Energy &

Environmental
Science
View Article Online
PAPER
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.

View Journal | View Issue

The growth of a CH3NH3PbI3 thin film using

Cite this: Energy Environ. Sci.,
2016, 9, 1486
simplified close space sublimation for efficient
and large dimensional perovskite solar cells†
Qiang Guo,a Cong Li,a Wenyuan Qiao,a Shuang Ma,a Fuzhi Wang,a Bing Zhang,a
Linhua Hu,b Songyuan Dai*a and Zhan’ao Tan*a

Received 1st December 2015,
Accepted 15th February 2016
DOI: 10.1039/c5ee03620h
www.rsc.org/ees
A way to fabricate pin-hole-free perovskite thin films with optimal crystalline morphology and full
coverage is crucial for highly efficient and large scale perovskite solar cells (PerSCs). Herein, we
demonstrate a new approach of simplified close space sublimation (CSS) deposition to grow a high
quality CH3NH3PbI3 thin film in a low-vacuum and even non-vacuum oven. In this method, a highly
crystallized and uniform perovskite film can be easily obtained by simply controlling the thickness of the
PbI2 and CH3NH3I films, the reaction time of CH3NH3I vapor with the PbI2 film, and the distance of the
sublimation space. With this approach, a large area of 100 mm2 can be easily obtained. The crystallinity of
the perovskite film was investigated using X-ray diffraction (XRD) and the morphology of the perovskite
thin film was examined using atomic force microscopy (AFM) and scanning electron microscopy (SEM).
The planar heterojunction PerSCs with a structure of FTO/PEDOT:PSS/CH3NH3PbI3/PCBM/Al were
fabricated and power conversion efficiencies of 16.2%, 12.9% and 10.8% were achieved for the devices
with a photoactive area of 4 mm2, 25 mm2 and 100 mm2, respectively.

Broader context
High-efficiency organic-inorganic hybrid perovskite solar cells (PerSCs) have been regarded as a promising candidate for clean energy applications. The quality
of the perovskite thin film for large dimensional PerSCs is one of the major challenges for the commercial application of PerSCs in the future. In this work, a
simplified close space sublimation method was proposed for high efficiency and large dimensional PerSCs. The PbI2 film and CH3NH3I film were put face to
face to form a close space, and then sublimated CH3NH3I reacted with PbI2 and grew into CH3NH3PbI3 in an isothermal low-vacuum oven. A high quality
perovskite film can be obtained even with large dimensions by simply controlling the distance of the close space, the thickness of the PbI2 and CH3NH3I films,
and the reaction time, which demonstrates that this approach would be a new option for the manufacture of high-efficiency and large dimensional PerSCs.

1. Introduction been dramatically improved from the initially reported 3.8% to

Organic–inorganic hybrid perovskite solar cells (PerSCs) have
been highlighted in the field of photovoltaic technology in recent
years. Because of the advantages of high carrier mobility, low
exciton binding energy, high absorption coefficient and suitable
band gap, perovskite materials (especially CH3NH3PbI3) have
been regarded as ideal photovoltaic materials.1–3 With great
contributions from the chemistry, materials and physics com-
munities, the power conversion efficiency (PCE) of PerSCs has
a
State Key Laboratory of Alternate Electrical Power System with Renewable Energy
Sources, Beijing Key Laboratory of Novel Thin Film Solar Cells,
North China Electric Power University, Beijing 102206, China.
E-mail: tanzhanao@ncepu.edu.cn, sydai@ncepu.edu.cn
b
Institute of Applied Technology, Hefei Institutes of Physical Science,
Chinese Academy of Sciences, Hefei 230088, China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ee03620h
20.1% in just six years.4,5
The quality of the perovskite film plays a crucial role in
the performance of perovskite devices. Several approaches
have been proposed to improve the quality of perovskite films
such as one-step spin-coating,6–10 sequential deposition,11–14
co-evaporation,15,16 and a vapor assisted solution process.17–19
For the one-step spin-coating method, precursors of the lead
source (PbI2, PbCl2, or Pb(Ac)2) and an alkyl halogenated amine
(typically CH3NH3I) are usually dissolved in the same polar
solvent (dimethylformamide, dimethyl sulphoxide or gamma-
butyrolactone). The crystal perovskite formed through solvent
evaporation when high-speed spin-coating the precursors onto
the substrate. In this case, the fast solvent evaporation and
rapid reaction of PbI2 with CH3NH3I make it hard to control the
growth of the perovskite crystal, inducing pin-holes within the
final perovskite film, leading to low coverage of the substrate.11,18

1486 | Energy Environ. Sci., 2016, 9, 1486--1494 This journal is © The Royal Society of Chemistry 2016

