See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268694505

Improved performance of organic solar cells by incorporating silica-coated

silver nanoparticles in buffer layer

Article · November 2014

DOI: 10.1039/C4TC01990C

CITATIONS READS

31 370

9 authors, including:

Qin-Jun Sun Yanxia Cui

Taiyuan University of Science and Technology Taiyuan University of Technology
43 PUBLICATIONS   201 CITATIONS    99 PUBLICATIONS   2,146 CITATIONS   

SEE PROFILE SEE PROFILE

Furong Zhu
Hong Kong Baptist University
231 PUBLICATIONS   4,373 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Nonfullerene organic solar cells View project

Top emitting OLED View project

All content following this page was uploaded by Furong Zhu on 07 February 2015.

The user has requested enhancement of the downloaded file.

 Journal of
Materials Chemistry C
View Article Online
PAPER View Journal | View Issue
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

Improved performance of organic solar cells by

incorporating silica-coated silver nanoparticles in
Cite this: J. Mater. Chem. C, 2015, 3,
1082 the buffer layer
Yang Hao,a Jingcheng Song,a Fan Yang,a Yuying Hao,*a Qinjun Sun,a Junjie Guo,b
Yanxia Cui,a Hua Wangc and Furong Zhu*d

It is demonstrated that the use of silica-coated silver nanoparticles (AgNPs) in the buffer layer improves the
performance of organic solar cells (OSCs). It is found that only large sized AgNPs are advantageous for
increasing the electric field distribution in the active layer, and therefore, increasing light absorption,
caused by the localized surface plasmonic resonance and far-field scattering. Furthermore, the
scattering of silica-coated AgNPs is more important to the light harvesting because of the existence of
the silica coating. It is also demonstrated that the silica coating is favorable for enhancing the exciton
dissociation because of the reduction of the exciton quenching that occurred at the interface between
the bare AgNPs and the active layer. Furthermore, silica-coated AgNPs also promote hole transport and
extraction, which is presumably explained by the introduction of “dopant” levels within the band gap of
the poly(3,4-ethylenedioxythiophene)–poly(styrenesulfonate) and reduction of hole trapping of a bare
silver surface. The combination of all these benefits results in a 25.4% improvement in photocurrent
density and an increase of 19.2% in power conversion efficiency. This work indicates that using silica-
Received 6th September 2014
Accepted 17th November 2014
coated AgNPs as light trapping elements is more efficient than using bare AgNPs in plasmonic organic
solar cells. The systematic exploration of the optical and electrical effects of silica-coated AgNPs
DOI: 10.1039/c4tc01990c
contributes to a more comprehensive understanding of the mechanism of performance improvement of
www.rsc.org/MaterialsC the plasmonic OSCs.

ways to effectively enhance light absorption in the active layer

1. Introduction without increasing its thickness. Different approaches have
Organic solar cells (OSCs) have great potential as a clean and been reported including incorporating metal nanoparticles
renewable energy source because of the solution-based fabri- (NPs) and periodic surface plasmonic structures in OSCs to
cation process, their light weight, transparency as well as their boost light absorption.4–6 In OSCs with metal NPs incorporated
compatibility with large area exible substrates.1–3 At the in them, the metal NPs function as sub-wavelength antennas;
moment, the power conversion efficiency (PCE) of OSCs is less the increase of the absorption in the active layer is attributed to
satisfactory when compared to their inorganic counterparts. the enhanced electromagnetic eld because of the excitation of
One of the limitations is insufficient light absorption in the thin localized surface plasmon resonance (LSPR).5 However, metal
organic photoactive layers. Absorption enhancement in OSCs NPs, serving as scattering centers, also increase the length of
cannot be simply realized by using a thicker active layer because the optical path contributing to light absorption enhancement
of the mismatch between the optical absorption length and the in the solar cells.5
charge transport scale, which is caused by the low carrier In research up to the present time, the metal NPs with
mobility in organic materials. Therefore, it is necessary to nd different materials, shapes, and sizes have been introduced into
various layers in the OSCs.7–22 The plasmonic OSCs seem
a
attractive, but there exist signicant challenges as well. There
Key Lab of Advanced Transducers and Intelligent Control System, Ministry of
Education and Shanxi Province, College of Physics and Optoelectronics, Taiyuan are some controversial reports, about their favorable10,14,17,18 and
University of Technology, Taiyuan 030024, China. E-mail: haoyuyinghyy@sina.com unfavorable23–26 effects on the performance of OSCs. Although
b
Material Science and Technology Division, Oak Ridge National Laboratory, Oak light absorption enhancement in OSCs based on metal NPs was
Ridge, Tennessee 37831, USA observed, the bare metal NPs in the active layer also induce
c
Key Laboratory of Interface Science and Engineering in Advanced Materials, Taiyuan charge recombination and exciton quenching near the vicinity
University of Technology, Taiyuan, China
d
of metal surface because of the dipole–dipole and charge-trap-
Department of Physics and Institute of Advanced Materials, Hong Kong Baptist
University, 224 Waterloo Road, Kowloon Tong, Hong Kong. E-mail: frzhu@hkbu.
ping coupling.27 Therefore the electrical losses because of the
edu.hk undesired exciton quenching and charge-trapping offset the

1082 | J. Mater. Chem. C, 2015, 3, 1082–1090 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry C

optical enhancement generated by the metal NPs.18 In order to used and the reaction temperature. Aer reaction at 120–130
C
avoid the exciton quenching and charge-trapping and maintain for 1 h, the AgNPs were collected by centrifugation and washed
the advantage of the absorption enhancement induced by the sequentially several times using ethanol and water. Next, the
metal NPs, the strategy of forming a thin dielectric layer coated AgNPs were then encapsulated with silica in the following way:
metal NP was proposed.28 The solution of placing the metal NPs the homogeneous Ag in ethanol solution was mixed with
outside the active layer7–16,19–21 is more effective for improving ammonium hydroxide with stirring, and then tetraethyl ortho-
the absorption in the cells, and thereby the PCE, resulting from silicate (TEOS) was slowly injected with continuous stirring. The
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

the LSPR and/or far-eld scattering of metal NPs. But Choy et al. reaction was continued for 6 h. The Ag@SiO2 nanoparticles
suggested that the strong near eld around gold nanoparticles were separated by centrifugation and washed sequentially
has a limited contribution to the light absorption enhancement several times using ethanol and water. The thickness of the
in the active layer as the LSPR effect is mainly distributed silica coating layer increases gradually with increase of the
laterally along the anode/buffer layer interface. The improve- TEOS doping ratio in the reaction.
ment in PCE originates from the enlarged contact area between
the active layer and the buffer layer and improved the conduc-
2.2. Fabrication of OSCs
tivity of the buffer layer.29
In this research, the large size silica-coated silver nano- The OSCs which have Ag@SiO2 incorporated in them have the
particles (Ag@SiO2NPs) were incorporated in the poly(3,4-ethyl- device structure of glass/indium tin oxide (ITO)/
enedioxythiophene)–poly(styrenesulfonate) (PEDOT:PSS) buffer PEDOT:PSS:Ag@SiO2NPs (30 nm)/poly(3-hexylthiophene):phenyl-
layer to improve the PCE. Indeed, incorporating Ag@SiO2NPs in C60-butyric acid methyl ester (P3HT:PC60BM, 220 nm)/lithium
the interface of the active layer/buffer layer for improving PCE uoride (LiF, 1 nm)/Al (150 nm), as shown in Fig. 1(a). ITO-coated
has been reported by other groups.30 The difference with the glass substrates were cleaned with acetone and isopropyl alcohol,
existing research is that here Ag@SiO2NPs are incorporated and then treated using ultraviolet (UV) ozone. An approximately
within the buffer layer and some large NPs are able to protrude 30 nm thick PEDOT:PSS layer with different Ag@SiO2 doping
partly into the active layer. This solution has advantages in ratios was spin-coated onto ITO at 3500 rpm for 60 s, and
improving the PCE by: (1) increasing light absorption in the annealed at 120
C for 30 min in air. Next, a 220 nm active layer
active layer by inducing simultaneously the far-eld scattering
effect and laterally distributed LSPR, caused by the large size
Ag@SiO2NPs; (2) reducing exciton quenching occurring at the
interface between bare metal and the active layer by coating
AgNPs using a SiO2 insulating thin layer and thus increasing
exciton dissociation; (3) promoting hole transport and extrac-
tion by introducing “dopant” levels within the band gap of the
PEDOT:PSS and reducing hole trapping in the bare silver
surface; and (4) the way that Ag@SiO2 is positioned is less dis-
turbing to the morphology of the active layer, which is also a key
factor for a high PCE. The combination of all these benets
results in 25.4% improvement in photocurrent density and
19.2% in PCE when compared to the optimized control cell
without NPs. These results conrm that such a positioning of
Ag@SiO2 leads to an excellent OSC compared to placing the
Ag@SiO2NPs in the interface active layer/buffer layer, which is
the method adopted in the literature.30 This work offers a more
optimized design for the Ag@SiO2NPs integrated OSC. Addi-
tionally, the optical and electrical effects of Ag@SiO2 NPs were
systematically explored to contribute to a more comprehensive
understanding of the mechanism of improvement of perfor-
mance for the Ag@SiO2NPs integrated OSC.

2. Experimental
2.1. Synthesis of metal nanoparticles
The AgNPs and the Ag@SiO2NPs were synthesized using a
simple and low cost wet chemical method.31,32 Firstly, the
ethylene glycol reduction agent was used to reduce the silver
ions (Ag+) in silver nitrate forming AgNPs in the presence of Fig. 1Cross-sectional view of different OSCs made with (a)
poly(vinyl pyrrolidone) (PVP) at a high temperature. The size of Ag@SiO2NPs and (b) AgNPs together with (c) a control cell without
the AgNPs could be tuned by controlling the amount of PVP NPs.

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. C, 2015, 3, 1082–1090 | 1083
 View Article Online

Journal of Materials Chemistry C Paper

was spin-coated onto PEDOT:PSS using the P3HT (9 mg ml1)

and PC60BM (9 mg ml1) blend solution in chlorobenzene at 600
rpm for 30 s. Finally, the devices were completed by deposition of
a top electrode (LiF (1 nm)/Al (100 nm)), which was then ther-
mally evaporated in a vacuum with a base pressure of 5  104
Pa. A set of structurally identical control OSCs with and without
AgNPs were also fabricated for comparison studies. The cross-
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

sectional view of the control cells is shown in Fig. 1(b) and (c).

2.3. Characterization
Scanning transmission electron microscope (STEM) images of
the AgNPs and Ag@SiO2NPs were measured using a h-order,
aberration corrected STEM (Nion UltraSTEM 100) with a cold
eld emission electron source. The absorption spectra of the
pristine AgNPs and Ag@SiO2NPs in ethanol solution were
measured using a UV-visible (UV-vis) spectrophotometer (Hita-
chi U-3900) at room temperature. The absorption spectra of the
multilayer lm (MLF) of ITO/PEDOT:PSS/P3HT:PC60BM with
and without NPs in the PEDOT:PSS layer were also recorded
using the same measurement system. Current density–voltage
(J–V) characteristics of devices in the dark and under illumi-
nation were measured using a programmable SourceMeter
(Keithley 2400). The cells were illuminated using an AM 1.5G
solar simulator (ABET Technologies Sun 3000 Solar Simulator)
with an irradiation intensity of 100 mW cm2, calibrated using a
standard silicon cell. External quantum efficiency (EQE) as a
function of wavelength was recorded using ZOLIX CSC1011,
where the light source is provided by an Ushio UXL-553 xenon
short arc lamp.

3. Results and discussion

STEM images of AgNPs and Ag@SiO2NPs are shown in Fig. 2(a) Fig. 2 (a) STEM images of AgNPs with a diameter in the range from 15
nm to 36 nm and (b) STEM images of Ag@SiO2NPs covered with a SiO2
and (b). It is found that the diameter size distribution of the
shell having an average layer thickness from 6 nm to 8 nm; (c)
majority of nanoparticles is over the range 15 nm to 36 nm, as absorption spectra of AgNPs and Ag@SiO2NPs in ethanol with the
shown in Fig. 2(a). But there exist some small sized NPs that LSPR peaks located at 404 nm and 414 nm, respectively.
have diameters in the range from 2 nm to 10 nm in the samples.
It is also clearly seen from Fig. 1(b) that the Ag@SiO2NPs were
successfully synthesized. The silica shell with an average layer
thickness from 6 nm to 8 nm is coated uniformly and almost a constant of about 0.6 V for all devices, while the short
completely on the surface of the individual AgNPs. Fig. 2(c) circuit current density (Jsc) increases dramatically aer incor-
shows that the UV-vis absorption spectra of Ag@SiO2NPs and porating Ag@SiO2NPs in the PEDOT:PSS layer. Jsc increases
AgNPs dispersed in ethanol solution. The absorption peak of gradually with the doping ratio of Ag@SiO2NPs increase from
AgNPs from the LSPR is localized at 404 nm, while that of 0 to 3 mg ml1, but begins to decrease as the doping ratio of
Ag@SiO2NPs is red shied to 414 nm. Ag@SiO2NPs increases to 4 mg ml1. For 3 mg ml1 Ag@SiO2
To investigate the effect of Ag@SiO2NPs on the photovoltaic NPs, Jsc is increased to 9.67 mA cm2 with a 25.4% enhance-
performance of OSCs, devices made with and without ment factor relative to the control cell without NPs. The exis-
Ag@SiO2NPs in the PEDOT:PSS layer were measured for tence of the optimal doping ratio is presumably because of the
comparison studies. Fig. 3(a) and (b) present J–V characteristics transformation of incident light to thermal loss caused by the
measured for OSCs made with different doping ratios of intrinsic absorption of Ag@SiO2NPs; alternatively, it could be
Ag@SiO2NPs in the PEDOT:PSS layer under illumination and in associated with the disturbed transport and extraction of charge
the dark. carriers by Ag@SiO2NPs, which will be discussed below. On the
Performance parameters of OSCs made with different whole, the ll factor (FF) decreases with the increase in the
doping ratios of Ag@SiO2NPs in PEDOT:PSS, and the optimized doping ratio of Ag@SiO2NPs. It has been well known that the FF
control cells with and without incorporating AgNPs are is associated with series and shunt resistances, which can be
summarized in Table 1. The open circuit voltage (Voc) remains obtained from the inverse slope of illuminated J–V curves at

