Supporting Information · where J is the measured current density, L is the film thickness of the active layer, is the μ mobility of charge carrier, εr is the relative dielectric
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Supporting Information
16.7%-Efficiency Ternary Blended Organic Photovoltaic Cells with PCBM as
the Acceptor Additive to Increase Open-Circuit Voltage and Phase Purity
Ming-Ao Pan,a,b,c Tsz-Ki Lau,d Yabing Tang,e Yi-Ching Wu,f Tao Liu,b,c Kun Li,b Ming-Chou
Chen,f Xinhui Lu,*d Wei Ma*e and Chuanlang Zhan*a,b
a College of Chemistry and Environmental Science, Inner Mongolia Normal University, Huhhot
010022, China.
b CAS key Laboratory of Photochemistry, CAS Key Laboratory of Photochemistry, Institute of
Chemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China.
c Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water
Bay, Kowloon, Hong Kong, China.
d Department of Physics, Chinese University of Hong Kong, New Territories, Hong Kong, China.
e State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an
710049, China.
f Department of Chemistry and Research Center of New Generation Light Driven Photovoltaic
Modules, National Central University, Taoyuan, 32001, Taiwan.
Materials. The materials of PBDB-TF (PM6), PBDB-T-2Cl (PM7), IT-4F, Y6, PC61BM and PC71BM were purchased from Companies of Solarmer, Eflexpv, Shenzhen Ruixun, and One-Material. The charge transport material PDINO was bought from Solarmer company. The purity of all the materials was checked with 1H NMR spectra provided by the company. In fact, this is not enough. We did more about the availability of the commercial materials. Before ordering these commercial materials and starting to use them for the fabrications of solar cells, we had doubly checked their availability by fabricating the solar cells with these materials. First, we got a small amount of a batch of the material from the company, then, we fabricated the solar cell. When the device performance reached the expectation, i.e. the device Voc, Jsc, FF and PCE reached the expected values, we then purchased this batch of the material. For example, for PM6 and IT-4F, the PCE of PM6:IT-4F device has to reach over 13%, and for Y6, the PCE of PM6:Y6 device has to be about 15.5%. After getting the batch of the material, we again checked the availability by fabricating the solar cells. After the double checking and when the materials met the expectations, we then started the studies using these materials. PEDOT:PSS (Baytron Clevios P VP AI 4083, Germany) was obtained from Heraeus Group, imported by Solarmer company.
UV-Vis absorption spectra. Absorption spectra of donor polymers and acceptors in solid thin films were prepared by spin-coating their solutions in either chlorobenzene or chloroform atop the quartz glass substrates and measured on a Hitachi U-3010 UV−vis spectrophotometer.
Measurements of the Frontier molecular orbital energy levels. The HOMO/LUMO energy levels of the donor polymer and nonfullerene and fullerene acceptors were measured with differential pulse voltammetry (DPV) in their solutions and with cyclic voltammetry (CV) in their thin films, respectively. Differential pulse voltammetry experiments were performed under N2 with a conventional three-electrode configuration (a platinum disk working electrode, an auxiliary platinum wire electrode, and a non-aqueous Ag reference electrode, with a supporting electrolyte of 0.1 M tetrabutylammoniumhexafluorophosphate (TBAPF6) in a dry o-C6H4Cl2 solution at the specified temperature using a CHI621C Electrochemical Analyzer (CH Instruments). All electrochemical potentials were referenced to an Fc+/Fc internal standard (at 0.6 V). For the CV experiments, the compound was fully dissolved in N2-degassed anhydrous CHCl3
with a concentration of 10-4 M and then the solution was deposited onto the work electrode surface to form a thin solid film. CV traces were measured on an electrochemical workstation (CHI 660) at a scan rate of 50 mV/s using tetrabutylammonium tetrafluoroborate (Bu4NBF4) as the supporting electrolyte. A glassy carbon electrode, a Pt wire and an Ag/AgCl electrode were used as the working, counter and reference electrodes, respectively.
