Journal of Power Sources 514 (2021) 230585

Contents lists available at ScienceDirect

Journal of Power Sources

journal homepage: www.elsevier.com/locate/jpowsour

Ultra-stable all-inorganic silver bismuth sulfide colloidal nanocrystal

photovoltaics using pin type architecture
Jae Taek Oh a, 1, Sung Yong Bae a, 1, Jonghee Yang a, Su Ryong Ha a, Hochan Song a,
Cheong Beom Lee a, Sanchari Shome a, Swarup Biswas c, Hyun-Mo Lee d, You-Hyun Seo e,
Seok-In Na e, Jin-Seong Park d, Whikun Yi a, Seungjin Lee f, Koen Bertens f, Bo Ram Lee g,
Edward H. Sargent f, Hyeok Kim c, *, Younghoon Kim b, **, Hyosung Choi a, ***
a
Department of Chemistry, Research Institute for Convergence of Basic Science, and Institute of Nano Science and Technology, Hanyang University, 222 Wangsimni-ro,
Seongdong-gu, Seoul, 04763, Republic of Korea
b
Department of Chemistry, Kookmin University, Seoul, 02707, Republic of Korea
c
School of Electrical and Computer Engineering, Institute of Information Technology, University of Seoul, 163 Seoulsiripdaero, Dongdaemun-gu, Seoul, 02504, Republic
of Korea
d
Division of Materials Science and Engineering, Hanyang University, Seoul, 04763, Republic of Korea
e
Professional Graduate School of Flexible and Printable Electronics LANL-CBNU Engineering Institute-Korea, Jeonbuk National University, 664-14, Deokjin-dong,
Deokjin-gu, Jeonju-si, Jeollabuk-do, 54896, Republic of Korea
f
Department of Electrical and Computer Engineering, University of Toronto, 10 King’s College Road, Toronto, ON, M5S 3G4, Canada
g
Department of Physics, Pukyong National University, Busan, 608-737, Republic of Korea

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Vulnerable polymer/molybdenum oxide

HTL was successively replaced with NiO
layer in device with pin architecture.
• All-inorganic pin type AgBiS2 device
was first reported.
• Combination of TMAI and MPA ligands
improved the device efficiency .
• All-inorganic pin type AgBiS2 device
showed superior device stability in all
air, heating, and immersing into water
conditions.

A R T I C L E I N F O A B S T R A C T

Keywords: Silver bismuth sulfide (AgBiS2) colloidal nanocrystals (NCs) have emerged as an environmentally friendly light
Silver bismuth sulfide absorber for next-generation photovoltaics. Classical AgBiS2 NC photovoltaics with nip-architecture have been
All-inorganic photovoltaics mandated to use a combination of polymer and molybdenum oxide as a hole transport layer (HTL), which are
Ligand engineering
vulnerable to oxygen, heat and water. In this work, we develop all-inorganic AgBiS2 NC photovoltaics with pin-
Environmental-friendly
Device stability
architecture, serving a nickel oxide (NiO) as HTL. We also employ a cascade-energy-level alignment by

* Corresponding author.
** Corresponding author.
*** Corresponding author.
E-mail addresses: [email protected] (H. Kim), [email protected] (Y. Kim), [email protected] (H. Choi).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.jpowsour.2021.230585
Received 5 June 2021; Received in revised form 18 September 2021; Accepted 28 September 2021
Available online 5 October 2021
0378-7753/© 2021 Elsevier B.V. All rights reserved.
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

introducing 3-mercaptopropionic acid-treated AgBiS2 NC layer, enabling enhanced hole collection with mini­
mized interfacial recombination. As a result, the pin type AgBiS2 NC photovoltaics demonstrate a power con­
version efficiency of 5.59% as well as excellent stability even under extreme conditions such as heat and water
exposures, attributed to superior chemical robustness of the inorganic HTL. This work is the first report on
AgBiS2 NC device with all-inorganic components and achieves the highest device efficiency in pin type AgBiS2
NC photovoltaics.

