The role of chlorine in the formation process of CH3NH3PbI3-xClx perovskite

Article abstract of DOI:10.1002/adfm.201401872

CH₃NH₃PbI_3-xCl_x is a commonly used chemical formula to represent the methylammonium lead halide perovskite fabricated from mixed chlorine- and iodine-containing salt precursors. Despite the rapid progress in improving its photovoltaic efficiency, fundamental questions remain regarding the atomic ratio of Cl in the perovskite as well as the reaction mechanism that leads to its formation and crystallization. In this work we investigated these questions through a combination of chemical, morphological, structural and thermal characterizations. The elemental analyses reveal unambiguously the negligible amount of Cl atoms in the CH₃NH₃PbI_3-xCl_x perovskite. By studying the thermal characteristics of methylammonium halides as well as the annealing process in a polymer/perovskite/FTO glass structure, we show that the formation of the CH₃NH₃PbI_3-xCl_x perovskite is likely driven by release of gaseous CH₃NH₃Cl (or other organic chlorides) through an intermediate organometal mixed halide phase. Furthermore, the comparative study on CH₃NH₃I/PbCl₂ and CH₃NH₃I/PbI₂ precursor combinations with different molar ratios suggest that the initial introduction of a CH₃NH₃⁺ rich environment is critical to slow down the perovskite formation process and thus improve the growth of the crystal domains during annealing; accordingly, the function of Cl^- is to facilitate the release of excess CH₃NH₃⁺ at a relatively low annealing temperatures.

Full text of DOI:10.1002/adfm.201401872

www.afm-journal.de
www.MaterialsViews.com
The Role of Chlorine in the Formation Process of
“CH3NH3PbI3-xClx” Perovskite
Hui Yu, Feng Wang, Fangyan Xie, Wenwu Li, Jian Chen, and Ni Zhao*
for PV applications, CH3NH3PbI3 is often
combined with mesostructured cells to
achieve good device performance, whereas
CH3NH3PbI3-xClx can yield high PCEs
in a planar device structure and through
a simple spin-coating or co-evaporation
film deposition process.[11–17] Very recently
there have been studies showing high-per-
formance planar CH3NH3PbI3 solar cells,
however, the device fabrication requires
more complicated solution processes.[18,19]
It has been broadly observed that the PCE
of perovskite solar cells is closely related
to the morphology and crystal structure
of the perovskite absorbers.[20–23] Accord-
ingly, the film growth parameters, such
as the molar ratio and concentration of
the precursor solutions as well as the
annealing temperature and duration, have
to be carefully designed.
CH3NH3PbI3-xClx is a commonly used chemical formula to represent the
methylammonium lead halide perovskite fabricated from mixed chlorine- and
iodine-containing salt precursors. Despite the rapid progress in improving its
photovoltaic efficiency, fundamental questions remain regarding the atomic
ratio of Cl in the perovskite as well as the reaction mechanism that leads to
its formation and crystallization. In this work we investigated these questions
through a combination of chemical, morphological, structural and thermal
characterizations. The elemental analyses reveal unambiguously the negli-
gible amount of Cl atoms in the CH3NH3PbI3-xClx perovskite. By studying the
thermal characteristics of methylammonium halides as well as the annealing
process in a polymer/perovskite/FTO glass structure, we show that the
formation of the CH3NH3PbI3-xClx perovskite is likely driven by release of
gaseous CH3NH3Cl (or other organic chlorides) through an intermediate
organometal mixed halide phase. Furthermore, the comparative study on
CH3NH3I/PbCl2 and CH3NH3I/PbI2 precursor combinations with different
molar ratios suggest that the initial introduction of a CH3NH3+ rich environ-
ment is critical to slow down the perovskite formation process and thus
improve the growth of the crystal domains during annealing; accordingly, the
function of Cl− is to facilitate the release of excess CH3NH3+ at a relatively low
annealing temperatures.