AFM characterizations. The AFM images were recorded using a Bruker multimode 8.0 instrument.
TEM characterizations. TEM experiments were performed on a JEM-2100 transmission electron microscope operated at 200 kV. For TEM experiments, the films were obtained by transferring the floated blend films from the water onto the Cu grid.
Solar cell fabrications and characterizations. Devices were fabricated on the Indium tin oxide (ITO) patterned glass with a conventional configuration of ITO/PEDOT:PSS/active layers/PDINO/Al. The ITO substrates were firstly cleaned by detergent, deionized water, acetone
S-2
and isopropanol in turn with sonication for 30 min respectively. The substrates were dried by nitrogen gas and then treated by UV-Ozone for 30 min before use. PEDOT:PSS was spin-coated onto the ITO substrates at 6000 rpm for 30 s, then the substrates were moved to oven and dried at 150 °C for 20 min. The PM6:IT-4F:PC71BM blends with different weight ratio were dissolved in chlorobenzene (CB) (the concentration of the donor was 10 mg mL-1). This solution was stirred at 50°C for about 10 hours. Different ratios of DIO (1,8-diiodooctane, 0.75%, 1.0%, 1.25%) were added to the solution before device fabrication. The blend solution were spin-coated on the top of PEDOT:PSS layer followed by a thermal annealing step (90°C-10min, 100°C-10min, 110°C-10min). The optimal method for active layer was to deposit the solution with 1.0% DIO at 2500 rpm and annealed at 100°C for 10 minutes. The PM6:Y6:PC71BM blends with different weight ratio were dissolved in chloroform (CF) (the concentration of the donor was 7 mg mL-1). This solution was stirred at 50°C for no less than 3 hours. Different ratios of CN (1-chloronaphthalene, 0.25%, 0.5%, 0.75%) were added to the solution before device fabrication. The blend solution were spin-coated on the top of PEDOT:PSS layer followed by a thermal annealing step (80°C-10min, 90°C-10min, 100°C-10min). The optimal method for active layer was to deposit the solution with 0.5% CN at 3000 rpm and annealed at 90°C for 10 minutes. PM7 was then used to replace the PM6 to fabricate the devices with the same procedure. The optimal thickness of the active layer measured by a Bruker Dektak XT stylus profilometer was about 100 nm. Atop the active layer, a thin electron transporting layer of PDINO (1.0 mg mL-1 in methanol, 3000rpm for 30 s, about 15 nm) was spin-coated, and the optimal rotating speed was about 3000 rpm. Finally, the Al electrodes (ca. 90 nm) were thermally deposited on the top of devices. The effective device area is 4 mm2 with a mask. The current density-voltage (J-V) curves were measured in a nitrogen glove box and were conducted on a computer-controlled Keithley 2400 source measure unit under AM 1.5G (calibrated to be 100 mW/cm2 with a reference silicon cell) using a solar illumination (AAA grade, XES-70S1). The EQE measurements were performed under the ambient condition with the as-fabricated solar cell using a QE-R3011 instrument (Enli Technology Co. Ltd., Taiwan).
SCLC measurements. The electron and hole mobility were measured by using the method called space-charge limited current (SCLC) for electron-only and hole-only devices. The structure of electron-only devices was ITO/titanium (diisopropoxide) bis(2,4-pentanedionate) (TIPD)[1]/active layer/PDINO/Al and the hole-only devices were fabricated with the structure of ITO/PEDOT:PSS/active layer/Au. The charge carrier mobility was determined by fitting the dark current to the model of a single carrier SCLC according to the Mott–Gurney law: J = 9ε0εrμV2/8L3, where J is the measured current density, L is the film thickness of the active layer, μ is the mobility of charge carrier, εr is the relative dielectric constant of the transport medium component, and ε0 is the permittivity of vacuum (8.85419×10–12 CV–1m–1), V is the difference of applied voltage (Vapp) and offset voltage (VBI). The mobility of charge carriers can be calculated from the slope of the J1/2 ~ V curves.