1. Introduction introduced 3-mercaptopropionic acid (MPA)-treated AgBiS2 NC (MPA-

AgBiS2) layer between the HTL and classical tetramethylammonium
Solution-processed next-generation thin film photovoltaics have iodide (TMAI)-treated AgBiS2 NC (TMAI-AgBiS2) active layer, thus
attracted considerable attention because of various advantages leading to a cascade-energy-level alignment for efficient charge trans­
including solution-processability, mechanical flexibility, and low cost port/collection as well as a minimized charge recombination in device.
for device manufacturing [1–4]. However, commercialization is held The best device exhibited a PCE of 5.59%, the first report and the highest
back by three major issues: device efficiency, environmental impact, and value in pin type AgBiS2 NC device to date, which is comparable to that
device stability. In terms of device efficiency, lead (Pb) halide-based of conventional nip type AgBiS2 NC devices. Remarkably, the pin type
perovskite-, quantum dots-, and organic semiconductors-based photo­ AgBiS2 device showed excellent stability under ambient air and even
voltaics have achieved high power conversion efficiencies (PCEs) of extreme conditions such as heat, and water exposures compared to those
25.2%, 13.3%, and 17.6%, respectively [5–7]. Despite their high effi­ of nip type AgBiS2 devices.
ciencies, perovskite and QD materials lead to harmful effect on the
human body as well as environment due to the presence of toxic Pb [8, 2. Experimental section
9]. Furthermore, enhancing the long-term stability of perovskite and
organic photovoltaics remains a challenge [10,11]. To solve this issue, 2.1. Materials
some endeavors have developed eco-friendly and air-stable light
absorbing semiconductors by replacing Pb with other transition metals All chemicals were purchased commercially and used without any
[12–15]. further purification. Silver acetate, 1-octadecene (ODE, technical
To this end, Konstantatos group has synthesized eco-friendly silver grade), oleic acid (OA, technical grade), hexamethyldisilathiane (HMS),
bismuth sulfide (AgBiS2) colloidal nanocrystals (NCs) as a light Nickel(II) acetate tetrahydrate, TMAI, MPA, 1,2-dichlorobenzene (DCB),
absorbing material and fabricated devices by employing nip type chloroform, methanol, and monoethanolamine were purchased from
structure with a configuration of indium tin oxide (ITO)/electron Sigma Aldrich. Bismuth(III) acetate, toluene, 2-methoxyethanol, and
transport layer (ETL)/AgBiS2 NCs/hole transport layer (HTL)/metal ethanol were obtained from Alfa Aesar. Zinc(II) acetate dihydrate was
electrode. This device showed PCE of 6.31% and retained ~82% of its received from Acros. Potassium hydroxide was purchased from DAE­
maximum PCE after 432 h under ambient conditions [16]. JUNG. Acetone was obtained from JUNEI CHEMICAL. PTB7 was
Though the successful demonstrations of AgBiS2 NCs as an active received from 1-material.
layer by adopting nip type architecture [17–19], however, several key
issues still remain toward its sustainable and eco-friendly realization. 2.2. Synthesis of AgBiS2 NCs
First, the use of polymeric semiconductor poly({4,8-bis[(2-ethylhexyl)
oxy]benzo[1,2-b:4,5-b′ ]dithiophene-2,6-diyl}{3-fluo­ AgBiS2 NCs was prepared with modified synthetic strategy [26]. In
ro-2-[(2-ethylhexyl)carbonyl]thieno[3,4-b]thiophenediyl}) (PTB7) can briefly, the mixture of silver acetate (0.8 mmol), bismuth (III) acetate (1
exclusively render high PCE over 6% in AgBiS2 NC photovoltaics while mmol), OA (6 mL), and ODE (4.5 mL) was degassed at 100 ◦ C to remove
other well-known polymeric HTLs are incompatible; even though other oxygen species and promote the formation of metal oleates. After 4 h,
polymeric HTLs provide more favorable band alignment at the NC/HTL ODE (0.5 mL) solution with HMS (1 mmol) was injected to metal pre­
interface and/or much higher carrier mobility for efficient extraction of cursor solution at 100 ◦ C. After injection of HMS/ODE solution, the
generated photocarriers, the devices employing these counterintuitively reaction vessel was immediately cooled down to room temperature by
exhibited low device efficiencies of 1–4% [16]. Although an elucidation immersing the flask in the ice water bath. Purification process of
on the underlying mechanism of the exclusive compatibility of PTB7 as-synthesized AgBiS2 NCs was performed by centrifugation with
with AgBiS2 NC devices that has not been proved yet, it is evident that toluene and acetone. After purification, the resultant AgBiS2 NCs were
this notably limits future development in device performance. finally dispersed in anhydrous toluene. The concentration of AgBiS2 NCs
Second, the use of PTB7 does not only mandate complicated fabri­ solution was adjusted to 20 mg mL− 1 to fabricate the device.
cation process such as inert condition and expensive cost of the resulting
device, but also provide poor device stability against ambient and humid 2.3. Synthesis of ZnO nanoparticles (NPs)
conditions since the material has to be used with labile molybdenum
oxide (MoO3) film to properly extract photocurrent [20–24]. Afore­ ZnO NPs solution was prepared based on the previously reported
mentioned problems regarding the HTLs used in nip type devices negate method [27]. Zinc(II) acetate dihydrate (14.4 mmol) was dissolved in
superior robustness of AgBiS2 NC film including water-resistivity, which methanol (125 mL) at 60 ◦ C. Potassium hydroxide (KOH, 26.4 mmol)
has partly confirmed in our previous work [25]. Thus, it is urgent to was dissolved in methanol (65 mL) at room temperature. The potassium
develop a cost-effective and air/water-stable HTL without compro­ hydroxide solution was slowly added in 10 min to the Zinc(II) acetate
mising the performance of AgBiS2 NC device, which is a challenging dihydrate solution under vigorous stirring. After addition of KOH solu­
issue to address when adhering the conventional nip type architecture. tion, the mixture was stirred at 60 ◦ C for 2 h, then the heater was turned
In this work, we demonstrate a simple, all-inorganic AgBiS2 NCs off and the mixture was kept stirring for 2 h. Crude ZnO NPs solution was
photovoltaics with pin architecture by employing a nickel oxide (NiO) purified by centrifugation with methanol for two times, and then
layer as an alternative to the existing HTL for nip device. Specifically, we dispersed in methanol and chloroform (1:1 by volume ratio) with con­
realized a simple and pin device architecture of ITO/NiO/AgBiS2 NCs/ centration of 20 mg mL− 1 for the device fabrication.
ZnO/Al. We deployed optical simulation to determine an optimum
thickness of the AgBiS2 NC layer in a pin type device. In addition, we

2
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

2.4. Characterizations charge-carrier lifetime and extraction time were extracted with
following bi-exponential decay function [28,29].
High-resolution transmission electron microscopy (HR-TEM) mea­
VOC or JSC A2 (− t / τ1 ) + A2 exp(− t / τ2 )
surement was performed with a FEI Tecnai G2 F20 S-TWIN TMP. X-ray
diffraction (XRD) measurement was carried out by using a Rigaku Model Where A1 and A2 are pre-factors, t is time, and τ1 and τ2 are decay
SmartLab high-resolution X-ray diffractometer. Ultraviolet–visible time constants.
(UV–Vis) absorption spectra of AgBiS2 NCs and PTB7 films were
measured using a Shimadzu UV-2600 UV–Vis spectrometer. The 2.7. Optical simulation
refractive index and extinction coefficient of NiO, AgBiS2 NC, ZnO films
were acquired with spectroscopic ellipsometry (SE, Ellipso technology). The 2-dimensional optical simulation of the solar cell (Fig. 1c) was
Thickness profiles were collected with atomic force microscopy (AFM), done by using Lumerical finite-difference time-domain (FDTD) solutions
Park Systems XE-100, in noncontact mode. ultraviolet photoelectron software. Due to the planar structure of the soalr cell, 2-dimensional
spectroscopy (UPS) measurement of AgBiS2 NC films was conducted FDTD was sufficient for simulating the light absorption in the solar
using a Thermo Scientific ESCALAB 250Xi analyzer. Cross-sectional cell [30]. define the optical property of different layers, experimentally
image of the pin type AgBiS2 NC device was obtained from scanning measured wavelength-dependent complex refractive indices values
electron microscopy (SEM; FEI Verios G4 UC). The X-ray photoelectron (Fig. S4) of each layer were incorporated in the software. Afterwards, we
spectroscopy (XPS) measurements were conducted by using an XPS in­ optimized the mesh density to minimize the error in the simulated re­
strument (Thetaprobe, Thermo Fisher). All spectra were calibrated with sults (mesh accuracy:3; min mesh step: 0.25 nm; dt stability factor: 0.99;
the C 1s peak at 284.8 eV. override x mesh: 0.0001; override y mesh: 0.0001; override z mesh:
0.0001). Then, we fixed periodic boundaries along the perpendicular
2.5. Device fabrications direction (i.e. x-axis) of the device axis (y-axis). Subsequently, we
applied two (below the bottom electrode and above the top electrode)
Patterned ITO glass substrates were sequentially sonicated with perfectly matched layers (PML) along with the solar cell axis confining
soapy water, deionized (DI) water, acetone, and isopropyl alcohol for 10 the simulation region. During the simulation, the software proceeds the
min and dried in an oven at 80 ◦ C. The nip type AgBiS2 NC device was calculation on the distribution of normalized electric field within the
identically fabricated according to previous reported method without device geometry. This was further utilized to calculate the amounts of
any additional modification [26]. In case of pin type AgBiS2 NC device, a absorbed photons and ideal values of short-circuit current density (JSC,
NiO precursor solution was prepared by dissolving Nickel(II) acetate ideal) in the solar cell considering 100% internal quantum efficiency
tetrahydrate (99.5 mg) in ethanol (4 mL) and ethanolamine (24.2 μL) (IQE) as this is an optical simulation and it cannot reproduce the
with stirring and heating the mixture at 70 ◦ C for 4 h. The resultant recombination effect [31]. During the simulation, we obtained the im­
emerald NiO precursor solution was then spin-coated at 4000 rpm for pulse response of the system by using continuous-wave (CW) normalized
40 s onto the cleaned ITO substrates and annealed at 275 ◦ C for 45 min source (detailed explanation for CW-normalization used in Lumerical,
to form the NiO thin film. AgBiS2 NCs solution (20 mg mL− 1 in toluene) FDTD solutions software is provided in previous report). For simulating
was spin-casted onto the NiO thin film at 2000 rpm for 10 s, and then JSC, ideal of the device, the source spectrum (AM 1.5G) was multiplied
ligand-exchanged with MPA and TMAI solution (0.1 v/v% and 1 mg with the impulse response. The wavelength range of the source light was
mL− 1 in methanol, respectively) for 20 s 2 times and washed with fixed within 300–900 nm. A further detail on the simulation procedure is
methanol. This coating process was repeated to obtain the desired described in our previous report [31,32]. As we are interested with the
AgBiS2 NC layer thickness. Next, ZnO NPs solution (20 mg mL− 1 in effect of active layer (AgBiS2) thickness on to different device perfor­
methanol/chloroform) was spin-coated onto the AgBiS2 NC layer at mance parameters (JSC,ideal, absorption spectra, number of absorbed
3000 rpm for 30 s. The resultant films were thermally-annealed at photons, electric field intensity distribution within the device), during
100 ◦ C for 10 min under ambient conditions. Finally, 120 nm thickness the simulation study we varied the thickness of the active layer from 10
of metal top electrode was deposited by thermal evaporation method. nm to 100 nm with an incrementing step size of 10 nm. Thickness of
For the water-resistant test, we used a gold electrode to exclude the other layers was considered as ITO: 150 nm, NiO: 10 nm, ZnO: 40 nm,
effect of Al oxidation on device performance instead of Al top electrode. and Al: 120 nm.
The Thermal annealing (TA) process (275 ◦ C for 45 min) was performed
before deposition of the metal top electrode (Fig. S2 and Table S1). 3. Results and discussion

