Surface Photovoltage Spectroscopy Study of Ultrasonically Sprayed-Aerosol CH3NH3PbI3 Perovskite Crystals

Article abstract of DOI:10.1002/pssa.201800133

A simple deposition process for preparing crystalline semiconductors with low degrees of disorder is of paramount interest for both device applications and research investigations. This study centers on the ultrasonically sprayed-aerosol based approach for preparing crystals of methylammonium lead trihalide perovskite (CH₃NH₃PbI₃). The surface photovoltage (SPV) spectra are found to depend strongly on the preparation conditions, with the SPV signals (below the material's band gap) providing information on the defect states. The measured values of tail states near the band gap (E_t) are found to be about 21 and 52 meV for the CH₃NH₃PbI₃ crystals prepared by the ultrasonically sprayed-on and spun casting approaches, respectively.

Full text of DOI:10.1002/pssa.201800133

a
ORIGINAL PAPER
Perovskite Crystals
www.pss-a.com
Surface Photovoltage Spectroscopy Study of
Ultrasonically Sprayed-Aerosol CH3NH3PbI3
Perovskite Crystals
Nakorn Henjongchom, Nopporn Rujisamphan, I-Ming Tang, and Thidarat Supasai*
mixed in a common solvent. The precipi-
A simple deposition process for preparing crystalline semiconductors with
low degrees of disorder is of paramount interest for both device applications
and research investigations. This study centers on the ultrasonically sprayed-
aerosol based approach for preparing crystals of methylammonium lead
trihalide perovskite (CH3NH3PbI3). The surface photovoltage (SPV) spectra
are found to depend strongly on the preparation conditions, with the SPV
signals (below the material’s band gap) providing information on the defect
states. The measured values of tail states near the band gap (Et) are found to
be about 21 and 52 meV for the CH3NH3PbI3 crystals prepared by the
ultrasonically sprayed-on and spun casting approaches, respectively.
tate arising from the interaction of these
two precursors would be the CH3NH3PbI3
perovskite. Burschka et al.[12] pointed out
that the uncontrolled precipitation of the
perovskite would result in large variations
in the morphology of the spun casted films
on various substrates. This would result in
a wide range of photovoltaic performances
of the devices employing these perovskite
films. Further improvements in the device
performance can also be achieved by
enhancing the formation of an active layer
in the perovskites, by which better surface
coverage,[13–16] higher crystallinity,[17–20]
larger domains/grain sizes,[21–24] and lower
defect states below the band gap[25–27] are obtained. For
examples, Shi et al.[28] employed the anti-solvent vapor-assisted
crystallization (AVC) method to grow relatively large single
crystals of the perovskite with low trap density, whose
performances when used in solar cells resemble those seen in
up-to-date Si based solar cells.
1. Introduction
As the emergence of the light absorbing materials, fabricated
from methylammonium lead halides CH3NH3PbX3 (where
X ¼ I and Br) perovskites for solar cell application,[1] much
research has been devoted to the continuous improvements in
the efficiencies[2–5] and the stability[6–10] of these organic-
inorganic hybrid perovskites based solar cells. Many of the
studies used a procedure similar to that described by Ball et al.[11]
to fabricate the material. In the procedure, a mixture of lead
iodide (PbI2) and methylammonium iodide (CH3NH3I) are
High crystalline semiconducting materials having low
degrees of disorder, i.e., evident by the presence of low energy
tail states (Et) near the band gap or deep defect states (Et, deep), are
of paramount interest for both device applications and research
investigations. Yang et al.[20] showed that the imperfections of
the perovskite crystals were greatly suppressed when using a
sequential spin-casting approach to fabricate the bilayer films
first, and then followed by a moderate heating in ambient air.
The passivation of the surface defects at CH3NH3PbI3 surface
was also reported after heating at 100 ꢀC.[29] Lian et al.[30] showed
that the large single crystal of the CH3NH3PbI3 perovskite
having a trap density of 7.6 ꢁ 108 cmꢂ3 and high carrier mobility
of 167 ꢃ 35 cm2 Vꢂ1 sꢂ1 could be achieved by using chloride to
meditate the solution based process. Dittrich et al.[31] reported
that the values of Et were found to be in the range of 14–24 meV
for the uncoated layers of the CH3NH3PbI3 perovskite and of
about 20 meV for the spun casting of the CH3NH3PbI3
perovskite onto nano porous TiO2. Wolf et al.[32] reported that
the value of the exponential trap states below the band gap
corresponding to an Urbach tail energy (Eu) was increased to
15 meV for the solution processed CH3NH3PbI3 perovskite
while Landi et al.[33] showed a correlation between Eu, (estimated
from the quantum efficiency spectra) and the grain size of the
CH3NH3PbI3. They observed that the lowest value of Eu was
found to be 31 meV in the relatively large grain size samples.
This was due to the fact that the larger grains provide a longer
N. Henjongchom, Dr. T. Supasai
Department of Materials Science
Faculty of Science
Kasetsart University
Bangkok 10900, Thailand
E-mail: fscitrs@ku.ac.th
Dr. N. Rujisamphan
Nanoscience and Nanotechnology Graduate Program
Faculty of Science
King Mongkut’s University of Technology Thonburi
Bangkok 10140, Thailand
Dr. N. Rujisamphan
Theoretical and Computational Science Center (TaCS)
Faculty of Science
King Mongkut’s University of Technology Thonburi
Bangkok 10140, Thailand
Prof. I-M. Tang
Computational and Applied Science for Smart Innovation Cluster
(CLASSIC) Faculty of Science
King Mongkut’s University of Technology Thonburi
Bangkok 10140, Thailand
DOI: 10.1002/pssa.201800133
Phys. Status Solidi A 2018, 1800133
1800133 (1 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
diffusion length and lower trap densities. As a side note, the
measured Et is not exactly the same as the Urbach tail as the
optical absorption considers the excitation from a localized
occupied into a localized unoccupied state whereas the SPV
signals arise from the charge separation where it occurs.