GIWAXS measurements. GIWAXS data were carried out with a Xeuss 2.0 SAXS/WAXS laboratory beamline using a Cu X-ray source (8.05 keV, 1.54 Å) and Pilatus3R 300K detector. The incidence angle is 0.2°.
S-3
Supporting figures
Figure S1. Traces of differential pulse voltammetry (DPV) of IT-4F, Y6, PC61BM, PC71BM, PM6, and PM7.
0.0 0.5 1.0 1.5 2.02.52.01.51.00.50.0
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
IT-4F Oxidation
Fc/Fc+
1.38
EHOMO= 4.2+1.38 = 5.58 eV
1.38
EHOMO= 4.2+1.38 = 5.58 eV
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
0.80.70.60.50.40.30.2
IT-4F Reduction
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
0.40
ELUMO= 4.2-0.40 = 3.80 eV
0.0 0.5 1.0 1.5 2.03.02.52.01.51.00.50.0
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
Y6 Oxidation
Fc/Fc+
EHOMO= 4.2+1.21 = 5.41 eV
1.21
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
0.80.70.60.50.40.30.2
Y6 ReductionCu
rrent
(x10
-6 A
)
Oxidation Potential (V)
0.44
ELUMO= 4.2-0.44 = 3.76 eV
0.0 0.5 1.0 1.5 2.0
43210
-1
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
PC61BM OxidationFc/Fc+
EHOMO= 4.2+1.73 = 5.93 eV
1.73
-2.0 -1.5 -1.0 -0.5 0.0
543210
-1
PC61BM ReductionCurre
nt (x
10-6 A
)
Oxidation Potential (V)
0.54
ELUMO= 4.2-0.54 = 3.66 eV
0.0 0.5 1.0 1.5 2.0
43210
-1
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
PC61BM OxidationFc/Fc+
EHOMO= 4.2+1.62 = 5.82 eV
1.62
-2.0 -1.5 -1.0 -0.5 0.0
543210
-1
PC61BM ReductionCurre
nt (x
10-6 A
)
Oxidation Potential (V)
0.56
ELUMO= 4.2-0.56 = 3.64 eV
0.0 0.5 1.0 1.5 2.06543210
-1
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
PM6 Oxidation
Fc/Fc+
EHOMO= 4.2+1.26 = 5.46 eV
1.26
0.0 0.5 1.0 1.5 2.0543210
-1
Curre
nt (x
10-6 A
)
Oxidation Potential (V)
PM7 Oxidation
Fc/Fc+
EHOMO= 4.2+1.26 = 5.46 eV
1.26
The HOMO energy was calculated from the first oxidation peak potentail and the LUMO energy is estimated from the first reduction peak potential. Ferrocene was used as the internal referecene compound which gives an oxidation peak at 0.6 V. For IT-4F, Y6, PC61BM, and PC71BM, the DPV experiments were performed in a o-C6H4Cl2 solution at the room temperature. While for polymers PM6 and PM7, the DPV experimets were carried out at 100°C (5 mg in 30 mL) of o-C6H4Cl2 since the polymers donot dissolve well at room temperature.
S-4
Figure S2. Traces of cyclic voltammetry (CV) of IT-4F, Y6, PC61BM, PC71BM, PM6, and PM7.
-0.5 0.0 0.5 1.0 1.5 2.0-4.0E-5
0.0
4.0E-5
8.0E-5
VOnset=1.35 V EHOMO=-5.71 eV
VOnset=-0.21 V ELUMO=-4.15 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Potential (V vs. Ag/AgCl (saturated KCl))
Cur
rent
(A)
IT4F
-0.5 0.0 0.5 1.0 1.5 2.0
-2.0E-4-1.0E-4
0.01.0E-42.0E-43.0E-44.0E-4
VOnset=1.26 V EHOMO=5.62 eV
Cur
rent
(A)
Potential (V vs. Ag/AgCl (saturated KCl))
Y6
VOnset=-0.25 V ELUMO=-4.11 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
-1.0E-4
-5.0E-5
0.0
5.0E-5
1.0E-4
VOnset=1.49 V EHOMO=-5.85 eVVOnset=-0.47 V ELUMO=-3.89 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Potential (V vs. Ag/AgCl (saturated KCl))
Cur
rent
(A)
PC61BM
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
-1.0E-4
0.0
1.0E-4
2.0E-4
3.0E-4
VOnset=1.60 V EHOMO=-5.96 eV
VOnset=-0.46 V ELUMO=-3.90 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Potential (V vs. Ag/AgCl (saturated KCl))
Cur
rent
(A)
PC71BM
-2 -1 0 1 2
-1.0E-4
0.0
1.0E-4
2.0E-4
3.0E-4
4.0E-4
5.0E-4
VOnset=-0.75 V ELUMO=-3.61 eV
VOnset=1.18 V EHOMO=-5.54 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Potential (V vs. Ag/AgCl (saturated KCl))
Cur
rent
(A)
PM6
-2 -1 0 1 2
-2.0E-4
0.0
2.0E-4
4.0E-4
6.0E-4
8.0E-4
1.0E-3
VOnset=1.16 V EHOMO=-5.52 eV
VOnset=-0.81 V ELUMO=-3.55 eV
EHOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Potential (V vs. Ag/AgCl (saturated KCl))
Cur
rent
(A)
PM7
Ferrocene was used as the reference. Its CV curve was measured under the same conditions. The reversible curve of the CV of ferrocene is shown in the middle of each figure. The LUMO and HOMO energy levels of each compound were calculated according to Eq. EHOMO/LUMO = e(Eox/red
+ 4.8 − EFc/Fc+), here, Eox/red is the oxidation/reduction onset obtained from the CV curve for each compound; EFc/Fc+
is 0.44 eV, the average of oxidation and reduction peak potentials of ferrocene; 4.8 eV is the energy difference between the energy level of ferrocene and the vacuum.
S-5
Figure S3. Scanning copy of the certificate report of the best device with PM6:Y6:PC71BM (1:1.2:0.2) as the active layer and PDINO as the ETL obtained from the National Institute of Metrology (NIM), China.
The certificated value is obtained with the encapsulation of the solar cell device and measured under the ambient condition. The value is lower than that measured from the unencapsulated device in the nitrogen-filling glove box. The decreased performance can be due to the effects from the encapsulation process.
S-6
Figure S4. Traces of cyclic voltammetry of the binary blends of IT-4F:PC61BM (a), IT-4F:PC71BM (b), and Y6:PC71BM (c) with the ratios changing from 1:0 to 1:0.1, 1:0.2, and 1:0.3, respectively.
-40.0μ0.0
40.0μ80.0μ
0.0
40.0μ
80.0μ
-40.0μ0.0
40.0μ80.0μ
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0
-50.0μ0.0
50.0μ100.0μ
HOMO=-5.72 eV
HOMO=-5.69 eV
HOMO=-5.71 eV
LUMO=--3.97 eV
LUMO=--4.09 eV
LUMO=--4.13 eV
HOMO=-5.73 eV
LUMO=--4.15 eV
HOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
IT-4F
Cur
rent
(A)
IT-4F:0.1PC61BM
(a)
IT-4F:0.2PC61BM
IT-4F:0.2PC61BM
Potential (V vs. Ag/AgCl (saturated KCl))
-40.0μ0.0
40.0μ80.0μ
-40.0μ0.0
40.0μ80.0μ120.0μ
-40.0μ0.0
40.0μ80.0μ120.0μ
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0-50.0μ
0.050.0μ100.0μ150.0μ
IT-4F
Cur
rent
(A)
IT-4F:0.1PC71BM
(b)
LUMO=--4.00 eV
LUMO=--4.01 eV
HOMO=-5.70 eV
HOMO=-5.71 eV
LUMO=--4.15 eVHOMO=-5.71 eV
HOMO=-5.72 eV
LUMO=--3.99 eV
HOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
IT-4F:0.2PC71BM
IT-4F:0.3PC71BM
Potential (V vs. Ag/AgCl (saturated KCl))
-200.0µ0.0
200.0µ400.0µ
-400.0µ0.0
400.0µ800.0µ
-400.0µ0.0
400.0µ800.0µ
1.2m
-1.0 -0.5 0.0 0.5 1.0 1.5 2.0
-400.0µ0.0
400.0µ800.0µ
1.2m
LUMO=--3.92 eV
LUMO=--3.91 eV
LUMO=--4.11 eV
HOMO=-5.66 eV
HOMO=-5.65 eV
HOMO=-5.62 eV
LUMO=--3.90 eV
HOMO=-5.66 eV
HOMO/LUMO=e(Eox/red+4.8-EFc/Fc+)
Y6
Curre
nt (A
)
Y6:0.1PC71BM
(c)
Y6:0.2PC71BM
Y6:0.3PC71BM
Potential (V vs. Ag/AgCl (saturated KCl))
Figure S5. The J0.5 – V plots for calculations of the electron (a, c, e, and g) and hole (b, d, f, and h) mobilities: PM6:IT-4F based (a and b), PM7:IT-4F based (c and d), and PM6:Y6 based (e and f).