2.6. Device characterization 3.1. Introduction of inorganic HTL to nip type AgBiS2 NC device

Current density− voltage (J− V) characteristics were measured using AgBiS2 NCs were prepared with our modified synthetic method [26].
a Keithley 2400 sourcemeter and an Air Mass 1.5 Global (AM1.5G) solar Particle size and crystal structure of AgBiS2 NCs were well matched with
simulator (Oriel Sol 2A Class AA) with an illumination intensity of 100 previous reports (Fig. S1) [26,33]. As an alternative to a PTB7, we first
mW cm− 2. The device area of solar cell is 0.055 cm2. A silicon reference tried to introduce a nickel oxide (NiO) film on to the AgBiS2 NC layer as
cell (VLSI standards, Oriel P/N 91150 V) was used to calibrate the an HTL because of its excellent p-type carrier transport properties as
illumination intensity. External quantum efficiency (EQE) spectra were well as energy levels, solution processability, and ambient stability [34,
estimated with a power source (Abet Technologies 150 W Xenon lamp, 35]. However, a high temperature TA process (>275 ◦ C) is required to
13014) containing a monochromator (DongWoo Optron, MonoRa500i) exclude organic residues used in the NiO precursor solution and achieve
and a CompactStat (Ivium Technologies; Eindhoven, The Netherlands). a crystalline NiO film, which can lead to deformation of the underlying
Mott− Schottky plots were acquired from capacitance− voltage (C–V) NCs [36,37]. The nip type AgBiS2 NC devices employing either NiO or
measurement using an impedance analyzer (CompactStat, Ivium Tech­ PTB7/MoO3 HTL were fabricated and the influence of TA treatment on
nologies B.V.) with modulation frequency of 1 kHz and amplitude of 50 the photovoltaic characteristics of the devices were characterized, as
mV under dark condition. The voltage sweep was performed between summarized in Fig. S2 and Table S1. The results revealed that, regardless
− 1 and 1 V with an AC signal. Transient photovoltage/photocurrent of HTL type, the devices upon TA treatment were not functional,
(TPV/TPC) measurements were carried out with a multi-functional attributed to the deformation of AgBiS2 NC (and PTB7) films evidenced
organic semiconductor parmeter system (T400, Mcscience) containing from absorption evolutions (Fig. S3). Based on these results, we
a Tektronix oscilloscope (DPO-2014B) with a 100 MHz bandwidth. The concluded that the introduction of a device-grade NiO HTL for nip type

3
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

Fig. 1. (a) Ideal JSC (JSC, ideal) as a function of AgBiS2 NC layer thickness in pin type device. (b) Absorption spectra and integral of absorbed photon flux of the AgBiS2
NC layer with different thickness (10 nm, 35, nm, and 60 nm). (c) Normalized electric field intensity distribution (E2) inside the pin type device with 35 nm AgBiS2
NC layer. (d) Device architecture of pin type AgBiS2 NC device. (e) J− V curves and (f) JSC and PCE dependence on the number of TMAI-AgBiS2 NC layers of pin type
AgBiS2 NC devices.

AgBiS2 NC device is not applicable, necessitating the pin type archi­ (Fig. 1b). For the AgBiS2 NC device with 35 nm thickness, the peak value
tecture for the use of high performance NiO as an HTL without inflicting of oscillating E2 was centered and much higher than those of the devices
damages to the other layers. with a AgBiS2 NC layer thickness of 10 and 60 nm (Figs. S5a and b).
These optical simulation results suggest that the optimum thickness of
the AgBiS2 NC layer in pin type device is ~35 nm.
3.2. Optical simulation

