Multifunctional Fullerene Derivative for Interface Engineering in Perovskite Solar Cells

Article abstract of DOI:10.1021/jacs.5b10614

In perovskite based planar heterojunction solar cells, the interface between the TiO₂ compact layer and the perovskite film is critical for high photovoltaic performance. The deep trap states on the TiO₂ surface induce several challenging issues, such as charge recombination loss and poor stability etc. To solve the problems, we synthesized a triblock fullerene derivative (PCBB-2CN-2C8) via rational molecular design for interface engineering in the perovskite solar cells. Modifying the TiO₂ surface with the compound significantly improves charge extraction from the perovskite layer. Together with its uplifted surface work function, open circuit voltage and fill factor are dramatically increased from 0.99 to 1.06 V, and from 72.2percent to 79.1percent, respectively, resulting in 20.7percent improvement in power conversion efficiency for the best performing devices. Scrutinizing the electrical properties of this modified interfacial layer strongly suggests that PCBB-2CN-2C8 passivates the TiO₂ surface and thus reduces charge recombination loss caused by the deep trap states of TiO₂. The passivation effect is further proven by stability testing of the perovskite solar cells with shelf lifetime under ambient conditions improved by a factor of more than 4, from ～40 h to ～200 h, using PCBB-2CN-2C8 as the TiO₂ modification layer. This work offers not only a promising material for cathode interface engineering, but also provides a viable approach to address the challenges of deep trap states on TiO₂ surface in planar perovskite solar cells.

Full text of DOI:10.1021/jacs.5b10614

Subscriber access provided by CARLETON UNIVERSITY
Article
A Multifunctional Fullerene Derivative for
Interface Engineering in Perovskite Solar Cells
Yaowen Li, Yue Zhao, Qi Chen, Yang (Michael) Yang, Yongsheng Liu, Ziruo
Hong, Zonghao Liu, Yao-Tsung Hsieh, Lei Meng, Yongfang Li, and Yang Yang
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b10614 • Publication Date (Web): 23 Nov 2015
Downloaded from http://pubs.acs.org on November 23, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
A Multifunctional Fullerene Derivative for Interface Engineer-
ing in Perovskite Solar Cells
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yaowen Li*,†,‡, Yue Zhao,† Qi Chen,‡ Yang (Michael) Yang,‡ Yongsheng Liu,‡ Ziruo Hong,*,‡
Zonghao Liu, ‡ Yao-Tsung Hsieh,‡ Lei Meng, ‡ Yongfang Li, *† Yang Yang*‡
†Laboratory of Advanced Optoelectronic Materials, College of Chemistry, Chemical Engineering and Materials Sci-
ence, Soochow University, Suzhou 215123, China
‡ Department of Materials Science and Engineering, University of California, Los Angeles, California 90095, United
States
KEYWORDS. Perovskite solar cell, Interface, Passivation, TiO2, Trap states, Fullerene derivative
ABSTRACT: In perovskite based planar heterojunction solar cells, the interface between TiO2 compact layer and perov-
skite film is critical for high photovoltaic performance. The deep trap states on TiO2 surface induce several challenging
issues, such as charge recombination loss and poor stability etc. To solve the problems, we synthesized a tri-block fuller-
ene derivative (PCBB-2CN-2C8) via rational molecular design for interface engineering in the perovskite solar cells. Modi-
fying TiO2 surface with the compound significantly improves charge extraction from perovskite layer. Together with its
uplifted surface work function, open circuit voltage and fill factor are dramatically increased from 0.99 V to 1.06 V, and
from 72.2% to 79.1%, respectively, resulting in 20.7% improvement in power conversion efficiency for the best performing
devices. Scrutinizing the electrical properties of this modified interfacial layer strongly suggests that PCBB-2CN-2C8 pas-
sivates TiO2 surface and thus reduces charge recombination loss caused by the deep trap states of TiO2. The passivation
effect is further proved by stability testing of the perovskite solar cells with shelf lifetime in ambient condition improved
by a factor of more than 4, from ~40 hours to ~200 hours, by using the PCBB-2CN-2C8 as TiO2 modification layer. This
work offers not only a promising material for cathode interface engineering, but also provides a viable approach to ad-
dress the challenges of deep trap states on TiO2 surface in planar perovskite solar cells.
layer, ETL) on ITO or FTO electrode.8 Only in some
INTRODUCTION
Organic-inorganic metal halide perovskite solar cells
have attracted growing interest recently, as its power con-
version efficiency (PCE) strived to 20% within several
years.1-7 Although the perovskite solar cells based on me-
so-porous TiO2 take the lead in efficiency competition,
planar structured perovskite solar cells is becoming popu-
lar recently due to its low temperature processing and
simple device architecture. Nevertheless, the interface
contacts in the devices are critically important for achiev-
ing high performance planar perovskite solar cells.8-11
The planar perovskite solar cells based on perovskite
film as photoactive layer can be classified as n-i-p device
structure4,7 with indium-tin-oxide (ITO) or fluorine-
doped tin-oxide (FTO) transparent electrode as cathode
or p-i-n device structure with ITO or FTO as anode12-14. In
the n-i-p structured perovskite solar cells, doped small
molecules and polymers (such as spiro-OMeTAD), as well
as polyelectrolytes, can be used as hole transport layers
(p-type anode buffer layer, HTL), and some metal oxides
(such as TiO2), fullerenes and their derivatives can be
used as electron transport layers (n-type cathode buffer
exceptional cases, either p- or n- type layers can be
skipped for simplicity, while the penalty is the inevitable
loss in charge recombination and build-in potential.13-15 To
effectively extract charge carriers, it is important for the
charge transport layers to have high selectivity. For
example, TiO2 has been regarded as one of the best
electron extraction and transport material due to its
proper conduction band compatible with those of hybrid
perovskites. At the meantime the deep valence band is
able to block the holes. Recent study also indicates that
low temperature processed TiO2 layer is able to play such
a critical role, which makes it an important electron
transport layer for planar n-i-p perovskite solar cells.15
However, there are two drawbacks of low temperature
processed TiO2 as electron transport layer in the planar
perovskite solar cells. Firstly, the low electron mobility
results in charge accumulation in TiO2 layer. That means
TiO2 is not an ideal electron acceptor and transport layer
for perovskite.16,17 Secondly, exposure of TiO2 to UV light
further induces formation of oxygen vacancies, which
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (a) Synthetic route and chemical structure of PCBB-2CN-2C8, and molecular structures of PCBM and PCBB-
2C8 for comparison. (b) The photographs of PCBM, PCBB-2C8 and PCBB-2CN-2C8 in CB or DMF solution with concen-
tration of 5 mg/mL. (c) The images of water and DMF droplet contact angles on different surfaces of fullerene derivatives
coated TiO2.