0 1 2 3 4 50
20
40
60
80
PM6:IT4FPM6:IT4F:0.2PC61BMPM6:IT4F:0.2PC71BMJ0.
5 (mA
0.5 /c
m)
Voltage (V)
(a)
0 1 2 3 4 5
0
20
40
60
80
100(b)
J0.5 (m
A0.
5 /cm
)
PM6:IT4FPM6:IT4F:0.2PC61BMPM6:IT4F:0.2PC71BM
Voltage (V)
0 1 2 3 4 5
0
20
40
60
80
100(c)
J0.5 (m
A0.
5 /cm
)
PM7:IT4FPM7:IT4F:0.2PC61BMPM7:IT4F:0.2PC71BM
Voltage (V)
0 1 2 3 4 5
0
20
40
60
80(d)
J0.5 (m
A0.
5 /cm
)
PM7:IT4FPM7:IT4F:0.2PC61BMPM7:IT4F:0.2PC71BM
Voltage (V)
S-7
Figure S6. Photocurrent-density (Jph) – effective voltage (Veff) curves of the optimized binary and ternary devices: PM6:IT-4F based (a), PM7:IT-4F based (b), PM6:Y6 based (c), and PM7:Y6 based (d).
0 1 2 3 4 50
20
40
60
80
PM6:Y6PM6:Y6:0.2PC71BM
J0.5 (m
A0.
5 /cm
)
Voltage (V)
(e)
0 1 2 3 4 50
10203040506070(f)
J0.5 (m
A0.
5 /cm
)
PM6:Y6PM6:Y6:0.2PC71BM
Voltage (V)
0.1 1
10
20
J ph (m
A/c
m2 )
Veff (V)
PM6:IT4F PM6:IT4F:0.2PC61BM PM6:IT4F:0.2PC71BM
(a)
0.1 1
10
20(b)
Veff (V)
J ph (m
A/c
m2 )
PM7:IT4F PM7:IT4F:0.2PC61BM PM7:IT4F:0.2PC71BM
0.1 1
10
20
30
J ph (m
A/c
m2 )
(c)
Veff (V)
PM6:Y6 PM6:Y6:0.2PC71BM
0.1 1
10
20
30
J ph (m
A/c
m2 )
(d)
Veff (V)
PM7:Y6 PM7:Y6:0.2PC71BM
S-8
Figure S7. The plots of short-circuit current-density (Jsc) (a, c, e, and g) and open-circuit voltage (Voc) (b, d, f, and h) vs. light intensity of the optimized binary and ternary devices: PM6:IT-4F based (a and b), PM7:IT-4F based (c and d), PM6:Y6 based (e and f), and PM7:Y6 based (g and h).
