Preparation of metal halide perovskite solar cells through a liquid droplet assisted method

Article abstract of DOI:10.1039/c4ta06392a

Solution-processed organometal trihalide-based perovskites have attracted attention in the field of solar energy due to their high absorption and low temperature fabrication. We demonstrated the droplet-assisted two-step process of spin coating lead iodide (PbI₂) followed by spray coating methylammonium iodide (CH₃NH₃I) to prepare a continuous lead iodide perovskite film. A simple airbrush gun was used to control the volume of CH₃NH₃I, in order to attain a uniform, stoichiometric and continuous perovskite film. An insufficient or excess volume of CH₃NH₃I gives poor crystallinity and morphology, which gradually reduces the device performance. A power conversion efficiency (PCE) of 11.66% was achieved for 100 nm of PbI₂ followed by 300 μl of CH₃NH₃I and annealing at 100 °C for 120 min. To address the reproducibility of the device performance, 50 devices were fabricated for statistical analysis and 80% of the devices showed the average PCE of 10-11% with reproducible J_SC, V_OC and FF.

Full text of DOI:10.1039/c4ta06392a

View Article Online
View Journal
Journal of
Materials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. M. BOOPATHI,
M. Ramesh, P. Perumal, Y. Huang, C. Tsao, Y. Chen, C. Lee and C. W. Chu, J. Mater. Chem. A, 2015, DOI:
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/materialsA

Page 1 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Preparation of Metal Halide Perovskite Solar cells through
Liquid Droplet Assisted Method
Karunakara Moorthy Boopathia,b,c, Mohan Rameshc, Packiyaraj Perumald, YuꢀChing Huange,
ChengꢀSi Tsaoe, YangꢀFang Chend, ChihꢀHao Leea, ChihꢀWei Chuc,*
aDepartment of Engineering and Systems Science, National Tsing Hua University, Hsinchu 30013,
Taiwan
bNano Science and Technology Program, Taiwan International Graduate Program, Academia
Sinica and National Tsing Hua University
cResearch Center for Applied Science, Academia Sinica, Taipei 115, Taiwan
d Department of Physics, National Taiwan University, Taipei 10617, Taiwan.
e Institute of Nuclear Energy Research, Longtan, Taoyuan 32546, Taiwan
*eꢀmail: gchu@gate.sinica.edu.tw
http://www.rcas.sinica.edu.tw/faculty/gchu.html
Abstract
Solutionꢀprocessed organometal trihalideꢀbased perovskites have attractive attention in the
field of solar energy due to high absorption and low temperature fabrication. We
demonstrated the dropletꢀassisted twoꢀstep process of spin coating the lead iodide (PbI2)
followed by spray coating the methylammonium iodide (CH3NH3I) to prepare continuous
lead iodide perovskite film. A simple airbrush gun was used to control the volume of
CH3NH3I, in order to attain a uniform, stoichiometric and continuous perovskite film.
Insufficient or excess volume of CH3NH3I gives poor crystallinity and morphology, which
gradually reduces the device performance. Power conversion efficiency (PCE) of 11.66%
was achieved at 100 nm of PbI2 followed by 300 ꢁl of CH3NH3I and annealed at 100°C for
120 min. To address the reproducibility of device performance, 50 devices were fabricated
for statistical analysis and 80% of the device showed the average PCE of 10–11% with
reproducible Jsc, Voc and FF.
Keywords: Perovskite, spray coating, planar heterojunction solar cells, hybrid, twoꢀstep
process

Journal of Materials Chemistry A
Page 2 of 22
View Article Online
DOI: 10.1039/C4TA06392A
1. Introduction:
Natural sources like sunlight, wind, tides and geothermal heat have abundant renewable
energy resources for safe environment. Photovoltaic (PV) energy conversion is one of
important techniques to accomplish our today’s energy demands. PV is the best way to
producing electrical energy directly from sunlight, without emitting the toxic substances and
greenhouse gas with necessitating less maintenance. Organic–inorganic hybrid solar cells
have unique properties with better efficiency than the only organic and inorganic solar cells.
Perovskiteꢀbased hybrid solar cells have been the most substantial development in recent
years with high efficiencies by simple solution process technique. Organic–inorganic
+
perovskite family were taken in the form of ABX3 (A= CH3NH3 , B= Pb2+ and X= I−, Cl− and
Br−) structure consists of organic components in cuboctahedral A site and inorganic
components in octahedral B site.1 Electronꢀhole diffusion length of methylammonium lead
triiodide perovskites (CH3NH3PbI3) is about 100 nm and also the mixedꢀhalide perovskite
CH3NH3PbI3−xClx is about 1 ꢁm.2, 3 These high values strengthen the hope for future hybrid
solar cells. In recent years, CH3NH3PbI3ꢀbased perovskites have achieved 15% efficiency in
both mesoporous4ꢀ10 and planar structures.11–16 By using mixed halide (CH3NH3PbI3ꢀxClx),
Yang Yang group reached highest efficiency of 19.3%.17 Several techniques were used to
form uniform perovskite film, including spin coating,18,19 thermal evaporation,15,20,21 dip
coating,14 vapourꢀassisted process,13 blade coating22 and spray coating.12,23
Oneꢀstep spin coating process mixed perovskite solution (CH3NH3I and PbI2) gives poor film
coverage due to quick reaction between PbI2 and CH3NH3I.24 In order to overcome pin hole
perovskite film formation, twoꢀstep process was identified and widely used, such as
sequential deposition of both PbI2 and CH3NH3I by thermal evaporation,16, 21 twoꢀstep spin

Page 3 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
coating of PbI2 and CH3NH3I,19 spin coating of PbI2 followed by dip coating of CH3NH3I14
and spin coating of PbI2 and vapour deposition of CH3NH3I.13 In the above technique, there
are few defects that appeared while using twoꢀstep processes such as PbI2 film peeling and
dissolving while dipping in CH3NH3I solution, dissolving the PbI2 layer while spinning
CH3NH3I on top layer, additionally need high vacuum and expensive technique for thermal
deposition. Some common disadvantages arise while using the abovementioned twoꢀstep
processes such as precursor wastage, difficult to control PbI2 and CH3NH3I stoichiometric
ratio. To overcome these problems, we introduce new twoꢀstep process combined with spin
and spray coating technique for pin hole free perovskite film formation. Spray coating the
CH3NH3I creates the path to fabricate large area rollꢀtoꢀroll process without wasting the
precursor which reduces the fabrication cost.
In this manuscript, we report a twoꢀstep process of spin coating PbI2 and spray coating of
CH3NH3I for continuous pin hole free perovskite film formation. The diffused CH3NH3I
droplets towards PbI2 provide wellꢀdefined grain size with full coverage. The stoichiometry
perovskite film could lead to a better crystallinity and improved device efficiency of 11.6%.
Controlling the thickness of PbI2 and volume of CH3NH3I precursor solution guides an
optimum perovskite thickness which yields better absorption and efficient charge
transformation.
2. Experimental
Preparation of methylammonium iodide
Aqueous solution of hydroiodic acid (HI) (57 wt.% in water), methylamine (CH3NH2) (40
wt.% in aqueous solution), lead iodide (PbI2) (99.998%), dimethyl sulfoxide (DMSO) and
diethyl ether were purchased from Alfa Aesar and used without further purification. Methyl
ammonium iodide (CH3NH3I) was synthesized using the method described by Qi Wang, et
al.18 reacting 15 ml of HI with 13.5 ml of CH3NH2 at 0°C for 2 h in threeꢀneck flask under

Journal of Materials Chemistry A
Page 4 of 22
View Article Online
DOI: 10.1039/C4TA06392A
nitrogen atmosphere with constant stirring. Solvent was evaporated from the mixture by a
rotary evaporator and CH3NH3I white precipitation was observed. The precipitated white
powder was collected and washed with diethyl ether for three times and dried under vacuum
at 60°C for overnight. Dried powder was stored inside the glove box for further use in our
whole experiment.
Device fabrication and characterization
ITOꢀcoated glass substrates (<10 ꢀ sq–1, RiTdisplay) were cleaned through sonication once
in detergent (20 min) and subsequently twice in deionized (DI) water (20 min each) and then
dried under N2 gas. The substrates were treated with ultraviolet (UV)/ozone for 15 min to
clean the surfaces and also to improve the surface adhesion. PEDOT:PSS was spinꢀcoated on
ITO surfaces at 4000 rpm spin speed for 60 sec followed by annealed at 130°C for 30 min.
The substrates were transferred to a glove box for preparing the perovskite active layer by
twoꢀstep method. PbI2 was dissolved in DMSO at 40 wt.% and CH3NH3I was dissolved in 2ꢀ
proponal at 3 wt.% and both solutions were kept on the hot plate at 70°C for overnight. PbI2
solution was spinꢀcoated on PEDOT:PSS at different spin speeds ( from 4k to 8k rpm) and
directly CH3NH3I solution was sprayꢀcoated (room temperature) on PbI2 film at different
volumes (200 µl to 500 µl). After spraying the CH3NH3I, sample was kept on the hot plate at
100°C for 60 to 240 min to form crystalline perovskite film. The optimized condition based
on device performance was 100°C annealing for 120 min. The [6,6]ꢀphenylꢀC61ꢀbutyric acid
methyl ester (PCBM) was dissolved in dichlorobenzine (CB) at a concentration of 20 mg/ml
and was spun on the perovskite layer at 6000 rpm for 60 sec followed by annealing at 100°C
for 30 min. The device structure (see fig. S1) was completed by sequential thermal
evaporation of C60 (30 nm), 2,9ꢀdimethylꢀ4,7ꢀdiphenylꢀ1,10ꢀphenanthroline (7 nm BCP) and
100 nm aluminium electrode through a shadow mask under vacuum (pressure: 1 × 10–6 torr).
The active area of each device was 10 mm2.