As the fine-tuning of the layer thickness consisting of the device is 3.3. Device fabrication with pin type architecture
also important to realize high efficiency of the thin AgBiS2 NC photo­
voltaics due to the interfacial reflection and/or interference of incident Based on optimum AgBiS2 thickness estimated from optical simula­
light that competing with absorption for photocarrier generation, opti­ tions, we fabricated pin type AgBiS2 NC devices with a configuration of
cal simulations were conducted. The optical simulations based on op­ ITO/NiO/AgBiS2 NCs/ZnO/Al (Fig. 1d) via solution processing. We
tical constants, refractive index (n) and extinction coefficient (k) of each employed a layer-by-layer (LBL) deposition method with TMAI ligand to
component provide valuable information for thickness optimization of precisely control the thickness of AgBiS2 NC film on the NiO layer. The
pin type AgBiS2 NC device. Ideal values of short-circuit current density detailed device fabrication process is described in the experimental
(JSC) of a pin type AgBiS2 NC device (JSC, ideal) were calculated based on section. Fig. 1e exhibits J− V characteristics of the device as a function of
n and k values of each layer (NiO, AgBiS2, and ZnO) that are measured TMAI-AgBiS2 layers. While the device with 3 TMAI-AgBiS2 layers
by spectroscopic ellipsometry (Fig. S4). Fig. 1a presents the JSC, ideal as a showed highest PCE of 4.74% with open-circuit voltage (VOC) of 0.46 V,
function of AgBiS2 NC thickness. In the range of 10–40 nm, increasing JSC of 17.53 mA cm− 2, and fill-factor (FF) of 0.59, the PCE of the device
the thickness of AgBiS2 NC layer led to rapid increase in JSC, ideal and with 4 layers began decrease. Fig. 1f clearly exhibits the dependence of
reached a peak value of 20.7 mA cm− 2 at 35 nm. Above this range, this JSC and PCE on the number of TMAI-AgBiS2 layers. The variation of PCEs
curve gradually decreased until 60 nm due to optical interference, in exclusively attributed to JSC differences rather than VOC and FF. The
good agreement with a previous report [16]. Fig. 1b exhibits the simu­ detailed photovoltaic parameters of the devices with different TMAI-
lated light absorption and integrated number of absorbed photons in the AgBiS2 layers are summarized in Table 1. The thicknesses of AgBiS2 NC
AgBiS2 NC layer as a function of AgBiS2 thickness. The peak positions of films with different number of layers were measured by using AFM
the absorption spectrum gradually red-shifted by increasing the thick­ (Fig. S6a− e). These results obviously showed that the thickness of the
ness of AgBiS2 NC layer. This behaviour would be attributed to the AgBiS2 NC film linearly increases with increasing number of TMAI-
geometric differences depending on the change of AgBiS2 NC layer AgBiS2 layers (Fig. S6f). The thickness of 3 TMAI-AgBiS2 layers was
thickness. It is worth noting that the AgBiS2 NC layer with 35 nm
thickness achieved the highest number of absorbed photons (2.55, 3.54, Table 1
and 3.25 × 1023 photons for 10, 30, and 60 nm of AgBiS2 NC layer Photovoltaic parameters of pin type AgBiS2 devices depending on the number of
thicknesses, respectively). These results could be correlated with a TMAI-AgBiS2 layers.
change in the distribution of normalized electric field intensity (E2) Number of TMAI-AgBiS2 layers VOC (V) JSC (mA cm− 2) FF PCE (%)
within the device (Fig. 1c and Fig. S5). Fig. 1c exhibits the 1 0.42 8.11 0.52 1.79
wavelength-dependent E2 distribution within the pin type device with 2 0.45 14.46 0.57 3.71
optimized AgBiS2 NC thickness. We utilized three different wavelength 3 0.46 17.53 0.59 4.74
values (365, 510, and 772 nm) of incident light, which were chosen from 4 0.46 16.46 0.58 4.40
5 0.47 15.15 0.59 4.14
the maximum points in the wavelength-dependent absorption spectrum

4
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

35.71 nm (and 34.8 nm from cross-sectional scanning electron micro­ combination of TMAI- and MPA-AgBiS2 layers (denoted as TMAI +
scopy (SEM) in Fig. S7, which is optimum thickness. MPA). The former device showed a PCE of 4.74%, with a VOC of 0.46 V,
for device performance also expected from optical simulation. The JSC of 17.53 mA cm− 2, and FF of 0.59. Interestingly, TMAI + MPA device
SEM image also exhibited the thicknesses of the other layers in the de­ outperformed all parameters compared to those of the TMAI device
vice (NiO: 10 ± 0.7 nm and ZnO: 37.7 ± 1.8 nm), suggesting that (VOC: 0.50 V, JSC: 18.54 mA cm− 2, and FF: 0.61), improved PCE of
reproducible thicknesses, well-matched with the values used in optical 5.59%, a comparable value with that of AgBiS2 NC device with nip type
simulations, could be made by current fabrication protocol. The structure (Fig. S2 and Table S1). Enhanced EQE also suggests that the
measured JSC trend of the pin type AgBiS2 NC devices was also well- insertion of MPA-AgBiS2 layer led to better charge collection in device
matched with simulation results (Fig. S8). (Fig. 2d). The PCE statistics based on 30 devices for each pin type AgBiS2
NC devices further justified a reproducible performance enhancement of
TMAI + MPA devices (Fig. 2e and Fig. S9). From these results, we
3.4. Cascade-energy-level alignment via combination of TMAI and MPA
concluded that the combination of TMAI- and MPA-AgBiS2 layer
ligands
induced cascade-energy-level alignment (Fig. 2c), which is favorable for
charge carrier transport and extraction, thereby leading to further in­
As there is a discrepancy in PCE between the pin type AgBiS2 NC
crease device efficiency of pin type AgBiS2 NC device.
device (4.74%) and our best nip type device (5.58%), we further
introduced a cascade-energy-level alignment strategy to enhance the
efficiency of pin type AgBiS2 NC device. Similar to the reported LBL 3.5. Charge transfer/recombination dynamics
protocols that have been utilized in PbS CQD-based devices, an MPA-
AgBiS2 layer was inserted between the NiO HTL and TMAI-AgBiS2 active To get more insights into the influence of cascade-energy-level
layer [38–40]. To investigate the energy level change of AgBiS2 NC films alignment on the interface in pin type AgBiS2 NC device, Mott‒
by different ligands, we performed UPS measurements. Fig. 2a shows the Schottky analysis was conducted, as shown in Fig. 3a. The results
UPS spectra of TMAI- and MPA-AgBiS2 films. The Fermi level (EF) and showed that TMAI + MPA device exhibited a higher built-in voltage
valence band maximum (VBM) of TMAI-AgBiS2 are − 4.71 eV and − 5.46 (0.51 V) than TMAI counterpart (0.43 V). It could be considered that
eV, respectively. The MPA-AgBiS2 film showed EF of − 4.79 eV and VBM more p-typed MPA-AgBiS2 layer would improve the junction charac­
of − 5.27 eV. These results imply that the ligand exchange with MPA can teristics at NiO/AgBiS2 NC interface, leading to an increase in VOC. Also,
induce relatively p-type character to AgBiS2 NC films, compared to the reduced capacitance in TMAI + MPA device compared to that of
TMAI-AgBiS2 NC. According to these UPS results and the bandgap of TMAI-AgBiS2 device implies that surface defects in the NCs would be
AgBiS2 NCs from a previous report [26], energy band diagrams of TMAI- reduced by MPA treatment, partly contributed to the VOC enhancement.
and MPA-AgBiS2 films along with other components are constructed, as Charge carrier dynamics in devices were further confirmed by
shown in Fig. 2b. The conduction band minimum (CBM) of TMAI- and measuring TPV/TPC analysis. Fig. 3b and c exhibit TPV and TPC profiles
MPA-AgBiS2 films are − 4.16 eV and − 3.97 eV, respectively. Thus, it can of pin type AgBiS2 NC devices with TMAI and TMAI + MPA, respec­
be expected that formation of cascade-energy-level alignment by com­ tively. The device with TMAI + MPA showed much longer charge carrier
bination of TMAI and MPA ligands leads to efficient charge transport lifetime of 970 μs and a faster charge carrier extraction time of 0.18 μs,
and extraction between ZnO and NiO layers. compared to those of the device with TMAI-AgBiS2 exhibiting 458 μs and
To confirm the effect of cascade-energy-level alignment on the de­ 0.96 μs, respectively. These results imply that the cascade-energy-level
vice performance, we fabricated pin type AgBiS2 NC devices with the alignment at HTL/active layer interface by inserting MPA-AgBiS2 layer
MPA-AgBiS2 layer. Fig. 2c presents the J‒V curves of pin type AgBiS2 NC in the device provides larger driving force for charge extraction and
devices that the active layer consists of solely TMAI-AgBiS2 layer and a suppression of charge recombination, leading to efficiency enhancement

Fig. 2. (a) UPS spectra of TMAI- and MPA-AgBiS2 NC films (left: secondary electron cut-off region, and right: valence band edge (b) Energy-level diagram of TMAI-
and MPA-AgBiS2 NC films within pin type device with NiO and ZnO. (c) J− V, (d) EQE curves, and device histograms of the device with TMAI and TMAI + MPA.

5
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

Fig. 3. (a) Mott− Schottky plots, (b) transient photovoltage (TPV) and (c) photocurrent (TPC) decays of pin type AgBiS2 NC devices with TMAI and TMAI + MPA.

of pin type AgBiS2 NC device, ultimately resulting in enhanced device min. Notably, the pin type AgBiS2 NC device maintained its initial PCE
performance in the pin type AgBiS2 NC device. even after water treatment while the nip type AgBiS2 NC device retained
just 13% of its initial PCE after water treatment (Fig. 4c).
3.6. Device stability test of nip and pin type AgBiS2 NC devices Given that the ZnO and AgBiS2 NCs have comparable chemical
robustness upon the harsh conditions as we checked above, these
Since photovoltaics are generally installed in ambient conditions and striking differences in device stability indicate that all-inorganic layers
suffer from elevated heat from device operation [41,42], device stability contribute to the superior water-resistant property of the pin type
is one of the key factors for future commercialization. We speculated AgBiS2 NC device. The marginally changed XPS spectra of the NiO films
that our pin type device with all-inorganic layers may demonstrate in both O 1s and Ni 2p regions also support that the chemically robust
better device stability compared to nip type devices. Therefore, we NiO HTL indeed leads to the decent stability even upon the exposure to
compared stability of the pin and nip type AgBiS2 NC devices under heat or water (Fig. S10). In contrast, although UV–Vis absorption spectra
various conditions: ambient (25 ± 5 ◦ C, relative humidity (RH) of 40 ± of PTB7 films before and after water immersion were nearly similar
5%), heating (80 ◦ C), and water immersion (2 min in deionized water). (Fig. S11a), decreased absorbance was observed according to heating
Schematic illustrations describe the robustness of all-inorganic pin type time (Fig. S11b). Furthermore, the XPS spectra of MoO3 films were
AgBiS2 NC devices against ambient, heat, and water conditions while notably changed upon exposure to extreme conditions. In Mo 3d
nip type AgBiS2 NC devices are more easily deteriorated in all condition photoemission region (Fig. S12a), it was evident that both heating and
due to the labile HTL (Fig. 4a− c). Fig. 4a presents device stability results water immersion increase the portion of Mo5+ species (Table S2) [43,
of both AgBiS2 NC devices with different structures under ambient 44]. Meanwhile, O 1s spectra (Fig. S12b) revealed that water immersion
condition. The device with nip structure showed 25% drops in PCE after induces adsorption of hydroxyl species via the unsaturated Mo sites (i.e.,
exposure to ambient condition for 250 h. In contrast, the pin type AgBiS2 O vacancies) and promote H2O molecule adsorption, while the change
NC device showed remarkable improvement in device stability. The by the thermal stress is negligible [45–47]. The XPS elemental O to Mo
device efficiency of this device retained the 88% of their initial effi­ ratio of the films (Table S3), suggests that (1) the prolonged heating
ciency even after 2500 h. Fig. 4b shows the device stability results under partially reduces the oxidation number of Mo by vanishing oxygen from
heating conditions at 80 ◦ C for 54 h in a nitrogen-filled glove box. While MoO3 lattice and (2) immersion of water does not only promote its
the PCE of the nip type device rapidly dropped to 61% of its initial value adsorption but also facilitates hydroxylation of the film. Both processes
after 6.6 h, the pin type device retained 94% of its initial PCE after were accompanied by the notable change in MoO3 stoichiometry, which
heating. We also performed the water-resistant test of the all-inorganic determines the energy level (particularly the work function) of the film
pin type AgBiS2 NC device by immersing the device in water for 2 [43,44].

Fig. 4. Schematic illustration and comparison of stability between nip type and pin type AgBiS2 NC device under different conditions: (a) ambient exposure (25 ±
5 ◦ C, 40 ± 5% RH), (b) heat (80 ◦ C for 54 h), and (c) immersing in water (2 min, in de-ionized water).

6
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

Indeed, the UPS analysis showed that the energy levels of the films device characterization. Jin-Seong Park: help to perform the device
were altered (Fig. S13). The work function of heated MoO3 was lowered characterization. Whikun Yi: help to perform the device characteriza­
to ~0.4 V than that of the pristine film, attributed to the increased Mo5+ tion. Seungjin Lee: Sargent help to perform data analysis and contribute
species [43,44]. The mismatched energy level subsequently leads to to discussion of the results and discussion. Koen Bertens: Sargent help
unfavorable band alignment, significantly blocking proper to perform data analysis and contribute to discussion of the results and
charge-carrier transport in the device [48]. Note that work function shift discussion. Bo Ram Lee: Sargent help to perform data analysis and
of the water-immersed MoO3 was negligible. Several factors would contribute to discussion of the results and discussion. Edward H. Sar­
concurrently account for this; excessive amounts of water adsorbates as gent: help to perform data analysis and contribute to discussion of the
well as binding of hydroxyl group, as observed from XPS analysis, could results and discussion. Hyeok Kim: Supervision. Younghoon Kim:
provide the complexed result. We also measured XRD patterns of the Supervision. Hyosung Choi: supervised the project.
MoO3 films (Fig. S14), revealing that the amorphous film (evidenced
from a broad peak centered at 25◦ ) is readily dissolved in water so that
results in thinner film. This would increase the contribution of ITO Declaration of competing interest
substrate to the work function of the film, further impeding appropriate
interpretation to assign its exact energy levels. Nevertheless, the results The authors declare that they have no known competing financial
unambiguously suggest the lability of the HTL used in nip type AgBiS2 interests or personal relationships that could have appeared to influence
NC device is responsible for the poor stability of the device while the the work reported in this paper.
chemical robustness of NiO serves a flawless platform for hole collection
in pin type device without the performance drops. Acknowledgements

