Multi-layered hybrid perovskites templated with carbazole derivatives: Optical properties, enhanced moisture stability and solar cell characteristics

Article abstract of DOI:10.1039/c8ta08019d

Research into 2D layered hybrid perovskites is on the rise due to the enhanced stability of these materials compared to 3D hybrid perovskites. Recently, interest towards the use of functional organic cations for these materials is increasing. However, a vast amount of the parameter space remains unexplored in multi-layered (n > 1) hybrid perovskites for solar cell applications. Here, we incorporate carbazole derivatives as a proof of concept towards the use of tailored functional molecules in multi-layered perovskites. Films of low-n carbazole containing perovskites show high photoconductivity half-lifetimes. Higher-n (?n? = 40) multi-layered perovskite films possess charge carrier diffusion lengths comparable to MAPI thin films. Solar cells containing these materials have comparable efficiencies to our MAPI and phenethylammonium (PEA)-containing multi-layered perovskite reference devices. Moisture stability tests were performed both at the material and device levels. In comparison to MAPI and PEA-based materials and solar cells, the addition of a small percentage of the carbazole derivative to the perovskite material significantly enhances the moisture stability.

Full text of DOI:10.1039/c8ta08019d

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Multi-layered hybrid perovskites templated with
carbazole derivatives: optical properties, enhanced
moisture stability and solar cell characteristics†
Cite this: J. Mater. Chem. A, 2018, 6,
22899
Roald Herckens, ‡a Wouter T. M. Van Gompel, ‡a Wenya Song,bc Marıa C. Gelvez-
´
´
d
ef
Rueda,
Arthur Maufort,a Bart Ruttens,e Jan D'Haen,
d
ae
Ferdinand C. Grozema,
and Dirk Vanderzande
Tom Aernouts,c Laurence Lutsen
ae
*
Research into 2D layered hybrid perovskites is on the rise due to the enhanced stability of these materials
compared to 3D hybrid perovskites. Recently, interest towards the use of functional organic cations for
these materials is increasing. However, a vast amount of the parameter space remains unexplored in
multi-layered (n > 1) hybrid perovskites for solar cell applications. Here, we incorporate carbazole
derivatives as a proof of concept towards the use of tailored functional molecules in multi-layered
perovskites. Films of low-n carbazole containing perovskites show high photoconductivity half-lifetimes.
Higher-n (hni ¼ 40) multi-layered perovskite films possess charge carrier diffusion lengths comparable
to MAPI thin films. Solar cells containing these materials have comparable efficiencies to our MAPI and
phenethylammonium (PEA)-containing multi-layered perovskite reference devices. Moisture stability
tests were performed both at the material and device levels. In comparison to MAPI and PEA-based
materials and solar cells, the addition of a small percentage of the carbazole derivative to the perovskite
material significantly enhances the moisture stability.
Received 17th August 2018
Accepted 25th October 2018
DOI: 10.1039/c8ta08019d
rsc.li/materials-a
Efforts to increase the stability have mostly focused on intro-
ducing other small cations such as formamidinium (FA) and
Introduction
caesium (Cs). Mainly the thermal stability of 3D perovskite
materials has improved signicantly via this methodology.2
A different approach to develop more resilient hybrid perov-
skite materials involves two-dimensional (2D) Ruddlesden–
Popper phase layered perovskites. Interest in these multi-layered
hybrid perovskites (A*2Anꢀ1PbnI3n+1, with A* a large organic
ammonium cation, A a small organic ammonium cation and n
the number of inorganic layers in the structure) for solar cell
applications increased due to reports about their enhanced
moisture stability compared to 3D hybrid perovskites.3–5
Although solar cells employing these materials are generally less
performant compared to 3D hybrid perovskites, their PCEs have
recently increased up to 12.5% by applying techniques such as
‘hot-casting’ (HC).6–12 One of the major advantages of the 2D
layered perovskite over the 3D perovskite is the structural
tolerance towards larger and more complex organic cations.12–14
Although it has been shown that a variety of organic cations13–15
can be used to obtain 2D-layered perovskites (A*2PbX4; n ¼ 1),
a vast amount of this parameter space remains unexplored in
multi-layered (n > 1) hybrid perovskites for solar cell applica-
tions. Until now mainly small alkylic and benzylic cations have
been applied in such multi-layered perovskites (e.g. buty-
lammonium,7 iso-butylammonium,16 phenethylammonium,17,18
anilinium,19 cyclohexylammonium,18 benzylammonium,18,20
Hybrid organic–inorganic perovskite solar cells have received
signicant attention over the past years due to their interesting
optoelectronic properties and their rapid increase in power
conversion efficiency (PCE) up to 23.3% today.1 Three-
dimensional (3D) perovskite materials have mainly been
studied due to the high efficiency of their solar cells. However,
these materials generally exhibit an intrinsic instability and can
degrade when exposed to moisture, heat, oxygen and/or light.2
aHasselt University, Institute for Materials Research (IMO-IMOMEC), Hybrid Materials
Design (HyMaD), Martelarenlaan 42, B-3500 Hasselt, Belgium. E-mail: dirk.
vanderzande@uhasselt.be
bimec, Thin-Film Photovoltaics, Energyville, Thor Park 8320, B-3600 Genk, Belgium
cKU Leuven, ESAT, B-3001 Leuven, Belgium
dDel University of Technology, Department of Chemical Engineering, Section
Optoelectronic Materials, Van der Maasweg 9, 2629 HZ Del, The Netherlands
eimec, Associated Laboratory IMOMEC, Wetenschapspark 1, B-3590 Diepenbeek,
Belgium
fHasselt University, Institute for Materials Research (IMO-IMOMEC), Electrical and
Physical Characterization (ELPHYC), Martelarenlaan 42, B-3500 Hasselt, Belgium
† Electronic supplementary information (ESI) available: Details of the carbazole
derivative synthesis, SEM pictures, additional UV-Vis absorption spectra,
photo-conductivity transients and optical pictures of lms and solar cells. See
DOI: 10.1039/c8ta08019d
‡ These authors contributed equally.