We report here on a modification of a technique/process
introduced by Balaji et al.[34] In their process, the films were
prepared by a two-step deposition method. The non-iodide lead
salts dissolved in a dimethylformamide (DMF) solution were
first spun casted onto a substrate. The CH3NH3PbI3 perovskites
were then slowly formed (in a few hours) after the spun-casting
layers were brought into contact with the methylammonium
iodide (CH3NH3I) vapors. This ultrasonically sprayed-aerosol
method is based on the ultrasonic nebulization of the precursor
solutions. This technique allows for the film preparation on a
large area of substrate area.[35–37] In this work, an ultra-sonic
vibration was used to generate ultra-fine aerosols of lead (II)
nitrate (Pb(NO3)2) and of methylammonium iodide (CH3NH3I).
Also, the dissolution of the non-iodide lead salts of lead (II)
nitrate in a nontoxic solvent would be an environmentally
friendly chemical process.[34,38] The time intervals, however,
before the CH3NH3PbI3 perovskite formation is completed, are
long enough to allow for the tracking of the changes in their
structural, optical and electronic properties. This allows one to
gain an understanding of the perovskite transformations
relevant to photo-induced charge generation and charge
separation behaviors during the sprayed-aerosol process. The
structural and optical properties of the CH3NH3PbI3 crystals
were measured with an X-ray diffraction spectroscope (XRD) and
UV-Vis spectrometer, respectively. Changes in the charge
separation and charge transport behaviors arising from the
local crystallization and/or during the evolution of a crystal
growth were probed using modulated surface photovoltage
(SPV) spectroscopy.
generated.[39] Then, the aerosols of the MAI dissolved in absolute
ethanol having a concentration of 0.3 mol Lꢂ1 were generated in
a similar manner to that of Pb(NO3)2 and sequentially sprayed
onto the as-sprayed-on Pb(NO3)2. In order to investigate a
formation of the CH3NH3PbI3 crystals and the evolution of
crystallization, various time durations (i.e., 5, 10, 15, 20 and
30 min.) for the spraying of MAI aerosol were used. After each
spraying, the individual samples were then heated at 100 ꢀC for
15 min inside the set-up where the nitrogen gas had flowed. This
was done before any characterizations. We note that before the
spraying of the MAI aerosols, the substrate temperature had to
be cooled down gradually to room temperature (30 ꢀC). It should
be noted that the substrate temperatures have a crucial effect on
the resulting crystal formation. For example, if the substrate
temperature is higher than the boiling point of the solvents
being used, some of the aerosols could be evaporated before
reaching the substrate and therefore longer reaction time for the
CH3NH3PbI3 crystal formation and conversion were needed.
For comparison, a series of CH3NH3PbI3 perovskite samples
prepared at various relative humidities (%RHs) were also
prepared by a conventional one-step spin casting approach using
a mixture of lead iodide (PbI2) (99.8%) (purchased from Sigma–
Aldrich) and MAI dissolved in 2 ml of γ-butyrolactone (purchased
from Sigma–Aldrich) in the equivalent molar amounts. The
mixed solution was stirred at 60 ꢀC overnight before use. The
CH3NH3PbI3 solution wasthenspuncasting ontofluorine-doped
tin oxide (FTO) coated glass substrates by spin coating done at
3000 rpm for 2 min. This was followed by a moderate heating at
100 ꢀC for 15 min to remove any excess solvent.
2.2. Sprayed-On CH3NH3PbI3 Characterization
The crystal structures of the sequentially sprayed-on
CH3NH3PbI3 perovskites were studied by X-ray diffraction
(XRD, Bruker D8 using a CuK radiation source). The
α
diffraction angles were gradually varied from 10ꢀ to 50ꢀ with
step increases of 0.02ꢀ. The surface morphology images of the
perovskites were obtained by a scanning electron microscope
(SEM) (Quanta 450 FEI), operating with an accelerating
voltage of 30 kV. Prior to taking images, a thin layer of Au
was deposited on the samples by sputtering to minimize the
charging effect. The absorbance spectra were collected in a
range of 300–850 nm using a UV–Vis spectrometer (Perkin
Elmer Lambda model 650) equipped with an integrating sphere
component. Modulated surface photovoltage spectroscopy
(SPV) measurements were performed under vacuum
(2 ꢁ 10ꢂ2 mbar) in a fixed capacitor arrangement where a very
thin mica sheet (30–40 μm thick) was used to form the
insulating spacer between the measurement electrode and the
sprayed-on perovskite samples. The samples were illuminated
through a SnO2:F electrode coated quartz cylinder. More details
of the SPV set-up and measurements can be found else-
where.[40] In this technique, the SPV signals which are
proportional to the photo-induced charge generation and charge
separation were measured with a double-phase lock-in amplifier
(from Elektron–Manufaktur Mahlsdorf, Germany), which was
connected through a high impedance buffer (10 GΩ). A halogen
light source of 100 W with a quartz prism monochromator
2. Experimental Results
2.1. Materials and Preparation
The CH3NH3I (MAI) was synthesized using the method
described elsewhere.[29] It was obtained by reacting 27.8 ml of
methylamine solution (CH3NH2, 40% in methanol, TCI) with
30 ml of hydroiodic acid (HI, 57% in water) (purchased from
Sigma–Aldrich) in a 250 ml round-bottom flask at 0 ꢀC with
stirring for 2 h. The white precipitates of the MAI were washed in
a diethyl ether (CH3OCH3) solvent and re-crystallized in ethanol
by adding diethyl ether. The crystals of the MAI were then dried
at 60 ꢀC in a vacuum oven for 24 h. Figure 1a shows a schematic
drawing of the ultrasonically sprayed-aerosol process for
preparing crystals of CH3NH3PbI3. First, a precursor solution
of Pb(NO3)2 (99%) (purchased from Sigma–Aldrich) having a
concentration of 0.3 mol Lꢂ1 was dissolved in deionized water.