3.7. Comparison of cost-effectiveness between HTL materials This work was supported by National Research Foundation of Korea
(NRF-2018R1C1B6001015). This work was also funded by National
In addition to the device stability, the cost-effectiveness of device Research Foundation (NRF) of Korea (Grants NRF-
fabrication is also a key factor for commercialization of next-generation 2020R1A4A1018163) under the program of Basic Research Laboratory
photovoltaics. The material cost of one component within the device can (BRL).
increase the total cost for device fabrication. As shown in Fig. S15, the
price of PTB7, which is indispensable for nip type AgBiS2 NC devices, is Appendix A. Supplementary data
$2800.22 per gram, whereas the price of nickel(II) acetate tetrahydrate,
which is a precursor for the NiO film in pin type AgBiS2 NC devices, is Supplementary data to this article can be found online at https://doi.
only $0.23 per gram. The PTB7 is approximately 10,000 times more org/10.1016/j.jpowsour.2021.230585.
expensive than the NiO precursor. It is meaningful that the realization of
all-inorganic pin type AgBiS2 NC device enables excellent device sta­ References
bility under various conditions as well as cost reduction for device
fabrication, while maintaining high device efficiency. [1] J. Yuan, Y. Zhang, L. Zhou, G. Zhang, H.-L. Yip, T.-K. Lau, X. Lu, C. Zhu, H. Peng, P.
A. Johnson, M. Leclerc, Y. Cao, J. Ulanski, Y. Li, Y. Zou, Single-junction organic
solar cell with over 15% efficiency using fused-ring acceptor with electron-
4. Conclusion deficient core, Joule 3 (2019) 1140–1151, https://doi.org/10.1016/j.
joule.2019.01.004.
We have demonstrated an all-inorganic AgBiS2 NC device with pin [2] A.K. Jena, A. Kulkarni, T. Miyasaka, Halide perovskite photovoltaics: background,
status, and future prospects, Chem. Rev. 119 (2019) 3036–3103, https://doi.org/
type device architecture by replacing the fragile organic HTL with a thin 10.1021/acs.chemrev.8b00539.
NiO layer. We predicted an optimum thickness of each layer in the pin [3] G.H. Carey, A.L. Abdelhady, Z. Ning, S.M. Thon, O.M. Bakr, E.H. Sargent, Colloidal
type AgBiS2 NC device via optical simulation study, which resulted in an quantum dot solar cells, Chem. Rev. 115 (2015) 12732–12763, https://doi.org/
10.1021/acs.chemrev.5b00063.
optimized PCE of 4.74% for a pin type AgBiS2 NC device with 35 nm of
[4] J. Yuan, A. Hazarika, Q. Zhao, X. Ling, T. Moot, W. Ma, J.M. Luther, Metal halide
TMAI-AgBiS2 layers. The introduction of a cascade-energy-level align­ perovskites in quantum dot solar cells: progress and prospects, Joule 4 (2020)
ment strategy using an extra MPA-AgBiS2 layer further improved the 1160–1185, https://doi.org/10.1016/j.joule.2020.04.006.
[5] J. Jeong, M. Kim, J. Seo, H. Lu, P. Ahlawat, A. Mishra, Y. Yang, M.A. Hope, F.
device efficiency up to 5.59%, attributed to enhanced charge transport/
T. Eickemeyer, M. Kim, Y.J. Yoon, I.W. Choi, B.P. Darwich, S.J. Choi, Y. Jo, J.
extraction properties with minimized charge recombination. Compared H. Lee, B. Walker, S.M. Zakeeruddin, L. Emsley, U. Rothlisberger, A. Hagfeldt, D.
to conventional nip type devices, the all-inorganic pin type AgBiS2 NC S. Kim, M. Gratzel, J.Y. Kim, Pseudo-halide anion engineering for alpha-FAPbI3
device exhibited excellent device stability in all ambient, heating, and perovskite solar cells, Nature 592 (2021) 381–385, https://doi.org/10.1038/
s41586-021-03406-5.
water immersion conditions as well as cost-effectiveness in device [6] B. Sun, A. Johnston, C. Xu, M. Wei, Z. Huang, Z. Jiang, H. Zhou, Y. Gao, Y. Dong,
fabrication by the chemical robustness and low precursor cost of the NiO O. Ouellette, X. Zheng, J. Liu, M.-J. Choi, Y. Gao, S.-W. Baek, F. Laquai, O.M. Bakr,
film, our work is a first report on an all-inorganic AgBiS2 NC device, as D. Ban, O. Voznyy, F.P. García de Arquer, E.H. Sargent, Monolayer perovskite
bridges enable strong quantum dot coupling for efficient solar cells, Joule 4 (2020)
well as the highest PCE in AgBiS2 NC photovoltaics with pin device 1542–1556, https://doi.org/10.1016/j.joule.2020.05.011.
structure. [7] X. Guo, Q. Fan, J. Wu, G. Li, Z. Peng, W. Su, J. Lin, L. Hou, Y. Qin, H. Ade, L. Ye,
M. Zhang, Y. Li, Optimized active layer morphologies via ternary copolymerization
of polymer donors for 17.6% efficiency organic solar cells with enhanced fill factor,
CRediT authorship contribution statement Angew. Chem. Int. Ed. 60 (2021) 2322–2329, https://doi.org/10.1002/
anie.202010596.
Jae Taek Oh: performed the main experiments, and characteriza­ [8] L. Jarup, Hazards of heavy metal contamination, Br. Med. Bull. 68 (2003) 167–182,
https://doi.org/10.1093/bmb/ldg032.
tion. Sung Yong Bae: performed the main experiments, and charac­ [9] L.C. Sorensen, A.M. Fox, H. Jung, E.G. Martin, Lead exposure and academic
terization. Jonghee Yang: help to perform the device characterization. achievement: evidence from childhood lead poisoning prevention efforts, J. Popul.
Su Ryong Ha: help to perform the device characterization. Hochan Econ. 32 (2018) 179–218, https://doi.org/10.1007/s00148-018-0707-y.
[10] C.C. Boyd, R. Cheacharoen, T. Leijtens, M.D. McGehee, Understanding degradation
Song: help to perform the device characterization. Cheong Beom Lee:
mechanisms and improving stability of perovskite photovoltaics, Chem. Rev. 119
help to perform the device characterization. Sanchari Shome: help to (2019) 3418–3451, https://doi.org/10.1021/acs.chemrev.8b00336.
perform the device characterization. Swarup Biswas: carried out opti­ [11] P. Cheng, X. Zhan, Stability of organic solar cells: challenges and strategies, Chem.
cal simulation for Pin-type AgBiS2 device. In addition. Hyun-Mo Lee: Soc. Rev. 45 (2016) 2544–2582, https://doi.org/10.1039/c5cs00593k.
[12] A.M. Ganose, C.N. Savory, D.O. Scanlon, Beyond methylammonium lead iodide:
help to perform the device characterization. You-Hyun Seo: help to prospects for the emergent field of ns2 containing solar absorbers, Chem. Commun.
perform the device characterization. Seok-In Na: help to perform the 53 (2016) 20–44, https://doi.org/10.1039/c6cc06475b.