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 22899–22908 | 22899

View Article Online
Journal of Materials Chemistry A
Paper
propylphenylammonium,20 guanidinium21 and allylammo- Scientic. Potassium phthalimide salt (95%) was purchased
nium22). The main advantages of these multi-layered hybrid from Fluorochem. The SnO2 water dispersion (15 wt%) was
perovskites over their n ¼ 1 counterparts are a lower exciton purchased from Alfa Aesar and was diluted to 2.5 wt% for use.
binding energy and a smaller band gap, which is benecial Spiro-OMeTAD was obtained from Lumtec. Lithium bis(tri-
towards their use in solar cell applications.7,17 Moreover, multi- uoromethanesulfonyl) imide, 4-tert-butylpyridine, dime-
layered perovskites possess tuneable properties – such as thylformamide, acetonitrile and chlorobenzene were obtained
absorption and emission – depending on the number of inor- from Sigma Aldrich.
ganic layers (n) in the structure.7,12,17 We incorporated bulkier,
more complex, carbazole (CA) derivatives containing an Synthesis of organic ammonium salts
extended p-conjugated system as
a rst proof-of-concept
The synthesis of the carbazole alkylammonium iodide deriva-
tive (CA-C4-NH3I, shortened to CAI) is detailed in the ESI
(Fig. S1†). Phenethylammonium iodide (PEAI) was synthesised
by neutralizing equimolar amounts of hydroiodic acid (HI, 57%
towards the introduction of tailored functional organic mole-
cules in multi-layered perovskites for solar cell applications.
Beyond the eld of perovskite materials, carbazole-based
polymers/semiconductors have primarily attracted attention
for their photoconductive properties.23 In the context of perov-
skite, carbazole derivatives have been used in a fundamental
study on the synthesis of 2D layered perovskites ((CA-Cm)2PbBr4),
where derivatives with different lengths of alkyl spacer (m) were
examined towards the formation of the n ¼ 1 layered struc-
ture.24,25 Although this n ¼ 1 layered perovskite is not suitable as
a material for solar cells, it exhibited an interesting increase of
the in-plane conductivity in comparison to both poly-
vinylcarbazole polymers and layered perovskites with hex-
ylammonium (an insulator) as an organic layer.24
ꢁ
w/w) and phenethylamine at 0 C in an ice bath. An off-white
precipitate was obtained by evaporation of the solvent using
rotary evaporation at 70 ꢁC under reduced pressure. The
resulting precipitate was recrystallised three times from
ethanol, washed with diethylether and dried overnight under
reduced pressure at room temperature.
Thin lm deposition and annealing
Precursor solutions were prepared according to the nominal
structural formula (CA-C4)2MAnꢀ1PbnI3n+1 (Table 1). Stoichio-
metric amounts of CAI, MAI and PbI2 were dissolved in dry
dimethylformamide (DMF) at 70 ꢁC for 1 h under constant
stirring. The resulting clear solutions were ltered through
a syringe lter (0.45 mm pore size).
Quartz substrates were cleaned through consecutive soni-
cation steps in a series of solvents (detergent water, deionised
water, acetone, isopropanol) for 15 min each, followed by a UV-
ozone treatment for 30 min.
Films for optical measurements, TRMC and XRD analysis
were deposited on quartz substrates using spin coating in
a glovebox with nitrogen atmosphere (<0.1 ppm O2, < 0.1 ppm
H2O) at 2000 rpm, 2000 rpm sꢀ1 for 20 s. Annealing was per-
formed via hot-casting (HC) or by post-annealing on a hotplate
in the glovebox, the annealing temperature differs for different
hni and is indicated in the relevant gure captions. Prepared
samples were kept in a glovebox and were only removed for
analysis.
Carbazole derivatives are therefore a prime choice of an
extended p-conjugated organic molecule to be introduced in
functional multi-layered (n > 1) perovskites for photovoltaic
applications.
In this work, carbazole-based organic cations were syn-
thesised with an alkyl-spacer length (m) of four CH2 groups.
These organic cations were used to obtain lead iodide (PbI2)
based multi-layered perovskites with methylammonium (MA) as
small organic cation ((CA-C4)2MAnꢀ1PbnI3n+1). The formation
and the structure of the multi-layered and quasi-3D perovskites
incorporating carbazole were investigated. Films of low-n
carbazole containing perovskites show high photoconductivity
half-lifetimes in comparison to lms of established buty-
lammonium based, low-n multi-layered perovskites. Higher-n
(hni ¼ 40) multi-layered perovskite lms possess charge carrier
diffusion lengths comparable to MAPI thin lms. It was found
that the moisture stability of the carbazole containing perov-
skites is superior compared to the reference MAPI and PEA-
based perovskites. Likewise, the moisture stability of the cor-
Device fabrication
responding solar cell devices was signicantly improved while Devices were made in an ITO/SnO2/perovskite/Spiro-OMeTAD/
maintaining a similar initial power conversion efficiency.
Au conguration. The ITO-coated glass substrates were
successively cleaned in ultrasonic baths containing detergent
water, deionised water, acetone and isopropanol for 5 min,
respectively. Aerwards, these cleaned substrates were treated
with oxygen plasma. A 2.5 wt% water dispersion of SnO2
Experimental
Chemicals and reagents
Lead iodide (PbI2, 99.999%) was either obtained from Lumtec or nanoparticles was then spin coated on the substrates at
from TCI (99.99%) and used without further purication. 2800 rpm and pre-heated at 110 ꢁC. Perovskite solution in DMF
Methylammonium iodide was obtained from GreatCell Solar. was spin coated on these heated substrates at 3000 rpm for 60ꢀs1,
Hydroiodic acid (HI, 57% w/w aq., distilled, unstabilised) and followed by annealing at 110 ꢁC for 15 min. An 80 mg mL
phenethylamine (PEA, 99%) were purchased from Acros Spiro-OMeTAD solution doped with 17.5 mL lithium bis(tri-
Organics and used as received. Carbazole (96%), potassium tert- uoromethanesulfonyl) imide (Li-TFSI) (520 mg mLꢀ1 in
butoxide (98+%), hydrazine monohydrate (65%) and all solvents acetonitrile) and 28.5 mL 4-tert-butylpyridine was then spin
used for synthesis (reagent grade) were purchased from Fisher coated onto the perovskite lms. These lms were exposed
22900 | J. Mater. Chem. A, 2018, 6, 22899–22908
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
Table 1 Composition of precursor solutions for different hni
PbI2 concentration
MAI concentration
CAI/PEAI concentration
hni
Nominal structural formula
(mol Lꢀ1
)
(mol Lꢀ1
)
(mol Lꢀ1
)
1
2
3
4
10
40
40
N
CA2PbI4
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0
0.15
0.2
0.225
0.27
0.2925
0.2925
0.3
0.6
0.3
0.2
0.15
0.06
0.015
0.015
N.A.