Paper
Using PbCl2 or Pb(Ac)2 as the lead source could result in
smoother and better coverage of the perovskite film.8,10 How-
ever, there are by-products released during the formation of
stoichiometric CH3NH3PbI3 crystals,20,21 and these by-products
can potentially deteriorate the purity of the perovskite film.
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
Therefore, Grätzel et al. developed the sequential deposition
method.11 This two-step sequential approach first involves the
deposition of the lead source PbI2, and then dipping the PbI2
film into a solution of CH3NH3I or spin-coating the CH3NH3I
solution onto the PbI2 film to yield a perovskite crystal. This
method has been successfully used in PerSCs with porous nano-
structured scaffolds, while it is hard to apply in planar PerSCs
without the scaffold structure, since the CH3NH3I molecules
would be much harder to infiltrate into the dense PbI2 film,
which would lead to the incomplete reaction of PbI2.3,18 The
co-evaporation method involves the simultaneous sublimation of
CH3NH3I and PbI2, co-depositing organic and inorganic species
under high vacuum conditions. This approach can obtain dense
and pin-hole-free high quality perovskite films. However, this
high vacuum based technique is time and energy consuming.15
To solve this problem, a vapor assisted solution process was
proposed, where CH3NH3I vapor reacts with the pre-deposited
PbI2 film to form a perovskite active layer, and a 1 micrometer
large perovskite crystal can be achieved using this method.18
However, for this approach it is hard to accurately control the
amount of sublimated CH3NH3I vapor originating from the
CH3NH3I powder, leading to the non-uniform reaction of PbI2,
especially for large dimension devices. So the photoactive areas of
PerSCs reported in the literature using this method to deposit
perovskite are less than 30 mm2.
Herein, we firstly demonstrate a novel simplified close space
sublimation (CSS) deposition for growing a high quality CH3NH3PbI3
thin film in a low-vacuum and even non-vacuum oven. In this
method, a highly crystallized and uniform perovskite film can
be easily obtained by simply controlling the thickness of the
PbI2 and CH3NH3I films, the reaction time of CH3NH3I vapor
with the PbI2 film, and the distance of the sublimation space.
With this approach, a large area of 100 mm2 can be easily
obtained. The CSS technique is characterized by placing the source
very close to the substrate, which was first proposed by Nicoll in
1963 for the heteroepitaxial growth of GaAs on a Ge substrate.22
Fig. 1 The procedures of CH3NH3PbI3 perovskite thin film deposition using the simplified CSS method.
This journal is © The Royal Society of Chemistry 2016
View Article Online
Energy & Environmental Science
In 1999, F. Rábago-Bernal developed an isothermal CSS technique,
which greatly decreased the cost for the epitaxial growth of thin
films.23 And now the CSS technique is widely used in all kinds
of inorganic film deposition for optoelectronic applications.
For the CSS technique, several interrelated parameters including
the temperature of the source and substrate, and the gap between
the source and substrate, as well as the temperature, pressure and
composition of the vapors in the deposition chamber significantly
influence the film growth rate.24
In our case, for CH3NH3PbI3 deposition the procedures are
greatly simplified and the conditions are much milder. One
source of PbI2 is pre-spin-coated on a modified fluorine-doped
tin oxide (FTO) glass slide, and another source of CH3NH3I is
also pre-spin-coated on a cleaned glass slide. The two slides are
then put face to face with a thin hollow aluminum foil gasket to
form the close space in an isothermal vacuum or non-vacuum
oven, as shown in Fig. 1. With increasing the temperature to
150 1C, the CH3NH3I film will sublimate to generate uniform
CH3NH3I vapor, which will react with the pre-deposited PbI2
layer to yield CH3NH3PbI3 perovskite on the FTO side. Since
almost all of the sublimated CH3NH3I vapor could react with
PbI2, the reaction can be easily handled by controlling the
thickness of the CH3NH3I film and the reaction time, and these
advantages are good for large dimensional perovskite film
growth. Based on this simplified CSS method, the planar
heterojunction PerSCs with a structure of FTO/poly(3,4-ethylene-
dioxythiophene) doped poly(styrene sulfonate) (PEDOT:PSS)/
CH3NH3PbI3/[6,6]-phenyl-C61-butyric acid methyl ester (PCBM)/Al
were fabricated and power conversion efficiencies of 16.2%,
12.9% and 10.8% were achieved for the small device of 4 mm2,
middle device of 25 mm2, and large device of 100 mm2, respectively.
2. Results and discussion
In order to monitor the reaction process of PbI2 with CH3NH3I
in a simplified CSS vacuum oven, perovskite films with differ-
ent annealing times were fabricated. The UV-Vis spectra of the
CH3NH3PbI3 films and precursor PbI2 film deposited on FTO/
PEDOT:PSS substrates are shown in Fig. 2. For the PbI2 film,
the exciton band is located at the absorption edge of 499 nm,
Energy Environ. Sci., 2016, 9, 1486--1494 | 1487

