Journal of Solid State Electrochemistry

https://doi.org/10.1007/s10008-019-04430-2

ORIGINAL PAPER

Spin-coated copper(I) thiocyanate as a hole transport layer

for perovskite solar cells
Utku Er 1,2 & Kerem Cagatay Icli 2,3 & Macit Ozenbas 1,2

Received: 17 April 2019 / Revised: 14 October 2019 / Accepted: 16 October 2019

# Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
Application of a low-cost and efficient p-type inorganic hole-transporting material, copper thiocyanate (CuSCN), on mesoporous n-
i-p-configurated perovskite-based devices was conducted in this study. Diethylsulfide was chosen for the preparation of precursor
solution in order to deposit CuSCN layer on perovskite without degrading it. Topographical, elemental, and electrical characteriza-
tions of spin-coated CuSCN layers were performed using XRD, AFM, SEM, XPS, UPS, and UV-Vis studies. A power conversion
efficiency exceeding 11.02% with an open-circuit voltage of 0.83 V was succeeded in the perovskite solar cells under full sun
illumination. Low-temperature solution process used for the deposition of CuSCN and a fast solvent removal method allowed the
creation of compact, highly conformal CuSCN layers that facilitate rapid carrier extraction and collection. The differences in series
and recombination resistances for CuSCN-free and CuSCN-containing cells were also determined using impedance spectroscopy
(IS) analysis. Moreover, the effect of TiO2 layer thickness on the cell performance was studied where these TiO2 layers were used
not only for electron extraction and transportation, but also as hole blocking layer in perovskite solar cells. The impedance
spectroscopy results were also consistent with the differently configurated cell performances. This work shows a well-defined n-
i-p perovskite cell with optimized layers which utilize low-cost and abundant materials for photovoltaic applications.

Keywords Perovskite solar cells . N-i-p mesoporous structure . Copper thiocyanide thin film . Hole transport material

Introduction cells on high illumination of solar radiation and expenditure

Solar energy is one of the most studied alternatives of clean
and renewable energy source worldwide, while current devel-
opments and applications are highly promising. Although cur-
rent applications with Si-based solar cells show higher effi-
ciency relative to other photovoltaic systems such as perov-
skite solar cells (PSCs), the necessity of the Si-based solar
Dedicated to the memory of Ivo Alexandre Hümmelgen
Utku Er and Kerem Cagatay Icli contributed equally to this work.
* Macit Ozenbas
ozenbas@metu.edu.tr
1
Department of Metallurgical and Materials Engineering, Middle East
Technical University, Dumlupinar Bulvari, 06800 Ankara, Turkey
2
Center for Solar Energy Research and Applications (GUNAM),
Middle East Technical University, Dumlupinar Bulvari,
06800 Ankara, Turkey
3
Micro and Nanotechnology Graduate Program, Middle East
Technical University, Dumlupinar Bulvari, 06800 Ankara, Turkey
of the production has pulled back the interest on the PSCs.
Due to their suitable electrical and optical properties, organo-
lead halide perovskites have allured much attention for solar
cell applications. Reaching outstanding power conversion ef-
ficiencies about 20%, perovskite-based solar cells have been
breakthrough photovoltaic materials recently [1, 2]. More eco-
nomical raw materials and easy production procedures for
fabrication are significant features of these devices which
make them attractive for researchers around the world and in
photovoltaic industry. In basic form, the architecture is deter-
mined by multilayer stack of light-absorbing perovskite layer
sandwiched between hole- and electron-transporting contact
materials. Many efforts have centered on the deposition of
dense perovskite layers, which are void and crack free with
maximum grain sizes, and several deposition methods, sol-
vent engineering [3, 4], additive-assisted deposition [5, 6],
and two-step deposition process [7, 8] have been improved,
yielding grain sizes up to a millimeter level. To achieve highly
efficient and stable solar cells, compositional engineering [9]
approaches were used in modifying the band structure of the
perovskite layers, and particular analysis of defect structures