Aerosolized forms of Pb(NO3)2 were generated by an ultrasonic
mist generator and then were transported using a nitrogen (N2)
carrier gas with a flow rate of 3 L minꢂ1 toward a FTO substrate
heated at 70 ꢀC for 15 min. Due to the nature of the vibrating
plate used in a mist maker, aerosolized forms (aerosols) having a
very small diameter in a range of 0.001 to 0.003 mm were
Phys. Status Solidi A 2018, 1800133
1800133 (2 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 1. a) Schematic of the sequentially ultrasonic sprayed-aerosol setup. b) X-ray diffraction (XRD) analysis of a sprayed Pb(NO3)2 layer, and of
sequentially sprayed MAI on the Pb(NO3)2 layer at different periods of time, i.e. 5, 10, 15, 20 and 30 min. Note that ꢀ, ꢄ, @, and # shown in the XRD
patterns correspond to the Pb(NO3)2, PbI2, CH3NH3PbI3 perovskite, and dihydrate (CH3NH3)4PbI6 ꢅ 2H2O characteristic peaks, respectively.
c) Photographs of the sprayed-on Pb(NO3)2 after exposing to aerosols of MAI at different periods of time. d) UV-Vis absorption spectra of the
sprayed-on Pb(NO3)2 and of sequentially sprayed MAI on the Pb(NO3)2 layer at various reaction periods of time.
(Bausch & Lomb) were used for the illumination. The light was
chopped with a modulation frequency of 15 Hz.
turned yellowish after the aerosolized MAI came into contact
with the Pb(NO3)2 surface (within 5 min (see Figure 1c)). The
XRD pattern of the resulting film after sequentially spraying
MAI for 5 min on the Pb(NO3)2 layers reveals the diffraction
peaks perfectly matching the characteristic peaks of the PbI2.
This indicated that the aerosolized MAI greatly enhanced the ion
exchange between the NO3ꢂ of Pb(NO3)2 and Iꢂ of MAI, needed
for the chemical interaction and caused the formation of an
intermediate state of the PbI2 (which is also evident by the
yellowish color). Formation of this intermediate state of the lead
thiocyanate (Pb(SCN)2) was also seen when it was exposed to
MAI vapors.[34]
Increasing the duration of the spraying time to 10 min,
resulted in the signature peaks of PbI2 to appear and for the
partial formation of sprayed-on CH3NH3PbI3 perovskite to
begin. Also there was a change in the color from yellowish to
blackish (see Figure 1c). A closer examination of the XRD
patterns revealed that the complete formation of the
CH3NH3PbI3 perovskite occurs within 15 min. While all peaks
which appeared in the XRD pattern of the sprayed-on
CH3NH3PbI3 perovskite remain unchanged, increasing the
spraying time of the MAI for another 20 or 30 min, caused a new
3. Results and Discussion
Figure 1b showstheXRDpatternsofthesprayed-onCH3NH3PbI3
perovskite when different deposition conditions were used.
Looking at the XRD patterns, we observe that the crystallographic
structures of the sprayed-on CH3NH3PbI3 perovskite went
through thefollowingstructuralphasetransitions:cubicstructure
of Pb(NO3)2, hexagonal 2H polytype of PbI2 and finally the
tetragonal structure of the CH3NH3PbI3 perovskite.
For an only sprayed-on layer of the Pb(NO3)2, the XRD pattern
shows characteristic peaks at 19.56ꢀ, 22.61ꢀ, 37.94ꢀ, and 39.69ꢀ
corresponding to reflections from the (111), (200), (311), and
(222) planes, respectively. Those peaks are in good agreement
with the material database (no. 36–1462) for the Pa-3 space
group. Calculation of the lattice parameters for Pb(NO3)2 having
̊
a cubic structure yields a ¼ b ¼ c ¼ 7.858 ꢃ 0.005 A, which also
agrees well with those report in the literature.[41] It is worth
mentioning that the color of the sprayed-on Pb(NO3)2 slowly
Phys. Status Solidi A 2018, 1800133
1800133 (3 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
peak to appear at 11.4ꢀ. This peak shows a presence of dihydrate
methylammonium lead iodide ((CH3NH3)4PbI6 ꢅ 2H2O). This
compound is due to the overabundance of MAI in the perovskite,
and has been reported by Halder et al.[42] and by Patel et al.[43] An
increase in the XRD peak intensity of the dihydrate phase is
clearly observed when the sample is brought into contact with
the MAI for 30 min. To future investigate the extent of this phase
mixing, quantitative analysis of the dihydrate percentage
times), the absorption edge begins to appear at 810 nm. This is
the signature of the band gap in CH3NH3PbI3 perovskite. The
complete formation of the CH3NH3PbI3 perovskite is observed
after spraying the MAI for 15 min by the appearance of a steep
absorption edge and a maximum intensity in the absorbance
spectra. Further increasing spraying time of MAI (i.e., for 20 and
30 min) causes the absorbance intensity to decrease abruptly
with another absorption edge located at around 370 nm
beginning to appear. This feature is due to the presence of
the dihydrate (CH3NH3)4PbI6 ꢅ 2H2O.[42] It is worth mentioning
that the absorption characteristics of the sprayed-on
CH3NH3PbI3 perovskite exhibited a non-monotonic behavior.