Page 5 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Xꢀray diffraction (XRD) patterns were recorded at room temperature using a Bruker D8 Xꢀ
ray diffractometer equipped with a diffracted beam monochromator set for Cu Ka radiation (λ
= 1.54056 Å)—in the 2θ range 10°–80° with a step size of 0.03939° and step time of 99.45
sec. Scanning electron microscopy (SEM) images were recorded using an FEI Noval 200
scanning electron microscope (15 kV). PEDOT:PSS spinꢀcoated glass substrates were used
for XRD and PEDOT:PSS spinꢀcoated ITO substrates were used for SEM image record.
Photoluminescence (PL) emission spectra of the samples were carried out by optically
excited Qꢀswitched Nd: YAG laser (266 nm, 3–5 ns pulse, 10 Hz) focused on a beam
diameter of about 0.5 mm. And the absorption spectra of the films were measured using a
Jacobs Vꢀ670 UV–Vis spectrophotometer. Xꢀray photoelectron spectroscopy (XPS) was
performed using a PHI 5000 Versa Probe equipped with an Al Kα Xꢀray source (1486.6 eV).
All measurements were performed at room temperature. The devices were illuminated with a
solar simulator (Thermal Oriel 1000 W), which provided a simulated AM 1.5 spectrum (100
mW cm–2), inside a glove box using a Xe lamp. The light intensity was calibrated using a
monoꢀsilicon photodiode with a KGꢀ5 colour filter (Hamamatsu).
Results and discussion
The preparation of perovskite film based on twoꢀstep process was schematically shown in
Fig. 1. At first, DMSOꢀbased PbI2 solution was spinꢀcoated on ITO/PEDOT:PSS substrate
and then 2ꢀpropanolꢀbased CH3NH3I solution was sprayed on top of PbI2. Air brush sprayer
(contains 0.3 mm size of nozzle head and 7cc capacity of fluid cup) was connected with air
compressor and pressure was maintained at 40 psi. Brush sprayer was kept 12 cm distance
from the substrate to make sure that the samples were fully covered while spraying. CH3NH3I
droplets were interꢀdiffuse into the PbI2 while spraying and the colour of film starts to change
from yellow to light brown. PbI2 has relatively low solubility in 2ꢀpropanol, when spraying it
will not damage and 2ꢀpropanol can evaporate faster. After CH3NH3I sprayed, immediately