In this work we focus on the growth pro-
cess of the MAPbX3 perovskite absorber
prepared from mixed chlorine- and iodine-
containing precursors, which is therefore
often referred as CH3NH3PbI3-xClx. Com-
monly, the CH3NH3PbI3-xClx fi lms are pre-
pared by spin-coating a precursor solution
1. Introduction
of CH3NH3I and PbCl2 in dimethylformamide (DMF) at a 3:1
molar ratio, followed by thermal annealing at around 100 °C.[16]
The PCEs of CH3NH3PbI3-xClx PVs have already exceeded
15%[14,24,25] thanks to the evolution of material and device engi-
neering, however, the underlying mechanism that leads to the
conversion of the precursors to the perovskite is not clear. In
particular, it is widely reported that the CH3NH3PbI3-xClx fi lms
show the same characteristic X-ray diffraction (XRD) peaks as
the CH3NH3PbI3 fi lms, suggesting that the two compounds
possess identical crystal lattices. Accordingly, only very low
chlorine (Cl) content should be present in CH3NH3PbI3-xClx,
because otherwise the lattice distortion induced by Cl atoms
will cause shift and broadening of the XRD peaks. Therefore
a fundamental question remains regarding the function of Cl
atoms during the formation process of the perovskite.
Recently organometal halide perovskites have emerged as a
promising photovoltaic (PV) material that exhibits high power
conversion efficiency (PCE) as well as low-temperature proces-
sibility.[1–4] Among various perovskites, the methylammonium
lead halide perovskites (MAPbX3), such as, CH3NH3PbI3,[5,6]
CH3NH3PbI3-xClx,[7,8] and CH3NH3PbBr3,[9,10] have been studied
most extensively due to their wide absorption range (visible-to-
near infrared) and excellent electronic properties. When used
H. Yu, Dr. F. Wang, W. Li, Prof. N. Zhao
Department of Electronic Engineering
The Chinese University of Hong Kong
New Territories, Hong Kong
E-mail: nzhao@ee.cuhk.edu.hk
Here we investigated this question through a combination
of chemical, morphological, structural and thermal analyses.
The energy dispersive X-ray spectrometry (EDX) and surface-
sensitive X-ray photoelectron spectroscopy (XPS) depth profile
reveal unambiguously the negligible amount of Cl atoms in the
CH3NH3PbI3-xClx compound. A polymer/perovskite/FTO glass
Prof. F. Xie, Prof. J. Chen
Instrumental Analysis and Research Center
Sun Yat-sen University
Guangzhou 510275, P. R. China
DOI: 10.1002/adfm.201401872
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1

www.afm-journal.de
www.MaterialsViews.com
testing structure is applied to study the thermal annealing pro-
cess, and the results show that the formation of the perovskite
is likely accompanied by release of gaseous CH3NH3Cl (or other
organic chlorides in a similar form) through an intermediate
organometal mixed halide phase. Based on the observation we
further compared the MAPbX3 fi lms produced from CH3NH3I/
PbCl2 and CH3NH3I/PbI2 precursor combinations with dif-
ferent molar ratios. The results demonstrate that the initial
an FTO substrate. The elemental composition of the annealed
film was analyzed by X-ray photoelectron spectroscopy (XPS).
A full spectrum scan from the top surface of the film shows
no trace of Cl in the perovskite (Figure 1b). Since the detec-
tion depth of XPS is limited to a few tens of nm, we further
performed XPS depth profile measurement by sputtering the
perovskite films with Ar ion gun (EX05 Ion Gun). As shown in
Figure 1c, no signal related to the Cl 2p electrons is found in
the bulk perovskite film (around 310 nm thick); very weak Cl
signal, which corresponds to 1% Cl, can be detected for the top
20 nm of the film.
The XPS analysis clearly suggests that there is negligible
amount of Cl atoms in the fully annealed perovskite film.
Therefore, one can speculate that Cl atoms escape from the
film during the annealing process, most likely in the form
of gaseous Cl-containing compounds. Although Cl does not
exist in the final product of the perovskite film, the use of the
Cl-containing precursor seems to play an important role in
determining the electronic properties of the perovskite thin
films. The time-resolved photoluminescence study by Snaith
et al.[26] shows that the MAPbX3 perovskite films prepared
from the CH3NH3I/PbCl2 precursors exhibit longer charge car-
rier diffusion lengths as compared to those prepared from the
CH3NH3I/PbI2. In order to understand how the Cl involved
reaction affects the properties (e.g., crystallinity and preferred
+
introduction of a CH3NH3 rich environment is critical to slow
down the perovskite formation process and thus improve the
growth of the crystal domains during annealing and that the
function of Cl is to facilitate the release of excess CH3NH3+ at a
relatively low annealing temperature.