The light absorption occurs when atoms/molecules absorb the
photon energy. The differences in the structural formation,
crystals packing, and surface coverage of the CH3NH3PbI3
perovskites can occur at any time before the surface is exposed to
the light rays. These variations can contribute to the non-
monotonic behavior.
C
in the sprayed-on CH3NH3PbI3 perovskite
ðCH3NH3Þ4 PbI6ꢅ2H2O
was characterized as follows:
ꢀ
I11:4
C
¼
;
ð1Þ
ðCH3NH3Þ4 PbI6ꢅ2H2O
ꢀ
ꢀ
I11:4 þ I14:04
ꢀ
ꢀ
where I11:4 and I14:04 are the peak intensity of the dihydrate and
perovskite phases. As deduced from the XRD patterns, the
percentage of the (CH3NH3)4PbI6 ꢅ 2H2O phase in the sprayed-
on CH3NH3PbI3 perovskite is about 32.44% after spraying MAI
for 20 min, but is greatly increased when spraying the MAI for
30 min to 56.20%.
The structural formation, crystals packing, and SEM
micrographs of the sprayed-on CH3NH3PbI3 perovskite
prepared using various spraying times are shown in Figure 2.
As seen in Figure 2a,b, at the early stages of the spraying of the
As reported in the literature,[34] the ions exchange between
NO3ꢂ and Iꢂ, occurring when the aerosols
of the Pb(NO3)2 and CH3NH3I chemically
interact makes it possible for the PbI2 phase
to form. This gives rise to a yellow color of
the intermediate state. The following
chemical reaction via the ion exchange is
proposed:
2CH3NH3I þ PbðNO3Þ2 ! PbI2
þ 2CH3NH3NO3;
PbI2 þ CH3NH3I ! CH3NH3PbI3:
As sprayed-on CH3NH3PbI3 perovskite
crystals are continuously being exposed to
the aerosols of MAI, the overabundant
supply of MAI suppresses the formation of
CH3NH3PbI3 perovskite and eventually
alters it into another structure, that of the
dihydrate (CH3NH3)4PbI6 ꢅ 2H2O. A previ-
ous report found that intercalation of the
water molecules into the perovskite struc-
ture forms hydrogen bonds between the
water and ammonium component, there-
fore giving rise to the formation of the
[PbI6]4ꢂ octahedral geometry.[42,44,45]
The absorbance spectra of the sprayed-
on CH3NH3PbI3 perovskite are shown in
Figure 1d. The sprayed-on Pb(NO3)2 does
not absorb in the ultraviolet to visible
regions. However, after exposure to MAI
for 5 min,
a new absorption edge at
wavelength of 525 nm begins to appear.
This absorption edge is related to the band
gap of PbI2. Further increasing the expo-
sure to MAI (by increasing the spraying
Figure 2. a) Top-view SEM images of sprayed-on Pb(NO3)2 and of sequentially sprayed MAI on the
Pb(NO3)2 for various time durations: b) 5 min, c) 10 min, d) 15 min, e) 20 min, and f) 30 min. Inset
shows the image at a 2 μm scale bar.
Phys. Status Solidi A 2018, 1800133
1800133 (4 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Figure 3. a) Schematic drawing of the electrode configuration for surface photovoltage measurements in the fix capacitor arrangement. b) Illustrations
of the origin of the X and Y signals collected from the double phase lock-in amplifier (showing the dark and illumination phases, sample voltage,
reference signals from in-phase and phase shifted by 90ꢀ, and X and Y signals after multiplication). c) A sketch showing where the charge carrier
dynamics can be explained with the SPV signal (i.e., trapping/detrapping and recombination mechanism).
MAI (i.e., 5 min), there is not much change in the morphology.
Changes in crystal structure (as shown in XRD results) and in
the morphologies, (appearance of small grain facets) begin to
appear when sequentially spraying of MAI is done for 10 min
(see Figure 2c). After spraying MAI for 15 min (Figure 2d), a
structural conversion into CH3NH3PbI3 perovskite with a large
crystal size of about 1 μm is observed. According to theoretical
considerations (based on Gibbs free energy) the grain size of
the CH3NH3PbI3 would be affected by the concentration of the
MAI used and the temperature used.[47] As seen, when spraying
the MAI for 20 and 30 min (Figure 2e,f), the cuboid crystals
collapse with the excess MAI remaining on the surface (see
Figure 2e,f).
separation in a space charge region and the trapping of charge
carriers at interface states.
Figure 4 shows the X and Y signals when the sequentially
sprayed-on MAI on the Pb(NO3)2 is carried out for various time
durations. As a remark, the X and Y signals provide evidence of
the fast and slow responses with respect to changes in the on
and off illumination, respectively, while the surface photo-
voltage amplitude (PV amplitude) is defined as the square root
of the sum of the squared X and Ysignals. When the X signal is
positive (negative) and the Y signal is negative (positive), the
preferential direction of the photo-generated electron carriers
The basic principles of modulated SPV and its measurement
are briefly summarized as follows:
The spectral dependent modulated SPV measurements were
performed in a fixed capacitor with a mica sheet between the
sample and electrode forming the capacitor (see also Figure 3a).