CA2MAPb2I7
CA2MA2Pb3I10
CA2MA3Pb4I13
CA2MA9Pb10 31
CA2MA39Pb40 121
PEA2MA39Pb40 121
MAPbI3
I
I
I
overnight to air with 26% relative humidity (RH) for oxygen such that the organic layers are oriented parallel to the
doping. Finally, the devices were completed by depositing an substrate.28 The reections are indexed tentatively based on
80 nm Au layer onto the Spiro-OMeTAD through a shadow similarity with literature 2D layered perovskites.28 An inter-
mask, dening an active area of 0.13 cm2.
planar spacing (d00l) of ꢂ25.5 A is obtained (Table S1†). This
˚
matches well with two times the length of the carbazole mole-
˚
cule (ꢂ12 A), indicating that a bilayer of these molecules is
Characterization
present in the structure as expected for a 2D layered perovskite.
When multi-layered perovskites are targeted, hni ¼ 2, hni ¼ 3,
hni ¼ 4, hni ¼ 10 and hni ¼ 40 will be used to denote the nominal
composition that is targeted based on the stoichiometry of the
precursor solution and n will be used to denote the multi-
layered (CA-C4)2MAnꢀ1PbnI3n+1 perovskite with that specic
number of layers (n) as identied. It is important to note
however that for a targeted hni, a mixture of layered perovskites
with a different number of layers (n) are generally present in
thin-lm. This has been frequently reported in literature for
multi-layered perovskites starting from different large organic
molecules (A*).29–31 For example, Sargent et al. showed, using
transient absorption spectroscopy, that a lm nominally
prepared as hni ¼ 3 using PEAI contains signatures of n ¼ 2, 3, 4
and 5 phases.29
X-ray diffraction measurements were performed at room
temperature in ambient atmosphere on a Bruker D8 Discover
diffractometer with CuKa radiation.
Optical absorption spectra were measured on a Cary 5000
UV-Vis-NIR spectrophotometer from Agilent Technologies,
a cleaned quartz substrate was used as calibration background.
Photoluminescence emission spectra were taken with
a Horiba-Jobin Yvon Fluorolog-3 spectrouorometer, equipped
with double-grating excitation and emission monochromators
and a 450 W Xe lamp as a light source. An excitation wavelength
of either 300 nm or 430 nm was used (as indicated in the cor-
responding gure captions).
SEM measurements were performed on a FEI Quanta 200F.
Laser induced time-resolved microwave conductivity (TRMC)
measurements26,27 were performed on thin lms of CA-C4 multi-
layered perovskites (with hni ¼ 1, 2, 3, 4, 10 or 40) deposited on
quartz substrates and placed in a sealed resonant cavity inside
a nitrogen-lled glovebox. The TRMC technique measures the
change in microwave (8–9 GHz) power upon pulsed excitation
(repetition rate 10 Hz) of the thin lms (the excitation wave-
lengths used are indicated in the relevant gure captions).
Before and during the photo-conductance measurements, the
samples were kept in an inert nitrogen environment to prevent
possible degradation by exposure to moisture.
Results and discussion
Formation and structure
The synthesis of the series of functionalized 2D multi-layered
perovskites featuring carbazole derivatives was achieved by
mixing stoichiometric ratios of CAI, MAI and PbI2 in DMF.
These solutions were hot-casted or spin-coated and post-
annealed to obtain the desired perovskites.
As a starting reference, the n ¼ 1 ((CA-C4)2PbI4) was obtained
as a thin-lm. The diffraction pattern (Fig. 1) consists of a series
of (0 0 l) reections characteristic for a n ¼ 1 layered 2D
Fig. 1 X-ray diffraction pattern of a (CA-C4)2PbI4 (n ¼ 1) film obtained
by post-annealing at 110 ꢁC for 15 min. The structure of the CA-C4
ammonium salt is shown as an inset. The reflections were indexed
perovskite with preferential growth along the (1 1 0) direction, tentatively based on ref. 28.
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 22899–22908 | 22901

View Article Online
Journal of Materials Chemistry A
Paper
In thin lms of multi-layered hni ¼ 2 perovskites, (0 k 0) perovskites in these higher hni lms. From the absorption and
reections are present, which are characteristic28 for a n ¼ 2 emission spectra of these lms (vide infra) it is clear that the
layered perovskite with the organic layer oriented parallel to the lms still contain multi-layered perovskites with optical prop-
˚
substrate (Fig. 2). An interplanar spacing (d0k0) of ꢂ30.8 A is erties different from bulk MAPbI3.
obtained for these reections (Table S2†). This is a difference of
For such quasi-3D perovskites it has been hypothesized that
˚
ꢂ5.3 A compared to the interplanar spacing of the n ¼ 1 the organic molecules are present at the lattice surface and at
perovskite, which approximately matches with the height of grain boundaries of 3D perovskite grains.41
a single PbI6 octahedron.33,34
For hni ¼ 3, the (1 1 1) and (2 0 2) reections of the perovskite
Optical properties
structure, respectively at ꢂ14ꢁ and at ꢂ28ꢁ, become much more
The excitonic absorption peaks and/or optical band gaps (Eg) of
pronounced, indicating that the preferential growth direction is
the (CA-C4)2MAnꢀ1PbnI3n+1 were investigated (Fig. 3). The
changing towards vertical growth.35 Small reections at lower
absorption spectrum for the n ¼ 1 CA-C4 perovskite contains
angles are also present indicative of the presence of other low-n
clear signatures corresponding to absorption by the carbazole
perovskites. For hni ¼ 4 different reections below 10ꢁ 2q are
molecules (Fig. S7†) at low wavelengths, as well as an excitonic
present, which points to the presence of a complex mixture of
absorption peak at ꢂ494 nm (2.51 eV). This is in the same
multiple n as can be deduced from the absorption and emission
energy region as the excitonic absorption peaks for other n ¼ 1
spectra (vide infra).
layered iodide hybrid perovskites reported in literature (e.g.
In literature, the hot-casting procedure has been applied for
ꢂ2.4 eV for (PEA)2PbI4).34 It has been shown that the excitonic
different multi-layered and 3D perovskites to improve the
absorption peak position is mainly dependent on the distortion
perovskite crystallinity, morphology and orientation.36,37 SEM
of the inorganic perovskite layer by the organic cation.42,43
pictures of thin lms of CA-C4 hni ¼ 40 show that the
Going to higher hni, the excitonic absorption peaks shi
morphology for post-annealed lms is needle-like38–40 and
changes to planar structures with less pin-holes for hot-casted
lms (Fig. S3†). The morphology of the hot-casted lms is ex-
pected to be more suitable for solar cell devices. This hot-
casting method was therefore applied on our material for the
solar cell fabrication (vide infra).
towards lower energy (Fig. 3, Table 2) due to decreased quantum
For lms of higher hni perovskites obtained using the hot-
casting method (hni ¼ 4, 10 and 40), the (1 1 1) and (2 0 2)
reections dominate the diffraction pattern (Fig. 2), indicating
that, combined with the results of optical measurements (vide
infra), mostly quasi-3D perovskites are obtained. The weak
reections between ꢂ20ꢁ and 25ꢁ 2q are similar to those of bulk
tetragonal MAPbI3, pointing to the presence of “quasi-3D”
Fig. 3 Tauc plots of CA-C4 films with different hni obtained by post-
annealing at 110 ꢁC (hni ¼ 1), 130 ꢁC (hni ¼ 3) and 150 ꢁC (hni ¼ 2) for
15 min. The hni ¼ 4, 10 and 40 films were obtained using hot-casting at
110 ꢁC. A zoomed-in version is shown in Fig. S5† to more clearly
distinguish the absorption onsets. The absorption spectra used to
make the Tauc plots are shown in Fig. S6.†
Table 2 Excitonic absorption peak energies for multi-layered perov-
skites with different n
Fig. 2 X-ray diffraction patterns of films based on CA-C4 with different
hni. The hni ¼ 2 film was obtained by post-annealing at 150 ꢁC for
15 min. The hni ¼ 3 film was obtained by post-annealing at 130 ꢁC for
15 min. The hni ¼ 4, 10 and 40 films were hot-cast at 110 ꢁC. The broad
diffraction band from ꢂ16–30ꢁ 2q corresponds to diffraction of the
underlying quartz substrate32 (Fig. S2†). The main reflections for the
multi-layered perovskites containing the carbazole derivative were
indexed tentatively based on ref. 28.
Excitonic absorption peak energy
(eV/nm)
CA-Cm
CA-C4
n
1
2
3
4
2.51/494
2.21/561
2.06/602
1.96/633
22902 | J. Mater. Chem. A, 2018, 6, 22899–22908
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
connement.44,45 The excitonic peak positions for the n ¼ 2, n ¼
3 and n ¼ 4 multi-layered perovskite are assigned on the
absorption spectrum of a CA-C4 hni ¼ 3 lm in Fig. S8.† For hni >
2 multiple excitonic peaks belonging to low-n perovskites are
clearly present in the spectra and for hni > 3 also a band gap at
much lower energy (ꢂ1.6 eV) is present (Table 3), indicative of
the formation of quasi-3D perovskites together with the low-n
phases. Note that the band gap (and emission peak; Fig. 4)
belonging to the quasi-3D perovskites that are formed together
with the low-n phases also shis to lower energy with increasing
hni (Table 3). This indicates that the quasi-3D part of the lms
contains progressively higher n CA-C4 perovskites when using
higher hni precursor stoichiometries. This matches well quali-
tatively with the results of the recent article by Quintero-
Bermudez et al.51
Table 4 Excitonic emission peak energies for multi-layered perov-
skites with different n. The emission peaks are assigned on the emis-
sion spectra in Fig. S11 for each n
Excitonic emission peak energy
CA-Cm
CA-C4
n
(eV/nm)
1
2
3
4
2.47/501
2.15/576
2.03/610
1.93/642
Table 5 Emission peak energies for quasi-3D perovskites obtained for
films from precursor solutions targeting different hni. The emission
peaks are assigned on the emission spectra in Fig. S11 as “quasi-3D”
Emission peak energy of quasi 3D
The photoluminescence (PL) emission spectra for these
materials show similar trends as the absorption spectra. Going
from lower to higher hni, the emission peaks shi towards lower
energy (Fig. 4, Tables 4 and 5). Looking at the PL emission in the
low wavelength region (320–500 nm) for CA-C4 n ¼ 1, one
notices the same emission peaks as for the CA-C4 precursor salt
(Fig. S10†). However, the relative intensities of these peaks are
different. This can be related to the incorporation of the
carbazole molecules into the perovskite structure. The excitonic
CA-Cm
CA-C4
hni
perovskites (eV/nm)
3
4
10
40
1.79/694
1.72/721
1.66/748
1.65/751
emission peaks for the n ¼ 1 is located at ꢂ501 nm (2.47 eV).
Hence, there is a small Stokes shi (0.04 eV) between the exci-
tonic absorption and emission peaks. For n ¼ 1, a broad
emission feature at lower energy can be noticed that disappears
for higher-n CA-C4 layered perovskites. Similar features in the
emission spectra of 2D layered perovskites have been assigned
to luminescence from self-trapped excitons or long-lived colour
centers.46–48 Determining the origin of this broad emission
feature goes beyond the scope of this article and is the subject of
further work. For higher hni, emission peaks at lower wave-
lengths corresponding to quasi-3D perovskites are present
(Table 5). From this analysis it is clear that we can tune the
absorption and emission properties of the carbazole containing
perovskites by changing the stoichiometry of the precursor
solutions.
Table 3 Optical band gap energies for quasi-3D perovskites obtained
for films from precursor solutions targeting different hni. A represen-
tative Tauc plot with tangent intersecting the energy axis is shown in
Fig. S9
Optical band gap energy
CA-Cm
CA-C4
hni
(eV/nm)
4a
10
40
1.688/735
1.626/763
1.617/767
a
Formed together with low-n perovskites (see Table 2).
Photoconductivity and charge-carrier dynamics
To study the charge carrier dynamics of the carbazole and PEA
multi-layered perovskites we performed photo-conductivity
measurements. Details on this technique and on the calcula-
tion of the charge carrier mobility, lifetime and diffusion
lengths can be found in ref. 26. In these measurements, the
laser-induced change in conductivity of the material, DG, is
measured as a function of time. Typical conductivity measure-
ments are shown in Fig. 5(a) for CA-C4 hni ¼ 40 (for other hni
this is shown in Fig. S12†). DG increases during the pulse as
charge carriers are generated and subsequently decays as the
charge carriers recombine or become trapped. The time reso-
lution of these measurements are limited by the width of the
laser pulse (3.5 ns full width at half maximum) and the response
Fig. 4 Photoluminescence emission spectra of CA-C4 films with
different hni obtained by post-annealing at 110 ꢁC (n ¼ 1), 130 ꢁC (hni ¼
3) and 150 ꢁC (hni ¼ 2) for 15 min. The hni ¼ 4, 10 and 40 films and the
MAPI film were obtained using hot-casting at 110 ꢁC. The samples
were excited at 430 nm.
time of the microwave cavity (ꢂ18 ns). The product of the yield
P
of free charges (4) and mobility of the charge carriers ( m ¼ me
+ mh) for CA-C4 (hni ¼ 1, 2, 3, 4, 10, 40) and PEA hni ¼ 40 are
shown Fig. 5(b). These values were calculated from the
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 22899–22908 | 22903

View Article Online
Journal of Materials Chemistry A
Paper
maximum change in photoconductivity, DGmax, according to
eqn (1). Where, I0 is the number of photons per unit area per
pulse, b is the ratio of the inner dimensions of the microwave
cell, e is the elementary charge, and FA is the fraction of light
absorbed by the sample at the excitation wavelength.
X
DGmax
4
m ¼
(1)
I0beFA
It is clear in Fig. 5(b and d) that the photo-conductivity of CA-
C4 hni ¼ 40 is larger than the photo-conductivity of CA-C4 hni ¼
1, 2, 3, 4, 10. A magnied version of hni ¼ 4 and 10 is shown in
Fig. S13.† A similar general trend is observed in the half-lifetime
of the charge carriers shown in Fig. 5(c, e). The photo-
conductivity and half-lifetime for PEA hni ¼ 40 are also large.
The low photo-conductivity and short lifetime of CA-C4 n ¼ 1 are
similar to previous measurements on other 2D perovskites with
the same experimental technique.49 This behaviour is explained
by the large exciton binding energy of 2D perovskites, which
decreases the yield of free charges (4) and increases the
recombination.49 In the case of our CA-C4 hni ¼ 1, 2, 3, 4, 10
samples the photoconductivity and half-lifetime do not follow
a clear increasing trend with n. This is contrary to what has been
shown previously for another 2D multi-layered material.49
A
probable reason for this is the presence of phases with different
n. It can be noticed that the expected increase in photocon-
ductivity and half-lifetime is present going from hni ¼ 1 to 2.
This ts with the XRD and absorption results of these lms.
Whereas the hni ¼ 2 lm is mostly pure phase, for hni > 2 phases
with different n are more clearly present. In conclusion, the
half-lifetimes for the novel CA-C4 based 2D multi-layered
perovskite thin lms with hni ¼ 2, 3 and 4 are longer than for
the reported butylammonium (BA) based multi-layered perov-
skite thin lms prepared with the same procedure.49 This is an
encouraging result showing the potential benet of using
organic cations with an extended p-system in multi-layered
perovskites.
Interestingly, the mobility of charge carriers of CA-C4 hni ¼
40 and PEA hni ¼ 40 is respectively ꢂ8 cm2 Vꢀ1 sꢀ1 and
16 cm2 Vꢀ1 sꢀ1 which is relatively high. We estimated the
diffusion length of charge carriers to be ꢂ2.4 mm and ꢂ4 mm;
which largely exceed the thickness of the lms (0.3 mm for CA-C4
hni ¼ 40 and 0.8 mm for PEA hni ¼ 40). The mobility and
diffusion lengths of the PEA hni ¼ 40 thin lms are comparable
to these measured for the 3D MAPI using the same technique.27
The mobility of charge carriers in MAPI thin lms measured by
photoconductivity TRMC varies between 8–40 cm2 Vꢀ1 sꢀ1
,
while the diffusion length varies between 1.5 and 10 mm.27 In
fact, the diffusion length is similar to that of planar MAPI lms
(ꢂ4.1 mm)27 in spite of having lower charge carrier mobilities
(30 cm2 Vꢀ1 sꢀ1 for planar MAPI thin-lms).27 These relatively
high mobilities and large diffusion lengths suggest that our CA-
C4 hni ¼ 40 and the PEA hni ¼ 40 lms are attractive candidates
for solar cell applications. The improved electronic properties of
CA-C4 hni ¼ 40 and PEA hni ¼ 40 over CA-C4 lms with lower
values of hni (hni ¼ 1, 2, 3, 4, 10) is most likely due to their quasi-
3D nature. This will increase the dielectric constant in the
Fig. 5 (a) Change of photo-conductivity as a function of time for CA-C4
hni ¼ 40. Excited at 633 nm. (b) Change of photo-conductivity and (c)
half-lifetime of the charge carriers as a function of hni for hni ¼ 1, 2, 3
excited at 493 nm, 561 nm and 601 nm, respectively. Prepared by spin
coating and post-annealing at 110 ꢁC, 150 ꢁC and 130 ꢁC, respectively.
(d) Change of photo-conductivity and (e) half-lifetime of the charge
carriers as a function of hni for hni ¼ 4, 10, 40 excited at 633 nm.
Prepared by hot-casting at 110 ꢁC.
22904 | J. Mater. Chem. A, 2018, 6, 22899–22908
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
inorganic octahedral layers and therefore decrease the exciton
binding energy and recombination rate. The comparable
diffusion lengths of the hni ¼ 40 materials with MAPI lms in
spite of having lower mobilities, could be caused by grain
boundary passivation by the carbazole and PEA molecules.
These “high-n” quasi-3D perovskites could actually be consid-
ered grains of 3D perovskites surrounded by the organic mole-
cules. The organic molecules passivate defects and might
improve the stability and performance of the solar cells.27 We
focus on high-hni materials for the preparation of solar cell
devices (vide infra). The solar cell stability and performance are
studied in the following sections.
Environmental stability of lms
Fig. 7 XRD patterns of a MAPI film hot-cast at 110 ꢁC, measured at
different times stored at 77% RH, showing the range of 11.5ꢁ till 35ꢁ 2q.
The label P is given to the main reflection of lead iodide located at
ꢂ12.6ꢁ 2q.
One of the objectives in the development of functional multi-
layered perovskites is to benet from their potential stability,
while maintaining as much as possible the PCE associated with
their parent 3D perovskites. The environmental stability of the
novel hni ¼ 40 multi-layered perovskite featuring CA-C4 was
therefore evaluated by XRD over time at 77% relative humidity
(RH) in the dark at room temperature. The stability was
compared to the references MAPI and the hni ¼ 40 multi-layered
perovskite with PEA.
For the MAPI lm, initially the monohydrate50 (Fig. 6, main
reections at ꢂ8.5ꢁ and ꢂ10.6ꢁ) is formed followed by slower
formation of lead iodide (Fig. 7, main reection at ꢂ12.6ꢁ). In the
case of PEA hni ¼ 40, there is no clear sign of hydrate formation
over the time of the experiment but lead iodide is slowly formed
over time (Fig. 8). The enhanced moisture stability of lms with
PEA was attributed in literature to the presence of the PEA
molecules at the perovskite grain boundaries.4,41 However, the
small amount of PEA used for the hni ¼ 40 stoichiometry is clearly
not sufficient to completely prevent degradation over the time
period of the experiment. For the novel CA-C4 hni ¼ 40, no
signicant changes in the diffraction pattern can be noticed over
the 62 days of the experiment (Fig. 9). In Fig. S4† we show that the
Fig. 8 XRD patterns of a PEA hni ¼ 40 film hot-cast at 110 ꢁC,
measured at different times stored at 77% RH. The label P is given to
the main reflection of lead iodide located at ꢂ12.6ꢁ 2q.
morphology of hot-casted PEA hni ¼ 40 lms is very similar to
that of hot-casted CA-C4 hni ¼ 40 lms, hence the enhanced
stability of the CA-C4 hni ¼ 40 lms can not be ascribed to
differences in lm morphology. The enhanced stability for CA-C4
compared to PEA could be related to the more bulky nature of the
carbazole molecules and their strong p–p stacking, which could
make water intrusion through the organic layer more difficult.
Although only a relatively small amount (ꢂ5 mol% compared to
molMA + molCA) of the large cation CA-C4 is incorporated,
a signicantly enhanced moisture stability is obtained for this
novel multi-layered perovskite material compared to the known
3D MAPI and PEA hni ¼ 40 (see ESI, Fig. S14† for optical pictures
of the lms). This result is a clear demonstration of the benet on
the intrinsic stability of the use of bulky cations containing
extended p-systems in multi-layered perovskites.
Fig. 6 XRD patterns of a MAPI film hot-cast at 110 ꢁC, measured at
different times stored at 77% RH. The label H is given to the two main
reflections of the monohydrate MAPbI3$H2O located at ꢂ8.5ꢁ and
ꢂ10.6ꢁ 2q. The label P is given to the main reflection of lead iodide
located at ꢂ12.6ꢁ 2q.
Solar cell behaviour and device stability under moisture
The novel multi-layered perovskite material CA-C4 hni ¼ 40 was
further integrated in photovoltaic devices and compared to the
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 22899–22908 | 22905

View Article Online
Journal of Materials Chemistry A
Paper
Table 6 Initial and final PCE (with standard deviation) of the devices
stored at 77% RH
Active layer
Initial PCE (%)
Final PCE (%)
CA-C4 hni ¼ 40
PEA hni ¼ 40
MAPI
6.6 ꢃ 0.2
7.6 ꢃ 0.3
8.0 ꢃ 0.2
5.4 ꢃ 0.2
2.7 ꢃ 0.3
2.9 ꢃ 0.2
contrast, the Jsc of carbazole hni ¼ 40 devices only slightly
decreases. Open circuit voltages of carbazole hni ¼ 40 devices
are also more stable than those of MAPI and PEA hni ¼ 40
devices (Fig. 10(c)). As a result, carbazole hni ¼ 40 devices
maintain more than 80% of their initial efficiencies, while MAPI
and PEA hni ¼ 40 devices lose more than 60% of their initial
efficiencies (Fig. 10(a) and Table 6). These results reveal that
adding a small percentage of the carbazole derivative to the
perovskite precursor solutions can remarkably enhance the
moisture stability of devices and that the stabilizing effect of the
carbazole derivatives signicantly outperforms that of PEA.
Fig. 9 XRD patterns of a CA-C4 hni ¼ 40 film hot-cast at 110 ꢁC,
measured at different times being stored at 77% RH.
3D perovskite MAPI and PEA hni ¼ 40 in terms of efficiency and
stability. Devices were made in an ITO/SnO2/perovskite/Spiro-
OMeTAD/Au conguration. The perovskite lms are around
200 nm thick, therefore the initial short circuit current (Jsc) of
the devices are between 12.5 and 14.5 mA cmꢀ2 (Fig. 10(b)). The
PCEs obtained for the devices are moderate (Table 6). We
deliberately chose to use the basic 3D perovskite MAPI as the
parent material of our CA-C4 hni ¼ 40 and not the advanced
mixed cations/mixed halides 3D perovskites which give higher
PCE in devices as our main goal here was only a comparative
study to evaluate directly the effect of using our carbazole
derivative. Therefore we kept the material system as simple as
possible for this rst proof-of-concept.
The devices were stored in 77% RH air for 264 h in the dark.
The colour of the carbazole hni ¼ 40 perovskite lms stays dark,
while MAPI and PEA hni ¼ 40 lms turn transparent (Fig. S15†).
The degradation of the active layer leads to a drastic decrease of
the Jsc for MAPI and PEA hni ¼ 40 devices (Fig. 10(b)). In
Conclusions
In summary, multi-layered perovskites templated with carba-
zole derivatives were synthesized. The absorption and emission
characteristics could be tuned by variation of the number of
inorganic layers (n) in the perovskite structure. Interestingly,
lms of low-n carbazole containing perovskites showed high
photoconductivity half-lifetimes in comparison to lms of
established low-n multi-layered perovskites. Higher-n (hni ¼ 40)
multi-layered perovskite lms possess charge carrier diffusion
lengths comparable to MAPI thin lms. Solar cells containing
these materials have comparable efficiencies to our MAPI and
phenethylammonium (PEA) hni
¼
40 reference devices.
Furthermore, moisture stability tests were performed both at
the material and device levels. These tests showed that adding
a small percentage of the carbazole derivative signicantly
enhances the moisture stability of the perovskite material and
of the solar cells, compared to their PEA counterparts. The
promising results obtained with carbazole as a rst example of
a large organic cation containing an extended p-system in
multi-layered perovskites open the way to other systems in
which similar tailored cations will be incorporated. Future work
includes applying this concept to state-of-the-art mixed cation/
mixed halide 3D perovskite formulations to combine the
higher PCE with enhanced stability.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
`
We thank Nadege Marchal for the work she did during her
internship at the HyMaD group. R. H. is a special research fund
(BOF) Doctoral (PhD) student at UHasselt/IMO. W. T. M. V. G. is
an SB PhD fellow at FWO (project number 1S17516N), the FWO
Fig. 10 Changes of device metrics after being stored in 77% RH air, in
the dark, for 264 h. Average value and standard deviation shown on
each data point are obtained from 12 devices.
22906 | J. Mater. Chem. A, 2018, 6, 22899–22908
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
Journal of Materials Chemistry A
is acknowledged for the funding of the research. The PVopMaat 18 L. Iagher and L. Etgar, ACS Energy Lett., 2018, 3, 366–372.
´
´
project funded by Interreg Vlaanderen-Nederland (01.01.2016– 19 J. Rodrıguez-Romero, B. C. Hames, I. Mora-Sero and
31.12.2018) and the M-ERA.NET project called PROMISES E. M. Barea, ACS Energy Lett., 2017, 2, 1969–1970.
(01.07.2016–30.06.2018) are acknowledged for their funding. 20 B.-E. Cohen, M. Wierzbowska and L. Etgar, Sustainable
The research leading to these results in the Del University of Energy Fuels, 2017, 1, 1935–1943.
Technology has received funding from the European Research 21 O. Nazarenko, M. R. Kotyrba, M. Worle, E. Cuervo-Reyes,
¨
Council Horizon 2020 ERC Grant Agreement no. 648433. This
work has been carried out in the context of the Solliance
S. Yakunin and M. V. Kovalenko, Inorg. Chem., 2017, 56,
11552–11564.
network (http://www.solliance.eu), from which imec, UHasselt 22 A. H. Proppe, R. Quintero-Bermudez, H. Tan, O. Voznyy,
and TUDel are members. Additionally, imec, KU Leuven and
UHasselt are partners in the EnergyVille consortium (http://
www.energyville.be/en/about-energyville).
S. O. Kelley and E. H. Sargent, J. Am. Chem. Soc., 2018, 140,
2890–2896.
23 J. V. Grazulevicius, P. Strohriegl, J. Pielichowski and
K. Pielichowski, Prog. Polym. Sci., 2003, 28, 1297–1353.
24 M. Era, T. Kobayashi, K. Sakaguchi, E. Tsukamoto and
Y. Oishi, Org. Electron., 2013, 14, 1313–1317.
References
1 National Renewable Energy Laboratory, Best Research-Cell 25 M. Era, T. Yasuda, K. Mori, N. Tomotsu, N. Kawano,
Efficiencies, https://www.nrel.gov/pv/assets/pdfs/pv-efficiencies-
07-17-2018.pdf, accessed 29 October 2018.
M. Koshimizu and K. Asai, J. Nanosci. Nanotechnol., 2016,
16, 3159–3167.
2 Z. Wang, Z. Shi, T. Li, Y. Chen and W. Huang, Angew. Chem., 26 T. J. Savenije, A. J. Ferguson, N. Kopidakis and G. Rumbles,
2017, 56, 1190–1212.
J. Phys. Chem. C, 2013, 117, 24085–24103.
3 I. C. Smith, E. T. Hoke, D. Solis-Ibarra, M. D. McGehee and 27 E. M. Hutter, G. E. Eperon, S. D. Stranks and T. J. Savenije,
H. I. Karunadasa, Angew. Chem., 2014, 126, 11414–11417. J. Phys. Chem. Lett., 2015, 6, 3082–3090.
4 D. S. Lee, J. S. Yun, J. Kim, A. M. Souani, S. Chen, Y. Cho, 28 D. H. Cao, C. C. Stoumpos, O. K. Farha, J. T. Hupp and
X. Deng, J. Seidel, S. Lim, S. Huang and A. W. Y. Ho-Baillie,
ACS Energy Lett., 2018, 3, 647–654.
5 H. Zheng, G. Liu, L. Zhu, J. Ye, X. Zhang, A. Alsaedi, T. Hayat,
X. Pan and S. Dai, Adv. Energy Mater., 2018, 8, 1800051.
6 X. Zhang, X. Ren, B. Liu, R. Munir, X. Zhu, D. Yang, J. Li,
M. G. Kanatzidis, J. Am. Chem. Soc., 2015, 137, 7843–7850.
29 M. Yuan, L. N. Quan, R. Comin, G. Walters, R. Sabatini,
O. Voznyy, S. Hoogland, Y. Zhao, E. M. Beauregard,
P. Kanjanaboos, Z. Lu, D. H. Kim and E. H. Sargent, Nat.
Nanotechnol., 2016, 11, 1–27.
Y. Liu, D.-M. Smilgies, R. Li, Z. Yang, T. Niu, X. Wang, 30 Q. Shang, Y. Wang, Y. Zhong, Y. Mi, L. Qin, Y. Zhao, X. Qiu,
A. Amassian, K. Zhao and S. Liu, Energy Environ. Sci., 2017,
10, 2095–2102.
7 C. M. M. Soe, W. Nie, C. C. Stoumpos, H. Tsai, J.-C. Blancon,
X. Liu and Q. Zhang, J. Phys. Chem. Lett., 2017, 8, 4431–4438.
31 J. Liu, J. Leng, K. Wu, J. Zhang and S. Jin, J. Am. Chem. Soc.,
2017, 139, 1432–1435.
F. Liu, J. Even, T. J. Marks, A. D. Mohite and M. G. Kanatzidis, 32 F. Barka-Bouaifel, K. Makaoui, P.-Y. Jouan, X. Castel, N. Bezzi,
Adv. Energy Mater., 2018, 8, 1700979.