Journal of Materials Chemistry A
Page 6 of 22
View Article Online
DOI: 10.1039/C4TA06392A
samples were moved to hot plate at 100°C to further increase the reaction rate or diffusion
process. The film colour was changed from light brown to dark brown with shining surface
indicating the perovskite formation.
The perovskite film was prepared on glass substrate by twoꢀstep method at different volumes
of CH3NH3I precursor for XRD characterization. Figure 2(a) shows the XRD spectra at
different volumes of CH3NH3I, peak at 2
θ
of 12.71° for 200 ꢁl was belonging to unconverted
of 14.19°, 28.51°
PbI2 which exactly matched with PbI2 spectra. Peak with low intensity at 2
θ
and 31.94° corresponding to (002), (004) and (310) confirms the partial formation of
tetrahedral perovskite structure. By increasing CH3NH3I volume to 250 and 300 ꢁl, peak
intensities at 2
unconverted PbI2. Weaker peak at 2
(004), (220) and (310) confirms the formation of tetrahedral perovskite structure. Further
increasing the volume to 350 ꢁl, XRD shows the peak at 2 of 11.27°, which represents the
θ
of 14.16⁰ become strong without any evidence in the presence of
θ
of 28.24°, 28.47° and 31.94° corresponding to (002),
θ
unreacted CH3NH3I19 besides the perovskite peak with lower intensity. Figure 2(b) presents a
UVꢀvisible absorbance spectrum of perovskite film fabricated at different volume of
CH3NH3I after annealing at 100°C for 120 min. The UVꢀvisible spectrum reveals the
absorption onsets at 800 nm with broad ranges from the visible to the nearꢀinfrared for all the
volume. Further the absorption spectra show stoichiometric perovskite formation at 745 nm
wavelength. Figure 2(b) insert shows photograph of corresponding perovskite film, the film
at 200ꢁl was yellow colour due to insufficient amount of CH3NH3I and the colour of film
becomes dark brown by increasing the CH3NH3I volume to 250 and 300 ꢁl. At 350 ꢁl, the
unreacted CH3NH3I forms the white colour cloudy on the perovskite and we can almost
distinguish unconverted with fully converted perovskite film using our naked eyes based on
the colour of perovskite film. PL measurement was carried for PEDOT:PSS/PbI2 film with
different volumes of sprayed CH3NH3I shown in Fig.2 (c). While increasing the volume of

Page 7 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
CH3NH3I from 200 to 350 ꢁl, PL peak intensity increased with blue shifted from 773 to 766
nm. The spectra for 250 and 300 ꢁl were 771.85 and 771.25 nm, respectively, which is close
to the previously reported stoichiometric perovskite.21 The PL intensity increased by
increasing the volume from 250 to 300 ꢁl because of more number of excitons were
generated in crystalline perovskite and start to reduce by further increasing to 350 ꢁl which
was due to poor crystallinity of perovskite film. The figure insert shows different volumes of
CH3NH3I on PEDOT:PSS/PbI2 film with PC61BM quenching layer. The crystalline
perovskite film (250 ꢁl and 300 ꢁl of CH3NH3I) shows better quenching compared to non
-
crystalline perovskite film due to strong interface between perovskite and quencher.
The morphology of perovskite film depends on the volume of CH3NH3I over the PbI2 film; to
investigate this, SEM were captured and shown in Fig. 3. Morphology of spinꢀcoated PbI2 on
PEDOT:PSS (Fig. 3(a)) shows cracked with nonꢀcontinuous film; cracks started to disappear
while spraying of CH3NH3I. Morphology and grain size of perovskite film were controlled by
controlling the volume of CH3NH3I. Figure 3(b) shows a uniform perovskite film with
smaller grain size for 200 ꢁl sprayed CH3NH3I and by increasing the volume (300 ꢁl) the
grain sizes were increased because of stoichiometric ratio of PbI2 and CH3NH3I. Larger grain
size enhances the way for efficient charge transfer and reduction of exciton recombination.24
There were excess of CH3NH3I present on the surface of perovskite while using 350 ꢁl,
which indicates the formation of nonꢀcontinuous rough film. For comparison we included
SEM image of spinꢀcoated mixed perovskite solution (CH3NH3I:PbI2 (1:0.75) molar ratio) in
Fig. S2 shows nonꢀcontinuous rough film with lots of microfibers.
To understand the diffusion of liquid droplet (CH3NH3I) on PbI2, the XPS depth profile was
analysed. XPS depth profile was measured using Ar gun with alternative sputtering mode for
100 min sputtering with 3 min of interval time. First, PbI2 in DMSO was spinꢀcoated on
ITO/PEDOT:PSS substrate and it was immediately moved for CH3NH3I spray. The strong