are energetically deep trap levels responsible for severe
recombination loss.18,19 As a result, the TiO2/perovskite
interface slows down the photoresponse of the photovol-
taic devices, causing strong hysteresis behaviors, and thus
making accurate efficiency measurement difficult.17 Snaith
et al proposed to replace TiO2 with other inert metal ox-
ides (Al2O3 etc.) through which highly stable perovskite
solar cells have been demonstrated.18 However it remains
challenging to identify a replacement that works as well
as TiO2, especially in planar heterojunction cells. As an
alternative approach, coating n-type materials like [6,6]-
Phenyl-C61-butyric acid methyl ester (PCBM) offers com-
petitive photovoltaic performance, however, since PCBM
is soluble in N,N-Dimethylformamide (DMF), it requires
deposition of PbI2 via thermal evaporation in high vacu-
um, which complexes the processing.20 Another effective
way to circumvent this challenge has been proposed to
lecular anchoring, not convenient for both lab research
and scale-up manufacturing. Taking these issues into
consideration, it is urgent to explore alternatives to engi-
neer the TiO2 film surface, in order to achieve: 1) efficient
electron collection and hole blocking; 2) simple and fast
processing; 3) high device stability against UV illumina-
tion. To satisfy the demand for surface modification on
TiO2 in perovskite solar cells, the interfacial materials
should meet these criteria.
In this work, we tried to deal with the multiple issues
via using a tri-block fullerene derivative [6,6]-phenyl-C61-
butyric acid-dioctyl-3,3'-(5-hydroxy-1,3-phenylene)-bis(2-
cyanoacrylate) ester (PCBB-2CN-2C8) (as shown in Figure
1a) as a cathode modification layer on TiO2. To have desir-
able electrical properties, we chose fullerene C60 as the
main building block of the molecule because of its high
electron affinity for efficient electron transfer. Electron-
deficient cyano-groups were implanted for pacifying oxy-
gen vacancy. Since multi-layer coating via solution pro-
cess in the device fabrication of perovskite solar cells re-
modify TiO2 surface with using self-assembly mono-layer
9,21
(SAM) of fullerene derivatives.
Yet the deposition of a
SAM layer usually takes more than 10 hours for the mo-
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
quires orthogonal solvents, we introduce dioctyloxy
chains and cyano-groups to minimize its solubility in po-
1
2
3
4
5
6
7
8
lar solvent, (such as DMF used for coating perovskite pre-
cursors) and to be highly soluble in non-polar solvents
like chlorobenzene (CB) for its solution-processing. The
selective solubility enables us to fabricate all the func-
tional layers via solution process except for the metal
electrodes. We then incorporated the PCBB-2CN-2C8 into
an n-i-p planar perovskite solar cells to modify TiO2 sur-
face, and obtained significantly enhancement in photo-
voltaic performance, as well as reduction of the hysteresis.
Moreover, the shelf stability of unencapsulated devices in
normal ambient condition was significantly improved
comparing with the TiO2 based control devices.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
The synthetic route of PCBB-2CN-2C8 is shown in Fig-
ure 1a. Reaction of fullerene acid (PCBA, compound 5)
Figure 2. Top-view SEM images of perovskite films grown on
(a) TiO2 and (b) TiO2/PCBB-2CN-2C8. The insets are respec-
tive AFM images. (c) XRD patterns of the perovskite films
grown on TiO2 and TiO2/PCBB-2CN-2C8. (d) Absorption
spectra of perovskite films grown on TiO2 and TiO2/PCBB-
2CN-2C8.
with
dioctyl
3,3'-(5-hydroxy-1,3-phenylene)-bis(2-
cyanoacrylate) (compound 4) in ortho-dichlorobenzene
(o-DCB) using 4-dimethylamino-pyridine (DMAP), p-
Toluenesulfonamide (PTSA) and diisopropylcarbinol
(DIPC) as catalysts gave the target PCBB-2CN-2C8, in
67% yield. In addition, PCBB-2CN-2C8 shows high ther-
mal stability with both high decomposition temperature
of ca. 356oC (5% weight loss from thermal gravimetric
analysis (TGA, in Supplementary Figure S1) and amor-
by using contact angle measurement, and PCBM and
PCBB-2C8 were used for comparison. In Figure 1c, water
droplets on TiO2/PCBM, TiO2/PCBB-2C8 and TiO2/PCBB-
2CN-2C8, have contact angles of 60o, 81o, 91o, respectively.
Both PCBB-2C8 and PCBB-2CN-2C8 show larger contact
angles than PCBM, indicating the octyloxy chains could
be responsible for poor wetting. The cyano-groups in
PCBB-2CN-2C8 makes it more hydrophobic than PCBB-
2C8. Similar trend holds for DMF, with slightly better
wetting than water, and the strong hydrophilic property
of PCBB-2CN-2C8 indicates its potential advantage for
PbI2 deposition from DMF solution.24
In attempt to integrate PCBB-2CN-2C8 in photovoltaic
cells, we investigated the effect of PCBB-2CN-2C8 coated
compact TiO2 on the growth of CH3NH3PbI3 film via a
two-step method. As shown in Figure S8 (Supporting In-
formation), pristine TiO2 surface has a root-mean-square
(rms) surface roughness of 4.43 nm. Coating PCBB-2CN-
2C8 on TiO2 slightly reduces the roughness to rms of 3.98
nm, which should not influence the growth of perovskite
crystal. Figure 2a and 2b shows top view images taken by
scanning electron microscope (SEM) and atomic force
microscope (AFM). We can see compact film and homog-
enous grain size distribution with an average value of 1097
nm (Figure S9 in Supporting Information). Without
PCBB-2CN-2C8, perovskite crystals grown directly on
TiO2 surface shows a slight smaller grain size of 1035 nm
with tiny cracks throughout the film. Both films have sim-
ilar surface roughness about 27.0 nm. The observation
suggests that high-quality perovskite film can be formed
on surface of TiO2/PCBB-2CN-2C8 with large grain size,
which is consistent with previous report.24,25
o
phous nature (Tg detected at 49.7 C, excluding obvious
endothermic peaks, shown in Supplementary Figure S2),
which are beneficial for enduring high temperature and
retaining a stable PCBB-2CN-2C8 film morphology at ele-
vated temperature, respectively.22 Besides, we measured
electron mobility (µe) of PCBB-2CN-2C8 using space-
charge limited current measurement as shown in Sup-
plementary Figure S3. It has µe of 4.8×10-3 cm2 V-1s-1 com-
parable to that of PCBM (2.5×10-3 cm2 V-1s-1). Thus PCBB-
2CN-2C8 is good for electron transport.15,23 Detailed syn-
thetic procedure and characterization are described in
experimental section and Supporting Information (Figure
S4-6). Molecular structures of PCBM and [6,6]-phenyl-
C61-butyric acid-3,5-bis(octyloxy)-benzyl ester (PCBB-
2C8) without cyan groups are also shown in Figure 1a for
comparison. Figure 1b shows the excellent solubility of
PCBB-2CN-2C8 in CB (more than 100 mg/mL), while it is
almost insoluble in DMF due to presence of dioctyloxy
chains, consistent with our design consideration. This is
important from processing point of view, since DMF is a
common solvent to synthesize lead iondide perovskite
films. In comparison, PCBB-2C8 is more soluble in DMF
than PCBB-2CN-2C8 due to the absence of cyano-groups.