Energy & Environmental Science
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
Fig. 2 The UV-Vis spectra of the PbI2 film and perovskite film deposited
on FTO/PEDOT:PSS substrates. Insets show the photos of the obtained
thin films.
which is consistent with previous reports.25 After reacting for
1 h, the characteristic CH3NH3PbI3 perovskite absorption shoulder
appears at around 780 nm,26 and the absorption intensity
increases sharply with increasing the annealing time from 1 h
to 2 h. When further increasing the annealing time to 3 h, there
is no obvious change in absorption shape and intensity,
indicating that the formation of CH3NH3PbI3 is almost com-
plete after annealing for over 2 h.
The crystallinity of the perovskite film is crucial for the perfor-
mance of PerSCs. To better understand the formation process and
the quality of the obtained perovskite film, the XRD characteriza-
tion was performed as shown in Fig. 3. Fig. 3(a) displays the XRD
pattern of the PEDOT:PSS modified FTO substrate. The peaks at
26.81 (110), 34.11 (101), 38.11 (200), 51.81 (211), and 54.81 (220)
are the characteristic diffractions of FTO.27 Fig. 3(b) shows the
diffraction pattern of the PbI2 film deposited on the PEDOT:PSS/
FTO substrate. Except for the substrate peaks, four diffraction
peaks at 12.81, 25.61, 38.71 and 52.41, corresponding to the c axis
orientation of the (001), (002), (003) and (004) lattice planes,
respectively, can be observed.11,28 Fig. 3(c)–(g) give the XRD
diffractions of the perovskite films grown on the PEDOT:PSS
modified FTO substrate with a 0.2 mm sublimation distance and
different annealing times. The strong diffraction peaks of 14.31,
28.61, 32.11 and 40.91, marked with an asterisk, attributed to the
(110), (220), (312) and (224) lattice planes, respectively, are the
characteristic peaks for an orthorhombic CH3NH3PbI3 perovskite
crystal.29,30 For comparison, the peaks of the substrate and PbI2
are marked with dashed lines. As shown in the XRD patterns, the
relative strength of the PbI2 peaks decline while the signals for
CH3NH3PbI3 increase with prolonging the annealing time. There
are obvious PbI2 peaks for the perovskite films annealed for less
than 2.5 h, which means that the reaction of PbI2 and CH3NH3I is
incomplete and unreacted PbI2 still remains in the perovskite
layer. This residual PbI2 in the perovskite film could deteriorate
the photovoltaic performance of the PerSCs, and even induce
serious hysteresis in the J–V curves.31,32 When prolonging the
annealing time to 2.5 h, the peaks of PbI2 disappear, leaving
1488 | Energy Environ. Sci., 2016, 9, 1486--1494
View Article Online
Paper
Fig. 3 X-ray diffraction (XRD) patterns of (a) FTO/PEDOT:PSS substrate,
(b) PbI2 and perovskite films deposited on the FTO/PEDOT:PSS substrate
reacted for (c) 1 h, (d) 1.5 h, (e) 2 h, (f) 2.5 h and (g) 3 h with a 0.2 mm
sublimation distance.
strong CH3NH3PbI3 characteristic peaks as shown in Fig. 3(f).
This infers that the sublimated CH3NH3I vapor can penetrate to
the bottom of the PbI2 film and react with the residual PbI2
when the annealing time is extended to 2.5 h. Upon further
increasing the annealing time to 3 h, there is no change in the
characteristic perovskite peaks (Fig. 3(g)). This means that all of
the PbI2 reacted with CH3NH3I to yield CH3NH3PbI3 crystals
when the film was annealed for 2.5 h, and the formation of the
CH3NH3PbI3 crystal is complete.
Besides the composition and the crystallinity of the perovskite
film, the surface topography of the film has a great influence on
the performance of PerSCs. The surface topography information
of the PbI2 film and perovskite films annealed for different
lengths of time was investigated using scanning electron micro-
scopy (SEM) and atom force microscopy (AFM), as shown in
Fig. 4. The surface topography changes with increasing the
reaction time. The root mean square (RMS) roughness of the
PbI2 film is 9.9 nm (Fig. 4(a)), while the RMS roughness
increased to 16.6 nm (Fig. 4(b)), 17.7 nm (Fig. 4(c)), 23.7 nm
(Fig. 4(d)), 22.4 nm (Fig. 4(i)) and 17.9 nm (Fig. 4(j)) when the
PbI2 film reacted with CH3NH3I vapor for 1 h, 1.5 h, 2 h, 2.5 h
and 3 h, respectively. The morphology change is ascribed to the
structural transformation from an edge sharing octahedral
structure to a corner-sharing octahedral structure when PbI2
grows into a CH3NH3PbI3 crystal.5,33 This structural transfor-
mation induces the increase in the surface roughness and film
thickness. Typically, a 160 nm PbI2 film could turn into a
290 nm perovskite film with this transformation. From the
SEM images as shown in Fig. 4(e), many pores are observed in
the plate-like PbI2 film. These pores allow CH3NH3I vapor to
penetrate into the PbI2 film and completely react with PbI2.
Fig. 4(f) shows the SEM images of a 1 h annealed perovskite film.
This journal is © The Royal Society of Chemistry 2016

Paper
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
Fig. 4 AFM (5 mm  5 mm) and top-view SEM (inset image with a scale bar of 1 mm) images of the PbI2 film (a and e), and perovskite films annealed for 1 h
(b and f), 1.5 h (c and g), 2 h (d and h), 2.5 h (i and k) and 3 h (j and l); (m) cross-section SEM image of the CH3NH3PbI3 film grown on the FTO/PEDOT:PSS
substrate annealed for 2.5 h.
In contrast to the PbI2 film as shown in Fig. 4(e), the pore size
and density are greatly decreased and rod-like large crystals
appear although there are many small particles of unreacted
PbI2. When extending the annealing time to 1.5 h (Fig. 4(g)), the
amount of small particles decreases, while the size of the crystal
is not obviously expanded. These observations suggest that at
the early stage of perovskite formation with annealing for 1 h to
1.5 h, CH3NH3I majorly reacts with the surface PbI2 film. When
the annealing time is increased to 2 h or longer, the grain size
of the CH3NH3PbI3 crystal is clearly enlarged, as shown in
Fig. 4(h), (k) and (l). In addition, there are many pin-holes
in the perovskite films annealed for 1 h and 1.5 h, while the pin-
holes and grain boundaries sharply decrease when the annealing
time is increased to 2 h or a much longer time. This means that
when the film is annealed for 2 h, the growth of CH3NH3PbI3
mainly arises from CH3NH3I molecules penetrating into the
PbI2 film and reacting with the residual PbI2. Fig. 4(m) shows
the cross-section SEM image of the perovskite film annealed for
2.5 h on a PEDOT:PSS/FTO substrate, and a dense perovskite
layer with uniform thickness can be easily identified. The
observed high quality of the perovskite film could contribute
to enhancing photovoltaic performance and enlarging the device
dimensions in PerSCs.
In simplified close space sublimation, the distance of sub-
limation space would also play an important role in the growth
of the CH3NH3PbI3 film. The crystal quality and morphology of
the CH3NH3PbI3 films fabricated under a different distance
(0.2, 0.4, and 0.6 mm) of sublimation space were also investigated
This journal is © The Royal Society of Chemistry 2016
View Article Online
Energy & Environmental Science
using XRD, and AFM and SEM, respectively. As shown in Fig. S1
(ESI†), with the annealing time set as 2.5 h and the distance
increasing to 0.4 mm or 0.6 mm, the peaks of PbI2 are observed in
the XRD pattern and the intensity of the PbI2 peaks is boosted as
the distance increases. There are no obvious PbI2 signals for the
perovskite film sublimated at a distance of 0.2 mm (Fig. 3(f)).
These observations indicate that increasing the sublimation dis-
tance results in more residual unreacted PbI2 and this is ascribed
to the limited mean free path of CH3NH3I compared to the space
between the PbI2 and CH3NH3I substrates. Fig. S2 (ESI†) demon-
strates the surface morphology of the obtained perovskite films.
From the AFM images, the RMS of the perovskite film with a 0.4
and 0.6 mm sublimation distance is 16 and 14 nm, respectively.
From the SEM images, the grain size decreases and the proportion
of smaller crystals increases with increasing the sublimation
distance. The decreased surface roughness and crystal size
are attributed to the fact that not enough CH3NH3I vapor can
reach the PbI2 substrate for crystal expansion at an increased
sublimation distance.
Based on the simplified-CSS grown perovskite films, typical
planar heterojunction PerSCs with the structure of FTO/
PEDOT:PSS/CH3NH3PbI3/PCBM/Al were fabricated, as shown
in Fig. 1. Two heterojunctions of CH3NH3PbI3/PCBM and
CH3NH3PbI3/PEDOT:PSS exist in this structure, and photo-
generated excitons with a low binding energy could be dissociated
into free holes and electrons at both heterojunction interfaces
when shining light from the FTO side.19,34 The function of the
PCBM and PEDOT:PSS is the collection and transportation of
Energy Environ. Sci., 2016, 9, 1486--1494 | 1489
 View Article Online