2. Results and Discussion
2.1. Elemental Analysis of Mixed Halide Perovskite
The mixed halide perovskite (CH3NH3PbI3-xClx) films were pre-
pared in a nitrogen filled glove box by spin-coating a precursor
solution of CH3NH3I and PbCl2 in DMF at a 3:1 molar ratio,
followed by thermal annealing at 100 °C for 50 min.
The scanning electron microscope (SEM) image in Figure 1a
demonstrates a full coverage of the annealed perovskite film on
Figure 1. a) Morphology of the mixed halide perovskite (CH3NH3PbClxI3-x) films on FTO substrates by SEM; b) XPS survey spectrum of the top surface
of the mixed halide perovskite (CH3NH3PbClxI3-x) film; *Atomic percentage was calculated from the high resolution XPS spectra; c) Cl2p core level XPS
spectra under Ar ion gun sputtering on the perovskite film, characteristic peak binding energy of Cl2p3/2 and Cl2p1/2 (198.9 eV and 200.5 eV) highlighted
by the arrows. "0 nm" means the top surface of the perovskite film before Ar ion gun sputtering, and other values (e.g., 20 nm, 50 nm, 100 nm, etc.)
refer to the gradually increased thickness of the removed perovskite during sputtering.
©
2
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872

www.afm-journal.de
www.MaterialsViews.com
Figure 2. Schematics and photos of a) perovskite/FTO and b) PMMA/perovskite/FTO structures before (top) and after (bottom) annealing. EDX
measured atomic percentages of the main elements (Pb, I and Cl) in the annealed perovskite films c) without and d) with the PMMA blocking layer.
orientation) of the MAPbX3 perovskite films, we further exam-
ined the formation process of the so-called mixed halide perovs-
kite through thermal and elemental analysis.
reaction from the precursor phase to perovskite phase requires
release of gaseous Cl-containing compound(s), a process that
is largely suppressed by the PMMA blocking layer. (We note
that PMMA is not a perfect gas barrier. However, the color of
the FTO/perovskite/PMMA structure changes little even after
6 hours’ annealing, suggesting that the release of the gas mol-
ecules is a very slow process.)
2.2. Formation Process of Mixed Halide Perovskite
Figure 2 illustrates a comparative study on the annealing pro-
cess of two MAPbX3 perovskite films, where one precursor
film was spun on an FTO substrate (Figure 2a) and the other
was sandwiched between FTO and a 200 nm thick polymethyl
methacrylate (PMMA) layer (Figure 2b). Color variation is a
straightforward indicator for the progress of the chemical reac-
tion occurring during the annealing process. For the perovs-
kite/FTO structure, a color change from yellow to dark brown
was observed after annealing at 100 °C for 50 min, and the full
formation of the perovskite phase was also confirmed by the
X-ray diffraction (XRD) measurement. In contrast, the PMMA/
perovskite/FTO structure exhibits only a slight color change,
indicating that the reaction to form perovs-
The above experiment clearly suggests that the release of the
Cl-containing compound(s) is closely linked to the formation of
crystalline MAPbX3 perovskite films. It is therefore important to
understand the possible chemical reactions involved in the per-
ovskite formation process. Dualeh et al. proposed that Reaction
(1) in Scheme 1 dominates the annealing process.[22] However,
we note that the initial CH3NH3I and PbCl2 compounds appear
to be white, while their DMF solution mixture produces a pre-
cursor film with light yellow color (Figure 2a), indicating that a
chemical reaction must occur to form a new precursor phase.
Hence, Reaction (1) may involve several intermediate steps.
Firstly, since the amount of CH3NH3I in molar is 3 times that
Reaction (1):
kite was suppressed. The elemental compo-
sition of both structures was analyzed using
energy dispersive X-ray (EDX) spectroscopy.