PCBM without long dioctyloxy chains and cyano-groups is
well dissolved in both DMF and CB. Therefore the films of
PCBB-2C8 and PCBM cannot stand DMF solution coating.
Surface wetting is an important factor for solution coat-
ing process. Generally, polar solvents have excellent wet-
ting properties on TiO2 surface, as confirmed by the ex-
tremely small contact angles of water and DMF (Figure S7
in Supporting Information). We tested if PCBB-2CN-2C8
in DMF solution was suitable for coating on TiO2 surface
The bulk properties of perovskite films strongly depend
on its crystal quality. X-ray diffraction (XRD) was con-
ducted to characterize the perovskite films grown on TiO2
with and without PCBB-2CN-2C8. In Figure 2c, both
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
that of galss/ITO/TiO2 control sample in the visible
region (Figure S10 in Supporting Information)
1
2
3
4
5
6
7
8
9
We then fabricated the planar heterojunction perov-
skite photovoltaic cell with a structure of glass/ITO/TiO2
(50 nm)/PCBB-2CN-2C8 (10 nm)/CH3NH3PbI3 (290
nm)/spiro-OMeTAD (170 nm)/Au (120 nm) as illustrated
in Figure 3a. Figure 3b provides the energy-level diagram.
We can see that PCBB-2CN-2C8 layer between TiO2 and
perovskite does not change the overall energetic land-
scape, still allowing a cascade electron transfer process.
Work function of TiO2 is 4.21 eV measured by Kelvin
probe force microscope (KPFM) (Supplementary Figure
S8), while that of PCBB-2CN-2C8 modified TiO2 is uplift-
ed by ~200 meV, reaching 4.01 eV.27,28 As the conduction
band bottom (ECB) of perovskite is 3.90 eV, the energy
offset is sufficient for electron extraction from perovskite
to the electron transport layer. In the cross-section SEM
image in Figure 3c, all functional layers can be clearly dis-
tinguished. The grain size are obviously larger than the
film thickness, giving average grain size/film thickness
aspect ratio of 3.8. The length of the grains is equal to the
thickness of the film and the film uniformity easily ex-
panding to several micrometers as displayed. Meanwhile,
combining the larger, compact and homogeneous grains
in perovskite film on TiO2/PCBB-2CN-2C8, the most of
the grain boundaries perpendicular to the substrate can
effectively minimize the grain boundary energy.24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (a) perovskite solar cells architecture employing
TiO2/PCBB-2CN-2C8 as ETL. (b) The corresponding energy-
level diagram. (c) Cross-section SEM images of complete
perovskite solar cell.
samples show similar crystal structures with distinctive
(110), (220) and (330) diffraction peaks at 14.2o, 28.4o and
43.2o, respectively, in accordance with the major phases of
CH3NH3PbI3.26 Similar to the film grown on TiO2 surface,
we cannot see peaks from impurity phase or significant
change of the diffraction peak ratio, indicating pure per-
ovskite phase on both the surfaces without any change in
crystal orientation. XRD peak at 14.2o becomes much
stronger and sharper, as the full width at half maximum
(FWHM) of the peak (110) decreases from 0.188o on TiO2
to 0.149o on TiO2/PCBB-2CN-2C8. It agrees well with the
improved perovskite film quality discussed in previous
section.
Figure 2d shows the ultraviolet-visible (UV-vis) absorp-
tion spectra of the perovskite films deposited on
glass/ITO/TiO2 and glass/ITO/TiO2/PCBB-2CN-2C8 sub-
strates. The sample with PCBB-2CN-2C8 shows a slightly
enhanced absorption in blue region due to the absorption
of fullerene. Similarly, except for slight drop in UV and
blue range, the transmittance spectrum of the
glass/ITO/TiO2 substrates with PCBB-2CN-2C8 assembles
J-V characteristics of the perovskite solar cells with or
without PCBB-2CN-2C8 modification layer were meas-
ured under simulated solar illumination of AM1.5G 100
mW/cm2. As shown in Figure 4a, while the TiO2 ETL
based device gave rise to a PCE of 13.35±0.85% (the best
PCE of 14.38%) from reverse scan,20,21,29 we achieved a
16.45±0.81% PCE (the best PCE of 17.35%) from the
TiO2/PCBB-2CN-2C8 based device. The key photovoltaic
parameters are summarized in Table 1. The device em-
ploying PCBB-2CN-2C8 as TiO2 modification layer yields
improved Voc from 0.98±0.015 V to 1.06±0.012 V, and FF
from 70.2±1.9% to 78.0±1.0%. Theoretically, Voc is de-
termined
by
the
quasi-Fermi
Table 1. Summary of photovoltaic parametersa of perovskite solar cells with and without PCBB-2CN-2C8 modification
layer.