for achieving defect-free perovskite layers by suitable addi-
tives was developed [10]. Modification of the electron trans-
port layers with dopants is promising for replacing the widely
used TiO2 layers with wider bandgap materials like SnO2,
which possess less hysteresis [11]. Dopants like cesium are
proven to not only modify the band structure of the perovskite
layers and enhance the tunability of bandgap or band posi-
tions, but also pave the way for all inorganic perovskite layers
for solving the stability issues related to metal-organic layers
[12]. In addition to well-known electrochemical spectroscopic
techniques, femto- and nanosecond transient absorption
methods have been developed and used in order to understand
charge injection and recombination dynamics for revealing
the mechanisms behind governing the cell function [13]. On
the side of commercialization, attempts have focused on re-
placing the organic layers causing stability problems with the
alternative inorganic counter parts and development of large-
area deposition techniques like screen printing or spray depo-
sition alternative to spin coating [14]. Although toxic Pb con-
taining perovskite layers may possess environmental issues
[15] and efforts are conducted to replace it with non-toxic
alternatives like Sn [16], recently, a perovskite mini-module
integrated with a CO2 to hydrocarbon reduction process has
been realized and demonstrated as a CO2 reduction and stor-
age device as a green technology application [17]. Other than
these engineering and feverish studies of the perovskite layer,
in turn for energy harvesting, charge transport contacts play a
significant role in extracting charge carriers from absorber
layer. Former studies proved that hole and electron transport
layers are important for efficient charge collection and attain-
ment of the maximum photovoltage [18–20]. In perovskite
solar cells, especially semiconducting metal oxides like ZnO
and TiO2 are n-type electron transport layers for n-i-p config-
urations when counter p-type hole transport layers are espe-
cially polymeric materials like spiro-OMeTAD, P3HT, and
MEH-PPV [21]. Although high efficiency values can be pro-
vided by organic conjugacy, the dilemma is higher costs of
these materials and the long-term instability. The p-type inor-
ganic metal oxide layers provide abundance and long-term
stability when collaborated in optoelectronic devices includ-
ing OLEDs [22] as well as organic solar cells [23].
Use of inorganic hole transport materials (HTM, such as
NiO, CuI, Cs2SnI6, and CuSCN) provides an opportunity for
stable power conversion efficiencies of 20% in perovskite
solar cells when subjected to light soaking under realistic op-
erational conditions [24–28]. Especially, CuSCN is a low-cost
and abundant p-type semiconductor which exhibits good ther-
mal stability, high hole mobility, and a well-aligned work
function among various inorganic HTMs [28]. CuSCN trans-
mits light across the entire visible and near-infrared spectral
region, and it is intrinsically p-doped material; therefore, it is
also charming for tandem solar cell applications where the
perovskite solar cell is positioned on top of a semiconductor
J Solid State Electrochem
with a lower bandgap value [29]. Doctor blading, electrode-
position, spin coating, and spray coating have been tried as
CuSCN film deposition methods [30–34]. The solution-based
bottom-up processes are more easy to produce; however, most
of the solvents in which CuSCN shows high solubility de-
grade the perovskite layer, so it is the most critical issue asso-
ciated with these techniques. The solvents that readily dis-
solve CuSCN, but not the perovskite, are rather limited. To
overcome this problem, an inverted device architecture or oth-
er deposition routes of CuSCN have been used in the literature
[29].
In this work, diethylsulfide was chosen for the preparation
of precursor solution in order to deposit CuSCN layer as HTM
layer on perovskite without degrading it. CuSCN is known to
have higher solubility in diethylsulfide compared with that in
dipropylsulfide, which is the most common solvent for depo-
sition of CuSCN layers. In addition, dipropylsulfide has a
polar character and it has the potential to damage the perov-
skite layer [35]. Due to higher solubility of CuSCN in
diethylsulfide, we could successfully achieve to deposit ho-
mogeneous and sufficiently thick CuSCN layers by a simple
spin-coating technique compared with unreproducible doctor
blading techniques employed in previous studies. In their
high-impact work, Arora et al. [36] investigated spin-coated
CuSCN layers, deposited on a hybrid CsFAMAPbI3–xBrx pe-
rovskite layer and achieved stable efficiencies exceeding 20%
by the aid of modification of the CuSCN layer in conjunction
with a reduced graphene oxide layer. They obtained high sta-
bility values along with a strong dependence of the CuSCN-
layer thickness on the cell efficiency, and the reason of this
dependency is unclear. We have deposited CuSCN layers by
spin coating on CH3NH3PbI3 perovskite layers without mod-
ification and investigated the effect of each layer with electro-
chemical impedance spectroscopic studies for providing a
deeper insight into contribution of the CuSCN layers to device
performance along with mesoporous electron transport layer
and the hole blocking layer for more reproducible and consis-
tent cell performances for future studies.
Experimental
Deposition of CuSCN films and cell preparation
N,N-Dimethylformamide (DMF), lead iodide (PbI 2 ),
methylammonium iodide (MAI), diethylsulfide, diethyl ether
anhydrous, dimethyl sulfoxide (DMSO), and copper thiocya-
nate (CuSCN) were purchased from Sigma-Aldrich. All
chemicals were used as received without further purification.
Firstly, fluorine-doped tin oxide (FTO)–coated glass sub-
strates (20 mm × 15 mm, Solaronix 22-7) were wet-etched
with zinc powder and hydrochloric acid. Etched FTO-coated
glass substrates were cleaned with DI water, detergent
J Solid State Electrochem
solution, acetone, and isopropanol ultrasonically. TiO 2
blocking layer was deposited on clean FTO substrates by
magnetron sputtering (Leybold, Uniwex 350) from a 3-in.
TiO2 sputtering target at 2 × 10−6 mbar and 250 W RF power,
until desired thicknesses (25, 50, 75, and 100 nm) were
achieved. Mesoporous (mp) TiO2 layers were deposited on
the blocking layers (BL) by spin-coating a solution of TiO2
nanoparticle paste diluted with ethanol with a ratio of 1:3.
Details of preparation of TiO2 paste are given elsewhere
[37]. Spin coating was conducted at 2000 rpm for 20 s. TiO2
layers with different thicknesses were achieved by repeating
the coating procedure after drying. Drying was performed at
125 °C for 5 min and substrates were fired at 500 °C for
15 min. After cooling down, TiO2 films were subjected to
TiCl4 treatment by immersing the substrates into 0.2 mM
TiCl4 aqueous solution at 90 °C for 10 min. For deposition
of perovskite layers, adduct method was followed [38]. A 40-
wt.% perovskite precursor was prepared in DMF including
PbI2, MAI, and DMSO with a molar ratio of 1:1:1. This pre-
cursor was spin-coated on TiO2-coated substrates at 4000 rpm
for 30 s. During spinning, 0.2 ml diethyl ether was poured at
once on the spinning substrate at the fifth second. Transparent
films were crystallized at 100 °C for 5 min and shiny black
films were obtained. For deposition of CuSCN layers,
diethylsulfide was chosen as the solvent where CuSCN is
known to have higher solubility compared with that in
dipropylsulfide which is the most common solvent for depo-
sition of CuSCN layers. A stock solution of 50 mg/ml CuSCN
in diethylsulfide was prepared. CuSCN layers with different
thicknesses were achieved by diluting this solution. Spin coat-
ing was performed at 2000 rpm for 30 s. Films were dried at
room temperature and 100-nm gold was evaporated on the
finished cells by e-beam evaporation (Leybold, Uniwex 350)
at 5 × 10−6 mbar. The cells were prepared on 15 × 20-mm-
sized FTO-coated glass substrates etched from the two sides
for making contacts during characterizations. Each substrate
contains 8 stripes of Au which were evaporated using a mask.
Active area of each cell is defined as the area between the edge
of the gold stripe and the etched FTO part, which corresponds
to 0.07 cm2. Measurements were taken from each contact
defined as the cell in this work by masking the cells using
the same mask for Au evaporation. J-V measurements were
recorded for the best performing cell on each substrate, and
impedance measurements were also conducted on these best
performing cells.
Characterizations
X-ray diffraction patterns of CuSCN and perovskite films
were recorded by Rigaku D/MAX 2200/PC using CuKα ra-
diation. Devices were investigated morphologically using
scanning electron microscopy (FEI Quanta 400 FEG equipped
with EDX analyzer). CuSCN films were characterized by X-
ray photoelectron spectroscopy (XPS) and ultraviolet photo-
electron spectroscopy (UPS) (PHI 5000 VersaProbe) for anal-
ysis of chemical and electronic properties. Morphological
properties of the films were also investigated by atomic force
microscopy (AFM) (Veeco CP II). Optical properties of the
films were determined by UV-Vis measurements (Ocean
Optics, USB-ISS), and visible light transmission of the films
was recorded between 300- and 900-nm wavelength. J-V
curves of the cells were recorded by a Yokogawa GS610
source measure unit under simulated AM 1.5 conditions using
Newport solar simulator 67005. Electrochemical impedance
spectroscopy was conducted using Gamry reference 3000 un-
der illumination with white LED light (60 mW/cm2), and
fitting was conducted by the Echem analyst software.
Results and discussion
In this work, studies have been conducted to construct perov-
skite solar cells employing copper thiocyanate (CuSCN) layer
as a hole transport layer, and characterizations were made in
order to examine structural and electronic properties of these
films and their effect on the performance of the cells. CuSCN
layers were characterized using XRD, AFM, and SEM studies
to identify structural properties like surface topography and
crystal structures.
The XRD pattern of the CuSCN thick layer on the glass
substrate is shown in Fig. 1. Observed diffraction peaks could
be well matched to the rhombohedral crystal orientations,
which is coherent with JCPDS card no. 29-0581. The promi-
nent peaks monitored at 2θ angles of 16.16°, 27.22°, 32.64°,
34.58°, 47.07°, 49.82°, 55.26°, 58.48°, 59.73°, 62.30°,
68.94°, and 75.44° are appointed scattering from (003),
(101), (006), (104), (110), (009), (021), (116), (024), (205),
(027), and (211) planes of the CuSCN crystal without
Fig. 1 XRD pattern of CuSCN thick film deposited on glass substrate