1 10 1000.1
1
10
PM6:IT4F PM6:IT4F:0.2PC61BM PM6:IT4F:0.2PC71BM
J sc (m
A/c
m2 )
Light Intensity (mW/cm2)
(a)
1 10 1000.66
0.72
0.78
0.84(b)
PM6:IT4FPM6:IT4F:0.2PC61BMPM6:IT4F:0.2PC71BM
Voc
(V)
Light Intensity (mW/cm2)
1 10 1000.1
1
10(c)
J sc (m
A/c
m2 )
PM7:IT4F PM7:IT4F:0.2PC61BM PM7:IT4F:0.2PC71BM
Light Intensity (mW/cm2)
1 10 1000.68
0.72
0.76
0.80
0.84
0.88(d)V
oc (V
)
PM7:IT4F PM7:IT4F:0.2PC61BM PM7:IT4F:0.2PC71BM
Light Intensity (mW/cm2)
1 10 1000.1
1
10(e)
PM6:Y6 PM6:Y6:0.2PC71BM
J sc (m
A/c
m2 )
Light Intensity (mW/cm2)1 10 100
0.68
0.72
0.76
0.80
0.84
0.88(f)
PM6:Y6 PM6:Y6:0.2PC71BM
Voc
(V)
Light Intensity (mW/cm2)
1 10 1000.1
1
10
(g)
Light Intensity (mW/cm2)
J sc ( m
A/c
m2 )
PM7:Y6 PM7:Y6:0.2PC71BM
1 10 1000.72
0.76
0.80
0.84
0.88(h)
PM7:Y6 PM7:Y6:0.2PC71BM
Voc
(V)
Light Intensity (mW/cm2)
S-9
Figure S8. Two-dimensional pictures of the GIWAXS data (a-d) and the linecut profiles (e-h) of
the pure Y6 (a and e), PC71BM (b and f), IT-4F (c and g), and PM6 (d and h) films.
S-10
Table S1. The Optimization of the Solar Cell Devices.
Entries IT4F:PC71BM DIO% Annealing
temperature (℃)
Voc (V)
PCE (%)
FF (%)
Jsc (mA/cm2)
1.1 1.0:0 1 100 0.843 13.06 75.16 20.63
1.2 1.0:0.1 1 100 0.850 13.51 75.56 21.03
1.3 1.0:0.2 1 100 0.858 13.92 76.74 21.15
1.4 1.0:0.3 1 100 0.865 13.61 74.66 21.07
2.1 1.0:0.2 0.75 100 0.863 13.41 74.66 20.89
2.2 1.0:0.2 1 100 0.857 13.92 76.72 21.11
2.3 1.0:0.2 1.25 100 0.853 13.10 73.05 21.03
3.1 1.0:0.2 1 90 0.866 12.97 72.13 20.77
3.2 1.0:0.2 1 100 0.859 13.90 76.59 21.13
3.3 1.0:0.2 1 110 0.851 13.77 76.07 21.28
Entries Y6:PC71BM CN% Annealing
temperature (℃)
Voc (V)
PCE (%)
FF (%)
Jsc (mA/cm2)
1.1 1.2:0 0.5 90 0.848 15.47 74.58 24.47
1.2 1.2:0.1 0.5 90 0.853 16.11 76.17 24.78
1.3 1.2:0.2 0.5 90 0.861 16.70 77.24 25.16
1.4 1.2:0.3 0.5 90 0.864 16.31 75.63 24.97
2.1 1.2:0.2 0.25 90 0.865 16.11 74.86 24.88
2.2 1.2:0.2 0.5 90 0.859 16.69 77.19 25.18
2.3 1.2:0.2 0.75 90 0.856 16.49 76.15 25.29
3.1 1.2:0.2 0.5 80 0.863 16.28 75.83 24.87
3.2 1.2:0.2 0.5 90 0.860 16.65 77.05 25.13
3.3 1.2:0.2 0.5 100 0.854 16.40 76.72 25.02
S-11
Reference [1] Z. Tan, W. Zhang, Z. Zhang, D. Qian, Y. Huang, J. Hou, Y. Li, Adv. Mater. 2012, 24, 1476.