1084 | J. Mater. Chem. C, 2015, 3, 1082–1090 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry C

short-circuit and open-circuit voltage, respectively.33 The

smaller the series resistance (Rs) and the larger the shunt
resistance (Rsh), the larger the FF. One can see from Table 1 that
the series resistances are slightly reduced, while the shunt
resistances are greatly decreased as the doping ratio of
Ag@SiO2NPs increases, leading, as a result, to the decreased FF.
But the enhancement of Jsc overwhelms the reduction of FF so
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

that a PCE of 3.35% with 19.2% enhancement factor was

obtained for the optimized Ag@SiO2NPs-integrated cell relative
to the control cell without NPs. Fig. 3(b) presents the J–V char-
acteristics of the OSCs made with different doping ratios of
Ag@SiO2NPs in the dark. It can be seen from Fig. 3(b) that the
current rectication ratio of OSCs decreases and leakage
current increases aer incorporating Ag@SiO2NPs. The highly
elevated leakage current is consistent with the signicant
decrease of the shunt resistance of the Ag@SiO2-integrated
OSCs. Incorporating the Ag@SiO2NPs into the PEDOT:PSS
buffer layer could induce some interface defects. Furthermore,
the small AgNPs existing in the sample could not be coated by
SiO2, which introduces the surface recombination centers.
These defects and impurities result in the recombination
current, which is responsible for the elevated leakage current
density at the driving voltage of less than 0.5 V and then the
reduced shunt resistance.
As a comparison, Fig. 3(c) and (d) present J–V characteristics
of the optimal OSCs with and without NPs under illumination
and in the dark. The Jsc of the optimized OSC with Ag@SiO2NPs
is noticeably higher than that of the optimized control cells with
AgNPs. It is also found from Fig. 3(d) that the AgNP integrated
OSCs possess a smaller shunt resistance and higher leakage
current than that of the cells with Ag@SiO2NPs. This may be
ascribed to the larger loss of charges because of recombination
and because they are trapped in the bare metal surface.
To better elucidate the increment of Jsc, the EQE spectra of
the optimized cells with and without NPs were measured, as
shown in Fig. 4(a). It was observed that the larger enhancement
in EQE is achieved by incorporating Ag@SiO2NPs rather than
AgNPs, which is in agreement with the change tendency of Jsc.
The EQE enhancement factor is calculated by subtracting the
EQE of the cell without NPs from that of the cell with
Ag@SiO2NPs and then dividing by the EQE of the cell without
NPs, as shown in the inset of Fig. 4(a). A broad increase in EQE
over a wavelength range from 300 to 700 nm is displayed for the
Ag@SiO2NP integrated OSCs relative to the control cell without
NPs.
To understand the enhancement in EQE, the optical effects
of NPs were investigated. The absorption spectra of the MLF of
glass/ITO/PEDOT:PSS/P3HT:PC60BM with and without NPs in
PEDOT:PSS were measured, as shown in Fig. 4(b). And the
absorption enhancement factor was obtained by subtracting the
absorption spectrum of the MLF without NPs from that of the
Fig. 3 Current density–voltage characteristics measured for solar
MLF with Ag@SiO2NPs and then dividing by the absorption
cells having different doping ratios of Ag@SiO2NPs in the PEDOT:PSS
layer under illumination (a) and in the dark (b). Current density–voltage spectrum of the MLF without NPs, as shown in the inset of
characteristics of the optimized solar cells without and with NPs in the Fig. 4(b). It was found that the EQE enhancement is highly
PEDOT:PSS layer under illumination (c) and in the dark (d), respectively. dependent on the absorption enhancement of the active layer,
but there also exists some mismatch between them. This means

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. C, 2015, 3, 1082–1090 | 1085
 View Article Online

Journal of Materials Chemistry C Paper

Table 1 Performance parameters of OSCs with different doping ratios of Ag@SiO2NPs in PEDOT:PSS under AM 1.5G illumination at 100 mW
cm2, and those of the optimized control cells with and without Ag NPs in PEDOT:PSS under the same condition

Device Jsc (mA cm2) Voc (V) FF (%) PCE (%) Rs (U cm2) Rsh (U cm2)

Ag@SiO2 1 mg ml1 8.02 0.599 62.2 2.99 11.55 1063.08

Ag@SiO2 2 mg ml1 9.04 0.600 60.4 3.28 11.14 950.30
Ag@SiO2 3 mg ml1 9.67 0.594 58.3 3.35 10.31 806.28
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

Ag@SiO2 4 mg ml1 8.77 0.600 55.2 2.91 12.47 588.23

Ag@NPs 2 mg ml1 9.13 0.592 57.0 3.08 11.69 664.61
Without NPs 7.71 0.600 60.9 2.81 12.65 1359.40

and AgNPs in PEDOT:PSS using the nite element method in

three-dimensional (3D) model, as shown in Fig. 5. The calcu-
lated device structure was completely the same as the experi-
mental device structure, where the AgNP has the diameter of 36
nm, while Ag@SiO2NPs have a 6 nm thick SiO2 shell and a core
with a diameter of 30 nm or 15 nm. These numerical methods
were veried by studying the problems discussed by Islam
et al.34 As shown in Fig. 5(a), the small sized Ag@SiO2NPs