The incident light was chopped at a frequency of 15 Hz and a
lock-in amplifier measured the signal from the sample as an in-
phase (X) signal and as a phase-shifted by 90ꢀ (Y) signal with
respect to the chopped light signal. Figure 3b shows the origin of
the X and Y signals collected from the lock-in amplifier.
Several processes can contribute to the SPV signals of
modulated charge separation, i.e., trapping/de-trapping and
polarization. Figure 3c shows a sketch where the charge carrier
dynamics can be explained with the SPV signal (i.e., trapping
(black arrows)/detrapping (red arrows) and recombination
mechanism at the surface). In the case that the charge
separation is caused by only one mechanism and that the
relaxation of SPV signals is caused by the recombination of the
charges, the X and Y signals have the opposite sign.[46] It should
be noted that mechanisms of charge separation with opposite
sign can also be caused by a combination of the charge
Figure 4. Modulated SPV spectra of the in-phase (filled circles) and
phase-shifted by 90ꢀ (open circles) signals of sequentially sprayed-on
MAI on the Pb(NO3)2 for various time durations: a) 10 min, b) 15 min,
c) 20 min, and d) 30 min.
Phys. Status Solidi A 2018, 1800133
1800133 (5 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
alsoseeninFigure4b,theXandYsignalsarenegativeandpositive,
respectively. This is experimental evidence for electron separation
and preferential migration toward the external surface with the
holes migrating preferably toward the internal interface.[29] The
relatively larger magnitude for the X signal (reaching maximum at
ꢂ380 mV at photon energy 1.6 eV) than for the Ysignal (reaching
maximum only at 112mV at photon energy 1.6 eV) provides
evidencethatthechargeseparationandtransportationshowsafast
response with respected to the modulation period.
After allowing the reaction time to reach 20 min, we noted
initially that the maximum X signals decreased dramatically
from ꢂ380 to ꢂ225 mV (at photon energy 1.6 eV) whereas the Y
signals increased sharply from 112 to 195 mV (at photon 1.6 eV),
as seen in Figure 4c. A strong decrease and increase in the X and
Ysignals, respectively, were clearly observed for the 30 min of the
sprayed-on CH3NH3PbI3 perovskite (see Figure 4d). This could
be attributed to the defects or impurities being induced by the
overabundance amount of MAI,[47,50] which causes a variation of
the orientation of organic MAI molecules resulting in a
distortion of PbI6 octahedral framework.[49] Therefore the
changes in the magnitude of the SPV signals could also be
linked to the structural formation of the sprayed-on perovskite.
The reduction of the SPV signals seen originates from defect
states below the band gap and so the SPV measurements can
provide information on the abundance of the defect states.[29]
Figure 5a shows the spectrum of the photovoltage amplitude
on a logarithm scale while Figure 5b shows that the spectrum
divided by the photon flux. Et can be obtained by the fitting the
SPV amplitude at the leading edge according to the equation as
Figure 5. SPV spectra (on a logarithm scale) of the PV amplitude of the
sequentially sprayed-on MAI on the Pb(NO3)2 for 15 min (a) and of that
amplitude divided by the photon flux (b). The solid lines in (b) show
slopes of the exponential tail states (Et) for the sequentially sprayed-on
method. Insert shows the value of Et for the CH3NH3PbI3 perovskite
prepared by the one-step spin casting method.
follows[51]
:
created by the charge separation is toward the internal
(external) interface and thus the opposite sign of the X and
Ysignals gives an evidence of only one dominant mechanism of
charge separation.[46]
As is seen, when the aerosols of MAI reacted with the Pb(NO3)2
for 10 min (Figure 4a), both the X and Y signals show an onset
energy of about 2.35 eV, which corresponds to the absorption edge
of the illuminating radiation in the PbI2 phase. Again, the results
agreewellwiththeXRDandUV–Vismeasurements. Wenotehere
that, for an as-sprayed Pb(NO3)2 layer (in an absence of any MAI
spraying), no SPV signals were detected in the whole range of the
photon energy studied. By allowing the reaction time between the
MAI and Pb(NO3)2 to be up to 15 min (see
ꢀ
ꢁ
hv
Et
SPVamplitude ¼ Aexp
;
where A is a proportionality constant and hv is a photon energy.
It was found that the value of Et is equal to 21 meV for the
sprayed-on perovskite, which is 57% lower than that obtained
from the one-step spin-casting sample (see inset of Figure 5b).
Figure 6a summarizes the optical band gap (Eg,Tauc) values
obtained from the absorption spectra and of the onset energy
(Eon,SPV) obtained from SPV measurements. The measured
Figure 4b), the onset energy at around 2.35 eV
disappearedandrelativelysharponsetenergyin
the SPV spectrum appeared at about 1.45 eV.
This onset energy could be connected to the
opticalbandgapoftheCH3NH3PbI3 perovskite.