impurity peaks, respectively. Sharpness of the diffraction
peaks in the pattern indicates the good crystallinity of the
deposited CuSCN layer.
The topography of CuSCN thin films on perovskite layer,
which is analyzed using atomic force microscopy (AFM) and
scanning electron microscopy (SEM), can be seen in Fig. 2a
and b, respectively. 4.5 μm × 4.5 μm 2D observation on the
AFM scans was used to estimate the roughness of the spin-
coated thin CuSCN layer. The dark brown to white contrast
displayed alongside the AFM scan image in Fig. 2a represents
the height scale of the AFM scans. The surface roughness
value determined by the 2D AFM scan is 19.46 nm, and the
observed average height in surface scan was 75.58 nm. The
average surface roughness value around 18 nm and presence
of nanoscale islands were also found in the former studies
proven by grazing incidence XRD measurements, from a so-
lution concentration of 35 mg/ml. The concentration of
50 mg/ml employed in this study gives similar results, and it
is consistent with the former study by Arora et al. and shows
the ability of the dynamic deposition method yielding highly
smooth surface characteristics [36]. 2D AFM surface scans
and SEM images support the allocation of the CuSCN layer
and both confirm the homogeneity of the fabricated layers.
In order to get accurate quantitative and qualitative infor-
mation about the chemical compositions and state of the ele-
ments in the prepared CuSCN film on glass substrates, high-
resolution XPS spectra of the characteristic regions were re-
corded for the CuSCN film. The XPS survey spectra given in
Fig. 3 specify the presence of the expected elements which are
copper (Cu), sulfur (S), carbon (C), and nitrogen (N).
The main peak of the copper Cu 2p3/2 spectrum at 931.6 eV
corresponds to Cu(I) with its shake-up satellite positioned at a
higher binding energy as shown in Fig. 3. The spectrum in the
sulfur 2p region has been fitted by a series of three doublets of
S 2p3/2 and S 2p1/2 each. The spectrum is dominated by a large
signal with 2p3/2 peak situated at 162.0 eV, which was as-
cribed to a sulfur atom in –S–C form. Observed XPS spectra
of C include C 1s aliphatic and C 1s, –SCN. The smaller peak,
Fig. 2 a 2D AFM image of
CuSCN layer. b SEM image of
CuSCN layer on perovskite
J Solid State Electrochem
which was used for the calibration of the whole spectrum, is
ascribed to the aliphatic carbon sp3 impurities present on the
surface, while the larger peak at around 285 eV corresponds to
carbon-nitrogen bond present in the CuSCN layer. The spec-
trum of N 1s has a major peak at 397.3 eV, which probably
corresponds to the nitrogen in a nitrile form (N≡C) with a
small shoulder [28].
Electronic structure of the CuSCN layer fabricated using
the spin-coating technique was studied by UPS using HeI
emission (21.21 eV), and spectrum is given in Fig. 4a. Work
function of the deposited films is defined by the cutoff energy
seen in Fig. 4a, which is located at 16.07 eV. Beyond this
value, no photoelectron emission is possible, and by
subtracting this value from the energy value of the HeI emis-
sion (21.21 eV), work function of the films or the Fermi level
(Ef) was found to be 5.14 eV with respect to vacuum level.
Onset energy value seen in Fig. 4a, which is 0.36 eV, defines
the energy difference between the valence band maximum
(VBM) and the Fermi level [39]. By adding this value to the
calculated Fermi level, the position of the VBM is found to be
5.50 eV. The optical bandgap value was calculated from the
Tauc plot given in Fig. 4b obtained using the light transmis-
sion spectrum of the spin-coated CuSCN layer on glass sub-
strates and found to be 3.86 eV. By simply subtracting the
bandgap value calculated from the Tauc plot, from the
VBM, conduction band maximum (CBM) of the films was
found to be 1.64 eV. The position of Fermi level was 0.36 eV
above the VBM, which is below the middle of the bandgap,
advocating the p-type character of the layer. The calculated
band diagram of the CuSCN layer in this work is described in
Fig. 4c with respect to the perovskite layer.
It is seen that the difference between VBMs of the CuSCN
and perovskite layers was found to be 0.07 eV and they are
close to each other. The large difference between CBMs of
CuSCN and perovskite layers ensures prevention of electron
transfer from perovskite to CuSCN and demonstrates an ef-
fective electron blocking feature of CuSCN in the cells. In the
work of Arora et al., they report faster extraction of charges
J Solid State Electrochem

Fig. 3 XPS spectra of the spin-

coated CuSCN film indicate the
presence of the expected
elements: copper, sulfur, carbon,
and nitrogen

Fig. 4 a UPS spectra of the CuSCN layer used in the fabrication of the of CuSCN films showing the conduction band minimum (CBM), valence
perovskite cells indicating the onset and cutoff energies. b UV-visible band maximum (VBM), and Fermi level (Ef) acquired from UPS
light transmission spectra of the CuSCN layer deposited on glass measurements with respect to that of perovskite
substrates and the Tauc plot is given in the inlet image. c Band diagram
 J Solid State Electrochem

Fig. 5 a Schematic illustration of

mesoscopic n-i-p configuration of
perovskite solar cell constructed
in this work. b Cross-sectional
SEM image of the cells. c J-V
curves of the hole conductor–free
and CuSCN-based cells of
varying thicknesses under
simulated AM 1.5 conditions. d
Efficiency distribution of 16 cells
constructed under optimized
conditions in this work

from the perovskite layer to CuSCN compared with the TiO2
electron transport layer using photoluminescence measure-
ments and we believe the small difference between the va-
lence band minima favors the efficient charge extraction be-
tween adjacent layers [36].
CuSCN employed as a HTM layer of perovskite solar cells
was characterized by a solar simulator using AM 1.5 simulat-
ed light, and J-V curves were obtained to compare layer thick-
ness on the performance of the devices. Stock solution of
CuSCN in diethylsulfide was used to deposit CuSCN layers
on top of a perovskite layer to be employed as a hole conduc-
tor layer. Structure consists of a blocking layer (BL) and a
mesoporous TiO2 layer for electron injection, a perovskite
layer embedded inside the mesoporous layer and as a separate
capping layer. CuSCN films were deposited on top of perov-
skite layer forming a mesoporous n-i-p architecture solar de-
vice (Fig. 5a).
For the fabrication of n-i-p-configurated cells, a stock so-
lution of CuSCN should be spinned directly on the perovskite
layer. However, this treatment provides a restriction on the
process because many solvents used to dissolve CuSCN also
dissolve the perovskite crystals and degrade the perovskite
layer. In addition, any solvents which require high-
temperature treatment for removal of them, also cannot be
used because the perovskite layer again limits high-
temperature heat treatment as it starts to degrade at tempera-
tures above 150 °C and residual solvent molecules inhibit
charge transfer between particles. From SEM image of the
cross section given in Fig. 5b, individual layers can be seen
and there is no evidence of physical damage related to perov-
skite degradation due to the use of diethylsulfide solution
employed in this study for the construction of CuSCN includ-
ed cells. Completed cells after Au evaporation were character-
ized under AM 1.5 conditions, and parameters of the cells are

Table 1 Performances of the cells

produced using different HTM Cell configuration Efficiency (%) Voc (V) Jsc (mA) FF
(CuSCN) layer thicknesses
50 nm BL + 250 nm mp TiO2 + HTM free 3.18 0.61 17.82 0.29
50 nm BL + 250 nm mp TiO2 + 250 nm CuSCN 6.89 0.75 17.96 0.51
50 nm BL + 250 nm mp TiO2 + 500 nm CuSCN 8.44 0.81 19.61 0.53
50 nm BL + 250 nm mp TiO2 + 750 nm CuSCN 11.02 0.83 22.15 0.60
J Solid State Electrochem

Fig. 6 a Nyquist plots of the CuSCN layer included perovskite solar cell transport resistances of two differently configurated cells upon increasing
structure under 0 V and 0.5 V applied bias. Equivalent circuit model used applied voltage. d Recombination resistances of two differently
to fit impedance data is also given in the inlet image. b Nyquist plots of configurated cells upon increasing applied voltage
hole conductor–free devices under 0 V and 0.5 V applied bias. c Charge

shown in Fig. 5c and tabulated in Table 1. Cells without a hole
conductor layer yield an efficiency value of 3.18% due to p-
type nature of the perovskite layer itself which can hold a
depletion layer together with an n-type mesoporous scaffold
layer and possess hole conducting ability as evidenced by
former studies [40, 41]. Addition of the CuSCN layer en-
hances both open-circuit voltage and short-circuit current si-
multaneously; however, fill factors highly depend on thick-
ness and homogeneity of the CuSCN layer. Thinner CuSCN
layers cause surface coverage problem. Otherwise, inhomoge-
neous deposited CuSCN layer occurred when the thickness
reached 1 μm. Therefore, thickness of CuSCN layer was reg-
ulated using the concentration of CuSCN particles in
diethylsulfide solution. Thinner CuSCN layer also causes re-
sistance for hole conduction revealing itself as an S-shaped J-
V characteristic which is mostly related to a problem of inter-
facial charge transport or excessive resistance [42]. Thickness
of the 750-nm CuSCN layer as a hole conductor in the cell
yielded the best results obtained in this work with an open-
circuit voltage of 0.83 V which outperforms hole conductor–
free cell. Fill factor rose up to 0.60, yielding an overall effi-
ciency of 11.02% for the best cell measured at 500 mV/s under
forward scan (from Jsc to Voc) conditions. A typical efficiency
distribution of 16 cells under optimized cell configuration of
50-nm blocking layer thickness, 250-nm mp-TiO2 mesopo-
rous layer thickness, and 750-nm CuSCN layer thickness

Table 2 Parameters for the fitting

procedure of the impedance Rseries RCT RREC nCT QCT nREC QREC
spectrum given in Fig. 6a and b
With CuSCN (0 V) 18 3652 10,120 0.72 2.93–7 0.74 2.72–6
With CuSCN (500 mV) 14 3571 12,243 0.66 5.72–7 0.65 9.2–6
Without CuSCN (0 V) 23 8116 50,668 0.92 1.39–8 0.64 9.6–6
Without CuSCN (500 mV) 16 6389 3621 0.89 1.61–8 0.78 2.04–5
 J Solid State Electrochem

Fig. 7 a J-V curves of CuSCN-based cells with different blocking layer thicknesses under simulated AM 1.5 conditions. b Charge transport resistances
of the cells having different blocking layer thicknesses at 0.4 V applied voltage

obtained in this work is given in Fig. 5d, with a mean efficien-
cy value of 10.62% and best cell efficiency of 11.02%. Further
decrease in the concentration of CuSCN for the same spin
rates causes inhomogeneous distribution of CuSCN particles,
and it is not possible to deposit continuous films. The strong
dependence of fill factor on CuSCN thickness and corre-
sponding efficiency of the cells can be explained by partial
agglomeration of the film which was highly dependent on the
concentration of CuSCN particles. A similar strong depen-
dence of layer thickness on the efficiency was also observed
by Arora et al., and they report an optimum concentration of
35-mg/ml solution concentration for deposition of CuSCN
films, which yields strongly reduced performances for both
lower and higher concentrations [36].
Recombination resistance and charge transfer resis-
tance are the most critical parameters of perovskite solar
cells. In order to determine the role of the CuSCN layer
on device operation in depth, both hole conductor–free
and CuSCN included cells were qualified using imped-
ance spectroscopy (IS). IS is a strong technique to explain
recombination dynamics and charge transport phenomena
in solar devices and has been used for characterization of
perovskite solar cells. In this study, impedance spectra of
the devices were registered under constant illumination of
60 mW/cm2 white LED light applied in different bias
voltages. There are two distinct arcs detected in Nyquist
plots for both structure of the cells at 0 mV and 500 mV
applied bias voltages, which are given in Fig. 6a and b.
There are no transmission lines or Gerischer feature de-
tected in mesoporous solar cells because of limited thick-
ness of TiO2 mesoporous layer of 250 nm used in this
work [43]. The impedance spectra were fitted by a simple
equivalent circuit model given in the inlet of Fig. 6a. The
impedance spectroscopic analyses of perovskite-based so-
lar cells have been studied extensively for planar or meso-
porous structure–based cells. The physical electronic pro-
cesses behind the IS response of the cells are attributed to
charge transfer, charge accumulation, or ionic motions
inside the cell and between the interfaces, and interpreta-
tion of each process is crucial. In general, high-frequency
responses in IS spectra are attributed to fast electronic
processes, where slow electronic processes like recombi-
nation events or ionic motions inside the cell lead to low-
frequency responses, with corresponding resistances and
capacitances. The RSERIES term is related to overall series
resistances of the cell due to conducting glass substrate
and wiring elements during measurements and contains
no capacitive component. Interfacial electronic transfer
events result in a resistance to electronic transport inside
the cell and appear at the high-frequency region (RCT).
This resistance gives rise to a charge accumulation be-
tween adjacent layers: the TiO2 layer, perovskite layer,
and CuSCN layer, accumulatively. Due to this accumula-
tion, the resistive element also possesses a capacitive ele-
ment (CPECT) and can be extracted from the first arc. The
second arc is related to relatively slow processes of

Table 3 Performances of the cells

produced using different blocking Cell configuration Efficiency (%) Voc (V) Jsc (mA) FF
layer thicknesses
25 nm BL + 250 nm mpTiO2 + 750 nm CuSCN 7.47 0.87 17.95 0.47
50 nm BL + 250 nm mpTiO2 + 750 nm CuSCN 11.02 0.83 22.15 0.60
75 nm BL + 250 nm mpTiO2 + 750 nm CuSCN 6.02 0.85 14.41 0.49
100 nm BL + 250 nm mpTiO2 + 750 nm CuSCN 7.68 0.83 19.56 0.47
J Solid State Electrochem

Fig. 8 Cross-sectional SEM images of cells produced using a 250-nm, b 500-nm, and c 750-nm mp-TiO2 layer

recombination events inside the perovskite layer or be-
tween perovskite layer and nearby TiO2 or CuSCN hole
conductor layer. The ability of the cell to resist recombi-
nation leads to recombination resistance (RREC) and the
capacitive behavior (CPEREC) due to charge separation,
which leads to the second arc. There is no third arc which
can be related to ionic motion or migration inside the cell.
The first arc emerging in high-frequency region is associ-
ated to charge transfer resistances (RCT) and related ca-
pacitances in the selective contacts (CuSCN layer and
TiO2 blocking and mesoporous layers) [25, 44]. For a
better fitting, capacitances are represented using a con-
stant phase element (CPECT) because both selective con-
tacts may not fully cover the cell and contain pin holes.
The low-frequency arc is associated with recombination
resistance (RREC) and capacitance (CPEREC) of the cell.
Charge transfer resistances of the cells at applied bias
voltages are shown in Fig. 6c. The series resistance,
charge transfer resistances, and recombination resistance
values are given in Table 2, together with the fitting pa-
rameters of the CPE. Here, the CPE factor is represented
by Q and n represents the exponent of the CPE factor,
according to previously established procedures [45].
Although it is possible to extract true capacitance from
the Q value for n values ranging between 0.9 and 1, the
exponent factor lies between 0.64 and 0.78 in this work.
The deviation of the n values from unity is a result of
non-ideal capacitive behavior and can be attributed to in-
homogeneities of individual layers of the cell, and the
interfaces are non-uniform.
As long as mesoporous and blocking TiO2 layers are the
same for both cell structures, charge transfer resistances are
explained by hole transfer medium. Addition of CuSCN layer
as a hole transport material introduces a net decrease in the
charge transfer resistance of the cells, almost 50% lower than
hole conductor–free cell. Hopping mechanism is the dominant
carrier transport pathway which is the main factor for electri-
cal resistance of inorganic p-type HTM included devices con-
trary to polymeric dense hole conductors. However, authentic
contribution of the CuSCN layer evinces itself as an increased
recombination resistance given in Fig. 6d. At 300 to 500 mV
applied voltages, recombination resistance of the CuSCN-
based cells is higher than the hole conductor–free devices.
Actually at 0 V, hole conductor–free cell has a higher recom-
bination resistance competing with CuSCN-based cell be-
cause of no current collection at low applied voltages; how-
ever, as voltage increases, a sharp drop in the recombination
resistance occurs for hole conductor–free cell. Meanwhile,

Fig. 9 a J-V curves of CuSCN-based cells with different mesoporous layer thicknesses under simulated AM 1.5 conditions. b Charge transport
resistances and recombination resistances of cells with different mesoporous layer thicknesses at 0.4 V applied voltage

Table 4 Performances of the cells
produced using different mp-TiO2 Cell configuration
layer thicknesses
50 nm BL + mp TiO2 free + 750 nm CuSCN
50 nm BL + 250 nm mpTiO2 + 750 nm CuSCN
50 nm BL + 500 nm mpTiO2 + 750 nm CuSCN
50 nm BL + 750 nm mpTiO2 + 750 nm CuSCN
CuSCN-based cells can resist recombination up to 0.7 V. This
ability of resistance to recombination mostly arises from low
electron affinity of CuSCN which makes it a suitable hole
blocking material for perovskite solar cells. Reduced recom-
bination in CuSCN-based cells is the major advantage and
leads to the enhancement in Voc, fill factor, and stability com-
pared with hole conductor–free devices.
The thicknesses of other layers such as mesoporous (mp)
and compact blocking TiO2 layers are also studied using the
same configuration. Compact TiO 2 layer produced by
sputtering acts as a hole blocking layer in the cell. J-V curves
of cells with different blocking layer thicknesses are given in
Fig. 7a, and cell performances are also tabulated in Table 3.
The best cell efficiency of 11.02% was obtained by deposition
of 50 nm BL due to its highest Jsc value where Voc values of
the cells with different BL thicknesses remain almost the
same.
Charge transfer resistances calculated from IS study at
0.4 V given in Fig. 7b are also consistent with cell perfor-
mances. The lowest charge transfer resistance was reached
by the construction of cell using 50 nm BL which provides
highest Jsc value. Increasing BL thicknesses more than
50 nm, the series resistance was added to the cell and the
charge transfer resistance is observed to increase. Fifty nano-
meters is the optimized thickness of this layer, and more or
less than this value of thickness gives undesirable effects in
the cell as explained by the IS study.
mp-TiO2 layers with different thicknesses were investigat-
ed using SEM and their images are given in Fig. 8. mp-TiO2
layer not only affects charge transfer resistance, but also influ-
ences recombination resistance of the cells. Due to lowest
charge transfer resistance with highest recombination resis-
tance that was calculated from the IS study, 250 nm of meso-
porous layer was optimized and gives the best performance
with an enhancement in Voc and Jsc values.
Enhancement in Jsc shows a decrease in charge transfer
resistance and an increase in Voc which is related to enhance-
ment in recombination resistances of the cells. J-V curves
which are given in Fig. 9a and cell performances tabulated
in Table 4 are consistent with the IS study. mp-layer-free cell
was not resistant to recombination and this is the main reason
of decrease in Voc. Although Voc values were nearly identical
with 250- and 500-nm mp-layer included cells, enhancement
in Jsc value was acquired using 250-nm mp-layer due to lower
charge transfer resistance. The high efficiency of the cell
J Solid State Electrochem
Efficiency (%) Voc (V) Jsc (mA) FF
5.65 0.71 17.84 0.44
11.02 0.83 22.15 0.60
7.52 0.89 18.82 0.45
4.25 0.75 15.00 0.38
produced using 250-nm-thick mp-layer depends on not only
low-charge transfer resistance, but also high recombination
resistance which is shown in Fig. 9b. The optimized value of
250-nm TiO2 mp-layer thickness is also consistent with the
former studies for highest efficiency value [36]; however, we
have shown the dependence and effect of mp-layer thickness
on the cell performance directly, while this information is
missing in the previous studies. It is obvious that for
CuSCN-based cells, careful optimization of the thickness
and morphology of the individual layers are essential to obtain
high performance from the CuSCN-based devices. A bal-
anced electron and hole extraction from the absorber perov-
skite layer is the key factor for maximized charge collection
and efficiency.
Conclusions
Diethylsulfide was used to solve CuSCN powder, and solution
deposition method was employed to produce CuSCN layer
without degrading the perovskite layer. Fabricated n-i-p-
configurated mesoporous perovskite solar cells using
CuSCN hole-transporting layer and CuSCN free were charac-
terized by recording J-V curves under illumination.
Efficiencies and fill factors of the cells were calculated using
J-V curves. CuSCN included cells were obtained as drastically
efficient compared with hole conductor–free cells. The cham-
pion cell obtained in this study with thick and dense CuSCN
layer reached 11.02% efficiency and 0.60 fill factor value. The
effects of mesoporous and dense blocking TiO2 layer thick-
nesses in cell performances were also studied, and results were
evaluated by charge transfer kinetics. Although there was no
change in potential of cells with the thickness of blocking
layer, which is dense and compact TiO2, the current flow in
the cell was different due to electron transportation. However,
the thickness of mesoporous TiO2 layer affects not only the
current, but also the potential of the cells. Indeed, mesoporous
layer is expected to prevent recombinations with efficient
charge transfer in perovskites. Results derived from J-V mea-
surements were consistent with the IS analysis. IS analysis
was also used to clarify recombination and charge transfer
resistances of the cells. In conclusion, this simple deposition
method of the CuSCN layer on perovskite allows to establish
n-i-p perovskite solar cells with CuSCN or other inorganic
J Solid State Electrochem
metal oxide materials as a hole-transporting layer, thus replac-
ing expensive and long-term instable polymeric structures.
Acknowledgments We would like to thank Ilker Yildiz from the Central
Laboratory of Middle East Technical University for his assistance through
XPS/UPS work. Sputtering of TiO2 blocking layer and e-beam evapora-
tion of gold contacts were carried out at NANOTAM, Bilkent University.
FE-SEM, XRD, and impedance spectroscopy measurements were carried
out at the laboratories of Metallurgical and Materials Engineering
Department of METU.
Funding information This work was supported by Research Fund of the
Middle East Technical University, Project Number: BAP-07-02-2017-
004-234.
References
1. Nie W, Tsai H, Asadpour R, Blancon J-C, Neukirch AJ, Gupta G,
Crochet JJ, Chhowalla M, Tretiak S, Alam MA, Wang HL, Mohite
AD (2015) High-efficiency solution-processed perovskite solar
cells with millimeter-scale grains. Science 347(6221):522–525
2. Ergen O, Gilbert SM, Pham T, Turner SJ, Tan MTZ, Worsley MA,
Zettl A (2016) Graded bandgap perovskite solar cells. Nat Mater
16:522–525
3. Jeon NJ, Noh JH, Kim YC, Yang WS, Ryu S, Seok SI (2014)
Solvent engineering for high-performance inorganic–organic hy-
brid perovskite solar cells. Nat Mater 13(9):897–903
4. Xiao M, Huang F, Huang W, Dkhissi Y, Zhu Y, Etheridge J, Gray-
Weale A, Bach U, Cheng Y-B, Spiccia L (2014) A fast deposition-
crystallization procedure for highly efficient lead iodide perovskite
thin-film solar cells. Angew Chem Int Ed 53(37):9898–9903
5. Zuo C, Ding L (2014) An 80.11% FF record achieved for perov-
skite solar cells by using the NH4Cl additive. Nanoscale 6:9935
6. Liang P-W, Liao C-Y, Chueh C-C, Zuo F, Williams ST, Xin X-K,
Lin J, Jen AK-Y (2014) Additive enhanced crystallization of
solution-processed perovskite for highly efficient planar-
heterojunction solar cells. Adv Mater 26(22):3748–3754
7. Im J-H, Jang I-H, Pellet N, Grätzel M, Park N-G (2014) Growth of
CH3NH3PbI3 cuboids with controlled size for high-efficiency pe-
rovskite solar cells. Nat Nanotechnol 9(11):927–932
8. Yang WS, Noh JH, Jeon NJ, Kim YC, Ryu S, Seo J, Seok SI (2015)
High-performance photovoltaic perovskite layers fabricated
through intramolecular exchange. Science 348(6240):1234–1237
9. Li X, Yang J, Jiang Q, Chu W, Zhang D, Zhou Z, Ren Y, Xin J
(2017) Enhanced photovoltaic performance and stability in mixed-
cation perovskite solar cells via compositional modulation.
Electrochim Acta 247:460–467
10. Zhang W, Pathak S, Sakai N, Stergiopoulos T, Nayak PK, Noel NK,
Haghighirad AA, Burlakov VM, deQuilettes DW, Sadhanala A, Li
W, Wang L, Ginger DS, Friend RH, Snaith HJ (2015) Enhanced
optoelectronic quality of perovskite thin films with
hypophosphorous acid for planar heterojunction solar cells. Nat
Commun 6:10030
11. Liu W, Ma Z, Wang S, Jiang J, Yuan N, Ding J (2018) Low-
temperature bromide modification of SnO2 for highly efficient pe-
rovskite solar cells. J Solid State Electrochem 22:3751
12. Bella F, Renzi P, Cavallo C, Gerbaldi C (2018) Caesium for perov-
skite solar cells: an overview. Chem Eur J 24:12183–12205
13. Pydzińska-Białek K, Szeremeta J, Wojciechowski K, Ziółek M
(2018) Insights into the femtosecond to nanosecond charge carrier
kinetics in perovskite materials for solar cells. J Phys Chem C 123:
110–119
14. Abate A, Correa-Baena J-P, Saliba M, Su’ait MS, Bella F (2017)
Perovskite solar cells: from the laboratory to the assembly line.
Chem Eur J 24:3083–3100
15. Bae S, Lee SY, Kim J, Umh HN, Jeong J, Bae S, Yi J, Kim Y, Choi J
(2019) Hazard potential of perovskite solar cell technology for po-
tential implementation of “safe-by-design” approach. Sci Rep 9:
4242
16. Ananthajothi P, Venkatachalam P (2016) Tin chloride perovskite-
sensitized core/shell photoanode solar cell with spiro-MeOTAD
hole transport material for enhanced solar light harvesting. J Solid
State Electrochem 20:2633
17. Huan TN, Corte DA, Lamaison S, Karapinar D, Lutz L, Menguy N,
Foldyna M, Turren-Cruz SH, Hagfeldt A, Bella F, Fontecave M,
Mougel V (2019) Low-cost high-efficiency system for solar-driven
conversion of CO2 to hydrocarbons. Proc Natl Acad Sci 116:9735–
9740
18. Hu Q, Wu J, Jiang C, Liu T, Que X, Zhu R, Gonq Q (2014)
Engineering of electron-selective contact for perovskite solar cells
with efficiency exceeding 15%. ACS Nano 8(10):10161–10167
19. Wang B, Zhang Z-G, Ye S, Gao L, Yan T, Bian Z, Huang C, Li Y
(2016) Solution-processable cathode buffer layer for high-
performance ITO/CuSCN-based planar heterojunction perovskite
solar cell. Electrochim Acta 218:263–270
20. Subbiah AS, Halder A, Ghosh S, Mahuli N, Hodes G, Sarkar SK
(2014) Inorganic hole conducting layers for perovskite-based solar
cells. J Phys Chem Lett 5:1748–1753
21. Heo JH, Im SH, Noh JH, Mandal TN, Lim C-S, Chang JA, Lee YH,
Kim H-J, Sarkar A, Nazeeruddin MK, Grätzel M, Seok SI (2013)
Efficient inorganic–organic hybrid heterojunction solar cells con-
taining perovskite compound and polymeric hole conductors. Nat
Photonics 7:486–491
22. Liu S, Liu R, Chen Y, Ho S, Kim JH, So F (2014) Nickel oxide hole
injection/transport layers for efficient solution-processed organic
light-emitting diodes. Chem Mater 26:4528–4534
23. Ryu MS, Jang J (2011) Enhanced efficiency of organic photovoltaic
cells using solution-processed metal oxide as an anode buffer layer.
Sol Energy Mater Sol Cells 95:3015–3020
24. Chen W, Wu Y, Yue Y, Liu J, Zhang W, Yang X, Chen H, Bi E,
Ashraful I, Grätzel M, Han L (2015) Efficient and stable large-area
perovskite solar cells with inorganic charge extraction layers.
Science 350(6263):944–948
25. Christians JA, Fung RCM, Kamat PV (2014) An inorganic hole
conductor for organo-lead halide perovskite solar cells improved
hole conductivity with copper iodide. J Am Chem Soc 136(2):
758–764
26. Chung I, Lee B, He J, Chang RPH, Kanatzidis MG (2012) All-
solid-state dye-sensitized solar cells with high efficiency. Nature
485(7399):486–489
27. Qin P, Tanaka S, Ito S, Tetreault N, Manabe K, Nishino H,
Nazeeruddin MK, Grätzel M (2014) Inorganic hole conductor-
based lead halide perovskite solar cells with 12.4% conversion
efficiency. Nat Commun 5:3834
28. Jaffe JE, Kaspar TJ, Droubay TC, Varga T, Bowden ME, Exarhos
GJ (2010) Electronic and defect structures of CuSCN. J Phys Chem
C 114:9111–9117
29. Ye S, Sun W, Li Y, Yan W, Peng H, Bian Z, Liu Z, Huang C (2015)
CuSCN-based inverted planar perovskite solar cell with an average
PCE of 15.6%. Nano Lett 15(6):3723–3728
30. Jung JW, Chueh CC, Jen AKY (2015) A low-temperature, solution-
processable, Cu-doped nickel oxide hole-transporting layer via the
combustion method for high-performance thin-film perovskite solar
cells. Adv Energy Mater 5:7874–7880
31. Jung M, Kim YC, Jeon NJ, Yang WS, Seo J, Noh JH, Il Seok S
(2016) Thermal stability of CuSCN hole conductor-based perov-
skite solar cells. ChemSusChem 9(18):2592–2596

32. Liu J, Pathak SK, Sabai N, Sheng R, Bai S, Wang Z, Snaith HJ
(2016) Identification and mitigation of a critical interfacial instabil-
ity in perovskite solar cells employing copper thiocyanate hole-
transporter. Adv Mater Interfaces 3:1600571
33. Ye S, Rao H, Yan W, Li Y, Sun W, Peng H, Liu Z, Bian Z, Li Y,
Huang C (2016) A strategy to simplify the preparation process of
perovskite solar cells by co-deposition of a hole-conductor and a
perovskite layer. Adv Mater 28(43):9648–9654
34. Madhavan VE, Zimmermann I, Roldan-Carmona C, Grancini G,
Buffiere M, Belaidi A, Nazeeruddin MK (2016) Copper thiocya-
nate inorganic hole-transporting material for high-efficiency perov-
skite solar cells. ACS Energy Lett 1:1112–1117
35. Yaacobi-Gross N, Treat ND, Pattanasattayavong P, Faber H,
Perumal AK, Stingelin N, Bradley DDC, Stavrinou PN, Heeney
M, Anthopoulos TD (2015) High-efficiency organic photovoltaic
cells based on the solution-processable hole transporting interlayer
copper thiocyanate (CuSCN) as a replacement for PEDOT:PSS.
Adv Energy Mater 5:1401529
36. Arora N, Dar MI, Hinderhofer A, Pellet N, Schreiber F,
Zakeeruddin SM, Grätzel M (2017) Perovskite solar cells with
CuSCN hole extraction layers yield stabilized efficiencies greater
than 20%. Science 358(6364):768–771
37. Kim HS, Park N-G (2014) Parameters affecting I−V hysteresis of
CH3NH3PbI3 perovskite solar cells: effects of perovskite crystal
size and mesoporous TiO2 layer. J Phys Chem Lett 5:2927–2934
38. Lee JW, Kim HS, Park N-G (2016) Lewis acid−base adduct ap-
proach for high efficiency perovskite solar cells. Acc Chem Res
49(2):311–319
J Solid State Electrochem
39. Cahen D, Kahn A (2003) Electron energetics at surfaces and inter-
faces: concepts and experiments. Adv Mater 15:271–277
40. Laban WA, Etgar L (2013) Depleted hole conductor-free lead ha-
lide iodide heterojunction solar cells. Energy Environ Sci 6:3249
41. Shi J, Dong J, Lv S, Xu Y, Zhu L, Xiao J, Xu X, Wu H, Li D, Luo Y,
Meng Q (2014) Hole-conductor-free perovskite organic lead iodide
heterojunction thin-film solar cells: high efficiency and junction
property. Appl Phys Lett 104:063901
42. Tress W, Petrich A, Hummert M, Hein M, Leo K, Riede M (2011)
Imbalanced mobilities causing S-shaped IV curves in planar
heterojunction organic solar cells. Appl Phys Lett 98:063301
43. Kim H-S, Mora-Sero I, Gonzalez-Pedro V, Fabregat-Santiago F,
Juarez-Perez EJ, Park N-G, Bisquert J (2013) Mechanism of carrier
accumulation in perovskite thin-absorber solar cells. Nat Commun
4:2242
44. Liu D, Yang J, Kelly TL (2014) Compact layer free perovskite solar
cells with 13.5% efficiency. J Am Chem Soc 136(49):17116–17122
45. Todinova A, Contreras-Bernal L, Salado M, Ahmad S, Morillo N,
Idígoras J, Anta JA (2017) Towards a universal approach for the
analysis of impedance spectra of perovskite solar cells: equivalent
circuits and empirical analysis. ChemElectroChem 4:2891–2901
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.