Fig. 4 (a) EQE spectra of the optimized OSCs with and without NPs.
Inset: EQE enhancement factor. (b) UV-vis absorption spectra of the
multilayer film of ITO/PEDOT:PSS/P3HT:PC60BM without and with
NPs in PEDOT:PSS. Inset: absorption enhancement factor.

that EQE is also related to the electrical effects such as carrier

transport and collection, except for the light harvesting.
One can see from the inset of Fig. 4(b) that a broad increase
in absorption of P3HT:PC60BM over a wavelength range from
300 to 700 nm is obtained by incorporating Ag@SiO2NPs into
the MLF, mismatching with the narrow plasmonic resonance
spectrum of Ag@SiO2NPs (see Fig. 2(c)). It is suggested that this
inconsistency could originate from the far-eld scattering effect
of Ag@SiO2NPs. To conrm this hypothesis, the distributions of
the electric eld, |E|, at 580 nm under transverse-magnetic (TM) Fig. 5 Distribution of the electric field, |E|, at 580 nm under TM
polarized light were investigated for the OSCs with Ag@SiO2NPs polarized light for the OSC with small (a) and large (b) size Ag@SiO2NPs
together with AgNPs (c) in the PEDOT:PSS layer.

1086 | J. Mater. Chem. C, 2015, 3, 1082–1090 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry C

(15 nm@6 nm) incorporated into the PEDOT:PSS layer do not more important contributor to the light harvesting than the
contribute to the absorption enhancement because of the plasmonic near-eld enhancement.
lateral eld distribution feature of the LSPR along the In order to further explore the effects of NPs, the maximum
PEDOT:PSS layer, which is consistent with the results reported exciton generation rates (Gmax) were determined for the opti-
by Fung et al.29 However, when large size Ag@SiO2NPs (30 mized OSCs with and without NPs. The current density–voltage
nm@6 nm) were incorporated in the PEDOT:PSS layer, as characteristics of the optimized OSCs without and with NPs
illustrated in Fig. 5(b), the strong electric eld round the were measured under illumination and in the dark when the
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

Ag@SiO2NPs clearly penetrates into the active layer. But one can devices were bias swept from +1 to 10 V. Following the
see from Fig. 6(c) that the plasmonic near-eld around AgNPs is analytical approach reported by Wu et al.,35 the dependence of
far stronger than that around Ag@SiO2NPs. It is well known that the photocurrent density (Jph) on the effective voltage (Veff) is
the probability of transition of a molecule from one ground revealed, as shown in Fig. 6. The values of Jph and Veff were
state to the excited state is proportional to the square of electric determined using the equations Jph ¼ JL  JD and Veff ¼ Vo  Va,
eld strength the molecule experiences, so it can be deduced respectively, where JL and JD are the current densities under
that introducing AgNPs should lead to larger enhancement in illumination and in the dark, respectively, Vo is the voltage
the light absorption of the active layer. But the experimental when Jph equals zero and Va is the applied voltage. It can be seen
results are the reverse of the theoretical analysis [see Fig. 4(b)]. from Fig. 6(a) that the value of Jph reaches a saturated level at a
Usually for a metal NP, two factors contribute to the light sufficiently high Veff. Therefore the values of the saturation
absorption enhancement: rstly, local enhancement of the photocurrent density (Jsat) were determined and then Gmax was
electromagnetic eld in the vicinity of the metal NP; secondly, calculated using the equation Jsat ¼ qGmaxL, where q is the
the scattering lengthening of the optical path in the active layer. electronic charge and L is the thickness of the active layer, as
This inconsistency means that Ag@SiO2NPs induces much shown in Table 2. It can be seen that incorporating Ag@SiO2NPs
stronger far-eld scattering because of the presence of the SiO2 leads to the largest Gmax. The enhanced Gmax is associated with
shell. Furthermore, the scattering of AgNPs@SiO2 should be a the increased light absorption, which is consistent with the
absorption spectra characteristic of the MLF.
Based on the maximum exciton generation rates (Gmax), the
exciton dissociation probabilities [P(E, T)] were calculated,
which are related to the electric eld (E) and temperature (T), for
the optimized OSCs with and without NPs. For OSCs, when the
excitons are photogenerated, only a portion of them can be
dissociated into free carriers. Jph can be expressed using the
equation Jph ¼ qGmaxP(E, T)L. As a result, the value of P(E, T) at
any bias can be obtained from the plot of the normalized
photocurrent density (Jph/Jsat) with respect to Veff. Under the
short-circuit conditions (Va ¼ 0 V), P(E, T) is 76%, 80% and 85%
for the OSCs with and without AgNPs together with Ag@SiO2
NPs, respectively. It was found that the exciton dissociation
probability is higher for the OSC with integrated Ag@SiO2 NPs.
This means that the exciton quenching occurred at the interface
between bare AgNPs and that the active layer is reduced by
coating the SiO2 insulating layer. The enhanced charge sepa-
ration is also responsible for the enhanced EQE and then for Jsc
except for the increased light absorption.
To further clarify the electrical effects of the NPs, the
current–voltage characteristics of a hole-only device of ITO/
PEDOT:PSS (100 nm)/molybdenum trioxide (MoO3, 5 nm)/Ag
with and without Ag@SiO2NPs in PEDOT:PSS were investigated
by adjusting the doping ratio of Ag@SiO2NPs. The results
indicated that the current–density of the hole-only device rst
increases and then reduces with an increase in the doping ratio
of Ag@SiO2NPs at the same voltage, as shown in Fig. 7(a), which
is consistent with the change trend of the series resistance. This
means that the electrical performance of PEDOT:PSS can be
improved by incorporating an appropriate amount of
Ag@SiO2NPs. The improved electrical performance is presum-
Fig. 6 (a) Photocurrent density (Jph) plotted against effective bias (Veff)
ably attributed to the introduction of “dopant” levels36 within
and (b) exciton dissociation probability [P(E, T)] plotted against effec- the band gap of the PEDOT:PSS by introducing Ag@SiO2NPs,
tive bias (Veff) for the optimized OSCs with and without NPs. which could increase the hole transport and reduce the energy