7
J.T. Oh et al. Journal of Power Sources 514 (2021) 230585

[13] M. Bernechea, Y. Cao, G. Konstantatos, Size and bandgap tunability in Bi2S3 organic solar cells through optimal design, Dyes Pigments 159 (2018) 306–313,
colloidal nanocrystals and its effect in solution processed solar cells, J. Mater. https://doi.org/10.1016/j.dyepig.2018.06.025.
Chem. 3 (2015) 20642–20648, https://doi.org/10.1039/c5ta04441c. [32] P. Vincent, D.-S. Song, H.B. Kwon, D.-K. Kim, J.-H. Jung, J.-H. Kwon, E. Choe, Y.-
[14] Z. Ran, X. Wang, Y. Li, D. Yang, X.-G. Zhao, K. Biswas, D.J. Singh, L. Zhang, R. Kim, H. Kim, J.-H. Bae, Towards maximizing the haze effect of electrodes for
Bismuth and antimony-based oxyhalides and chalcohalides as potential high efficiency hybrid tandem solar cell, Appl. Surf. Sci. 432 (2018) 262–265,
optoelectronic materials, npj Comput. Mater. 4 (2018) 14, https://doi.org/ https://doi.org/10.1016/j.apsusc.2017.07.204.
10.1038/s41524-018-0071-1. [33] I. Burgués-Ceballos, Y. Wang, M.Z. Akgul, G. Konstantatos, Colloidal AgBiS2
[15] R. Tang, X. Wang, W. Lian, J. Huang, Q. Wei, M. Huang, Y. Yin, C. Jiang, S. Yang, nanocrystals with reduced recombination yield 6.4% power conversion efficiency
G. Xing, S. Chen, C. Zhu, X. Hao, M.A. Green, T. Chen, Hydrothermal deposition of in solution-processed solar cells, Nanomater. Energy 75 (2020) 104961, https://
antimony selenosulfide thin films enables solar cells with 10% efficiency, Nat. doi.org/10.1016/j.nanoen.2020.104961.
Energy 5 (2020) 587–595, https://doi.org/10.1038/s41560-020-0652-3. [34] J. You, L. Meng, T.B. Song, T.F. Guo, Y.M. Yang, W.H. Chang, Z. Hong, H. Chen,
[16] M. Bernechea, N.C. Miller, G. Xercavins, D. So, A. Stavrinadis, G. Konstantatos, H. Zhou, Q. Chen, Y. Liu, N. De Marco, Y. Yang, Improved air stability of perovskite
Solution-processed solar cells based on environmentally friendly AgBiS2 solar cells via solution-processed metal oxide transport layers, Nat. Nanotechnol.
nanocrystals, Nat. Photonics 10 (2016) 521–525, https://doi.org/10.1038/ 11 (2016) 75–81, https://doi.org/10.1038/nnano.2015.230.
nphoton.2016.108. [35] J.H. Lee, Y.W. Noh, I.S. Jin, S.H. Park, J.W. Jung, A solution-processed cobalt-
[17] V.A. Öberg, M.B. Johansson, X. Zhang, E.M.J. Johansson, Cubic AgBiS2 colloidal doped nickel oxide for high efficiency inverted type perovskite solar cells, J. Power
nanocrystals for solar cells, ACS Appl. Nano Mater. 3 (2020) 4014–4024, https:// Sources 412 (2019) 425–432, https://doi.org/10.1016/j.jpowsour.2018.11.081.
doi.org/10.1021/acsanm.9b02443. [36] J.R. Manders, S.W. Tsang, M.J. Hartel, T.H. Lai, S. Chen, C.M. Amb, J.R. Reynolds,
[18] S. Ming, X. Liu, W. Zhang, Q. Xie, Y. Wu, L. Chen, H.-Q. Wang, Eco-friendly and F. So, Solution-processed nickel oxide hole transport layers in high efficiency
stable silver bismuth disulphide quantum dot solar cells via methyl acetate polymer photovoltaic cells, Adv. Funct. Mater. 23 (2013) 2993–3001, https://doi.
purification and modified ligand exchange, J. Clean. Prod. 246 (2020) 118966, org/10.1002/adfm.201202269.
https://doi.org/10.1016/j.jclepro.2019.118966. [37] R. Wang, X. Wu, K. Xu, W. Zhou, Y. Shang, H. Tang, H. Chen, Z. Ning, Highly
[19] M.Z. Akgul, A. Figueroba, S. Pradhan, Y. Bi, G. Konstantatos, Low-cost RoHS efficient inverted structural quantum dot solar cells, Adv. Mater. 30 (2018)
compliant solution processed photovoltaics enabled by ambient condition 1704882, https://doi.org/10.1002/adma.201704882.
synthesis of AgBiS2 nanocrystals, ACS Photonics 7 (2020) 588–595, https://doi. [38] R.W. Crisp, D.M. Kroupa, A.R. Marshall, E.M. Miller, J. Zhang, M.C. Beard, J.
org/10.1021/acsphotonics.9b01757. M. Luther, Metal halide solid-state surface treatment for high efficiency PbS and
[20] Y.W. Soon, H. Cho, J. Low, H. Bronstein, I. McCulloch, J.R. Durrant, Correlating PbSe QD solar cells, Sci. Rep. 5 (2015) 9945, https://doi.org/10.1038/srep09945.
triplet yield, singlet oxygen generation and photochemical stability in polymer/ [39] S.-W. Baek, S.-H. Lee, J.H. Song, C. Kim, Y.-S. Ha, H. Shin, H. Kim, S. Jeong, J.-
fullerene blend films, Chem. Commun. 49 (2013) 1291–1293, https://doi.org/ Y. Lee, A hydro/oxo-phobic top hole-selective layer for efficient and stable
10.1039/c2cc38243a. colloidal quantum dot solar cells, Energy Environ. Sci. 11 (2018) 2078–2084,
[21] J. Kettle, Z. Ding, M. Horie, G.C. Smith, XPS analysis of the chemical degradation of https://doi.org/10.1039/c7ee03184j.
PTB7 polymers for organic photovoltaics, Org. Electron. 39 (2016) 222–228, [40] C.H. Chuang, P.R. Brown, V. Bulovic, M.G. Bawendi, Improved performance and
https://doi.org/10.1016/j.orgel.2016.10.016. stability in quantum dot solar cells through band alignment engineering, Nat.
[22] A. Perthué, T. Gorisse, H. Santos Silva, D. Bégué, A. Rivaton, G. Wantz, Influence of Mater. 13 (2014) 796–801, https://doi.org/10.1038/nmat3984.
traces of oxidized polymer on the performances of bulk heterojunction solar cells, [41] K. Domanski, E.A. Alharbi, A. Hagfeldt, M. Grätzel, W. Tress, Systematic
Mater. Chem. Front. 3 (2019) 1632–1641, https://doi.org/10.1039/c9qm00191c. investigation of the impact of operation conditions on the degradation behaviour of
[23] H. Ding Irfan, Y. Gao, C. Small, D.Y. Kim, J. Subbiah, F. So, Energy level evolution perovskite solar cells, Nat. Energy 3 (2018) 61–67, https://doi.org/10.1038/
of air and oxygen exposed molybdenum trioxide films, Appl. Phys. Lett. 96 (2010) s41560-017-0060-5.
243307, https://doi.org/10.1063/1.3454779. [42] M.O. Reese, S.A. Gevorgyan, M. Jørgensen, E. Bundgaard, S.R. Kurtz, D.S. Ginley,
[24] K.E. Lee, L. Liu, T.L. Kelly, Effect of molybdenum oxide electronic structure on D.C. Olson, M.T. Lloyd, P. Morvillo, E.A. Katz, A. Elschner, O. Haillant, T.
organic photovoltaic device performance: an X-ray absorption spectroscopy study, R. Currier, V. Shrotriya, M. Hermenau, M. Riede, K.R. Kirov, G. Trimmel, T. Rath,
J. Phys. Chem. C 118 (2014) 27735–27741, https://doi.org/10.1021/jp508972v. O. Inganäs, F. Zhang, M. Andersson, K. Tvingstedt, M. Lira-Cantu, D. Laird,
[25] J.T. Oh, S.Y. Bae, S.R. Ha, H. Cho, S.J. Lim, D.W. Boukhvalov, Y. Kim, H. Choi, C. McGuiness, S. Gowrisanker, M. Pannone, M. Xiao, J. Hauch, R. Steim, D.
Water-resistant AgBiS2 colloidal nanocrystal solids for eco-friendly thin film M. DeLongchamp, R. Rösch, H. Hoppe, N. Espinosa, A. Urbina, G. Yaman-
photovoltaics, Nanoscale 11 (2019) 9633–9640, https://doi.org/10.1039/ Uzunoglu, J.-B. Bonekamp, A.J.J.M. van Breemen, C. Girotto, E. Voroshazi, F.
c9nr01192g. C. Krebs, Consensus stability testing protocols for organic photovoltaic materials
[26] J.T. Oh, H. Cho, S.Y. Bae, S.J. Lim, J. Kang, I.H. Jung, H. Choi, Y. Kim, Improved and devices, Sol. Energy Mater. and Sol. 95 (2011) 1253–1267, https://doi.org/
size distribution of AgBiS2 colloidal nanocrystals by optimized synthetic route 10.1016/j.solmat.2011.01.036.
enhances photovoltaic performance, Int. J. Energy Res. 44 (2020) 11006–11014, [43] M.T. Greiner, L. Chai, M.G. Helander, W.-M. Tang, Z.-H. Lu, Metal/metal-oxide
https://doi.org/10.1002/er.5695. interfaces: how metal contacts affect the work function and band structure of
[27] J. Choi, Y. Kim, J.W. Jo, J. Kim, B. Sun, G. Walters, F.P. Garcia de Arquer, MoO3, Adv. Funct. Mater. 23 (2013) 215–226, https://doi.org/10.1002/
R. Quintero-Bermudez, Y. Li, C.S. Tan, L.N. Quan, A.P.T. Kam, S. Hoogland, Z. Lu, adfm.201200993.
O. Voznyy, E.H. Sargent, Chloride passivation of ZnO electrodes improves charge [44] F. Xie, W.C. Choy, C. Wang, X. Li, S. Zhang, J. Hou, Low-temperature solution-
extraction in colloidal quantum dot photovoltaics, Adv. Mater. 29 (2017) 1702350, processed hydrogen molybdenum and vanadium bronzes for an efficient hole-
https://doi.org/10.1002/adma.201702350. transport layer in organic electronics, Adv. Mater. 25 (2013) 2051–2055, https://
[28] Y.-M. Sung, Y.-C. Lai, M.-F. Tsai, H.-H. Hsieh, M.-H. Yang, P.P. Wei, C.-S. Yeh, F.- doi.org/10.1002/adma.201204425.
C. Hsu, Y.-F. Chen, Broad band plasmonic nanomaterials for high performance [45] A.A. Bortoti, A.d.F. Gavanski, Y.R. Velazquez, A. Galli, E.G. de Castro, Facile and
solar cells, J. Mater. Chem. C 4 (2016) 513–520, https://doi.org/10.1039/ low cost oxidative conversion of MoS2 in α-MoO3 : synthesis, characterization and
C5TC02329G. application, J. Solid State Chem. 252 (2017) 111–118, https://doi.org/10.1016/j.
[29] X. Zhang, J. Zhang, D. Phuyal, J. Du, L. Tian, V.A. Öberg, M.B. Johansson, U. jssc.2017.05.006.
B. Cappel, O. Karis, J. Liu, H. Rensmo, G. Boschloo, E.M.J. Johansson, Inorganic [46] A.R. Head, C. Gattinoni, L. Trotochaud, Y. Yu, O. Karslioglu, S. Pletincx,
CsPbI3 perovskite coating on PbS quantum dot for highly efficient and stable B. Eichhorn, H. Bluhm, Water (Non-)Interaction with MoO3, J. Phys. Chem. C 123
infrared light converting solar cells, Adv. Energy Mater. 8 (2018) 1702049, (2019) 16836–16842, https://doi.org/10.1021/acs.jpcc.9b03822.
https://doi.org/10.1002/aenm.201702049. [47] A.R. Head, R. Tsyshevsky, L. Trotochaud, Y. Yu, L. Kyhl, O. Karslıoǧlu, M.
[30] P. Vincent, J.W. Shim, J. Jang, I.M. Kang, P. Lang, J.-H. Bae, H. Kim, The crucial M. Kuklja, H. Bluhm, Adsorption of dimethyl methylphosphonate on MoO3: the
role of quaternary mixtures of active layer in organic indoor solar cells, Energies 12 role of oxygen vacancies, J. Phys. Chem. C 120 (2016) 29077–29088, https://doi.
(2019) 1838, https://doi.org/10.3390/en12101838. org/10.1021/acs.jpcc.6b07340.
[31] P. Vincent, S.-C. Shin, J.S. Goo, Y.-J. You, B. Cho, S. Lee, D.-W. Lee, S.R. Kwon, K.- [48] R. Ji, J. Cheng, X. Yang, J. Yu, L. Li, Enhanced charge carrier transport in spray-cast
B. Chung, J.-J. Lee, J.-H. Bae, J.W. Shim, H. Kim, Indoor-type photovoltaics with organic solar cells using solution processed MoO3 micro arrays, RSC Adv. 7 (2017)
3059–3065, https://doi.org/10.1039/c6ra27647d.