R. Boukherroub and S. Szunerits, RSC Adv., 2012, 2, 12482.
8 J.-C. Blancon, H. Tsai, W. Nie, C. C. Stoumpos, L. Pedesseau, 33 A. Vassilakopoulou, D. Papadatos, I. Zakouras and
C. Katan, M. Kepenekian, C. M. M. Soe, K. Appavoo,
I. Koutselas, J. Alloys Compd., 2017, 692, 589–598.
M. Y. Sfeir, S. Tretiak, P. M. Ajayan, M. G. Kanatzidis, J. Even, 34 W. Peng, J. Yin, K. Ho, O. Ouellette, M. De Bastiani,
J. J. Crochet and A. D. Mohite, Science, 2017, 355, 1288–1292.
9 J.-F. Liao, H.-S. Rao, B.-X. Chen, D.-B. Kuang and C.-Y. Su,
J. Mater. Chem. A, 2017, 5, 2066–2072.
B. Murali, O. El Tall, C. Shen, X. Miao, J. Pan, E. Alarousu,
J. He, B. S. Ooi, O. F. Mohammed, E. Sargent and
O. M. Bakr, Nano Lett., 2017, 17, 4759–4767.
10 Z. Wang, Q. Lin, F. P. Chmiel, N. Sakai, L. M. Herz and 35 X. Gan, O. Wang, K. Liu, X. Du, L. Guo and H. Liu, Sol. Energy
H. J. Snaith, Nat. Energy, 2017, 2, 17135.
Mater. Sol. Cells, 2017, 162, 93–102.
11 A. R. b. M. Yusoff and M. K. Nazeeruddin, Adv. Energy Mater., 36 W. Nie, H. Tsai, R. Asadpour, J.-C. Blancon, A. J. Neukirch,
2017, 1702073.
G. Gupta, J. J. Crochet, M. Chhowalla, S. Tretiak,
M. A. Alam, H.-L. Wang and A. D. Mohite, Science, 2015,
347, 522–525.
12 L. Etgar, Energy Environ. Sci., 2018, 11, 234–242.
13 B. Saparov and D. B. Mitzi, Chem. Rev., 2016, 116, 4558–4596.
14 J. V. Passarelli, D. J. Faireld, N. A. Sather, M. P. Hendricks, 37 H. Tsai, W. Nie, J.-C. Blancon, C. C. Stoumpos, R. Asadpour,
H. Sai, C. L. Stern and S. I. Stupp, J. Am. Chem. Soc., 2018,
140, 7313–7323.
15 M. I. Saidaminov, O. F. Mohammed and O. M. Bakr, ACS
Energy Lett., 2017, 2, 889–896.
B. Harutyunyan, A. J. Neukirch, R. Verduzco, J. J. Crochet,
S. Tretiak, L. Pedesseau, J. Even, M. A. Alam, G. Gupta,
J. Lou, P. M. Ajayan, M. J. Bedzyk, M. G. Kanatzidis and
A. D. Mohite, Nature, 2016, 536, 312–316.
16 Y. Chen, Y. Sun, J. Peng, W. Zhang, X. Su, K. Zheng, 38 A. Sharenko and M. F. Toney, J. Am. Chem. Soc., 2016, 138,
T. Pullerits and Z. Liang, Adv. Energy Mater., 2017, 7, 1700162.
463–470.
17 L. N. Quan, M. Yuan, R. Comin, O. Voznyy, E. M. Beauregard, 39 R. A. Kerner, L. Zhao, Z. Xiao and B. P. Rand, J. Mater. Chem.
S. Hoogland, A. Buin, A. R. Kirmani, K. Zhao, A. Amassian, A, 2016, 4, 8308–8315.
D. H. Kim and E. H. Sargent, J. Am. Chem. Soc., 2016, 138, 40 J. C. Hamill, J. Schwartz and Y.-L. Loo, ACS Energy Lett., 2018,
2649–2655.
3, 92–97.
This journal is © The Royal Society of Chemistry 2018
J. Mater. Chem. A, 2018, 6, 22899–22908 | 22907

View Article Online
Journal of Materials Chemistry A
Paper
41 N. Li, Z. Zhu, C.-C. Chueh, H. Liu, B. Peng, A. Petrone, X. Li, 47 I. Neogi, A. Bruno, D. Bahulayan, T. W. Goh, B. Ghosh,
L. Wang and A. K.-Y. Jen, Adv. Energy Mater., 2017, 7, 1601307.
42 K. Du, Q. Tu, X. Zhang, Q. Han, J. Liu, S. Zauscher and
D. B. Mitzi, Inorg. Chem., 2017, 56, 9291–9302.
R. Ganguly, D. Cortecchia, T. C. Sum, C. Soci,
N. Mathews and S. G. Mhaisalkar, ChemSusChem, 2017,
10, 3765–3772.
43 M. E. Kamminga, H.-H. Fang, M. R. Filip, F. Giustino, 48 M. D. Smith, A. Jaffe, E. R. Dohner, A. M. Lindenberg and
J. Baas, G. R. Blake, M. A. Loi and T. T. M. Palstra, Chem.
Mater., 2016, 28, 4554–4562.
H. I. Karunadasa, Chem. Sci., 2017, 8, 4497–4504.
´
49 M. C. Gelvez-Rueda, E. M. Hutter, D. H. Cao, N. Renaud,
44 B. Traore, L. Pedesseau, L. Assam, X. Che, J.-C. Blancon,
H. Tsai, W. Nie, C. C. Stoumpos, M. G. Kanatzidis, S. Tretiak,
A. D. Mohite, J. Even, M. Kepenekian and C. Katan, ACS
Nano, 2018, 12, 3321–3332.
45 C. C. Stoumpos, D. H. Cao, D. J. Clark, J. Young,
J. M. Rondinelli, J. I. Jang, J. T. Hupp and M. G. Kanatzidis,
Chem. Mater., 2016, 28, 2852–2867.
C. C. Stoumpos, J. T. Hupp, T. J. Savenije,
M. G. Kanatzidis and F. C. Grozema, J. Phys. Chem. C,
2017, 121, 26566–26574.
50 A. M. A. Leguy, Y. Hu, M. Campoy-Quiles, M. I. Alonso,
O. J. Weber, P. Azarhoosh, M. van Schilfgaarde,
M. T. Weller, T. Bein, J. Nelson, P. Docampo and
P. R. F. Barnes, Chem. Mater., 2015, 27, 3397–3407.
46 X. Wu, M. T. Trinh, D. Niesner, H. Zhu, Z. Norman, 51 R. Quintero-Bermudez, A. Gold-Parker, A. H. Proppe,
J. S. Owen, O. Yaffe, B. J. Kudisch and X.-Y. Zhu, J. Am.
Chem. Soc., 2015, 137, 2089–2096.
R. Munir, Z. Yang, S. O. Kelley, A. Amassian, M. F. Toney
and E. H. Sargent, Nat. Mater., 2018, 17, 900–907.
22908 | J. Mater. Chem. A, 2018, 6, 22899–22908
This journal is © The Royal Society of Chemistry 2018