DOI: 10.1039/C3CC48615J
ChemComm
COMMUNICATION%
Journal)Name%
HRMS (Figure S3) and infrared (IR) spectrum (Figure S4). The monochromatic photon-to-electron conversion efficiency (IPCE)
IR band of the trisilanol silsesquioxane cage appears at 1118 as a function of wavelength. The IPCE values in he visible
cm which is accompanied by CH3 stretching (2951 cm ) and spectral range were the lowest for the POSS-8EsPImI/I2
-1
-1
-1 7
the corresponding deformational mode (1234 cm ). IR band at among the three electrolytes. The IPCE increased significantly
-
1
1
1
752 cm is attributed to C=O in the ester group. The band at upon addition of LiI to POSS-8EsPImI/I2, which decreased to
561 cm is assigned to the C=C stretching in the imidazolium some extent upon further addition of NMBI. The IPCE order is
-1
-1
ring, while the bands at 3143 and 3082 cm are attributed to C- the same as that of Jsc.
H stretching in the imidazolium ring. The decomposition
8
o
temperature of POSS-8EsPImI is determined to be 120 C by
(
a)
thermogravimetric analysis (TGA, Figure S5a). POSS-
EsPImI is amorphous as revealed by the differential scanning
calorimetry (DSC) curve (Figure S5b), where glass transition
8
1
5
0
5
0
o
takes place in the range of 52 C. This feature avoids
crystallization of the ionic conductor during electrolyte
injection followed by drying and facilitates pore filling of TiO2
with the solid electrolyte. The pore filling and interfacial
contact between solid electrolyte and TiO2 are expected to
improve owing to the amorphous feature and the presence of
Si-O bonds.
The conductivity of the solid electrolyte, which is crucial to
the performance of ssDSSCs, was measured by electrochemical
impedance spectroscopy (Figure S6). The conductivity of
1
POSS-8EsMImI/I2
POSS-8EsMImI/I /LiI
2
POSS-8EsMImI/I /LiI/NMBI
2
-1
POSS-8EsPImI was 0.061 mS cm , which was enhanced
0.0 0.1
0.2 0.3 0.4 0.5
Voltage / V
0.6 0.7
-1
significantly to 0.17 mS cm when doping iodine to POSS-
EsPImI (molar ratio, POSS-8EsPImI/I2 = 1.5/1). This
remarkable increase in ionic conductivity originates from the
8
−
(
b)
formation of polyiodides such as I3 ions by means of reaction
80
60
40
20
0
between iodide and iodine. The formation of polyiodides
facilitates the charge transfer along the polyiodide chain by the
Grotthus-type exchange mechanism, which is the major reason
for the remarkable enhancement of conductivity. Introduction
of LiI to the above mixture (molar ratio, POSS-8EsPImI/I2/LiI
9
-
1
= 1.5/1/3) further increased the conductivity to 0.78 mS cm .
This is attributed to the increased concentration of iodide and
+
-1
Li ions. The conductivity of 8EsPImI/I2/LiI was 0.77 mS cm ,
comparable to that of POSS-8EsPImI/I2/LiI. This suggests that
linking EsPImI to POSS hardly influenced the conductivity of
the ionic liquid. When NMBI, which like 4-tertbutylpyridine is
a useful additive in electrolyte for voltage enhancement, was
added to the above mixture electrolyte (molar ratio, POSS-
POSS-8EsMImI/I2
POSS-8EsMImI/I /LiI
2
10
POSS-8EsMImI/I /LiI/NMBI
2
300
400
500
600
λ / nm
700
800
8
EsPImI/I2/LiI/NMBI = 1.5/1/3/10), the conductivity changed
negligibly as shown in Figure S6. It is noted that the above-
mentioned molar ratios in the solid electrolyte were determined
by optimizing the solar cell performance.
!
Figure%2.%J<V%curves%(a)%and%IPCE%spectra%(b)%for%the%ssDSSCs%
The ssDSSCs based on the solid electrolytes with various
compositions were tested under illumination of AM1.5G
simulated solar light (100 mW cm ). For POSS-8EsPImI/I2
The filling of pores in TiO2 film with solid electrolytes is
crucial to achieve good performance. Figure 3 shows the
scanning electronic microscopy (SEM) images for TiO2 films
before and after solid electrolyte filling. The surface of bare
TiO2 film is porous, which becomes non-porous and smooth
when the solid electrolyte is filled in the film. This indicates
that the pores in TiO2 are filled with the solid electrolyte very
well, which accounts for the good photocurrent and efficiency.
However, for the solid ionic liquid (EsPImI, Figure S1) that is
not linked to the POSS, pore filling is not good due to the
serious ionic crystallization (Figure S7). For this reason,
EsPImI only produces η of 2.07% without adding the crystal
growth inhibitor and produces 3.81% efficiency with small
amount of crystal growth inhibitor. Despite the comparable
conductivity, the solid electrolyte based on POSS-8EsPImI
yields much higher efficiency than the corresponding solid
electrolyte based on EsPImI without or with crystal growth
inhibitor. It is evident the combination of POSS and ionic liuid
can improve photovoltaic performance remarkably.
-2
electrolyte (molar ratio, 1.5/1), the DSSC produced short-circuit
−2
photocurrent density (Jsc) of 11.08 mA cm , open-circuit
photovoltage (Voc) of 0.60 V, and fill factor (FF) of 0.62,
corresponding to power conversion efficiency (η) of 4.12%
(
Figure 2a). The η was improved to 5.81% (Jsc = 15.63 mA
−2
cm , Voc = 0.63 V, FF = 0.59) using the POSS-8EsPImI/I2/LiI
solid electrolyte (molar ratio, 1.5/1/3), as shown in Figure 2a.
The increase in η is mainly attributed to the enhancement of Jsc
because of the positive shift of conduction band edge of TiO2
caused by the adsorption of Li ions on TiO2 surface and the
increased conductivity as well. When NMBI was added to the
above electrolyte (molar ratio of POSS-8EsPImI/I2/LiI/NMBI
was 1.5/1/3/10), Voc increased to 0.74 V but Jsc decreased to
1
Jsc is likely attributed to the negative shift of conduction band
edge of TiO2 caused by the basic NMBI. As a result, the η
was improved to 7.11%. Figure 2b displays the incident
+
11
−2
4.12 mA cm . The observed increase in Voc and decrease in
10
2
%|%J.!Name.,%2012,%00,%1<3%
This%journal%is%©%The%Royal%Society%of%Chemistry%2012%