Energy & Environmental Science Paper

Table 1 The photovoltaic parameters of PerSCs averaged from 20

individual devices

Jsc Voc FF PCE

Variables Scan (mA cm 2) (V) (%) (%)
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.

Annealing time 1h Forward 7.5 0.62 50 2.3

Reverse 7.5 0.64 53.3 2.6
1.5 h Forward 12.8 0.81 53.5 5.6
Reverse 13.5 0.79 54.7 5.9
2h Forward 16 0.88 66.5 9.3
Reverse 15.5 0.87 67.6 9.1
2.5 h Forward 19.6 1.03 77.3 15.6
Reverse 19.6 1.03 80.2 16.2
3h Forward 17.1 0.95 69.3 11.3
Reverse 16.7 0.95 69.3 11

Perovskite 320 nm Forward 19.3 0.90 67.4 11.7

thickness Reverse 18.5 0.91 71.1 12
290 nm Forward 19.6 1.03 77.3 15.6
Reverse 19.6 1.03 80.2 16.2
270 nm Forward 18.3 0.97 68.3 12.1
Reverse 18 0.98 74.2 13.1
240 nm Forward 17.6 0.93 65.9 10.8
Reverse 17.3 0.93 70.3 11.3

Photoactive 4 mm2 Forward 19.6 1.03 77.3 15.6

area Reverse 19.6 1.03 80.2 16.2
25 mm2 Forward 19.2 0.92 69.5 12.2
Reverse 19.4 0.92 72.3 12.9
100 mm2 Forward 18.5 0.93 58.8 10.1
Reverse 18.9 0.93 61.7 10.8

15.3 mA cm 2, 19.1 mA cm 2 and 17.1 mA cm 2, respectively,

Fig. 5 (a) Current–voltage (J–V) characteristic curves of perovskite devices
annealed for different lengths of time under AM 1.5G, 100 mW cm 2
illumination with forward and reverse bias. (b) External quantum efficiency
(EQE) curves of the perovskite devices annealed for different lengths of time.
the electrons and holes to the Al and FTO electrode, respec-
tively. Fig. 5(a) shows the J–V curves of the PerSCs based on the
different annealing times with a 0.2 mm sublimation distance.
At the initial stage (1–2 h), the PCE of the perovskite device is
lower than 10%. Contrastingly, when the annealing time is
increased to 2.5 h, the PCE of the devices is greatly enhanced
from 2.6% (1 h) to 16.2% (2.5 h), with simultaneous improve-
ment of the short-circuit current density (Jsc), open-circuit
voltage (Voc) and filling factor (FF). The Jsc, Voc, and FF increase
from 7.5 mA cm 2, 0.62 V, and 50.0% to 19.6 mA cm 2, 1.03 V,
and 80.2%, respectively, for devices annealed for 1 h and 2.5 h.
The photovoltaic parameters (averaged from 20 individual
devices) of Jsc, Voc, FF, and PCE are listed in Table 1, and the
champion cell performance is shown in Fig. S3 (ESI†), demon-
strating a high PCE of 17.1% (Jsc = 23.5 mA cm 2, Voc = 0.96 V
and FF = 75.8%). The light response property of the devices was
further investigated using external quantum efficiency (EQE)
measurements, and Fig. 5(b) displays the EQE curves of the
perovskite devices annealed for different lengths of time.
The integrated current of the perovskite devices annealed for
1 h, 1.5 h, 2 h, 2.5 h, and 3 h is 7.3 mA cm 2, 12.8 mA cm 2,
which is in good agreement with the Jsc derived from the J–V
measurement. It can be concluded that the annealing time
greatly influences the formation and topography of the perovs-
kite film. If the annealing time is too short, there are many PbI2
impurities remaining in the perovskite film. The PbI2 impurities
in the device would deteriorate the performance of the PerSCs
and aggravate the J–V hysteresis.31,32 However, excessive annealing
would make unreacted CH3NH3I deposit on the surface of per-
ovskite, which could hinder the excitons dissociation and charge
transportation at the interface of CH3NH3PbI3 and PCBM. So an
appropriate annealing time can greatly improve the quality of the
perovskite film with higher coverage, larger grain size, less grain
boundaries and fewer crystal defects, which are indispensable for
high performance PerSCs.
We also studied the effect of the distance of sublimation
space on the device performance. As shown in Fig. S4 (ESI†),
the device with a 0.4 mm sublimation distance shows a PCE of
13.1%, with a Jsc of 17.5 mA cm 2, a Voc of 0.96 V, and an FF of
78.3%. When increasing the sublimation distance to 0.6 mm, the
performance greatly declines, especially the Jsc (14.1 mA cm 2)
and Voc (0.92 V), and the overall PCE decreases to 10.1%. The
deterioration of the Jsc and Voc is ascribed to the crystallinity and
morphology change (more residual PbI2 and a smaller crystal
size) of the perovskite film with an increased sublimation
distance as show in Fig. S1 and S2 (ESI†).
Besides the annealing time and the sublimation distance,
the thickness of the perovskite layer is another important factor
in improving the performance of PerSCs.35 To investigate the
influence of the perovskite layer thickness on the photovoltaic

1490 | Energy Environ. Sci., 2016, 9, 1486--1494 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Paper Energy & Environmental Science

area devices with good performance. In order to examine the

feasibility of fabricating large area PerSCs with a simplified CSS
deposited perovskite film, devices with photoactive areas of
25 mm2 and 100 mm2 were fabricated. Fortunately, the PCE of
the devices with the middle area (25 mm2) and large area
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.

(100 mm2) does not sharply decrease compared with the small
area (4 mm2) device. As listed in Table 1, the device with a
photoactive area of 25 mm2 gives a PCE of 12.9% (Voc = 0.92 V,
Jsc = 19.4 mA cm 2 and FF = 72.3%) and the PCE for the device
with a photoactive area of 100 mm2 is 10.8% (Voc = 0.93 V, Jsc =
18.9 mA cm 2 and FF = 61.7%), indicating that the quality of the
perovskite film does not decline with the increase in the photo-
active area. Fig. 7(a) shows the J–V curves of the devices with
different photoactive areas under AM 1.5G solar irradiation
(100 mW cm 2) in forward and reverse bias. The decrease of
the PCE for large area devices is mainly ascribed to the decline
in the FF, since the Jsc and Voc just slightly decrease when
increasing the photoactive area. The corresponding EQE curves
for the devices (Fig. 7(b)) show a broad plateau over the whole of
the visible range. Compared to the small size device (4 mm2,
19.1 mA cm 2), there is less than a 5% decrease in the integrated
current for the large size device (100 mm2, 18.4 mA cm 2).

Fig. 6 (a) Current–voltage (J–V) characteristic curves of perovskite devices

with different perovskite thickness under AM 1.5G, 100 mW cm 2 illumina-
tion with forward and reverse bias. (b) External quantum efficiency (EQE)
curves of the PerSCs with different perovskite thicknesses.

performance of PerSCs, perovskite devices with different per-

ovskite thicknesses annealed for 2.5 h were fabricated. Fig. 6(a)
shows the J–V curves of the devices with perovskite films of 240,
270, 290, and 320 nm. The thickness of the PEDOT:PSS layer,
PbI2 and the perovskite film on the PEDOT:PSS modified FTO
substrate were measured using a Dektak XT profilometer as
shown in Fig. S5 (ESI†). In the range of 240–290 nm, as the
thickness increases, the Jsc of the PerSCs increases from 17.3 to
19.6 mA cm 2 and the PCE increases from 11.3% to 16.2%.
However, the PCE begins to decrease when the thickness of the
perovskite film is increased from 290 to 320 nm. Generally speaking,
with more light-harvesting, more excitons can be generated, indu-
cing more free charge carriers, with increasing perovskite thickness.
Nevertheless, with an excessively thick perovskite film, charge
recombination will dominate during transportation and this is
evidenced by the decrease in the FF and Voc.15,36,37 As observed in
the external quantum efficiency (EQE) test (Fig. 6(b)), the integrated
current dropped from 19.1 to 18.3 mA cm 2. The detailed photo-
voltaic parameters of these PerSCs are also listed in Table 1. Fig. 7 (a) Current–voltage (J–V) characteristic curves under AM 1.5G one
Nowadays, many approaches are successful for fabricating sun (100 mW cm 2) illumination and (b) external quantum efficiency (EQE)
small area PerSCs, but it is still challenging to fabricate large curves of the perovskite devices with different photoactive areas.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1486--1494 | 1491

Energy & Environmental Science
In our experiment the resistance of FTO is 15 O sq 1, and
the series resistance of the device should sharply increase with
the enlargement of the photoactive area, which will inevitability
decrease the conductivity of the substrate and induce a
decrease in the FF of the PerSCs.38 In order to improve the
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
charge transport and study the effect of FTO resistance on the
performance of large dimensional devices, we fabricated
devices (100 mm2) on the FTO with a resistance of 7 O sq 1.
As shown in Fig. S6 (ESI†), the devices exhibit a PCE of 10.9%
for the forward scan and 11.4% for the reverse scan with a Voc
around 0.95 V, a Jsc around 18 mA cm 2, and an FF around
65%. Compared to the devices based on 15 O sq 1 FTO, the FF
of the devices really increases when using a lower resistance
FTO of 7 O sq 1, while there is a slight drop in the Jsc as the
higher conductivity FTO bares poor light transmission.
To further simplify the processing procedures, we even tried to
grow the perovskite film in a non-vacuum (ambient air filled) oven.
Fig. S7 (ESI†) displays the J–V curves of the devices (4 mm2) with the
perovskite film grown in a non-vacuum oven. The device still shows
a PCE of 12.7% with a Jsc of 18.4 mA cm 2, a Voc of 0.96 V, and an
FF of 72%, although all the parameters (Voc, Jsc, FF, and PCE) are
decreased in comparison with devices based on a low vacuum
( 0.1 MPa) grown perovskite film. The deteriorated performance is
ascribed to the diffusion-limited CH3NH3I vapor transport in non-
vacuum conditions,24 where the mean free path of CH3NH3I vapor
is short compared to the space between the PbI2 and CH3NH3I
substrates, and the CH3NH3I vapor molecules will collide several
times with air before they condense on and then react with the PbI2
substrate, thus decreasing the kinetic energy of CH3NH3I vapor and
inducing an incomplete reaction with PbI2.
3. Conclusion
In conclusion, we demonstrated a novel approach for the growth of
a CH3NH3PbI3 thin film using simplified close space sublimation
for efficient and large dimension perovskite solar cells. By simply
controlling the annealing time and the thickness of the PbI2 and
CH3NH3I layer, a pin-hole free, uniform and highly crystallized
perovskite thin film can be achieved. The planar heterojunction
perovskite solar cells based on this approach were fabricated and
the dependence of the photovoltaic performance on the annealing
time, perovskite thickness and photoactive area were investigated.
Under optimized conditions, a PCE of 16.2% can be achieved for
devices with a photoactive area of 4 mm2, by thermally annealing at
150 1C for 2.5 h to yield a 290 nm perovskite film. A promising PCE
of 10.8% was also obtained for devices with a photoactive area of
100 mm2, and there is no obvious J–V hysteresis in the forward and
reverse scan. Our findings supply a new option for the fabrication of
perovskite solar cells with a high performance and large dimension.
4. Experimental
Materials
1
Patterned FTO glass with a sheet resistance of 15 O sq was
purchased from Wuhan Geao (China). PEDOT:PSS with brand
1492 | Energy Environ. Sci., 2016, 9, 1486--1494
View Article Online
Paper
Clevious P VP AI 4083 was purchased from H. C. Stark.
Methylamine solution (33 wt% in absolute ethanol), hydriodic
acid (57 wt% in water) and isopropanol (99.8%, extra dry) were
purchased from Acros Organics. N,N-Dimethylformamide
(DMF) and lead(II) iodide (99.999%) were purchased from Alfa
Aesar. PCBM was purchased from Solarmer Energy, Inc. All
these commercially available materials were used as received
without further purification.
CH3NH3I synthesis
CH3NH3I was synthesized according to the previous literature
with some modification.12 Typically, 34 mL of methylamine
(33 wt% in absolute ethanol) and 38 mL of hydroiodic acid
(57 wt% in water) were mixed in a 150 mL three-necked flask,
and then stirred at 0 1C for 2 h in an ice-water bath. After that,
the mixtures were rotary evaporated at 50 1C for 1 h to remove
the solvent and white CH3NH3I precipitates were recovered.
Finally, the product was washed with diethyl ether three times
and dried at 60 1C overnight in a vacuum oven.
Device fabrication
The patterned FTO was sequentially ultrasonic cleaned twice with
detergent, pure water, deionized water, acetone and isopropyl
alcohol. The pre-cleaned substrate was ultraviolet ozone treated for
10 min. After that, the filtered PEDOT:PSS aqueous solution was
spin-coated onto the pre-treated FTO-glass substrates at 3000 rpm
for 35 s and then dried at 150 1C for 15 min. A 350 mg mL 1 PbI2
DMF solution was spin-coated on the FTO/PEDOT:PSS substrate and
then annealed at 100 1C for 10 min. The CH3NH3I thin layer was
deposited by spin-coating 40 mg mL 1 CH3NH3I isopropyl solution
at 1000–3000 rpm for 20 s and then annealed at 80 1C for 5 min. The
PbI2 film and CH3NH3I film were pressed together face to face with a
0.2, 0.4, and 0.6 mm hollow aluminum foil gasket between them,
and the space is sufficient for the thickness increase of the PbI2 film
when it grows into a CH3NH3PbI3 crystal. The whole sets were then
annealed at 150 1C in a vacuum oven under a pressure of 0.1 MPa
or non-vacuum oven, and the perovskite film was obtained on the
PbI2 side. The process of simplified CSS to fabricate perovskite
is shown in Fig. 1. After that, PCBM solution (30 mg mL 1 in
chlorobenzene) was spin-coated on the perovskite layer at 3000 rpm
for 30 s. Finally, a 100 nm Al cathode was deposited through thermal
evaporation under a pressure of 5  10 5 Pa.
Device characterization
The current density–voltage curves were measured under
AM 1.5G illumination (100 mW cm 2) using a solar simulator
(SAN-EI, AAA grade) with a Keithley 2400 Source Meter under an
N2 atmosphere. The light intensity was calibrated with a Si solar
cell for 1 sun. The external quantum efficiency (EQE) was
measured using QE-R systems (Enli Tech.). The light intensity
at each wavelength was adjusted with a standard single-crystal
Si photovoltaic cell.
Instrumentation
The thickness of the films was recorded using a DektakXT
profilometer (Bruker Nano, Inc.). The surface morphologies of
This journal is © The Royal Society of Chemistry 2016