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. C, 2015, 3, 1082–1090 | 1087
 View Article Online

Journal of Materials Chemistry C Paper

Table 2 Comparison of other parameters of the optimized OSCs with and without NPs

Device Jsat (A m2) Gmax (m3 s1) P(E, T) (%) Ideality factor (n)

Without NPs 91 2.585  1027 76 3.18

With Ag NPs 2 mg ml1 97 2.755  1027 80 2.07
With Ag@SiO2 NPs 3 mg ml1 105 2.982  1027 85 2.17
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

Ag@SiO2NPs was inserted at the ITO/PEDOT:PSS interface to

promote hole injection or extraction. However, too large an
amount of Ag@SiO2 NPs at the ITO/PEDOT:PSS interface
inhibits hole injection or extraction because of too large a
contact area between ITO and SiO2.
The current–voltage characteristics of a hole-only device of
ITO/PEDOT:PSS (100 nm)/MoO3 (5 nm)/Ag with Ag NPs and
Ag@SiO2NPs in PEDOT:PSS with the same doping ratio as the
optimized OSCs were also compared, as shown in Fig. 8. One
can see that the current–density of the hole-only device with
Ag@SiO2NPs is higher at the same voltage than that with
AgNPs. This means that the conductivity of PEDOT:PSS is lower
when AgNPs are introduced than if Ag@SiO2 NPs are used. This
unexpected result could originate from the higher degree of
charge loss because of surface recombination of bare metal.
In other words, the presence of the SiO2 shell could inhibit
hole-trapping and reduce the space charge accumulation and
thus increase the hole transport.
Another electrical parameter, the ideality factor (n), of the
optimized OSCs with and without NPs is derived from the J–V
characteristics in the dark by using the relationship:6
vV KT
¼n þ IRs
v ln I q
where q is the electronic charge, T is the absolute temperature, K
is the Boltzmann constant, n is the ideality factor, and Rs is the
series resistance. The n values were determined from the y-axis
intercept of the dV/d ln J–J plot. The values of n obtained are
Fig. 7 (a) Current–voltage characteristics of a hole-only device of presented in Table 2. For an ideal diode, the value of n is one,
ITO/PEDOT:PSS (100 nm)/MoO3 (5 nm)/Ag with and without
Ag@SiO2NPs in PEDOT:PSS at different doping ratios. (b) Current–
voltage characteristics of hole-only devices of ITO/Ag@SiO2/
PEDOT:PSS (100 nm)/MoO3 (5 nm)/Ag with different amounts of
Ag@SiO2NPs in the ITO/PEDOT:PSS interface.

barrier of hole injection from ITO to PEDOT:PSS. However, if

the doping ratio is too large, the possibility of contact between
Ag@SiO2 and ITO is increased, which leads to enhanced contact
resistance and thus, inhibition of the hole injection. In order to
conrm this concept, the current-voltage characteristics of hole-
only devices of ITO/Ag@SiO2NPs/PEDOT:PSS (100 nm)/MoO3(5
nm)/Ag were investigated, as shown in Fig. 7(b), where
Ag@SiO2NPs are located at the ITO/PEDOT:PSS interface by
spin-coating the ethanol solution of Ag@SiO2NPs with different
concentrations of 2, 6 and 10 mg ml1. One can see that the
conductivity of the hole-only devices is rst increased and then
Current–voltage characteristics of a hole-only device of ITO/
Fig. 8
decreased as the amount of Ag@SiO2NPs at the ITO/PEDOT:PSS PEDOT:PSS (100 nm)/MoO3 (5 nm)/Ag with and without NPs in
interface increases. This means that an appropriate amount of PEDOT:PSS with the same doping ratio as the optimized OSCs.

1088 | J. Mater. Chem. C, 2015, 3, 1082–1090 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry C

Table 3 Comparison of the performance of OSCs with Ag@SiO2NPs incorporated in different positions, within the buffer layer and at the
interface of the buffer layer/active layer

OSCs with Ag@SiO2 NPs Voc (V) Jsc (mA cm2) FF (%) PCE (%) Rs (U cm2) Rsh (U cm2)

At interface 0.592 9.47 60.0 3.28 9.45 533.72

Within PEDOT:PSS 0.594 9.67 58.3 3.35 10.31 642.07
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

but experimentally obtained value of n are greater than unity. A (2012DFR50460), Shanxi Natural Science Foundation
departure from unity gives an indication of the imperfections in (2010021023-2, 2011021022-2, 2012011020-4), New Teachers'
a diode. It was found that the degree of deviation of the Fund (20121402120017), Hong Kong Scholar Program
behaviour of the NP-integrated OSC is less than the control cell (XJ2013002), and Talents for Early Career Scheme – Tertiary
without NPs. Education Division, Shanxi Province.
Finally, we compared the performance of OSCs with
Ag@SiO2NPs incorporated in different positions including
within the buffer layer and at the interface of buffer layer/active
layer. The results conrmed that the performance of OSCs with Notes and references
Ag@SiO2NPs in the buffer layer is better than in any other 1 G. Li, R. Zhuand and Y. Yang, Nat. Photonics, 2012, 6, 153–
position, as shown in Table 3, because in such a position the 161.
Ag@SiO2NPs disturb the morphology of the active layer less 2 Z. He, C. Zhong, S. Su, M. Xu, H. Wu and Y. Cao, Nat.
than at the interface of buffer layer/active layer, which is also a Photonics, 2012, 6, 593–597.
key factor for obtaining a high PCE with OSCs. 3 G. M. Ng, G. M. Ng, E. L. Kietzke, T. Kietzke, L. W. Tan,
P. K. Liew and F. R. Zhu, Appl. Phys. Lett., 2007, 90, 103505.
4. Conclusions 4 Q. Gan, F. J. Bartoli and Z. H. Kafa, Adv. Mater., 2013, 25,
2377.
In summary, it was demonstrated that the performance of OSCs 5 H. A. Aatwater and A. Polman, Nat. Mater., 2010, 9, 205–213.
is improved by incorporating Ag@SiO2NPs within the 6 X. Z. Wang, X. Z. Wang, J. W. Ho, Q. Y. Yang, H. L. Tam,
PEDOT:PSS layer. The short-circuit current increases from 7.71 G. X. Li, K. W. Cheah and F. R. Zhu, Org. Electron., 2011,
mA cm2 for the optimized control cell without NPs to 9.67 mA 12, 1943.
cm2 for the optimized Ag@SiO2NPs-integrated cells, with an 7 S. S. Kim, S. I. Na, J. Jo, D. Y. Kim and Y. C. Nah, Appl. Phys.
enhancement factor of 25.4%. The current gain gives a rise of Lett., 2008, 93, 073307.
the conversion efficiency of 3.35%, with an enhancement factor 8 A. J. Morfa, K. L. Rowlen, T. H. Reilly, M. J. Romero and
of 19.2%. The enhancement is attributed to three factors: (1) the J. V. D. Lagemaat, Appl. Phys. Lett., 2008, 92, 013504.
increased light absorption by the near-eld enhancement effect 9 N. Kalfagia, P. G. Karagiannidis, C. Pitsalidis,
and the far-eld scattering of the large size Ag@SiO2NPs N. T. Panagiotopoulos, C. Gravalidis, S. Kassavetis,
because of the LSPR, thus, the scattering is a more important P. Patsalas and S. Logothetidis, Sol. Energy Mater. Sol. Cells,
contributor to the light harvesting; (2) the increased hole 2012, 104, 165–174.
transport and extraction, which is presumably explained by 10 H. S. Noh, E. H. Cho, H. M. Kim, Y. D. Han and J. Joo, Org.
introducing “dopant” levels within the band gap of the Electron., 2013, 14(1), 278–285.
PEDOT:PSS and reducing hole trapping in bare silver surface; 11 K. Leonard, Y. Takahashi, J. You, H. Yonemura, J. Kurawaki
(3) the enhanced exciton dissociation because of the reduction and S. Yamada, Chem. Phys. Lett., 2013, 584, 130–134.
of exciton quenching that occurred at the interface between the 12 B. Wu, T. Z. Oo, X. Li, X. Liu, X. Wu, E. K. L. Yeow, H. J. Fan,
bare AgNPs and the active layer by SiO2 coating. This research N. Mathews and T. C. Sum, J. Phys. Chem. C, 2012, 116,
not only offers a more optimized design for the Ag@SiO2NPs 14820–14825.
integrated OSC, but also contributes to a more comprehensive 13 X. Li, W. C. H. Choy, H. Lu, W. E. I. Sha and A. H. P. Ho, Adv.
understanding of the mechanism of the improvement of Funct. Mater., 2013, 23, 2728–2735.
performance. 14 J. H. Lee, J. H. Park, J. S. Kim, D. Y. Lee and K. Cho, Org.
Electron., 2009, 10, 416–420.
Acknowledgements 15 F. C. Chen, J. L. Wu, C. L. Lee, Y. Hong, C. H. Kuo and
M. H. Huang, Appl. Phys. Lett., 2009, 95, 013305.
This research work was nancially supported by National 16 D. H. Wang, D. Y. Kim, K. W. Choi, J. H. Seo, S. H. Im,
Natural Science Foundation of China (61274056, 11204205, J. H. Park, O. O. Parkand and A. J. Heeger, Angew. Chem.,
61205179, 21101111, 21071108, 11204202, 91233208), Key 2011, 123, 5633–5637.
Laboratory of Advanced Display and System Applications, 17 C. C. D. Wang, W. C. H. Choy, C. Duan, D. D. S. Fung,
Ministry of Education, Shanghai University, International W. E. I. Sha, F.-X. Xie, F. Huang and Y. Cao, J. Mater.
Science and Technology Cooperation Program of China Chem., 2012, 22, 1206–1211.