PCE (%)
ETL
Voc (V)
FF (%)
Jsc (mA/cm2)
Max. power outputb
Average (Best)
TiO2 (RS)
0.98±0.015
(0.99)
70.2±1.9
(72.2)
19.23±0.83
(20.18)
13.35±0.85
(14.38)
12.58
TiO2 (FS)
0.95±0.010
61.0±2.0
19.10±0.76
10.95±0.80
(0.95)
1.06±0.012
(62.1)
78.0±1.0
(20.24)
19.85±0.75
(11.95)
16.45±0.81
TiO2/PCBB-2CN-2C8 (RS)
(1.06)
(79.1)
(20.68)
(17.35)
16.81
TiO2/PCBB-2CN-2C8 (FS)
1.05±0.08
(1.04)
73.5±1.2
(74.4)
19.7±0.70
(20.54)
15.25±0.60
(15.90)
aAverage and standard deviation values were obtained based on over 20 cells from 3 different batches for each ETL. FS: for-
ward scan, scan rate 200 mV/s, delay time 100 ms; RS: reverse scan, scan rate 200 mV/s, delay time 100 ms. Parameters of the
best cell are reported in brackets. bMaximal steady-state power output, which is equal the PCE of maximum power output.
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
1
gram of the PCE of 50 perovskite solar cells based on TiO2
2
3
4
5
6
7
8
9
and TiO2/PCBB-2CN-2C8 ETLs in Figure 4c. It shows the
top 1/3 of TiO2/PCBB-2CN-2C8 ETL based devices deliv-
ered PCE over 16.5%, comparing with 13.5% TiO2 ETL
based devices. The improved device performance and
encouraging reproducibility unambiguously revealed
tremendous potential of molecular engineering of the
cathode interface materials in perovskite solar cells. To
avoid overestimation of photovoltaic efficiency caused by
hysteresis, we took maximal steady-state power and pho-
tocurrent output in Figure 4c. TiO2/PCBB-2CN-2C8 based
cell delivered 19.55 mA/cm2 at 0.86 V, corresponding to
PCE of 16.81%, very close to the value from reverse bias
scan (defined as from positive bias to negative). From the
TiO2 based control device, a PCE of 12.58% was deter-
mined. This experiment gives solid evidence for efficiency
enhancement due to the TiO2 ETL interfacial modifica-
tion.
To further look into the hysteresis issue, we then took
scan in both reverse and forward direction to compare the
photoresponse behavior of the perovskite solar cells (see
detail in Supplementary Figure S12). Under forward and
reverse scan, TiO2 based control device gives PCE of
11.95% and 14.38%, respectively, and the hysteresis index
of 0.184 as calculated according to equation (1) in Sup-
porting Information.34 The perovskite solar cells with
TiO2/PCBB-2CN-2C8 ETL shows a hysteresis index of
0.052, much less hysteresis than that of the control de-
vice. Consistently, as shown in Figure 4c, the device with
PCBB-2CN-2C8 modification layer shows a faster rising
photocurrent to maximum value during immediate light
illumination than the control device. Since the electrons
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
have
to
flow
through
Figure 4. (a) J-V curves of perovskite solar cells with struc-
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ture of glass/ITO/TiO2/with or without PCBB-2CN-
2C8/MAPbI3/spiro-OMeTAD/Au under illumination of
AM1.5G 100 mW/cm2 simulated solar light. (b) Maximal
steady-state photocurrent output at the maximum power
point (for TiO2 based device at 0.84 V, and TiO2/PCBB-2CN-
2C8 based device at 0.86 V) and their corresponding power
output. (c) Histograms of device PCE measured for 50 devic-
es based on TiO2 and TiO2/PCBB-2CN-2C8 ETLs.
level splitting in photoactive layer, and limited by the
work function difference of two electrode contacts, i.e.
TiO2/PCBB-2CN-2C8 and spiro-MeOTAD.3 The higher Voc
could be attributed to the lower work function of
TiO2/PCBB-2CN-2C8 (4.01 eV) than that of TiO2 (4.21 eV).
As discussed above, the well-defined perovskite film with
high crystallinity, homogeneous crystal size distribution
and equal film thickness with grain size also can effective-
ly depress bulk recombination and reduce shunting
pathways.30 It shall, at least partially, account for the im-
proved FF of 79.1% of the best device, which is high for
perovskite photovoltaic devices.15,31-33 External quantum
efficiency (EQE) spectra of the perovskite solar cell is
shown in Figure S11. Since the reproducibility of device
efficiency has been a concern, we further show a histo-
Figure 5. (a) The steady-state PL spectra of perovskite film
on TiO2 and TiO2/PCBB-2CN-2C8. (b) TRPL decay transient
spectra of ITO/TiO2/Perovskite and ITO/TiO2/PCBB-2CN-
2C8/Perovskite.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
two time constants, i.e. τ1 and τ2, as summarized in Table
1
S1 (Supporting Information). τ1 represents the bulk re-
combination in perovskite bulk film in relatively short
time scale of a few tens of nanosecond, and the slow one,
τ2 ranging in hundred nanosecond, comes from the de-
layed recombination of trapped charges.36,37 For
TiO2/Perovskite, τ1 is 9.7 ns with a 96.82% ratio and the τ2
was 172.0 ns with a 3.18% ratio, suggesting that the de-
population of photogenerated charges was dominated by
charge collection through the TiO2/Perovskite interface,
accompanying with weak delayed radiative recombina-
tion. As for TiO2/PCBB-2CN-2C8/Perovskite, both τ1 and
τ2 were shortened to 3.6 ns and 63.2 ns, respectively, and
interestingly τ2 is almost overwhelmed by τ1. It is not only
because the modified TiO2 induces fast electron transfer
from perovskite film, but also very likely PCBB-2CN-2C8
reduces the trap state density of TiO2 via passivation. This
is consistent with enhanced Jsc and FF from modified TiO2
based device. As one can expect, the surface passivation
on TiO2 shall also be responsible for the less hysteresis
behaviors, since the trap sites enable long existing charge
carriers and thus slow down the device response speed.
To further understand the charge extraction and re-
combination processes, we characterized transient photo-
current decay (see experimental detail in Supporting In-
formation) as shown in Figure 6a. The TiO2/PCBB-2CN-
2C8 ETL based device shows a faster decay compared with
that of TiO2 based control device, which confirms the
strong electron accepting properties of PCBB-2CN-2C8.
We further investigate how PCBB-2CN-2C8 affects carrier
recombination dynamics in the photovoltaic device. Tran-
sient photovoltage (TPV) decay was performed at open
circuit condition under white light bias equal to 0.5 sun
intensity (see experimental detail in Supporting Infor-
mation). The photovoltage signal is triggered by a pulse
light that pumps up the carrier density and thus quarsi-
Fermi level splitting, and the charge carriers through re-
combination is directly measured via monitoring the pho-
tovoltage decay.38,39 The photovoltage decay curves of the
perovskite solar cells based on TiO2 and TiO2/PCBB-2CN-
2C8 ETLs are plotted in Figure 6b. In both cases, the pho-
tovoltage decay constants are on the scale of microsec-
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. (a) Transient photocurrent decay and (b) Transient
photovoltage decay of the optimized devices employing the
TiO2and TiO2/PCBB-2CN-2C8 as ETLs.
the TiO2 layer, no matter modified or not, the fast pho-
toresponse should not be related to electron transport
properties. However, considering the large amount of
defects and interband states on TiO2 surface, the prompt
photocurrent saturation could be related to the filling of
these states. Therefore we believe the passivation of the
trap states by PCBB-2CN-2C8 is the key to reduce the hys-
teresis as discussed later.35
In order to further look into the effect of PCBB-2CN-
2C8 modification on the improvement of photovoltaic
performance and reduction of hysteresis behavior, we
investigated charge dynamics by measuring photolumi-
nescence (PL) and time-resolved PL (TRPL) of perovskite
film on TiO2 and TiO2/PCBB-2CN-2C8. Figure 5a shows
the steady-state PL spectra of perovskite films on TiO2
and TiO2/PCBB-2CN-2C8, respectively. PL intensity of
TiO2/PCBB-2CN-2C8/perovskite was reduced by 3-fold of
magnitude compared with that of TiO2/Perovskite, while
the absorbance of the two films are essentially the same.
This result shows fast electron injection from perovskite
into PCBB-2CN-2C8. It is well known that fullerenes and
their derivatives are strong electron acceptors, and the C60
carbon ball is able to accommodate even multiple elec-
trons at the same time.9 Thus in spite of less energy offset
for the electron transfer at TiO2/PCBB-2CN-
2C8/perovskite interface as shown in Figure 3b, the signif-
icant PL quenching confirms the well electronic coupling
between conduction band of perovskite and LUMO of
PCBB-2CN-2C8. Furthermore, the TRPL was measured by
monitoring the peak emission at 768 nm as shown in Fig-
ure 5b. Fitting the data with bi-exponential decay yields
onds,
in
Figure 7. Capacitance-voltage characteristics of the opti-
mized perovskite solar cells employing the TiO2 and
TiO2/PCBB-2CN-2C8 as ETLs.
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
accordance to the recent report.6 The fast component of
TPV decay is typically associated with non-germinate re-
1
2
3
4
5
6
7
8
combination.38,39 The carrier lifetime in the PCBB-2CN-
2C8 based device was measured to be ~1.5 times longer
than that of TiO2 based device, consistent with the higher
Voc, Jsc and FF we observed due to insertion of PCBB-2CN-
2C8. Again our observation supported the argument that
the passivation of the trap states on TiO2 surface can effi-
ciently suppress the charge recombination.
9
In view of similar film quality of perovskite on both
TiO2 and TiO2/PCBB-2CN-2C8 surface, we speculate that
the increase in carrier lifetime could be due to two possi-
ble reasons. According to the previous report,40,41 TiO2 is
known to contain many oxygen vacancies (or Ti3+ sites),
particularly on the surface, corresponding to the trap
states. In low temperature processed TiO2 compact layer
used in this work, the mixing nature of crystal TiO2 and
amorphous TiOX gives even higher trap density. Lewis
base groups like –CN and -SCN40 have been proved to
occupy the oxygen vacancies (or Ti3+ sites), and thus to
pacify oxygen vacancies.40 In such a scenario, the adsorp-
tion of electron-withdrawing groups, considering both –CN
and carbon ball, on oxygen vacancies (Ti3+) shall dramati-
cally reduce the charge trapping effect. Consequently the
passivation of TiO2 surface by PCBB-2CN-2C8 could be
responsible for the long carrier lifetime. Similar explana-
tions have been invoked to describe reductions in carrier
recombination due to changes in electrode chemistry in
perovskite,42 and organic solar cells.43
The electronic trap states are able to delocalize charge
carriers and to induce high capacitance at the interface,
which can be readily detected using impedance spectros-
copy. To further confirm the hypothesis that PCBB-2CN-
2C8 pacifies the trap states on the TiO2 surface, we per-
formed impedance measurements on the perovskite solar
cells of ITO/TiO2/with or without PCBB-2CN-
2C8/CH3NH3PbI3/spiro-OMeTAD/Au at 1 kHz frequency.
We extracted capacitance-voltage characteristics as
shown in Figure 7. At this frequency obvious variation of
capacitance occurs with increasing bias voltage, which is
indicative of charge accumulation at the compact layer
and can represent its capacitance.21 With PCBB-2CN-2C8
modification, the capacitances of devices are unambigu-
ously reduced, confirming the passivation effect of PCBB-
2CN-2C8 as discussed above.40,41 It leads us to the conclu-
sion that higher current extracting ability and longer life
time in TiO2/PCBB-2CN-2C8 based device is predominat-
ed by the reduced TiO2 trap state.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. (a) The efficiency and (b) short current density
evolution of the unencapsulated devices stored on windowsill
in ambient air with the humidity of 45-50%.
PCEs and Jsc vs time from ten devices for each structure
are summarized in Figure 8. TiO2 based control devices
showed 5.6% of initial PCE after 120 hours exposure, ac-
companying the significantly decreased Jsc to 30% of their
initial values. At the meantime, the device based on
PCBB-2CN-2C8 coated TiO2 showed a significant im-
provement in stability under prolonged illumination, re-
taining more than 60% and 90% of their initial efficiency
and photocurrent, respectively, after 500 hours
exposure.44 In term of chemistry, the origin of this effect
is possibly caused by the TiO2 photocatalyst degredation
of perovskite film,19,45 and electrically the creation of a
large density of deep trap sites on ultraviolet excitation on
TiO2 surface.18 The main point of the stability test is to
further confirm passivation of surface trap states of TiO2.
In summary, we have synthesized a tri-block function-
alized fullerene derivative PCBB-2CN-2C8 for ETL inter-
face engineering on low temperature-processed TiO2 in
the planar n-i-p perovskite solar cells. The molecule de-
sign theme allows us simultaneously address several ma-
jor challenges, including processing compatibility for
multi-layer coating, interface recombination loss, work
function tuning of electrodes, the notorious hysteresis
and interface induced device degradation, via simple and
fast solution spin-coating process. We proposed that the
passivation of TiO2 surface by PCBB-2CN-2C8 on it is an
effective approach to enhance the photovoltaic perfor-
mance, as supported by both materials characterization
and devices analysis. Interestingly, the encouraging stabil-
ity data imply that surface passivation is a promising
strategy to avoid the fast degradation caused by UV sensi-
tive TiO2 surface.