Paper
the thin films were examined using an AC Mode III (Agilent5500)
atomic force microscope (AFM) operated in the tapping mode
under ambient atmosphere. The surface and cross-section
morphologies of the thin films were further characterized using
a Scanning Electron Microscope (SEM; HITACHI SU8010, Japan)
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
operated at an accelerating voltage of 5.0 kV and 30 kV, respec-
tively. The crystallinity of the perovskite films were measured
using an X-ray diffractometer (Rigaku miniflex 600) in the
2y range of 5–901 at a scanning rate of 51 min 1. The optical
absorption of the films was measured using a Shimadzu
UV-2450 UV-visible spectrophotometer.
Acknowledgements
This work was supported by the National Key Basic Research
Program of China (973 project, 2015CB932201), Program for
New Century Excellent Talents in University (NCET-12-0848),
The Science and Technology Commission of Beijing Municipality,
China (Z141100003314003), State Key Laboratory of Physical
Chemistry of Solid Surfaces (Xiamen University, 201404), Jiangsu
Province Science and Technology Support Program, China
(BE2014147-4) and Fundamental Research Funds for the Central
Universities, China (JB2015RCJ02, 2015ZZD06).
Notes and reference
1 G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam,
M. Grätzel, S. Mhaisalkar and T. C. Sum, Science, 2013,
342, 344–347.
2 H. J. Snaith, J. Phys. Chem. Lett., 2013, 4, 3623–3630.
3 F. Z. Wang, Z. A. Tan, S. Y. Dai and Y. F. Li, Acta Phys. Sin.,
2015, 64, 038401.
4 A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am.
Chem. Soc., 2009, 131, 6050–6051.
5 W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo and
S. I. Seok, Science, 2015, 348, 1234–1237.
6 S. Y. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd,
G. C. Xing, T. C. Sum and Y. M. Lam, Energy Environ. Sci.,
2014, 7, 399–407.
7 M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and
H. J. Snaith, Science, 2012, 338, 643–647.
8 H. Zhou, Q. Chen, G. Li, S. Luo, T. B. Song, H. S. Duan,
Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345,
542–546.
9 W. J. Ke, G. J. Fang, J. W. Wan, H. Tao, Q. Liu, L. B. Xiong,
P. L. Qin, J. Wang, H. W. Lei, G. Yang, M. C. Qin, X. Z. Zhao
and Y. F. Yan, Nat. Commun., 2015, 6, 6700.
10 W. Zhang, M. Saliba, D. T. Moore, S. K. Pathak, M. T.
Horantner, T. Stergiopoulos, S. D. Stranks, G. E. Eperon,
J. A. Alexander-Webber, A. Abate, A. Sadhanala, S. Yao,
Y. Chen, R. H. Friend, L. A. Estroff, U. Wiesner and
H. J. Snaith, Nat. Commun., 2015, 6, 6142.
11 J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao,
M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499,
316–319.
This journal is © The Royal Society of Chemistry 2016
View Article Online
Energy & Environmental Science
12 J.-H. Im, I.-H. Jang, N. Pellet, M. Grätzel and N.-G. Park, Nat.
Nanotechnol., 2014, 9, 927–932.
13 C. H. Chiang, Z. L. Tseng and C. G. Wu, J. Mater. Chem. A,
2014, 2, 15897–15903.
14 J. Xi, Z. X. Wu, H. Dong, B. Xia, F. Yuan, B. Jiao,
L. X. Xiao, Q. H. Gong and X. Hou, Nanoscale, 2015, 7,
10699–10707.
15 M. Z. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501,
395–398.
16 O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas,
M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Nat.
Photonics, 2014, 8, 128–132.
17 H. Hu, D. Wang, Y. Zhou, J. Zhang, S. Lv, S. Pang, X. Chen,
Z. Liu, N. P. Padture and G. Cui, RSC Adv., 2014, 4,
28964–28967.
18 Q. Chen, H. P. Zhou, Z. R. Hong, S. Luo, H. S. Duan,
H. H. Wang, Y. S. Liu, G. Li and Y. Yang, J. Am. Chem.
Soc., 2014, 136, 622–625.
19 C. Li, F. Z. Wang, J. Xu, J. X. Yao, B. Zhang, C. F. Zhang,
M. Xiao, S. Y. Dai, Y. F. Li and Z. A. Tan, Nanoscale, 2015, 7,
9771–9778.
20 W. Zhang, M. Saliba, D. T. Moore, S. K. Pathak, M. T.
Horantner, T. Stergiopoulos, S. D. Stranks, G. E. Eperon,
J. A. Alexander-Webber, A. Abate, A. Sadhanala, S. Yao,
Y. Chen, R. H. Friend, L. A. Estroff, U. Wiesner and
H. J. Snaith, Nat. Commun., 2015, 6, 6142.
21 Y. Z. Xu, L. F. Zhu, J. J. Shi, S. T. Lv, X. Xu, J. Y. Xiao, J. Dong,
H. J. Wu, Y. H. Luo, D. M. Li and Q. B. Meng, ACS Appl.
Mater. Interfaces, 2015, 7, 2242–2248.
22 F. N. Nicoll, J. Electrochem. Soc., 1963, 11, 1165.
23 O. de Melo, E. Sánchez, H. Rodrı́guez, S. De Roux,
F. Rábago-Bernal and J. Ruiz-Garcı́a, Mater. Chem. Phys.,
1999, 59, 120–124.
24 S. N. Alamri, Phys. Status Solidi A, 2003, 200, 352–360.
25 E. M. S. dos Santos, L. S. Pereira and G. J. F. Demets, J. Braz.
Chem. Soc., 2011, 22, 1595–1600.
26 C. Roldan-Carmona, P. Gratia, I. Zimmermann, G. Grancini,
P. Gao, M. Graetzel and M. K. Nazeeruddin, Energy Environ.
Sci., 2015, 8, 3550–3556.
27 K. T. Lee and S. Y. Lu, J. Mater. Chem., 2012, 22,
16259–16268.
28 S. Y. Ye, W. H. Sun, Y. L. Li, W. B. Yan, H. T. Peng,
Z. Q. Bian, Z. W. Liu and C. H. Huang, Nano Lett., 2015, 15,
3723–3728.
29 H. Choi, C. K. Mai, H. B. Kim, J. Jeong, S. Song, G. C. Bazan,
J. Y. Kim and A. J. Heeger, Nat. Commun., 2015, 6, 7438.
30 S. R. Raga, M. C. Jung, M. V. Lee, M. R. Leyden, Y. Kato and
Y. B. Qi, Chem. Mater., 2015, 27, 1597–1603.
31 J. H. Heo, H. J. Han, D. Kim, T. K. Ahn and S. H. Im, Energy
Environ. Sci., 2015, 8, 1602–1608.
32 J. H. Heo, D. H. Song, H. J. Han, S. Y. Kim, J. H. Kim, D. Kim,
H. W. Shin, T. K. Ahn, C. Wolf, T. W. Lee and S. H. Im, Adv.
Mater., 2015, 27, 3424–3430.
33 L. Hu, J. Peng, W. W. Wang, Z. Xia, J. Y. Yuan, J. L. Lu,
X. D. Huang, W. L. Ma, H. B. Song, W. Chen, Y. B. Cheng
and J. Tang, ACS Photonics, 2014, 1, 547–553.
Energy Environ. Sci., 2016, 9, 1486--1494 | 1493