For the perovskite film without the PMMA
blocking layer, there is a negligible trace of Cl
in the film, as shown in Figure 2c. (Note that
the measured atomic percentage of 0.7% for
Cl is lower than the detection limit of ≈1%
for the EDX setup.) For the FTO/perovskite/
PMMA structure, however, the atomic per-
centage for Cl is 6.7%, and the atomic ratio
of Cl to (Cl +I) is around 0.4 (Figure 2d),
which is the same as the ratio in the pre-
cursor solution. Therefore it is likely that the
PbCl2 +3CH3NH3I CH NH PbI +2CH NH Cl g
( )
3
3
3
3
3
Reaction (2):
(2-1)
PbCl2 +3CH3NH3I PbI2 +CH3NH3I +2CH3NH3Cl
PbI2 +CH3NH3I +2CH3NH3Cl CH NH PbI +2CH NH Cl g
(2-2)
( )
3
3
3
3
3
(2-3) PbI2 + xCH3NH3I + yCH3NH3Cl
CH NH
PbI2+xCly (Intermediate phase)
CH NH PbI +CH NH Cl g
(
)
3
3
x+y
( )
3
3
3
3
3
Scheme 1. Possible chemical reactions involved in the perovskite formation process.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3

www.afm-journal.de
www.MaterialsViews.com
of PbCl2, Reaction (2–1) is proposed, where part of CH3NH3I
reacts with PbCl2 to form PbI2. As a result of Reaction (2–1),
the as-spun film contains a mixture phase of PbI2 (appears in
light yellow), CH3NH3I, CH3NH3Cl, and possibly some unre-
acted PbCl2. During annealing two reaction processes may
occur: For Reaction (2–2), CH3NH3I reacts with PbI2 to form
the dark brown CH3NH3PbI3 perovskite; meanwhile, the excess
CH3NH3Cl escapes from the film through either sublimation
or decomposition into other gaseous Cl-containing mixtures.
For Reacton (2–3), an intermediate phase forms during spin-
coating or the early stage of the annealing process. In this case,
the CH3NH3PbI3 perovskite crystal domains grow during the
decomposition of the intermediate phase, which is driven by
the release of gaseous CH3NH3Cl (or other organic chlorides).
In order to find out whether Reaction (2–2) or Reaction
(2–3) dominates the formation of CH3NH3PbI3 perovskite,
we performed XRD measurement on 5- and 15-min annealed
precursor films to observe the intermediate products of the
reaction. The XRD measurement was performed in ambient
conditions (60% humidity level), and a 300 nm CYTOP (a
fluorinated polyether) layer was covered on the samples to
slow down the penetration of water molecules in air. As shown
in the XRD spectra in Figure 3, two perovskite peaks at 14.2°
and 28.5° can be readily observed in the freshly made 5- and
15-minutes annealed films, although the peak intensity is rela-
tively small.
We therefore suspect that the 15.7° and 31.5° peaks may origi-
nate from another (CH3NH3)x+yPbI2+xCly phase or a mixture
of several phases. To unambiguously identify the intermediate
reaction products, a combination of in situ XRD and XPS meas-
urements in vacuum would be critical. Such study is currently
undergoing. It's worth mentioning that there have been several
reports recently to study the intermediate phase structures.[29,30]
2.3. Comparison Between CH3NH3PbI3 Perovskites Formed
from CH3NH3I/PbCl2 and CH3NH3I/PbI2 Precursor Mixtures
Since the amount of Cl in CH3NH3PbI3-xClx is negligible,
it is more precise to write the formula of the perovskite as
CH3NH3PbI3. Although at present the exact formation reac-
tion of CH3NH3PbI3 cannot be identified, one can still explore
the role of Cl by comparing the formation processes of the
perovskites from the CH3NH3I/PbCl2 and CH3NH3I/PbI2 pre-
cursor combinations. Firstly, we note that due to the better
solubility of PbI2, the CH3NH3I/PbI2 precursor solution can
be made at a 1:1 molar ratio. (In contrast, a 3:1 molar ratio is
required for CH3NH3I/PbCl2). A CH3NH3PbI3 perovskite film
is obtained by spin coating the precursor solution on an FTO
substrate, followed by only 10 minutes’ annealing at 100 °C. In
this case, however, there are large voids formed in the perovs-
kite film (Figure S1, Supporting Information), and changing
the annealing time and temperature cannot improve the film
quality. We note that previous studies have demonstrated
high performance planar PVs with closely packed crystalline
CH3NH3PbI3 fi lms fabricated from the CH3NH3I/PbI2 pre-
cursor combination. For instance, PCEs of over 15% and 12%
are achieved through a sequential deposition method[18] and
vapor-assisted solution process,[19] respectively. In both cases,
a CH3NH3I-rich environment was provided, suggesting that
excess CH3NH3I could be beneficial to the formation of high
quality CH3NH3PbI3 fi lms.
The peaks at 12.7° and 17.5° correspond to PbI2 and
CH3NH3Cl, respectively. Two strong peaks at 15.7° and 31.5°
can be seen initially but drops quickly after exposure of the
films in air for a few minutes, accompanied by the growth of
the perovskite peaks. In previous studies, the 15.7° and 31.5°
peaks are correlated to the (100) and (200) diffractions of a
CH3NH3PbCl3 single crystal phase grown in an aqueous solu-
tion.[27,28] However, such assignment could not explain the dis-
appearance of the “CH3NH3PbCl3” phase after air exposure.
We therefore monitored the film morphology as a func-
tion of the CH3NH3I:PbI2 molar ratio. Notably, as the molar
ratio increases to 2:1 or above, the annealing condition has
to be adjusted to 150 °C for 20 minutes to ensure full con-
version of the precursors to the perovskite phase (Analysis of
the incompletely converted precursor films annealed at lower
temperatures is provided in Figure S2,S3 in Supporting Infor-
mation). Only by reducing the film thickness to < 200 nm, the
annealing temperature can be lowered to 130 °C. (Note that
the thickness of the CH3NH3I:PbCl2 produced films is around
310 nm). Figure 4 shows the SEM images of the annealed per-
ovskite films obtained from the CH3NH3I:PbI2 precursor mix-
tures with different molar ratios.
It can be seen that uniform and compact perovskite films are
obtained when the molar ratio exceeds 2:1. Furthermore, the 2:1
and 3:1 films also exhibit stronger preferred orientation along
(110), as revealed by the XRD measurement (Figure 5). Inter-
estingly, these structural characteristics are very similar to those
of the perovskite phase prepared from the Cl-containing precur-
sors (CH3NH3I:PbCl2) with 3:1 molar ratio. Nevertheless, the PV
performances of the two perovskite phases are very different in
a planar device structure (FTO/compact TiO2/ Perovskite/P3HT/
Ag) (Figure S4, Supporting Information): the 100 °C annealed
Figure 3. XRD patterns for the precursor films annealed for 5 mins and
15 mins. ♦, ◊ and # denote the peaks of CH3NH3PbI3 perovskite, PbI2,
and CH3NH3Cl, respectively; the unlabelled peaks are attributed to the
intermediate phase.
©
4
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872

www.afm-journal.de
www.MaterialsViews.com
Figure 4. SEM images of annealed perovksite films obtained from
CH3NH3I:PbI2 precursor mixtures with a) 1:1, b) 1.5:1, c) 2:1, d) 3:1 molar
ratios. Scar bar: 3 µm.
Figure 5. XRD patterns for the fully annealed perovksite films obtained
from the CH3NH3I:PbI2 precursor mixtures with different molar ratios.
Small peaks at 20–30° (highlighted by arrows) correspond to (200), (211),
and (202) diffractions in the films with CH3NH3I:PbI2 ratios of 1:1 and
1.5:1, indicating more randomly oriented crystal domains.
perovskite formed from CH3NH3I:PbCl2 exhibits a PCE of
≈12%, whereas the 150 °C annealed perovskite formed from 2:1
and 3:1 CH3NH3I:PbI2 with a comparable thickness yields <2%
PCE in the same device configuration. The 130 °C annealed thin
perovskite formed from 2:1 CH3NH3I:PbI2 (optimized for PV
performance) yields ≈8% PCE. In this case, although the reduc-
tion in the annealing temperature leads to a better PV perfor-
mance, the thin film thickness limits the light absorption and
thus the generation of photocurrent in the perovskite.
The above morphological and structural studies of the
perovskite formation process suggest that the use of excess
CH3NH3X (X: Cl or I) can slow down the formation and crys-
tallization of the perovskite, possibly through the introduction
of certain intermediate phases. This principle works for both
pure iodine precursors (CH3NH3I/PbI2) as well as mixed halide
precursors (CH3NH3I/PbCl2), and can be used to obtain dense,
uniform and highly crystalline perovskite films. On the other
hand, the species of the precursors could largely affect the
annealing temperature. The CH3NH3I/PbCl2 (3:1 molar ratio)
combination only requires 100 °C annealing to achieve full
conversion to perovskite; while in the case of CH3NH3I/PbI2
(3:1 molar ratio), the annealing temperature can be as high as
150 °C, due possibly to a higher energy required for the forma-
tion (or decomposition) of the intermediate phase or release of
the gaseous I-containing compounds (e.g., CH3NH3I).
we further characterized the thermal property of CH3NH3I and
CH3NH3Cl via thermal gravimetric analysis (TGA) and differ-
ential scanning calorimetry (DSC) measurements. The reac-
tions related to a weight loss of CH3NH3I and CH3NH3Cl could
be monitored by the TGA measurement.
As shown in Figure 6a, the TGA curves show nearly 100%
weight loss between 250 °C and 360 °C for CH3NH3Cl and
between 290 °C and 370 °C for CH3NH3I, respectively. The
lower onset of the weight loss temperature of CH3NH3Cl sug-
gests that it is easier to turn into gaseous phase as compared
to CH3NH3I. Note that the TGA measurement was performed
on CH3NH3I and CH3NH3Cl powders. In the case of around
300 nm thick perovskite precursor films, a much larger surface-
to-volume ratio is provided; therefore the sublimation temper-
ature could be considerably lower than 250–300 °C. (Directly
performing the TGA measurement on thin film perovskite
samples cannot yield reliable results due to the extremely low
mass loss of the material as compared to the substrate.)
We further performed DSC measurement to understand in
what manner CH3NH3Cl and CH3NH3I could possibly escape
from the precursor films during the annealing process. In gen-
eral, a sublimation process is endothermic, while a decomposi-
tion process is exothermic. As shown in Figure 6b, the endo-
thermic peaks in the DSC curves suggest that the release of
CH3NH3Cl and CH3NH3I is related to a sublimation process.
Due to the uncertainty of the intermediate product in Reaction
(2–3), we can only speculate that the release of the Cl atoms
is in the form of gaseous CH3NH3Cl. Both the TGA and DSC
2.4. Thermal Behavior of CH3NH3I and CH3NH3Cl
Since the formation of the CH3NH3PbI3 perovskites is signifi-
cantly influenced by the release of methylammonium halides,
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5

www.afm-journal.de
www.MaterialsViews.com
Figure 6. a) TGA and b) DSC curves for CH3NH3I and CH3NH3Cl powders.
CH3NH3PbI3 precursor solutions were prepared by dissolving 0.9 M
lead iodide (99.999%, Sigma-Aldrich) and 0.9 M for 1:1, 1.35M for 1.5:1,
1.8 M for 2:1, 2.7 M CH3NH3I for 3:1 in anhydrous DMF. The precursor
solutions were filtered though a 0.2 µm PTFE filter before spin-coating.
Fabrication of Polymer/Perovskite/FTO Glass Sandwiched Structure Test
Samples: The whole process was carried out in a nitrogen filled glovebox.
CH3NH3PbI3-xClx precursor solution was first spin-coated on a FTO
glass. After drying over-night, a 40 mg/mL poly(methyl methacrylate)
(PMMA, Mw of 120000, purchased from Sigma-Aldrich) solution in butyl
acetate (anhydrous, 99%, Sigma-Aldrich) was then spin-coated on the
perovskite film either before or after annealing.
results confirm that the release of CH3NH3I occurs at higher
temperature as compared to CH3NH3Cl, which may explain the
higher annealing temperature required for the CH3NH3I/PbI2
precursor mixtures. Therefore we propose that the function of
Cl (or more precisely, Cl−) is to facilitate the release of excess
CH3NH3+ at a relatively low annealing temperature.
3. Conclusion
Fabrication of Planar Solar Cell Devices: Solar cells were fabricated
as follows: FTO (DHS-FTO22–15N, HeptaChroma) substrates were
sequentially cleaned with deionized water, acetone, 2-propanol for
30 mins each, followed with oxygen plasma treatment for 1.5 mins. A
compact TiO2 hole-block layer was deposited on cleaned FTO by spin-
coating a mildly acidic solution of titanium butoxide (97%, Sigma-
Aldrich) in ethanol (consisting of 0.5 mL titanium butoxide and 50 µL
37% HCl solution in 6 ml ethanol) at 5000 rpm for 25s, and annealed
at 500 °C for 45 min in an oven. Then TiO2 coated FTO substrates
were transferred into a nitrogen-filled glovebox and pre-heated at
100 °C for 10 min. After the substrates were cooled to room
temperature, the perovskite precursor solution was spin-coated at
3000 r.p.m. for 60 s. After drying for more than 10 mins, the as spun films
were annealed for 50 mins at 100 °C in the case of CH3NH3PbI3-xClx and
at 130 °C for 20 min (or 150 °C for 20 min) in the cases of CH3NH3I/PbI2
at molar ratios of 2:1 and 3:1. 30nm P3HT (Poly(3-hexylthiophene-
2,5-diyl), LT-S909 from Lumtec.) hole transport layer was deposited on
the annealed perovskite film by spin-coating 15 mg/ml P3HT solution
in chlorobenzene at 3000 r.p.m. for 60 s. Finally, 100 nm Ag electrodes
were thermally evaporated through a shadow mask under a vacuum of
1 × 10−6 Torr to complete the device fabrication.
Characterizations: Morphology of the perovskite films were observed
by HR-FESEM (FEI, Quanta 400). The elemental composition in the
films was measured by EDX with Standardless Quantification Model
and by XPS (Thermo Fisher Scientific, ESCALAB 250). The crystalline
structure for the perovskite films were identified by XRD (Rigaku,
SmartLab). Thermal analyses for CH3NH3I and CH3NH3Cl powders were
carried out by Thermogravimetric Analyzer (PerkinElmer, TGA 6) and
DSC (PerkinElmer, TGA 6). Sample thicknesses were measured using an
Alpha step 500 Surface profiler. The current density–voltage (J–V) curves
were measured (Keithley Instruments, 2612 Series SourceMeter) under
simulated AM 1.5 sunlight at 100 mW cm−2 irradiance generated by an
94011A-ES Sol series Solar Simulator.
In summary, we have shown by EDX and XPS depth-profile that
only negligible amount of Cl atoms exist in the fully annealed,
so-called “mixed halide” perovskite (CH3NH3PbI3-xClx) films.
By monitoring the annealing products in the perovskite/FTO
and PMMA/perovskite/FTO testing structures we found that
the slow formation of the CH3NH3PbI3-xClx perovskite is likely
driven by release of gaseous CH3NH3Cl (or other organic chlo-
rides in a similar form) through an intermediate organolead
mixed halide phase. Furthermore, by comparing the perovs-
kite films produced from CH3NH3I/PbCl2 and CH3NH3I/
PbI2 precursor combinations with different molar ratios, we
+
demonstrated that the initial introduction of a CH3NH3 rich
environment is critical to slow down the perovskite formation
process and thus improve the growth of the crystal domains
during annealing; accordingly, the function of Cl is to allow
+
the removal of excess CH3NH3 at a relatively low annealing
temperature. Our study reveals the importance of controlling
the formation reaction of perovskites. The method of using
mixed halide precursors to tune the annealing process can be
extended to other organometal perovskite material systems.
4. Experimental Section
Preparation of Perovskite Precursor: CH3NH3I was synthesized in an ice
bath for 2 hrs by reacting methylamine (CH3NH2, 33 wt% in ethanol from
Sigma-Aldrich) with hydroiodic acid (HI, 57 wt% in water from Sigma-
Aldrich). The white powders were precipitated by drying at 60 °C and
washed for 3 times with diethyl ether (Sigma-Aldrich) before further dried
out to be stored in nitrogen-filled glove-box. CH3NH3Cl was synthesized
in the same way as CH3NH3I by reacting methylamine with hydrochloric
acid (HCl, 36.5–38 wt% in water from Sigma-Aldrich). CH3NH3PbI3-xClx
precursor solution was prepared by dissolving 0.88M lead chloride
Supporting Information
(99.999%, Sigma-Aldrich) and 2.64
N,N-Dimethylformamide (DMF, 99.8%, Sigma-Aldrich). Similarly,
M CH3NH3I in anhydrous
Supporting Information is available from the Wiley Online Library or
from the author.
©
6
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872

www.afm-journal.de
www.MaterialsViews.com
[14] M. Liu, M. B. Johnston, H. J. Snaith, Nature 2013, 501, 395.
Acknowledgements
[15] J. B. You, Z. R. Hong, Y. Yang, Q. Chen, M. Cai, T. B. Song,
C. C. Chen, S. R. Lu, Y. S. Liu, H. P. Zhou, Y. Yang, ACS Nano 2014,
8, 1674.
[16] B. Conings, L. Baeten, C. De Dobbelaere, J. D’Haen, J. Manca,
H.-G. Boyen, Adv. Mater. 2014, 26, 2041.
[17] P.-W. Liang, C.-Y. Liao, C.-C. Chueh, F. Zuo, S. T. Williams, X.-K. Xin,
J. Lin, A. K. Y. Jen, Adv. Mater. 2014, 26, 3748.
[18] D. Y. Liu, T. L. Kelly, Nat. Photonics 2014, 8, 133.