As mentioned earlier, UV irradiation causes fast release
of O2 on TiO2 surface, and induces large amount of deep
trap states. According to the report by Snaith’s group, it is
responsible for the fast degradation of the perovskite solar
cells. Therefore we looked into the stability issue, in order
to confirm the passivation effect. The devices using TiO2
as ETL with and without PCBB-2CN-2C8 were investigat-
ed for comparison. Stored on windowsill in ambient air
without encapsulation allows 6 hours exposure to natural
sunlight per day at humidity of 45-50%. The normalized
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
was suspended in methanol to give PCBB-2CN-2C8in
EXPERIMENTAL SECTION
Material synthesis.
66.5% yield as a brown solid.1H NMR (400 MHz, CDCl3):
δ(ppm) 8.23 (s, 2H), 8.17 (s, 1H), 7.96 (d, J = 8.5 Hz, 4H),
7.57 (t, J = 7.6 Hz, 2H), 7.50 (t, J = 7.3 Hz, 1H), 4.34 (t, J =
6.7 Hz, 4H), 3.02 (t, 2H), 2.85 (t, J = 7.2 Hz, 2H), 2.34 (m,
2H), 1.82 – 1.70 (m, 4H), 1.42 (m, 4H), 1.29 (m, J = 11.8 Hz,
16H), 0.88 (t, J = 6.9 Hz, 6H)13C NMR (400 MHz, CDCl3) δ
170.80, 161.54, 151.99, 151.43, 148.66, 147.65, 145.14, 145.02,
144.75, 144.61, 144.46, 144.39, 143.71, 143.00, 142.95, 142.87,
142.14, 142.08, 140.99, 140.72, 137.96, 137.55, 136.52, 133.65,
132.06, 130.69, 128.54, 128.36, 126.83, 114.53, 106.19, 79.71,
67.29, 51.58, 33.98, 33.49, 31.74, 29.13, 28.43, 25.73, 22.63,
22.14, 14.12.MALDI-TOF MS: Calcd. forC113H50N2O6:
1
2
3
4
5
6
7
8
Dimethyl 5-hydroxyisophthalate (1).To the solution
of 5-hydroxyisophthalic acid (10 g, 0.054 mol) in CH3OH
(200 mL), 10 drops H2SO4 was added, and the mixture was
refluxed for 5h. Then the reaction mixture was poured
into saturated sodium bicarbonate solution and stirred
for 0.5 h. Then the solution was filtered to give dimethyl
5-hydroxyisophthalatein 84.1% yield as a white solid. 1H
NMR (400 MHz, CDCl3): δ(ppm) 8.26 (s, J = 1.4 Hz, 1H),
7.74 (s, J = 1.4 Hz, 2H), 5.83 (s, 1H), 3.95 (s, 6H).
(5-hydroxy-1,3-phenylene)dimethanol (2). LiAlH4
(1.448 g, 38.095 mmol) was added in the THF (100 mL)
solution of dimethyl 5-hydroxyisophthalate (2 g, 9.524
mmol) under the protection of Ar in ice-bath and stirred
at room temperature for 24 h. Then the reaction mixture
was quenched by HCl solution. The product was extracted
with ethyl acetate three times. The combined organic
layers were dried over anhydrous Na2SO4. After filtration
and evaporation, the residue was purified by silica gel
chromatography with petroleum ether: ethyl ester (1/5
v/v). The yield is 66.5% as a colorless oil.1H NMR (400
MHz, DMSO): δ(ppm) 9.21 (s, 1H), 6.66 (s, 1H), 6.59 (s,
2H), 5.08 (t, J = 5.8 Hz, 2H), 4.39 (d, J = 5.8 Hz, 4H).
5-hydroxyisophthalaldehyde (3). To the solution of
(5-hydroxy-1,3-phenylene)dimethanol (1 g, 6.494 mmol)
in THF (20 mL) and CH2Cl2 (30 mL) was added SiO2
(4.169 g, 69.481 mmol) and PCC (4.199 g, 19.481 mmol),
the reaction mixture was stirred in room temperature for
12 h. Then the product was purified by silica gel chroma-
tography with petroleum ether: ethyl ester (5/1, v/v). The
yield is 50.5% as a white solid. H NMR (400 MHz,
DMSO): δ(ppm)10.53 (s, 1H), 10.03 (s, 2H), 7.91 (s, 1H),
7.56 (s, 2H).
Dioctyl-3,3'-(5-hydroxy-1,3-phenylene)-bis(2-
cyanoacrylate) (4). 5-hydroxyisophthalaldehyde (1 g,
6.667 mmol), octyl 2-cyanoacetate (2.629 g,13.333 mmol)
was added to ethanol (5 mL) and water (2 mL), and then
several drops of (NH4)2HPO3 was put in. The reaction
mixture was stirred at room temperature for 12 h. After
reaction, the solution was filtrated and the filter residue
was purified by silica gel chromatography with petroleum
ether: ethyl ester (3/1 v/v) to give in 89.4% yield as a kelly
solid. 1H NMR (400 MHz, CDCl3): δ(ppm) 8.21 (s, 2H), 7.79
(s, 3H), 6.32 (s, 1H), 4.33 (t, J = 6.7 Hz, 4H), 1.82 – 1.72 (m,
4H), 1.42 (m, 4H), 1.30 (m, J = 10.8 Hz, 16H), 0.88 (t, J = 7.0
Hz, 6H).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1386.37; Found: 1386.48
.
Device fabrication.
Methylammonium iodide (CH3NH3I, MAI) and me-
thylammonium chloride (CH3NH3Cl, MACl) were synthe-
sized using the method described elsewhere6. The TiO2
nanocrystals were obtained from a non-hydrolytic sol-gel
approach6. ITO glass substrates were sequentially washed
with isopropanol, acetone, distilled water and ethanol.