Energy & Environmental Science
34 J. Y. Jeng, Y. F. Chiang, M. H. Lee, S. R. Peng, T. F. Guo,
P. Chen and T. C. Wen, Adv. Mater., 2013, 25, 3727–3732.
35 K. Wang, C. Liu, P. C. Du, L. Chen, J. H. Zhu, A. Karim and
X. Gong, Org. Electron., 2015, 21, 19–26.
36 O. Malinkiewicz, C. Roldan-Carmona, A. Soriano, E. Bandiello,
Published on 15 February 2016. Downloaded by Benemerita Universidad Autonoma de Puebla on 06/06/2016 17:38:14.
L. Camacho, M. K. Nazeeruddin and H. J. Bolink, Adv. Energy
Mater., 2014, 4, 1400345.
1494 | Energy Environ. Sci., 2016, 9, 1486--1494
View Article Online
Paper
37 Z. G. Xiao, C. Bi, Y. C. Shao, Q. F. Dong, Q. Wang, Y. B. Yuan,
C. G. Wang, Y. L. Gao and J. S. Huang, Energy Environ. Sci.,
2014, 7, 2619–2623.
38 J. You, L. Meng, T. B. Song, T. F. Guo, Y. M. Yang,
W. H. Chang, Z. Hong, H. Chen, H. Zhou, Q. Chen, Y. Liu,
N. De Marco and Y. Yang, Nat. Nanotechnol., 2016, 11,
75–81.
This journal is © The Royal Society of Chemistry 2016
Electronic Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © The Royal Society of Chemistry 2016

The growth of CH3NH3PbI3 thin film by simplified close space

sublimation for efficient and large dimensional perovskite solar cells

Qiang Guo1, Cong Li1, Wenyuan Qiao1, Shuang Ma1, Fuzhi Wang1, Bing Zhang1,

Linhua Hu2, Songyuan Dai1,*, Zhan’ao Tan1,*

1 State Key Laboratory of Alternate Electrical Power System with Renewable Energy

Sources, Beijing Key Laboratory of Novel Thin Film Solar Cells, North China

Electric Power University, Beijing 102206, China.

2 Institute of Applied Technology, Hefei Institutes of Physical Science, Chinese

Academy of Sciences, Hefei 230088, China

Corresponding Authors

*E-mail: tanzhanao@ncepu.edu.cn (Z. A. Tan); sydai@ncepu.edu.cn (S. Y. Dai)

Figure S1. X-ray diffraction (XRD) pattern of perovskite films deposited on the

FTO/PEDOT:PSS substrate (annealed for 2.5 h) with the sublimation distance of 0.4

and 0.6 mm, respectively.

Figure S2. AFM (5µm×5µm) and top-view SEM (inset image with scale bar of 1µm)

images of perovskite films annealed for 2.5 h with the sublimation distance of 0.4 (a,

c) and 0.6 mm (b, d), respectively.

Figure S3. Current-voltage (J-V) characteristic curve of the champion cell (4 mm2).

Figure S4. Current-voltage (J-V) characteristic curves of perovskite devices with the

sublimation space of 0.4 and 0.6 mm.

Figure S5. Thickness of PEDOT:PSS layer, PbI2 film and CH3NH3PbI3 film

deposited on PEDOT:PSS substrate measured by Dektak XT profilometer.

Figure S6. Current-voltage (J-V) characteristic curves of perovskite device (100

mm2) fabricated with FTO glass with sheet resistance of 7 Ω/sq.

Figure S7. (a) Current-voltage (J-V) characteristic curves and (b) External quantum

efficiency (EQE) curve of perovskite devices with the CH3NH3PbI3 film grown in

standard atmosphere.