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. C, 2015, 3, 1082–1090 | 1089
 View Article Online

Journal of Materials Chemistry C Paper

18 J. L. Wu, F. C. Chen, Y. S. Hsiao, F. C. Chien, P. Chen, 27 M. Salvador, B. A. MacLeod, A. Hess, A. P. Kulkarni,

C. H. Kuo, M. H. Huang and C. S. Hus, ACS Nano, 2011, 5, K. Munechika, J. I. L. Chen and D. S. Ginger, ACS Nano,
959–967. 2012, 6(11), 10024–10032.
19 A. Kirkeminde, M. Retsch, Q. Wang, G. W. Xu, D. R. Q. Hui, 28 V. Jankovi, Y. (Michael) Yang, J. You, L. Dou, Y. Liu,
J. Wud and S. Q. Ren, Nanoscale, 2012, 4, 4421. P. Cheung, J. P. Chang and Y. Yang, ACS Nano, 2013, 7(5),
20 L. Lu, Z. Luo, T. Xu and L. Yu, Nano Lett., 2013, 13, 59–64. 3815–3822.
21 L. Qiao, D. Wang, L. Zuo, Y. Ye, J. Qian, H. Chen and S. He, 29 D. D. S. Fung, L. Qiao, W. C. H. Choy, C. Wang, W. E. I. Sha,
Published on 24 November 2014. Downloaded by Taiyuan University of Technology on 23/01/2015 05:16:52.

Appl. Energy, 2011, 88, 848–852. F. Xie and S. He, J. Mater. Chem., 2011, 21(41), 16349–16356.
22 V. Kochergin, L. Neely, C.-Y. Jao and H. D. Robinson, Appl. 30 H. Choi, J. P. Lee, S. J. Ko, J. W. Jung, H. Park, S. Yoo, O. Park,
Phys. Lett., 2011, 98, 133305. J. R. Jeong, S. Park and J. Y. Kim, Nano Lett., 2013, 13, 2204–
23 B. Wu, X. Y. Wu, C. Guan, K. F. Tai, E. K. L. Yeow, H. J. Fan, 2208.
N. Mathews and T. C. Sum, Nat. Commun., 2013, 4(2004), 1– 31 F. Zhang, G. B. Braun, Y. F. Shi, Y. C. Zhang, X. H. Sun,
7. N. O. Reich, D. Y. Zhao and G. Stucky, J. Am. Chem. Soc.,
24 K. Topp, H. Borchert, F. Johnen, A. V. Tunc, M. Knipper, 2010, 132(9), 2850–2851.
E. von Hauff, J. Parisiand and K. Al-Shamery, J. Phys. Chem. 32 J. P. Yang, F. Zhang, Y. R. Chen, S. Qian, P. Hu, W. Li,
A, 2010, 114, 3981–3989. Y. H. Deng, Y. Fang, L. Han, M. Luqmanb and D. Y. Zhao,
25 W. J. Yoon, K. Y. Jung, J. Liu, T. Duraisamy, R. Revur, Chem. Commun., 2011, 47, 11618–11620.
F. L. Teixeira, S. Sengupta and P. R. Berger, Sol. Energy 33 A. Moliton and J.-M. Nunzi, Polym. Int., 2006, 55, 583–600.
Mater. Sol. Cells, 2010, 94, 128–132. 34 K. Islam, A. Alnuaimi, E. Battal, A. K. Okyay and A. Nayfeh,
26 M. Xue, L. Li, B. J. T. d. Villers, H. Shen, J. Zhu, Z. Yu, Sol. Energy, 2014, 103, 263–268.
A. Z. Stieg, Q. Pei, B. J. Schwartz and K. L. Wang, Appl. 35 J.-L. Wu, F.-C. Chen, Y.-S. Hsiao, F.-C. Chien, P. Chen,
Phys. Lett., 2011, 98, 253302. C.-H. Kuo, M. H. Huang and C.-S. Hsu, ACS Nano, 2011,
5(2), 959–967.
36 K. Kim and D. L. Carroll, Appl. Phys. Lett., 2005, 87, 203113.

1090 | J. Mater. Chem. C, 2015, 3, 1082–1090 This journal is © The Royal Society of Chemistry 2015

View publication stats