The TiO2 ETL was subsequently spin-coated on ITO sub-
strates with a TiAcac-stabilized TiO2 solution at 3000 rpm
o
and annealed at 150 C for 30 min in air. The second TiO2
ETL was fabricated according to the above process. Then
PCBB-2CN-2C8 (dissolved in CB, 5 mg/ml) was spin-
coated on TiO2 at 3000 rpm for 30 s, and dried in vacuum
o
chamber for 1 hour. A 70 C PbI2 solution (dissolved in
DMF, 460 mg/ml) was spin-coated on top of
ITO/TiO2/PCBB-2CN-2C8 substrate at 3000 rpm for 30 s,
and kept on hot plane at 75 oC for 10 min. Then a mixture
of MAI/MACl (50 mg/ml:5 mg/ml, dissolved in 2-
propanol) was spin-coated on top of the dried PbI2 layer
at room temperature at 4000 rpm for 40 s in nitrogen
filled glove box. All of the films were annealed in the air
1
o
at 135 C for 20 min. A hole transport layer solution was
coated on the perovskite film at 3000 r.p.m. for 30 s,
where a spiro-OMeTAD/chlorobenzene (90 mg/ml) solu-
tion was employed with addition of 45 ml Li-
TFSI/acetonitrile (170 mg/ml) and 10 ml tBP. Finally, the
counter electrode was deposited by thermal evaporation
of gold under a pressure of 5×10-5 Torr. The active area
was 0.108 cm2. The details for characterization are de-
scribed in Supporting Information.
SUPPORTING INFORMATION.
Detailed material and device characterization. Additional
figures. This material is available free of charge via the Inter-
net at http://pubs.acs.org.
[6,6]-phenyl-C61-butyric
hydroxy-1,3-phenylene)-bis(2-cyanoacrylate)ester
(PCBB-2CN-2C8). To a flask, PCBA (0.1 g, 0.112 mmol),
acid-dioctyl-3,3'-(5-
dioctyl
3,3'-(5-hydroxy-1,3-phenylene)-bis(2-
cyanoacrylate)(0.057 g, 0.112 mmol), DMAP(0.014 g, 0.112
mmol), PTSA(0.021 g, 0.112 mmol) was added. Then
DIPC(0.020 g, 0.157 mmol) was washed into it by dichlo-
robenzene(30mL), stirred at room temperature for 12 h
with the protection of Ar. Then the solution was filtered
and the filtrate was purified by silica gel chromatography
with petroleum ether: toluene(1/3 v/v). The crude product
AUTHOR INFORMATION
Corresponding Author
*(Y. W.L.) E-mail: ywli@suda.edu.cn.
*(Z. H.) E-mail: zrhong@ucla.edu.
* (Y. F. L.) E-mail: liyongfang@suda.edu.cn;
ACS Paragon Plus Environment

Page 9 of 11
*(Y.Y.) E-mail: yangy@ucla.edu.
Journal of the American Chemical Society
(20) Tao, C.; Neutzner, S.; Colella, L.; Marras, S.; Srimath
1
2
3
Kandada, A. R.; Gandini, M.; Bastiani, M. D.; Pace, G.;
Manna, L.; Caironi, M.; Bertarelli, C.; Petrozza, A. Energy
Environ. Sci. 2015, 8, 2365.
ACKNOWLEDGMENT
Yaowen Li acknowledged for financial support from Jiangsu
Government Scholarship for Overseas Studies. Vishal Bharti
for SCLC test. This work was partially supported by the Na-
tional Natural Science Foundation of China (Grant No.
21204057). Priority Academic Program Development of Jiang-
su Higher Education Institutions. National Science Founda-
tion (grant number ECCS-1202231, Program Director Dr Paul
Werbos), Air Force Office of Scientific Research (grant num-
ber FA9550-12-1-0074, Program Manager Dr Charles Lee).
4
5
6
7
8
9
(21)
Wojciechowski, K.; Stranks, S. D.; Abate, A.;
Sadoughi, G.; Sadhanala, A.; Kopidakis, N.; Rumbles, G.;
Li, C.-Z.; Friend, R. H.; Jen, A. K. Y.; Snaith, H. J. ACS Nano
2014, 8, 12701.
(22) Zhang, Y.; Yip, H.-L.; Acton, O.; Hau, S. K.; Huang, F.;
Jen, A. K. Y. Chem. Mater. 2009, 21, 2598.
(23) Leijtens, T.; Lim, J.; Teuscher, J.; Park, T.; Snaith, H. J.
Adv. Mater. 2013, 25, 3227.
(24) Bi, C.; Wang, Q.; Shao, Y.; Yuan, Y.; Xiao, Z.; Huang, J.
Nat. Commun. 2015, 6, 7747.
(25) Liu, C.; Wang, K.; Du, P.; Meng, T.; Yu, X.; Cheng, S.
Z. D.; Gong, X. ACS Appl. Mater. & Interfaces 2015, 7, 1153.
(26) Chen, Q.; Zhou, H.; Fang, Y.; Stieg, A. Z.; Song, T.-B.;
Wang, H.-H.; Xu, X.; Liu, Y.; Lu, S.; You, J.; Sun, P.; McKay,
J.; Goorsky, M. S.; Yang, Y. Nat. Commun. 2015, 6, 7269.
(27) Shimizu, H.; Cojal González, J. D.; Hasegawa, M.;
Nishinaga, T.; Haque, T.; Takase, M.; Otani, H.; Rabe, J. P.;
Iyoda, M. J. Am. Chem. Soc. 2015, 137, 3877.
(28) Zhou, H.; Zhang, Y.; Mai, C.-K.; Collins, S. D.; Bazan,
G. C.; Nguyen, T.-Q.; Heeger, A. J. Adv. Mater. 2015, 27,
1767.
(29) Docampo, P.; Hanusch, F. C.; Stranks, S. D.; Dö-
blinger, M.; Feckl, J. M.; Ehrensperger, M.; Minar, N. K.;
Johnston, M. B.; Snaith, H. J.; Bein, T. Adv. Energy Mater.
2014, 4, 1400355.
(30) Bi, C.; Shao, Y.; Yuan, Y.; Xiao, Z.; Wang, C.; Gao, Y.;
Huang, J. J. Mater. Chem. A 2014, 2, 18508.
(31) Liu, D.; Kelly, T. L. Nat. Photon. 2014, 8, 133.
(32) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker,
R.; Gao, P.; Nazeeruddin, M. K.; Gratzel, M. Nature 2013,
499, 316.
(33) Malinkiewicz, O.; Yella, A.; Lee, Y. H.; Espallargas, G.
M.; Graetzel, M.; Nazeeruddin, M. K.; Bolink, H. J. Nat.
Photon. 2014, 8, 128.
(34) Kim, H.-S.; Park, N.-G. J. Phys. Chem. Lett. 2014, 5,
2927.
(35) Yu, J. C.; Kim, D. B.; Baek, G.; Lee, B. R.; Jung, E. D.;
Lee, S.; Chu, J. H.; Lee, D.-K.; Choi, K. J.; Cho, S.; Song, M.
H. Adv. Mater. 2015, 27, 3492.
(36) Liang, P.-W.; Liao, C.-Y.; Chueh, C.-C.; Zuo, F.; Wil-
liams, S. T.; Xin, X.-K.; Lin, J.; Jen, A. K. Y. Adv. Mater. 2014,
26, 3748.
(37) Zuo, L.; Gu, Z.; Ye, T.; Fu, W.; Wu, G.; Li, H.; Chen, H.
J. Am. Chem. Soc. 2015, 137, 2674.
(38) Abate, A.; Saliba, M.; Hollman, D. J.; Stranks, S. D.;
Wojciechowski, K.; Avolio, R.; Grancini, G.; Petrozza, A.;
Snaith, H. J. Nano Lett. 2014, 14, 3247.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCES
(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am.
Chem. Soc. 2009, 131, 6050.
(2) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.;
Snaith, H. J. Science 2012, 338, 643.
(3) Kim, H.-S.; Lee, C.-R.; Im, J.-H.; Lee, K.-B.; Moehl, T.;
Marchioro, A.; Moon, S.-J.; Humphry-Baker, R.; Yum, J.-H.;
Moser, J. E.; Grätzel, M.; Park, N.-G. Sci. Rep. 2012, 2, 591.
(4) Liu, M.; Johnston, M. B.; Snaith, H. J. Nature 2013, 501,
395.
(5) Lee, J.-W.; Seol, D.-J.; Cho, A.-N.; Park, N.-G. Adv. Ma-
ter. 2014, 26, 4991.
(6) Zhou, H.; Chen, Q.; Li, G.; Luo, S.; Song, T.-b.; Duan,
H.-S.; Hong, Z.; You, J.; Liu, Y.; Yang, Y. Science 2014, 345,
542.
(7) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.;
Seo, J.; Seok, S. I. Science 2015, 348, 1234.
(8) Shi, J.; Xu, X.; Li, D.; Meng, Q. Small 2015, 11, 2472.
(9) Abrusci, A.; Stranks, S. D.; Docampo, P.; Yip, H.-L.;
Jen, A. K. Y.; Snaith, H. J. Nano Lett. 2013, 13, 3124.
(10) Kuang, C.; Tang, G.; Jiu, T.; Yang, H.; Liu, H.; Li, B.;
Luo, W.; Li, X.; Zhang, W.; Lu, F.; Fang, J.; Li, Y. Nano Lett.
2015, 15, 2756.
(11) Azimi, H.; Ameri, T.; Zhang, H.; Hou, Y.; Quiroz, C. O.
R.; Min, J.; Hu, M.; Zhang, Z.-G.; Przybilla, T.; Matt, G. J.;
Spiecker, E.; Li, Y.; Brabec, C. J. Adv. Energy Mater. 2015, 5,
1401692.
(12) Liu, X.; Yu, H.; Yan, L.; Dong, Q.; Wan, Q.; Zhou, Y.;
Song, B.; Li, Y. ACS Appl. Mater. & Interfaces 2015, 7, 6230.
(13) Docampo, P.; Ball, J. M.; Darwich, M.; Eperon, G. E.;
Snaith, H. J. Nat. Commun. 2013, 4, 2761.
(14) Wang, Q.; Shao, Y.; Dong, Q.; Xiao, Z.; Yuan, Y.;
Huang, J. Energy & Environmental Science 2014, 7, 2359.
(15) Wojciechowski, K.; Saliba, M.; Leijtens, T.; Abate, A.;
Snaith, H. J. Energy Environ. Sci. 2014, 7, 1142.
(16) Kim, H.-S.; Mora-Sero, I.; Gonzalez-Pedro, V.; Fa-
bregat-Santiago, F.; Juarez-Perez, E. J.; Park, N.-G.; Bis-
quert, J. Nat Commun 2013, 4, 2242.
(39) Credgington, D.; Durrant, J. R. J. Phys. Chem. Lett.
2012, 3, 1465.
(17) Snaith, H. J.; Abate, A.; Ball, J. M.; Eperon, G. E.;
Leijtens, T.; Noel, N. K.; Stranks, S. D.; Wang, J. T.-W.;
Wojciechowski, K.; Zhang, W. J. Phys. Chem. Lett. 2014, 5,
1511.
(18) Leijtens, T.; Eperon, G. E.; Pathak, S.; Abate, A.; Lee,
M. M.; Snaith, H. J. Nat Commun 2013, 4, 2885.
(19) Ito, S.; Tanaka, S.; Manabe, K.; Nishino, H. J. Phys.
Chem. C 2014, 118, 16995.
(40) Schwanitz, K.; Weiler, U.; Hunger, R.; Mayer, T.; Jae-
germann, W. J. Phys. Chem. C 2007, 111, 849.
(41) Schwanitz, K.; Mankel, E.; Hunger, R.; Mayer, T.; Jae-
germann, W. CHIMIA Int. J. Chem. 2007, 61, 796.
(42) Nagaoka, H.; Ma, F.; deQuilettes, D. W.; Vorpahl, S.
M.; Glaz, M. S.; Colbert, A. E.; Ziffer, M. E.; Ginger, D. S. J.
Phys. Chem. Lett. 2015, 6, 669.
ACS Paragon Plus Environment

Journal of the American Chemical Society
(43) Knesting, K. M.; Ju, H.; Schlenker, C. W.; Giordano,
A. J.; Garcia, A.; Smith, O. N. L.; Olson, D. C.; Marder, S.
R.; Ginger, D. S. J. Phys. Chem. Lett. 2013, 4, 4038.
(44) Stranks, S. D.; Snaith, H. J. Nat Nano. 2015, 10, 391.
(45) Fujishima, A.; Rao, T. N.; Tryk, D. A. J. Photochem.
Photobiol., C 2000, 1, 1.
Page 10 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
1
2
3
Table of Contents Graphic
4
5
6
7
8
9
A Multifunctional Fullerene Derivative for Interface Engineering in
Perovskite Solar Cells
Yaowen Li*,†,‡, Yue Zhao,† Qi Chen,‡ Yang (Michael) Yang,‡ Yongsheng Liu,‡ Ziruo Hong,*,‡ Zonghao Liu, ‡ Yao-Tsung Hsieh,‡ Lei
Meng, ‡ Yongfang Li, *† Yang Yang*‡
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment