. Varlamov, K. R.
prepared composite materials of PEDOT:PSS buffer layer-metal
Catchpole, P. Campbell, and M. A. Green, Effective light trapping in
nanoparticles and their application to organic solar cells. Nanoscale
polycrystalline silicon thin-lm solar cells by means of rear localized
Res. Lett. 7, 1 (2012).
surface plasmons. Appl. Phys. Lett. 96, 261109 (2010).
H. M. Stec and R. A. Hatton, Plasmon-active nano-aperture win69. M. A. Ruderer, S. Guo, R. Meier, H.-Y. Chiang, V. Krstgens,
dow electrodes for organic photovoltaics. Adv. Energy Mater. 3, 193
J. Wiedersich, J. Perlich, S. V. Roth, and P. Mller-Buschbaum,
(2013).
Solvent-induced morphology in polymer-based systems for organic
R. S. Kim, J. Zhu, J. H. Park, L. Li, Z. Yu, H. Shen, M. Xue,
photovoltaics. Adv. Funct. Mater. 21, 3382 (2011).
K. L. Wang, G. Park, T. J. Anderson, and Q. Pei, E-beam deposited
70. M. Xue, L. Li, B. J. T. de Villers, H. Shen, J. Zhu, Z. Yu, A. Z. Stieg,
Ag-nanoparticles plasmonic organic solar cell and its absorption
Q. Pei, B. J. Schwartz, and K. L. Wang, Charge-carrier dynamics
enhancement analysis using FDTD-based cylindrical nano-particle
in hybrid plasmonic organic solar cells with Ag nanoparticles. Appl.
optical model. Optic. Express 20, 12649 (2012).
Phys. Lett. 98, 253302 (2011).
A. J. Morfa, K. L. Rowlen, T. H. R. Iii, M. J. Romero, and J. van
71. K. Kim and D. L. Carroll, Roles of Au and Ag nanoparticles
de Lagemaat, Plasmon-enhanced solar energy conversion in organic
in efciency enhancement of poly(3-octylthiophene)/C60 bulk hetbulk heterojunction photovoltaics. Appl. Phys. Lett. 92, 13504
erojunction photovoltaic devices. Appl. Phys. Lett. 87, 203113
(2008).
(2005).
B. V. K. Naidu, J. S. Park, S. C. Kim, S.-M. Park, E.-J. Lee, K.-J.
72. K. Topp, H. Borchert, F. Johnen, A. V. Tunc, M. Knipper, E. von
Hauff, J. Parisi, and K. Al-Shamery, Impact of the incorporation of
Yoon, S. J. Lee, J. Wook Lee, Y.-S. Gal, and S.-H. Jin, Novel hybrid
1118
Uddin and Yang
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85. J. J. Mock, D. R. Smith, and S. Schultz, Local refractive index

Au nanoparticles into polymer/fullerene solar cells. J. Phys. Chem. A
dependence of plasmon resonance spectra from individual nanopar114, 3981 (2009).
ticles. Nano Lett. 3, 485 (2003).
J.-Y. Lee and P. Peumans, The origin of enhanced optical absorption
86. H. Kuwata, H. Tamaru, K. Esumi, and K. Miyano, Resonant
in solar cells with metal nanoparticles embedded in the active layer.
Optic. Exp. 18, 10078 (2010).
light scattering from metal nanoparticles: Practical analysis beyond
Rayleigh approximation. Appl. Phys. Lett. 83, 4625 (2003).
J. R. Nagel and M. A. Scarpulla, Design principles for light trapping
87. H. Kong, J. S. Moon, N. S. Cho, I. H. Jung, M.-J. Park, J.-H. Park,
in thin silicon lms with embedded dielectric nanoparticles. Prog.
S. Cho, and H.-K. Shim, Thermal annealing induced bicontinuous
Photovoltaics: Res. Applicat. 19, 1 (2011).
Z. Ouyang, X. Zhao, S. Varlamov, Y. Tao, J. Wong, and S. Pillai,
networks in bulk heterojunction solar cells and bipolar eld-effect
Prog. Photovoltaics: Res. Applicat. 19, 917 (2011).
transistors. Appl. Phys. Lett. 95, 173301 (2009).
W.-J. Yoon, K.-Y. Jung, J. Liu, T. Duraisamy, R. Revur, F. L.
88. P. Spinelli, C. van Lare, E. Verhagen, and A. Polman, Controlling
Teixeira, S. Sengupta, and P. R. Berger, Sol. Energy Mater. Sol. Cells
Fano lineshapes in plasmon-mediated light coupling into a substrate.
94, 128 (2010).
Optic. Exp. 19, A303 (2011).
D. D. Evanoff, R. L. White, and G. Chumanov, J. Phys. Chem. B
89. F. Zhou, Z.-Y. Li, Y. Liu, and Y. Xia, Quantitative analysis of dipole
108, 1522 (2004).
and quadrupole excitation in the surface plasmon resonance of metal
G. Mie, Beitrge zur Optik trber Medien, speziell kolloidaler Metnanoparticles. J. Phys. Chem. C 112, 20233 (2008).
allsungen. Annalen der Physik, 330, 377 (1908).
90. J. Bellessa, C. Bonnand, J. C. Plenet, and J. Mugnier. Phys. Rev.
B. P. Rand, P. Peumans, and S. R. Forrest, Long-range absorption
Lett. 93, 36404 (2004).
enhancement in organic tandem thin-lm solar cells containing silver
91. P. Kik and M. Brongersma, Springer Series in Optical Sciences 31, 1
nanoclusters. J. Appl. Phys. 96, 7519 (2004).
(2007).
92. T. D. Heidel, J. K. Mapel, M. Singh, K. Celebi, and M. A. Baldo,
C. C. D. Wang, W. C. H. Choy, C. Duan, D. D. S. Fung, W. E. I. Sha,
F.-X. Xie, F. Huang, and Y. Cao, Optical and electrical effects of
Appl. Phys. Lett. 91, 93506 (2007).
gold nanoparticles in the active layer of polymer solar cells. J. Mater.
93. D. Duchea, P. Torchioa, L. Escoubasa, F. Monestiera, J.-J. Simona,
Chem. 22, 1206 (2012).
F. Floryb, and G. Mathian. Sol. Energy Mater. Sol. Cells 93, 1377
Z. Ouyang, X. Zhao, S. Varlamov, Y. Tao, J. Wong, and S. Pillai,
(2009).
Nanoparticle-enhanced light trapping in thin-lm silicon solar cells.
94. D. D. Evanoff, Jr, and G. Chumanov, J. Phys. Chem. B 108, 13957
Prog. Photovoltaics: Res. Applicat. 19, 917 (2011).
(2004).
M. Dante, J. Peet, and T. Q. Nguyen, Nanoscale charge trans95. S. W. Baek, J. Noh, C. H. Lee, B. S. Kim, M. K. Seo, and J. Y. Lee,
port and internal structure of bulk heterojunction conjugated polyPlasmonic forward scattering effect in organic solar cells: A powmer/fullerene solar cells by scanning probe microscopy. J. Phys.
erful optical engineering method. Sci. Rep. 3 (2013), article:1726;
Chem. C 112, 7241 (2008).
DOI: 10.1038/Srep01726.
D. Mhlbacher, M. Scharber, M. Morana, Z. Zhu, D. Waller,
96. L. L. Zhao, K. L. Kelly, and G. C. Schatz, The extinction spectra of
University
of New South Wales
R. Gaudiana, and C. Brabec, High photovoltaic performance of a
silver nanoparticle arrays: Inuence of array structure on plasmon
IP:
149.171.67.164
On:
Mon,
27
Jan
2014
22:47:18
low-bandgap polymer. Adv. Mater. 18, 2884 (2006).
resonance wavelength and width. J. Phys. Chem. B 107, 7343 (2003).
Publishers
J. Kettle, M. Horie, L. A. Majewski, B.Copyright:
R. Saunders, S.American
Tuladhar, Scientific
97. M. Xue,.
L. Li, B. J. Tremolet de Villers, H. Shen, J. Zhu, Z. Yu,
J. Nelson, and M. L. Turner, Optimisation of PCPDTBT solar cells
A. Z. Stieg, Q. Pei, B. J. Schwartz, and K. L. Wang, Charge-carrier
dynamics in hybrid plasmonic organic solar cells with Ag nanoparusing polymer synthesis with Suzuki coupling. Sol. Energy Mater.
Sol. Cells 95, 2186 (2011).
ticles. Appl. Phys. Lett. 98, 253302 (2011).
Received: 6 July 2013. Revised/Accepted: 25 July 2013.
1119

Surface Plasmon Effects On Organic Solar Cells

Uploaded by

Copyright:

Available Formats

Surface Plasmon Effects On Organic Solar Cells

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Surface Plasmon Effects On Organic Solar Cells

Uploaded by

Copyright:

Available Formats

Review

Copyright 2014 American Scientic Publishers

Vol. 14, 10991119, 2014

Surface Plasmonic Effects on Organic Solar Cells

Author to whom correspondence should be addressed.

J. Nanosci. Nanotechnol. 2014, Vol. 14, No. 2

Organic semiconductors are currently being considered

Surface Plasmonic Effects on Organic Solar Cells

Uddin and Yang

commercial production of primary OPV products and a

J. Nanosci. Nanotechnol. 14, 10991119, 2014

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

2.1. Localized Surface Plasmons

where m is the relative permittivity of the metal particle.

(for P3HT:PC60 BM (1:1)) the refractive index is 1.87

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

Figure 3. Schematic diagram of the scattering process for particles on

5 and 15 nm gold nanospheres (Au NSs) and doped them

2.4. Effect of Material Properties

2.5. Effect of Dielectric Medium

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

metal particles, and the peak can be further red-shifted by

Since = n + ik2 where n and k are the real and

Figure 5 shows the observed red-shift of the normalized

3. CHEMICALS AND MATERIALS

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

TEM images of 20 nm and 40 nm Ag nano-spheres in

As an example, Figure 8 shows the TEM images of

J. Nanosci. Nanotechnol. 14, 10991119, 2014

10 nm Ag NSs (Peak: 408 nm)

Spectral Irradiance (Wm2nm1)

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

Active layer system

J. Nanosci. Nanotechnol. 14, 10991119, 2014

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

Energy Level (eV)

4.1. Device Fabrication

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

Film Thickness (nm)

800 1000 1200 1400 1600 1800 2000 2200

Spin Speed (rpm)

Then, the device need to be thermally annealed at 130 C

the resonance position and scattering properties of the

J. Nanosci. Nanotechnol. 14, 10991119, 2014

Uddin and Yang

Surface Plasmonic Effects on Organic Solar Cells

NPs exhibit a decrease in quantum efciency at 450 nm

Surface Plasmonic Effects on Organic Solar Cells

Uddin and Yang

Figure 17. Effects of Au NPs concentration on the hole and electron

Since = n + ik2 where n and k are the real and

1 + 22 2 + 22

1 + 22 2 + 22 = 0