The band gap of CH3NH3PbI3 (varied from 1.5
to 1.7 eV) was previously reported and was
dependent on the preparation conditions and
measurement techniques.[47–49] It should be
noted that while the SPV signals could be
originated from different processes caused to a
separation in space of photon-exited charge
carriers, i.e. trapping/de-trapping process at the
shallow or deep defect states or injection into a
Figure 6. a) Correlation of the optical band gap energy (filled circles), modulated SPV
absorption onset energy (open circles), the crystalline sizes (filled triangles), and the
exponential tail states (open triangles) for the sprayed-on CH3NH3PbI3; and b) Structural
phase transitions as a function of the reaction time.
neighboring layer, an optical absorption
accounts only for the excitation from a localized
occupied into a localized unoccupied state. As is
Phys. Status Solidi A 2018, 1800133
1800133 (6 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
values of the Eon,SPV vary from 1.46 to 1.51 eV, which is 0.02–0.1
lower than that of the Eg,Tauc, which are in a range of 1.53–
1.56 eV. The difference is due to the fact that the SPV signals
arise from the charge separation where it occurs, i.e., from the
excitation of electrons from the tail states to the localized
unoccupied states or from the localized occupied to the tail states
whereas the band gap energy relates to the optical absorption
corresponding to the excitation from a localized occupied state to
a localized unoccupied state. The average crystallite size of
sprayed-on CH3NH3PbI3 calculated from the diffraction peak at
14.04ꢀ (by using the Scherrer equation[52]) varies in the range of
27–30 nm independent of the reaction times. It is well accepted
that the structural, morphological and electronic properties of
the sample depend strongly upon preparation technologies. The
Et seems to depend on the phase evolution (see Figure 6b) from
PbI2 to CH3NH3PbI3. The value of Et for the CH3NH3PbI3 with
the reaction time for 10 min is about 34 meV and abruptly
decreases to 21 meV in 15 min. As shown in the inset of
Figure 5b, the spun-casting perovskite sample prepared in a 75%
relative humidity (RH) without the use of toluene dripping
exhibits an Et of 52 meV. To see how the preparation technology
affects the values of Et, a series of spun casted samples prepared
at various %RH were also fabricated. Figure 7 shows the values
of Et and their surface morphologies for the CH3NH3PbI3
perovskite samples prepared in various %RH. As seen, the set of
samples fabricated by the spin casting approach presents the
values of Et ranging from 26 to 35 meV, depending on the %RH
conditions. It is also worth mentioning that the Et is responsible
for the recombination kinetics, and therefore in perovskite solar
cells the recombination kinetics are greatly influenced by the
electron-phonon interactions.[33]
It should be remembered that the sum of the thermal and
static terms is often responsible for the energy disorder in a
system.[55] Recently, Singh et al.[56] reported that at room
temperature, the main contribution to the energy disorder of the
tetragonal phase of CH3NH3PbI3 is related to the thermal
disorder whereas the static contribution is the dominant one at
lower temperatures (T < 60 K). Based on the measured SPV
results, the observed energy disorder would be related to the
thermal disorder. In literature the measured values of Et have
been reported to depend largely on the preparation conditions
and the layer structures. For example, an Et of 38 meV was found
for the CH3NH3PbI3 perovskite deposited on a layer of the TiO2
whereas the value of 21 meV was reported for a double layer
design by having TiO2 with the organic fullerene derivatives.[57]
Awino et al.[58] showed the dependence of PMMA concentrations
for stabilizing the CH3NH3PbI3. The values of Et were found to
be in a range of 17–45 meV and were strongly dependent on the
concentrations of PMMA being used.
4. Conclusions
In sum, using an ultrasonically sprayed-on based technique, we
were able to prepare crystals of CH3NH3PbI3 perovskite. In this
method, nebulized forms of Pb(NO3)2 were sprayed-onto a
substrate, followed by a sequential spraying of a nebulized form
of MAI. The structural and optical properties of the resulting
CH3NH3PbI3 perovskite were found to be greatly influenced by
the reaction time (spraying time of MAI). A complete conversion
to form a CH3NH3PbI3 structure having cuboid facets was
achieved after spraying the MAI onto Pb(NO3)2 for 15 min. The
proof is seen in several measurements. The conclusion is
supported by the presence of an onset wavelength at 810 nm, in
the UV-Vis absorbance spectrum having an extended absorption
intensity band, and further confirmed by the XRD measure-
ments. While the SPV measurements allow for a systematic
determination of the photo-induced charge generation and
charge separation in the (sprayed-on) perovskite, analysis of the
SPV spectra not only provides information on conversion of the
structure but also determines the dependences on the properties
on the preparation conditions. Also, SPV measurements could
be performed over a short period of time and can therefore be
used for in-line control of the preparation of other metal halide
perovskite compounds.
As clearly seen, the obtained morphological and electronic
properties of the CH3NH3PbI3 perovskite depend very much on
the preparation technologies. Also the small grain boundaries
and defects in polycrystalline sample hinder the overall device
performance; a perovskite layer with large grains and pin-hole
free is highly desired for achieving high-efficiency device.[53,54]
Acknowledgements
This work was supported by a co-funding research program from EGAT-
NSTDA (Grant #P-15-50560 phase 2). T.S would like to acknowledge the
Thailand Research Fund (Grant# MRG 5980035). The authors would like
to acknowledge the financial support provided by Kasetsart University
Research and Development Institute (grant no. P-3.2 (D) 228.61) and by
King Mongkut’s University of Technology Thonburi through the
“KMUTT 55th Anniversary Commemorative Fund.”
Figure 7. The obtained Et for CH3NH3PbI3 samples prepared in various %
RH conditions by the one-step spin casting method. Inset shows the top-
view SEM images.
Conflict of Interest
The authors declare no conflict of interest.
Phys. Status Solidi A 2018, 1800133
1800133 (7 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
www.advancedsciencenews.com
www.pss-a.com
Keywords
P. A. Dowben, O. F. Mohammed, E. H. Sargent, O. M. Bakr, Science
2015, 347, 519.
[29] T. Supasai, N. Rujisamphan, K. Ullrich, A. Chemseddine, T. Dittrich,
Appl. Phys. Lett. 2013, 103, 183906.
CH3NH3PbI3, surface photovoltage, trap states, ultrasonically sprayed-on
based technique
[30] Z. Lian, Q. Yan, T. Gao, J. Ding, Q. Lv, C. Ning, Q. Li, J. L. Sun, J. Am.
Chem. Soc. 2016, 138, 9409.
[31] T. Dittrich, C. Awino, P. Prajongtat, B. Rech, M. C. Lux-Steiner, J. Phys.
Chem. C 2015, 119, 23968.
Received: February 19, 2018
Revised: April 13, 2018
Published online:
[32] S. De Wolf, J. Holovsky, S.-J. Moon, P. Löper, B. Niesen, M. Ledinsky,
F.-J. Haug, J.-H. Yum, C. Ballif, J. Phys. Chem. Lett. 2014, 5, 1035.
[33] G. Landi, H. C. Neitzert, C. Barone, C. Mauro, F. Lang, S. Albrecht,
B. Rech, S. Pagano, Adv. Sci. 2017, 4, 1700183.
[1] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 2009,
131, 6050.
[34] G. Balaji, P. H. Joshi, H. A. Abbas, L. Zhang, R. Kottokkaran,
M. Samiee, M. Noack, V. L. Dalal, Phys. Chem. Chem. Phys. 2015, 17,
10369.
[35] J. E. Bishop, D. K. Mohamad, M. Wong-Stringer, A. Smith,
D. G. Lidzey, Sci. Rep. 2017, 7, 7962.
[36] J. Qiu, Z. Jin, J. Qian, Y. Shi, W. Wu, J. Cryst. Growth 2005, 282, 421.
[37] C.-H. Fischer, N. A. Allsop, S. E. Gledhill, T. Köhler, M. Krüger,
R. Sáez-Araoz, Y. Fu, R. Schwieger, J. Richter, P. Wohlfart, P. Bartsch,
N. Lichtenberg, M. C. Lux-Steiner, Sol. Energy Mater. Sol. Cells 2011,
95, 1518.
[2] M. Liu, M. B. Johnston, H. J. Snaith, Nature 2013, 501, 395.
[3] M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami, H. J. Snaith,
Science 2012, 338, 643.
[4] J.-W. Lee, D.-J. Seol, A.-N. Cho, N.-G. Park, Adv. Mater. 2014, 26, 4991.
[5] X. Zhang, J. Ye, L. Zhu, H. Zheng, G. Liu, X. Liu, B. Duan, X. Pan,
S. Dai, Nanoscale 2017, 9, 4691.
[6] S. R. Raga, L. K. Ono, Y. Qi, J. Mater. Chem. A 2016, 4, 2494.
[7] C. Liu, W. Ding, X. Zhou, J. Gao, C. Cheng, X. Zhao, B. Xu, J. Phys.
Chem. C 2017, 121, 6546.
[8] F. Wang, A. Shimazaki, F. Yang, K. Kanahashi, K. Matsuki,
Y. Miyauchi, T. Takenobu, A. Wakamiya, Y. Murata, K. Matsuda, J.
Phys. Chem. C 2017, 121, 1562.
[38] T.-Y. Hsieh, T.-C. Wei, K.-L. Wu, M. Ikegami, T. Miyasaka, Chem.
Commun. 2015, 51, 13294.
[39] T. Supasai, N. Henjongchom, I. M. Tang, F. Deng, N. Rujisamphan,
Sol. Energy 2016, 136, 515.
[40] V. Duzhko, V. Timoshenko, F. Koch, T. Dittrich, Phys. Rev. B 2001, 64,
75204.
[41] G. K. Bichile, R. G. Kulkarni, Acta Crystallogr. Sect. A 1975, 31, 446.
[42] A. Halder, D. Choudhury, S. Ghosh, A. S. Subbiah, S. K. Sarkar, J.
Phys. Chem. Lett. 2015, 6, 3180.
[43] J. B. Patel, R. L. Milot, A. D. Wright, L. M. Herz, M. B. Johnston, J.
Phys. Chem. Lett. 2016, 7, 96.
[9] W. Chen, Y. Wu, Y. Yue, J. Liu, W. Zhang, X. Yang, H. Chen, E. Bi,
I. Ashraful, M. Grätzel, L. Han, Science 2015, 350, 944.
[10] F. Bella, G. Griffini, J.-P. Correa-Baena, G. Saracco, M. Gratzel,
A. Hagfeldt, S. Turri, C. Gerbaldi, Science 2016, 354, 6309.
[11] J. M. Ball, M. M. Lee, A. Hey, H. J. Snaith, Energy Environ. Sci. 2013, 6,
1739.
[12] J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao,
M. K. Nazeeruddin, M. Grätzel, Nature 2013, 499, 316.
[13] Q. Zhao, G. R. Li, J. Song, Y. Zhao, Y. Qiang, X. P. Gao, Sci. Rep. 2016,
6, 38670.
[14] F. Arabpour Roghabadi, V. Ahmadi, K. Oniy Aghmiuni, RSC Adv.
2016, 6, 112677.