Journal of Materials Chemistry A
Page 8 of 22
View Article Online
DOI: 10.1039/C4TA06392A
interaction between DMSO and Pb2+ can retard the crystal structure of PbI2, which directly
allows the CH3NH3I to diffuse and react fully with PbI2.25 The diffusion of CH3NH3I with
two different volumes (200 ꢁl and 300 ꢁl) in PbI2 film was analysed. While using the 200 ꢁl
volume of CH3NH3I, curve describes the distribution of higher iodine ion than lead ion on the
top surface and starts to drop before reaching the PEDOT:PSS film. Insufficient volume of
CH3NH3I produces incomplete conversion and guide to nonstoichiometric perovskite
formation as shown in Fig. 4(a). For 300 ꢁl of CH3NH3I, distributions of lead and iodine ion
were similar throughout the film and both started to drop once it reaches PEDOT:PSS. The
curve clearly shows the diffusion of iodine ion throughout the film and reaches the bottom of
PbI2 layer as shown in Fig. 4 (b). Carbon 1s peak starts to arise at the same time the iodine
and lead peaks start to drop after 45 min of sputtering. Complete formation of perovskite film
throughout the whole layer was observed while using 300 ꢁl volume of CH3NH3I precursor
with annealing at 100°C for 120 min. A volume of 300 ꢁl of CH3NH3I (3 wt.%) was enough
to fully convert the 100 ± 5 nm of PbI2 film into perovskite film and insufficient or excess
volume provide nonstoichiometric and low crystalline perovskite film.
We
fabricated
devices
having
the
architecture
of
glass/ITO/PEDOT:PSS/Perovskite/PC61BM/C60/BCP/Al (see Fig. S1) for investigating the
PV performance. Here, C60 was used to reduce the trap densities. Figure 5 shows the J–V
characteristics of the device with 100 ± 5 nm of PbI2 thickness with different volumes of
CH3NH3I under simulated air mass 1.5 global (AM 1.5G) solar irradiation. No delay time
between measurement voltage points in all J–V measurements. All the solar cells
performance were measured using aperture metal mask with 10 mm2 active area. Table 1 lists
the corresponding device parameters. The best performance device was monitored at 300 ꢁl
of CH3NH3I followed by annealing (100°C) for 60 min. The device showed the shortꢀcircuit
current density (JSC) of 20.84 mA/cm2, an openꢀcircuit voltage (VOC) of 0.91 V and a fill

Page 9 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
factor (FF) of 58.64%, leading to a high efficiency (η) of 11.12% and average η of 10.32 %.
Device performance of volume with 200 ꢁl (CH3NH3I) shows lower VOC, JSC, FF and η were
0.87 V, 16.45 mA/cm2, 55.69% and 7.97%, respectively, because of unconverted PbI2 still
present in perovskite film. This unconverted PbI2 suppresses the charge transport and
absorption of the device performance. While increasing volume CH3NH3I more than 300 ꢁl
also shows lower device efficiency of 10.38% with VOC of 0.91 V, JSC of 18.59 mA/cm2 and
FF of 61.36%, because of excess amount CH3NH3I lead, the poor crystalline perovskite film.
Perovskite film thickness varied from 175 ±5 nm to 230 ±5 nm while varying CH3NH3I
volume. We further optimize the thickness effect on perovskite film by varying the 40 wt.%
of PbI2 film with different volumes of CH3NH3I. All the devices were annealed for 60 min at
100°C and device performance parameters were listed in Table S2. Different spin speeds (4k,
5k, 6k, 7k and 8k rpm) of PbI2 film show the variation in thickness from 135 ± 5 nm to 80 ± 5
nm. For thicker film of PbI2 required more volume of CH3NH3I, vice versa for thinner film
required less volume to form stoichiometric perovskite film. Thick perovskite film absorbs
more light compared to thinner film expected to give improved JSC with better device
performance, but exciton recombination was higher in thicker film because of limited
diffusion length in iodineꢀbased perovskite film2,3. Best device performance was
accomplished at a thickness of 220 ± 5 nm perovskite film by 100 ± 5 nm (6000 rpm) of PbI2
film with 300 ꢁl of CH3NH3I gives 220 ± 5 nm perovskite film (Table 1). For comparison, J–
V characteristics and device performance parameters of spin (PbI2) / spray (CH3NH3I), spin (PbI2) /
spin (CH3NH3I) and oneꢀstep mixed solution (CH3NH3I:PbI2 (1:0.75) molar ratio) spin coating were
also included in Fig. S3 and Table S1. The corresponding device shows less efficiency of
7.38% with poorer JSC and FF due to nonꢀuniform film formation in single step preparation.
Optimizing the thickness of PC61BM is also important to improve the device
performance26,27, the J–V curve of perovskite layer with different thicknesses of PC61BM

Journal of Materials Chemistry A
Page 10 of 22
View Article Online
DOI: 10.1039/C4TA06392A
layers were shown in Fig. S4 and detailed device parameters are summarized in Table S3. For
130 nm thick PC61BM layer gives lower JSC and FF leads to poor device performance. This
arises due to the high series resistance when a dissociated charge carrier takes longer distance
to reach electrode. In the case of thinner PC61BM thickness, the charge transportation will be
easier and lead to higher JSC and FF. Too thinner PC61BM (~75 nm) shows decreased device
performance because of surface defects in PCBM layer. The optimal device performance was
obtained for 85 nm of thin PC61BM layer yields a η of 11.55%.
Device performance depends on annealing duration also investigated, the J–V characteristics
were monitored at different annealing durations which are shown in Fig. 6(b) and
summarized in Table 2. Device with 120 min annealing period gives better performance
compared to lower and higher annealing durations. While spraying CH3NH3I on top of PbI2,
the colour of perovskite had slightly changed from yellow to pale brown which represents the
less reaction happen between CH3NH3I and PbI2. Colour of the film starts to become dark
brown during prolonged annealing time and it plays a key role in crystalline perovskite
formation19,28,29. Shorter annealing duration may lead to unconverted perovskite film and too
longer annealing duration may degrade the perovskite film. Optimum annealing 120 min
shows the better VOC of 0.91 V, JSC of 21.06 mA/cm2 and FF of 60.84% guide to best η of
11.66% and average η of 11.17%. To check the reproducibility of device performance, we
repeated the device fabrication of best condition for more than 50 samples in four batches. As
shown in Fig. (c) –(f), the maximum number of devices display the PCE around 11% with the
average fill factor of 62%, shortꢀcircuit current of 19 mA cm−2 and openꢀcircuit voltage of
0.91 V.
Forward and reverse sweep of J–V curve for the champion device was shown in Fig. 6(a),
which gives negligible J–V hysteresis that were observed regardless of sweep direction.
Origin of J–V hysteresis was mainly attributed to the traps formation in lead halide perovskite

Page 11 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
materials. Optimized device structure and film formation significantly reduced the J–V
hysteresis.18 Here we used continuous perovskite film by two step method (spin / spray) and
PC60BM as a passivation layer to reduce the trap densities.30 The best device showed broad
external quantum efficiency (EQE) from 350 to 800 nm and attains maximum value of 88%
at 510 nm as shown in Fig. 6b. There was dimple at the wavelength of ~600 nm in EQE
spectrum, which is due to strong reflection of glass/ITO at this wavelength. The Jsc value
integrated from the EQE (18.87 mA cm−2) agreed well with that measured by solar simulator
(20.24 mA cm−2). The integrated Jsc from EQE spectrum was slightly lower than Jsc
measured from solar simulator, because of barrier for the photocurrent was large under
monochromatic (or low intensity) illumination and large number of micro shunts present
while the cell is irradiated only on a limited area31.
Conclusion
In summary, we demonstrated novel twoꢀstep preparation of combining spin and spray
coating technique to form continuous, crystalline and stoichiometric lead iodide perovskite
for highꢀefficiency hybrid solar cells. Simple airbrush sprayer was used to spray CH3NH3I,
which produces efficient perovskite film with best device performance of 11.66%. Complete
diffusion of CH3NH3I into PbI2 took place while providing sufficient amount of CH3NH3I,
and leads to high crystalline and stoichiometric perovskite formation. This combined
technique provides new route for fabrication of lowꢀtemperature solutionꢀprocessed large
area rollꢀtoꢀroll perovskite solar cells.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version.

Journal of Materials Chemistry A
Page 12 of 22
View Article Online
DOI: 10.1039/C4TA06392A
Acknowledgement
We thank Mr. Tzuꢀyen Huang, Mr. Mohmoud Farahat for their helpful discussions and Mr.
Senthil kumaran for thickness measurement. Dr. Chu thanks the Ministry of Science and
Technology (MOST), Taiwan (102ꢀ2221ꢀEꢀ001 ꢀ029 ꢀMY2), and the Career development
award of Academia Sinica, Taiwan (103ꢀCDAꢀM01) for financial support.
References:
1.
2.
3.
4.
5.
6.
7.
8.
9.
A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J Am Chem Soc, 2009, 131,
6050–6051.
S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens,
L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341–344.
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.
J.ꢀH. Im, I.ꢀH. Jang, N. Pellet, M. Grätzel and N.ꢀG. Park, Nat Nano, 2014, 9, 927–
932.
J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Energ Environ Sci, 2013, 6, 1739–
1743.
J. J. Choi, X. H. Yang, Z. M. Norman, S. J. L. Billinge and J. S. Owen, Nano Lett,
2014, 14, 127–133.
K. Wojciechowski, M. Saliba, T. Leijtens, A. Abate and H. J. Snaith, Energ Environ
Sci, 2014, 7, 1142–1147.
K. C. Wang, J. Y. Jeng, P. S. Shen, Y. C. Chang, E. W. G. Diau, C. H. Tsai, T. Y.
Chao, H. C. Hsu, P. Y. Lin, P. Chen, T. F. Guo and T. C. Wen, Sci RepꢀUk, 2014, 4.
P. Qin, S. Tanaka, S. Ito, N. Tetreault, K. Manabe, H. Nishino, M. K. Nazeeruddin
and M. Gratzel, Nat Commun, 2014, 5, DOI:10.1038/ncomms4834.

Page 13 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
10.
11.
J. Y. Jeng, K. C. Chen, T. Y. Chiang, P. Y. Lin, T. D. Tsai, Y. C. Chang, T. F. Guo, P.
Chen, T. C. Wen and Y. J. Hsu, Adv Mater, 2014, 26, 4107ꢀ4113.
P. Docampo, F. Hanusch, S. D. Stranks, M. Döblinger, J. M. Feckl, M. Ehrensperger,
N. K. Minar, M. B. Johnston, H. J. Snaith and T. Bein, Advanced Energy Materials,
2014, DOI: 10.1002/aenm.201400355.
12.
13.
A. T. Barrows, A. J. Pearson, C. K. Kwak, A. D. F. Dunbar, A. R. Buckley and D. G.
Lidzey, Energ Environ Sci, 2014, 7, 2944–2950.
Q. Chen, H. Zhou, Z. Hong, S. Luo, H.ꢀS. Duan, H.ꢀH. Wang, Y. Liu, G. Li and Y.
Yang, J Am Chem Soc, 2013, 136, 622–625.
14.
15.
16.
D. Liu and T. L. Kelly, Nat Photon, 2014, 8, 133–138.
M. Z. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
O. Malinkiewicz, C. RoldánꢀCarmona, A. Soriano, E. Bandiello, L. Camacho, M. K.
Nazeeruddin and H. J. Bolink, Advanced Energy Materials, 2014, 1–9.
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.
17.
18.
19.
20.
21.
22.
Q. Wang, Y. Shao, Q. Dong, Z. Xiao, Y. Yuan and J. Huang, Energ Environ Sci,
2014, 7, 2359–2365.
Z. Xiao, C. Bi, Y. Shao, Q. Dong, Q. Wang, Y. Yuan, C. Wang, Y. Gao and J. Huang,
Energ Environ Sci, 2014, 7, 2619–2623.
C.ꢀW. Chen, H.ꢀW. Kang, S.ꢀY. Hsiao, P.ꢀF. Yang, K.ꢀM. Chiang and H.ꢀW. Lin, Adv
Mater, 2014, 26, 6647–6652.
J. Burschka, N. Pellet, S. J. Moon, R. HumphryꢀBaker, P. Gao, M. K. Nazeeruddin
and M. Gratzel, Nature, 2013, 499, 316–319.
J. H. Kim, S. T. Williams, N. Cho, C.ꢀC. Chueh and A. K. Y. Jen, Advanced Energy
Materials, 2014, 1–6.

Journal of Materials Chemistry A
Page 14 of 22
View Article Online
DOI: 10.1039/C4TA06392A
23.
24.
25.
26.
27.
28.
29.
M. Ramesh, K. M. Boopathi, T.ꢀY. Huang, Y.ꢀC. Huang, C.ꢀS. Tsao and C.ꢀW. Chu,
Acs Appl Mater Inter, 2015, 10.1021/am506886d
E. Edri, S. Kirmayer, A. Henning, S. Mukhopadhyay, K. Gartsman, Y. Rosenwaks, G.
Hodes and D. Cahen, Nano Lett, 2014, 14, 1000–1004.
Y. Wu, A. Islam, X. Yang, C. Qin, J. Liu, K. Zhang, W. Peng and L. Han, Energ
Environ Sci, 2014, 7, 2934–2938.
L. Dou, J. You, Z. Hong, Z. Xu, G. Li, R. A. Street and Y. Yang, Adv Mater, 2013,
25, 6642–6671.
J. Seo, S. Park, Y. Chan Kim, N. J. Jeon, J. H. Noh, S. C. Yoon and S. I. Seok, Energ
Environ Sci, 2014, 7, 2642–2646.
A. Dualeh, N. Tétreault, T. Moehl, P. Gao, M. K. Nazeeruddin and M. Grätzel, Adv
Funct Mater, 2014, 24, 3250–3258.
C. Bi, Y. Shao, Y. Yuan, Z. Xiao, C. Wang, Y. Gao and J. Huang, J Mater Chem A,
2014, 2, 18508–18514.
30.
31.
Y. Shao, Z. Xiao, C. Bi, Y. Yuan and J. Huang, Nat Commun, 2014, 5,
doi:10.1038/ncomms6784
R. Scheer and H.ꢀW. Schock, Chalcogenide Photovoltaics: Physics, Technologies,
and Thin Film Devices, WileyꢀVCH Verlag GmbH & Co. KGaA, 2011.

Page 15 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Table 1: Device performance parameters of PbI2 thickness of 100 ± 5 nm while varying
different volumes of CH3NH3I at a concentration of 3 wt.%.
PbI2
thickness
CH3NH3I (3 Perovskite VOC (V) JSC
FF (%)
η
(%)
wt.%)
thickness
(mA/cm2)
volume
100 ± 5 nm 200 ꢁl
(6000 rpm)
175 ± 5
nm
0.87
0.89
0.91
0.91
16.45
7.97a
55.69
58.39
58.64
61.36
(7.62)b
9.37a
250 ꢁl
200 ± 5
nm
18.03
(8.81)b
11.12a
(10.32)b
10.38a
(9.86)b
300 ꢁl
350 ꢁl
220 ± 5
nm
245 ± 5
nm
20.84
18.59
a – best device efficiency, b – average device efficiency for 32 samples
Table 2: Device performance parameters of perovskite thickness of 220 ± 5 nm while varying
different annealing durations.
Different annealing VOC (V)
JSC (mA/cm2)
FF (%)
η
(%)
durations
60 min
0.91
0.91
0.90
20.84
21.06
20.01
11.12a (10.32)b
11.66a (11.17)b
10.77a (10.06)b
58.64
60.84
59.80
120 min
180 min
a – best device efficiency, b – average device efficiency for 32 samples

Journal of Materials Chemistry A
Page 16 of 22
View Article Online
DOI: 10.1039/C4TA06392A
Figure 1: Schematic representations of perovskite formation from twoꢀstep process (spin
coated PbI2 and sprayꢀcoated CH3NH3I). (a) First, PbI2 solution was spinꢀcoated on
PEDOT:PSS and then secondly CH3NH3I solution was sprayꢀcoated using normal air brush
sprayer, (b) diffusion of sprayed CH3NH3I into PbI2 film and forms perovskite upon
annealing.

Page 17 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Figure 2: (a) XRD pattern of PbI2 and perovskite at different volumes of sprayed CH3NH3I
solution, (b) Absorption of perovskite film at different volumes of sprayed CH3NH3I solution
and (c) PL measurement of perovskite film at different volumes of sprayed CH3NH3I
solution, Q represents film with quenching layer (PC61BM), insert shows perovskite with
quencher.

Journal of Materials Chemistry A
Page 18 of 22
View Article Online
DOI: 10.1039/C4TA06392A
Figure 3: SEM image of (a) spinꢀcoated PbI2 film, spinꢀcoated PbI2 film with (b) 200 ꢁl, (c)
300 ꢁl, (d) 350 ꢁl volume of CH3NH3I stacked film, with annealing temperature of 100°C for
120 min. Insert shows the magnified image of same film. Scale bars correspond to 2 ꢁm.
(Insert scale bars represent 500 nm.)

Page 19 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Figure 4: XPS depth profile analysis of perovskite film on the top of PEDOT:PSS (a) 100 ±
5 nm of PbI2 with 200 ꢁl of CH3NH3I and (b) 100 nm of PbI2 with 300 ꢁl of CH3NH3I, after
annealing at 100°C for 120 min.

Journal of Materials Chemistry A
Page 20 of 22
View Article Online
DOI: 10.1039/C4TA06392A
Figure 5: J–V characteristics of (a) device with PbI2 thickness of 100 ± 5 nm while varying
different volume of CH3NH3I at a concentration of 3 wt.%, (b) device with 220 ± 5 nm
perovskite film while varying different annealing durations, (c) to (f) shows statistical
analysis of device performance for the best condition.

Page 21 of 22
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C4TA06392A
Figure 6: (a) J–V curve measured at forward and reverse sweep directions and (b) EQE
spectrum of the device employing 220 ± 5 nm perovskite film.

Journal of Materials Chemistry A
Page 22 of 22
View Article Online
DOI: 10.1039/C4TA06392A
Continuous and crystalline metal halide perovskite films for efficient solar cells using liquid
droplet assisted method