[19] Q. Chen, H. Zhou, Z. Hong, S. Luo, H.-S. Duan, H.-H. Wang, Y. Liu,
G. Li, Y. Yang, J. Am. Chem. Soc. 2014, 136, 622.
The authors gratefully acknowledge the fundings from Research Grants
Council of Hong Kong (Grants No. CUHK419311 and T23–407/13-N),
National Natural Science Foundation of China (Grants No. 61205036
and 51303217) and Shun Hing Institute of Advanced Engineering (Grant
no. 8115041).
Received: June 8, 2014
Revised: August 7, 2014
Published online:
[20] G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely, H. J. Snaith,
Adv. Funct. Mater. 2014, 24, 151.
[1] N. G. Park, J. Phys. Chem. Lett. 2013, 4, 2423.
[2] H. J. Snaith, J. Phys. Chem. Lett. 2013, 4, 3623.
[3] R. F. Service, Science 2014, 344, 458.
[4] S. Kazim, M. K. Nazeeruddin, M. Grätzel, S. Ahmad, Angew. Chem.
Int. Ed. 2014, 53, 2812.
[5] L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu,
M. K. Nazeeruddin, M. Grätzel, J. Am. Chem. Soc. 2012, 134, 17396.
[6] O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel,
M. K. Nazeeruddin, H. J. Bolink, Nat. Photonics 2014, 8, 128.
[7] M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami, H. J. Snaith,
Science 2012, 338, 643.
[21] J. J. Choi, X. Yang, Z. M. Norman, S. J. L. Billinge, J. S. Owen, Nano
Lett. 2014, 14, 127.
[22] A. Dualeh, N. Tétreault, T. Moehl, P. Gao, M. K. Nazeeruddin,
M. Grätzel, Adv. Funct. Mater. 2014, 24, 3250.
[23] M. Saliba, K. W. Tan, H. Sai, D. T. Moore, T. Scott, W. Zhang,
L. A. Estroff, U. Wiesner, H. J. Snaith, J. Phys. Chem. C 2014, 118,
17171.
[24] J. T.-W. Wang, J. M. Ball, E. M. Barea, A. Abate,
J. A. Alexander-Webber, J. Huang, M. Saliba, I. Mora-Sero,
J. Bisquert, H. J. Snaith, R. J. Nicholas, Nano Lett. 2014, 14,
724.
[8] J. M. Ball, M. M. Lee, A. Hey, H. J. Snaith, Energy Environ. Sci. 2013,
6, 1739.
[9] J. H. Noh, S. H. Im, J. H. Heo, T. N. Mandal, S. I. Seok, Nano Lett.
2013, 13, 1764.
[10] E. Edri, S. Kirmayer, M. Kulbak, G. Hodes, D. Cahen, J. Phys. Chem.
Lett. 2014, 5, 429.
[11] E. Edri, S. Kirmayer, A. Henning, S. Mukhopadhyay, K. Gartsman,
Y. Rosenwaks, G. Hodes, D. Cahen, Nano Lett. 2014, 14, 1000.
[12] E. Edri, S. Kirmayer, S. Mukhopadhyay, K. Gartsman, G. Hodes,
D. Cahen, Nat. Commun. 2014, 5.
[25] J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao,
M. K. Nazeeruddin, M. Grätzel, Nature 2013, 499, 316.
[26] S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou,
M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza, H. J. Snaith,
Science 2013, 342, 341.
[27] O.-Y. Noriko, T. Matsuo, H. Suga, J. Phys. Chem. Solids 1990, 51,
1383.
[28] O.-Y. Noriko, T. Matsuo, H. Suga, J. Phys. Chem. Solids 1992, 53,
935.
[29] K. W. Tan, D. T. Moore, M. Saliba, H. Sai, L. A. Estroff, T. Hanrath,
H. J. Snaith, U. Wiesner, ACS Nano 2014, 8, 4730.
[30] Y. Zhao, K. Zhu, J. Phys. Chem. C 2014, 118, 9412.
[13] V. Gonzalez-Pedro, E. J. Juarez-Perez, W.-S. Arsyad, E. M. Barea,
F. Fabregat-Santiago, I. Mora-Sero, J. Bisquert, Nano Lett. 2014, 14, 888.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401872
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7