[15] Y. Ding, X. Yao, X. Zhang, C. Wei, Y. Zhao, J. Power Sources 2014, 272,
351.
[16] Y. Li, X.-L. He, B. Ding, L.-L. Gao, G.-J. Yang, C.-X. Li, C.-J. Li, J. Power
Sources 2016, 320, 204.
[17] B. Wang, K. Y. Wong, S. Yang, T. Chen, J. Mater. Chem. A 2016, 4, 3806.
[18] F. Zhang, J. Song, M. Chen, J. Liu, Y. Hao, Y. Wang, J. Qu, P. Zeng,
Phys. Chem. Chem. Phys. 2016, 18, 32903.
[44] B. R. Vincent, K. N. Robertson, T. S. Cameron, O. K. Can, J. Chem.
1987, 65, 1042.
[45] A. M. A. Leguy, Y. Hu, M. Campoy-Quiles, M. I. Alonso, O. J. Weber,
P. Azarhoosh, M. van Schilfgaarde, M. T. Weller, T. Bein, J. Nelson,
P. Docampo, P. R. F. Barnes, Chem. Mater. 2015, 27, 3397.
[46] P. Prajongtat, T. Dittrich, J. Phys. Chem. C 2015, 119, 9926.
[47] X. Ke, J. Yan, A. Zhang, B. Zhang, Y. Chen, Appl. Phys. Lett. 2015, 107,
91904.
[48] S. A. Bretschneider, J. Weickert, J. A. Dorman, L. Schmidt-Mende,
APL Mater. 2014, 2, 40701.
[49] C. Motta, F. El-Mellouhi, S. Kais, N. Tabet, F. Alharbi, S. Sanvito, Nat.
Commun. 2015, 6, 7026.
[50] W.-J. Yin, T. Shi, Y. Yan, Adv. Mater. 2014, 26, 4653.
[51] A. Owino Juma, A. Azarpira, A. Steigert, M. Pomaska, C.-H. Fischer,
I. Lauermann, T. Dittrich, J. Appl. Phys. 2013, 114, 53711.
[52] A. Patterson, Phys. Rev. 1939, 56, 978.
[19] A. Ishii, A. K. Jena, T. Miyasaka, APL Mater. 2014, 2, 91102.
[20] B. Yang, O. Dyck, J. Poplawsky, J. Keum, A. Puretzky, S. Das, I. Ivanov,
C. Rouleau, G. Duscher, D. Geohegan, K. Xiao, J. Am. Chem. Soc.
2015, 137, 9210.
[21] Z. Liang, S. Zhang, X. Xu, N. Wang, J. Wang, X. Wang, Z. Bi, G. Xu,
N. Yuan, J. Ding, RSC Adv. 2015, 5, 60562.
[22] M. K. Kim, T. Jeon, H. Il Park, J. M. Lee, S. A. Nam, S. O. Kim,
CrystEngComm 2016, 18, 6090.
[53] J. Chen, T. Shi, X. Li, B. Zhou, H. Cao, Y. Wang, Appl. Phys. Lett. 2016,
108, 53302.
[23] M. Yang, T. Zhang, P. Schulz, Z. Li, G. Li, D. H. Kim, N. Guo,
J. J. Berry, K. Zhu, Y. Zhao, Nat. Commun. 2016, 7, 1.
[24] Z. Bi, Z. Liang, X. Xu, Z. Chai, H. Jin, D. Xu, J. Li, M. Li, G. Xu, Sol.
Energy Mater. Sol. Cells 2017, 162, 13.
[54] W. Nie, H. Tsai, R. Asadpour, J.-C. Blancon, A. J. Neukirch, G. Gupta,
J. J. Crochet, M. Chhowalla, S. Tretiak, M. A. Alam, H.-L. Wang,
A. D. Mohite, Science 2015, 347, 522.
[55] G. D. Cody, T. Tiedje, B. Abeles, T. D. Moustakas, B. Brooks,
Y. Goldstein, Le J. Phys. Colloq. 1981, 42, C4.
[56] S. Singh, C. Li, F. Panzer, K. L. Narasimhan, A. Graeser, T. P. Gujar,
A. Köhler, M. Thelakkat, S. Huettner, D. Kabra, J. Phys. Chem. Lett.
2016, 7, 3014.
[57] L. Kegelmann, C. M. Wolff, C. Awino, F. Lang, E. L. Unger, L. Korte,
T. Dittrich, D. Neher, B. Rech, S. Albrecht, ACS Appl. Mater. Interfaces
2017, 9, 17245.
[25] S. D. Stranks, P. K. Nayak, W. Zhang, T. Stergiopoulos, H. J. Snaith,
Angew. Chemie ꢂ In t. Ed. 2015, 54, 3240.
[26] F. Si, F. Tang, H. Xue, R. Qi, J. Semicond. 2016, 37, 72003.
[27] S. Heo, G. Seo, Y. Lee, D. Lee, M. Seol, J. Lee, J.-B. Park, K. Kim, D.-
J. Yun, Y. S. Kim, J. K. Shin, T. K. Ahn, M. K. Nazeeruddin, Energy
Environ. Sci. 2017, 10, 1128.
[28] D. Shi, V. Adinolfi, R. Comin, M. Yuan, E. Alarousu, A. Buin, Y. Chen,
S. Hoogland, A. Rothenberger, K. Katsiev, Y. Losovyj, X. Zhang,
[58] C. Awino, V. Odari, T. Sakwa, Coatings 2017, 7, 115.
Phys. Status Solidi A 2018, 1800133
1800133 (8 of 8)
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim