The effect of organic cations on the electronic, optical and luminescence properties of 1D piperidinium, pyridinium, and 3-hydroxypyridinium lead trihalides

Article abstract of DOI:10.1039/c9dt04543k

We present a structural and optoelectronic study of 1D piperidinium, pyridinium, and 3-hydroxypyridinium lead trihalides. In contrast to the piperidinium and pyridinium species whose single inorganic chains [PbX₃^1-]_n are separated by organic cations, the 3-hydroxypyridinium compound is characterized by double inorganic chains. According to DFT the valence and conduction bands of the piperidinium lead trihalides are composed of occupied p-orbitals of the halogen anions and unoccupied p-orbitals of the Pb²⁺ cations. In contrast, the pyridinium species feature low-lying cationic energy levels formed from the cation's π?-orbitals. Thus, electronic transitions between the cationic energy levels and valence bands require less energy than valence to conduction band transitions in the case of piperidinium lead trihalides. The presence of an OH group in the pyridinium ring leads to a bathochromic shift of the cationic energy levels resulting in a decreased energy of transitions from the cationic energy levels to the valence band. Electronic transitions predicted by DFT are observable in experimental optical absorption and luminescence spectra. This study paves the way for creation of 1D perovskite-like structures with desired optoelectronic properties.

Full text of DOI:10.1039/c9dt04543k

View Article Online
View Journal
Dalton
Transactions
An international journal of inorganic chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: N. I. Selivanov, J.
A. Rozhkova, R. Kevorkyants, A. Emeline and D. W. Bahnemann, Dalton Trans., 2020, DOI:
10.1039/C9DT04543K.
Volume 46
Number 10
14 March 2017
Pages 3073-3412
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
An international journal of inorganic chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
rsc.li/dalton
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
ISSN 0306-0012
COMMUNICATION
Douglas W. Stephen et al.
shall the Royal Society of Chemistry be held responsible for any errors
N-Heterocyclic carbene stabilized parent sulfenyl, selenenyl,
and tellurenyl cations (XH+, X = S, Se, Te)
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
rsc.li/dalton

Page 1 of 26
Dalton Transactions
The effect of an organic cation on electronic, optical and luminescent properties of 1D
View Article Online
DOI: 10.1039/C9DT04543K
piperidinium, pyridinium, and 3-hydroxypyridinium lead trihalides.
N.I. Selivanov1*, Yu.A. Rozhkova1, R. Kevorkyants1*, A.V. Emeline1, and D.W. Bahnemann1,2
1Laboratory of Photoactive Nanocomposite Materials, St. Petersburg State University,
Ulyanovskaya 1, St.Petersburg, 198504 Russia
2Leibniz University of Hannover, Callinstrasse 3, Hannover 30167, Germany
*Corresponding authors: selivanov_chem@mail.ru; ruslan.kevorkyants@gmail.com
Abstract
We present structural and optoelectronic study of 1D piperidinium, pyridinium, and 3-
hydroxypyridinium lead trihalides. In contrast to the piperidinium and pyridinium species whose
1-
single inorganic chains [PbX3 ]n are separated by organic cations, the 3-hydroxypyridinium
compound is characterized by doubled inorganic chains. According to DFT the valence and
conduction bands of the piperidinium lead trihalides are composed of occupied p-orbitals of
halogen anions and unoccupied p-orbitals of Pb2+ cations. In contrast, the pyridinium species
feature low-laying cationic energy levels formed from the cation’s π*-orbitals. Thus, electronic
transitions between the cationic energy levels and valence bands require less energy than valence to
conduction band transitions in the case of piperidinium lead trihalides. The presence of OH group in
pyridinium ring leads to a batochromic shift of the cationic energy levels resulting in a decreased
energy of transitions from the cationic energy levels to a valence band. Electronic transitions
predicted by DFT are observable in experimental optical absorption and luminescence spectra. This
study paves a way to a creation of 1D perovskite-like structures with desired optoelectronic
properties.
Keywords: halide perovskites, 1D organic-inorganic lead halides, absorption and luminescence
spectra, DFT, self-trapped excitons, π-system.
Introduction
Organic-inorganic halide perovskites is a large class of photoactive materials attractive due
to unique properties enabling their applications in photonic, optoelectronic, and photovoltaic
devices. The best investigated organo-inorganic halide perovskites are 3D systems containing
methylammonium or formamidinium cations. Recently, a number of studies devoted to perovskites
of a lower dimensionality including 1D systems has been made public [1-3]. Owing to a quantum-

Dalton Transactions
Page 2 of 26
size effect 1D perovskites demonstrate peculiar excited states’ properties: quantum size effect, large
View Article Online
DOI: 10.1039/C9DT04543K
exciton binding energies and oscillator strengths, and the related non-linear optical properties [2-4].
Similar behavior is also demonstrated by low-dimensional organic-inorganic metal halide structures
where the basic structural units, octahedra MX6, form either face-sharing or edge-sharing quasi-1D
chains. To distinguish them from corner-sharing low-dimensional perovskites Stoumpos et al. name
such structures as “perovskitoids” [5]. Regardless the inorganic chain structure all these low-
dimensional compounds have become attractive materials from both fundamental and applied
points of view [6 - 11].
At present, the studies on 1D perovskite-like structures mostly concern methods of synthesis
and crystal structure. Only a few works report on the luminescent properties of such materials and
how they are affected by the type of organic cation [2, 9]. Moreover, there is no general approach
yet to choose an organic cation appropriate for a synthesis of desired 1D materials. 1D systems
known so far contain organic cations of different structures and compound classes [12 - 18]. The
only similarity between them is a presence of protonatable nitrogen atom. Therefore, a synthesis of
such materials is by no means trivial.
Several studies report on a synthesis of 1D perovskite-like compounds with saturated and
unsaturated cyclic amines. The former ones include reports on piperidinium and morpholinium
based 1D structures [19 - 28], while the latter describe pyrrolinium [6], pyridinium [29 - 35], and
tropylium [36] containing compounds. 1D systems with unsaturated amines are particularly
interesting since π-system of amines may interact with inorganic electronic subsystem affecting
band structures of the materials [35, 36]. Our earlier study on 1D pyridinium lead trihalides [35] has
shown an existence of energy levels formed from unoccupied π*-orbitals of aromatic pyridinium
cation within the compound bandgaps. The results reported in [36] on 1D material with tropylium
cation evidence that π-system of organic cation in these compounds interacts with inorganic
subsystem due to a significant contribution from carbon atoms into the conduction band. There also
exist few reports concerning the possibility of strong interaction between the electronic subsystem
of organic component (π-system) and inorganic chains in 2D perovskites [37 – 40].
In this study, we describe a preparation route of 1D piperidinium, pyridinium and 3-
hydroxypyridinium lead trihalides and report on their experimental crystal structures compared to
computed DFT data. Particularly, we have demonstrated a difference in the structural and optical
properties of solvated 2PipPbI3·DMFA and pure PipPbI3, that infers that previously reported results
[21, 22, 26] most likely related to solvated form of the crystals. We also report a new structure of 3-
OH-PyPbBr3 indicating that modification of pyridinium cation with OH group results in formation
1-
of doubled inorganic chains [PbX3 ]n and dimmers of 3-OH-Py+. Thereafter, the results of
experimental studies of the optical properties with stress on luminescence characteristics are

Page 3 of 26
Dalton Transactions
compared and correlated with those predicted by theoretical simulation. The effect of organic cation
View Article Online
DOI: 10.1039/C9DT04543K
type on the crystal structure, electronic and optical properties of the studied 1D lead trihalide
materials is analyzed and clarified.
Experimental techniques
Reactants
PbBr2, PbI2, pyridine, piperidine, and 3-hydroxypyridine were purchased from Sigma-
Aldrich. Amine hydrohalides were prepared using 46% HBr and 56% HI obtained from Reachim.
For the synthesis the following solvents were utilized: 96% ethanol, dimethylformamide, acetone,
and distilled water. The organic solvents were of a spectroscopic purity. A silicate gel was prepared
from a sodium metasilicate solution (Na2SiO3·9H2O) upon addition of acetic acid. Piperidinium
lead triiodide was prepared using lead acetate and KI (special purity grade).
Synthesis of amine hydrohalides
Pyridine hydrohalides
1 ml of pyridine was dissolved in 5 ml of ethanol. Concentrated hydrogen halogenides were
added drop-wise with constant stirring to the ethanol solutions. Hydrogen halogenides were taken in
an excess amount of 5% with respect to pyridine. The resulting solutions were kept at a temperature
0
0
of 3 C to 5 C for 12 h. Pyridine hydrohalide precipitates were filtrated and rinsed with ethanol.
Finally, they were dried over P2O5 in a desiccator for 12 h.
3-hydroxypyridine hydrobromide
5 g of 3-hydroxypyridine was dissolved in equimolar amount of concentrated HBr. After
evaporation of the solution the obtained precipitate was rinsed with acetone. The precipitate was
dried over P2O5 in a desiccator for 12 h.
Piperidine hydrohalides
10 ml of piperidine was dissolved in 20 ml of ethanol. Concentrated hydrogen halogenides
were added drop-wise with constant stirring to the ethanol solutions. Hydrogen halogenides were
taken in an excess amount of 5% with respect to pyridine. Hydrobromide solution was evaporated
to a volume of 5 ml. The solutions were kept in the temperature interval between 3 0C and 5 0C for
12 h. Piperidine hydrohalogenide precipitates were filtrated and rinsed with ethanol. Finally, they
were dried over P2O5 in a desiccator for 12 h.
Synthesis of 1D organic-inorganic lead trihalides
Hybrid lead trihalide crystals were prepared using the dimethylformamide (DMFA) [21 -
28]. Lead halide and cyclic amine hydrohalide (molar ratio 1:1) were dissolved in DMFA. The
resulting solutions underwent slow evaporation in the temperature interval between 50 0C and 60 0C

Dalton Transactions
Page 4 of 26
for 1 to 2 weeks. The obtained crystals were of a sufficiently good quality to be used in further
View Article Online
DOI: 10.1039/C9DT04543K
studies. As evident from the further characterization, using this preparation route all but the
piperidinium lead triiodide compound formed solvent-free crystals. In contrast, the piperidinium
lead triiodide formed the solvate 2C5H11NH2PbI3·DMFA.
Solvent-free piperidinium lead triiodide crystals were obtained applying a synthesis route in
silicate gel [19], which is used to prepare water-insoluble compounds [41, 42]. This route lacks
disadvantages of the DMFA method such as the use of toxic solvents and a preparation of amine
hydrohalides. Here, at first, a silicate sol must be prepared by mixing equal volumes of 0.4 M
sodium metasilicate and 2 M acetic acid. Prior to that piperidine was added to an acetic acid
solution such as to ensure its concentration of 0.2 М with respect to the silicate sol. To prevent
untimely gel formation sodium metasilicate was added drop-wise with intense stirring. The
prepared sol was poured into U–shaped tube. The sol to homogeneous gel transition took 24 h. In
one tube’s side a solution of 0.75 M lead acetate was added. The gel surface in another tube’s side
was covered with KI solution of the same concentration. 1D piperidinium lead triiodide crystals
(transparent, bright yellow, prism-shaped) form in the gel owing to a diffusion of Pb2+ and I- ions.
0
The crystal’s growth occurs at a temperature of 30 C and accomplishes within 3 weeks. The
crystals could be easily extracted from the gel. Prior to their use in subsequent studies the crystals
were rinsed in isopropyl alcohol.
Structural characterization
Single crystal X-ray diffraction
X-ray diffraction study of the obtained single crystals was carried out using the Agilent
Technologies diffractometer «Xcalibur» (Oxford Diffraction). The apparatus includes an anode –
the X-ray tube MoKα and a detector of reflected X-rays – the high-speed position-sensitive detector
CCD series «Atlas». XRD measurements were performed at a temperature of 100 К.
Powder X-ray diffraction
Powder X-ray diffraction studies were conducted with the high-resolution X-ray
diffractometer Bruker «D8 Discover» using the long focus X-ray tube CuKα anode. Reflected X-
rays were detected by the solid position-sensitive detector LYNXEYE. Measurements were
conducted at room temperature.
Temperature dependent X-ray diffraction
Phase transitions in powdered samples were studied using the high-resolution X-ray
diffractometer «Ultima IV» Rigaku with the low-temperature chamber Rigaku «R-300», X-ray tube
CuKα anode, and high-speed energy-dispersive detector of reflected X-rays DTEX/ULTRA.

Page 5 of 26
Dalton Transactions
0
0
Measurements were conducted in the temperature interval between -180 C and 200 C. The
View Article Online
temperature step of 20 0C and the speed of 5 0C/min were applied.
DOI: 10.1039/C9DT04543K
Diffuse reflectance spectroscopy
The diffuse reflectance spectroscopic studies of the powders were conducted using the
spectrometer Cary 5000 UV-Vis-NIR equipped with a diffuse reflectance accessory. Powders were
prepared by grinding the prepared crystals in an agate mortar.
Photoluminescence spectroscopy
Photoluminescence spectra of the synthesized lead trihalides were recorded at a temperature
of 77 K using the spectrofluorometer Agilent Cary Eclipse.
Computational approach
DFT calculations with periodic boundary conditions were performed using the PBE [43, 44]
density functional as implemented in the ABINIT 8.6.3 program [45]. The basis set in the form of
Troullier-Martins norm-conserving pseudopotentials was adopted [46]. Kinetic energy cut-off of 30
Hartree was set. For the purpose of geometry optimization and electronic band structure calculation
the Brillouin Zone (BZ) was sampled over the automatically generated Γ-point-centered 2x2x2 and
4x4x4 Monkhorst–Pack grids of k-points, respectively [47]. Band diagrams were plotted as follows.
Piperidinium based structures: at 161 k-points along the high symmetry path Г-A-C-D-D1-E-Y-Y1-
Z (P21/c; #14). Hydroxy-pyridinium based structures: at 141 k-points along the high symmetry path
Г-L-M-N-R-X-Y-Z (P-1; #2). Optical absorption spectra were simulated based on the Bethe-
Salpeter Equation [48].
Results and discussion
Experimental crystal structure
The analysis of X-ray data shows that the prepared single crystals possess 1D structure
1-
comprising of the inorganic chains [PbX3 ]n surrounded and therefore, separated from each other
by organic cations. According to the earlier report, piperidinium lead tribromide (PipPbBr3) belongs
to the monoclinic space group P21/c(n) [24]. The structure of PipPbBr3 determined in the present
work matches the earlier reported data. Inorganic chains of the bromide are directed toward
crystallographic axis b and isolated from each other by non-planar piperidinium cations. The
4-
octahedra [PbX6 ] share sides containing three Br- anions (face-sharing). Owing to the presence of
1-
inorganic structural units [PbX3 ], the synthesized materials resemble classic perovskites.
1-
However, they do not form a 3D network of corner-sharing [PbX3 ] octahedra. For this reason, we
will address them as “perovskitoids” following the term introduced elsewhere [5]. In contrast, upon
crystallization of the piperidinium lead triiodide from DMFA solution the solvent molecules

Dalton Transactions
Page 6 of 26
penetrate into crystal structure forming the solvate 2PipPbI3·DMFA. This leads to an alteration of
View Article Online
DOI: 10.1039/C9DT04543K
the lattice cell parameters and transforms the crystal structure to the orthorhombic space group
Pbca. Ball-and-stick models of the PipPbBr3 and 2PipPbI3·DMFA perovskitoids constructed on the
basis of the experimental XRD patterns are illustrated in Figure 1.
XRD data analysis of the powdered samples (Figure S1a) shows that the crystals of
PipPbBr3 are monophasic. Its experimental diffractogram perfectly matches the simulated one. The
simulation of XRD data for powdered samples was done based on the analysis and transformation
of a single crystal XRD data using the Mercury software. The solvate 2PipPbI3·DMFA crystals are
monophasic as well. However, its experimental diffractograms do not correspond to the simulated
one obtained from experimental XRD data for single crystals (Figure S1b). This implies changes in
a crystal structure taking place upon grinding of the compound single crystals. Presumably, during
this process DMFA molecules leave the 2PipPbI3·DMFA solvate crystal lattice forming solvent-
free PipPbI3. This is confirmed by the match of experimental diffractograms of “solvated” and
solvent-free PipPbI3 powdered samples.
Figure 1. Ball-and-stick models of PipPbBr3 (a and b) and 2PipPbI3·DMFA (c and d). Brown,
violet, light-grey, dark-grey, black, blue and red balls are Br, I, H, C, Pb, N and O atom,
respectively.

Page 7 of 26
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT04543K
Crystal structures of the pyridinium lead tribromide and triiodide (PyPbBr3, PyPbI3) were
studied by us earlier [35]. The compounds are isostructural and belong to the orthorhombic space
group Pna21. They have smaller crystal lattice parameters and higher symmetry as compared to 1D
piperidinium led trihalides. The compound’s ball-and-stick model exemplified by PyPbBr3 is
depicted in Figure 2 (a and b). Our structural data agree well with results reported elsewhere [29,
34].
The structure of 3-hydroxypyridinium lead tribromide (3-OH-PyPbBr3) belongs to the
triclinic space group P-1 and features the smallest crystal cell size comparing to all other studied
perovskitoids. In contrast to the piperidinium and pyridinium compounds it is characterized by
1-
4-
doubled inorganic chains [PbX3 ]n, in which octahedra [PbX6 ] share edges containing two Br-
anions. Moreover, the 3-hydroxypyridinium cations form hydrogen bonded dimers (Figure 2c). The
dimers are arranged into π-stacked layers (Figure 2d).
Figure 2. Ball-and-stick models of PyPbBr3 (a and b) and 3-ОН-PyPbBr3 (c and d). Brown, light-
grey, dark-grey, black, blue and red balls are Br, H, C, Pb, N and O atom, respectively.

Dalton Transactions
Page 8 of 26
Thus, one may conclude that in spite of a minor structural difference between the organic
View Article Online
DOI: 10.1039/C9DT04543K
cations (planar aromatic Py and 3-OH-Py vs. non-planar saturated Pip) their interaction with each
1-
other and with the inorganic chains [PbX3 ]n plays a significant role in what perovskitoid structure
is to be formed.
Our earlier XRD data analysis of the powdered sample PyPbI3 has revealed that it is
monophasic [35]. In contrast, the sample PyPbBr3 contained two phases. We attributed the presence
of an additional phase to mechanochemical effects arisen upon grinding. In the case of powdered
sample 3-OH-PyPbBr3 (Figure S1c) there is a good agreement between experimental and simulated
diffractograms. This implies no changes in crystal structure occurring upon grinding of the
compound.
Low-temperature X-ray diffraction studies on the perovskitoid powders reveal that the
compounds are stable in the temperature interval between -180o С and 130o С. At higher
temperatures they decompose giving rise to a lead halide phase. At a temperature of 200o С the
diffractograms demonstrate lead halide peaks and show no evidence of an original perovskitoid
phase. Additionally, in this temperature interval both PipPbI3 and PyPbI3 undergo a second-order
phase transition. In the case of PipPbI3 it occurs in the temperature interval between 0o С and 20o С
(Figure S2). As a result, the primary phase completely turns into the higher temperature one, which
remains stable up to a temperature of 140o С. In the case of PyPbI3 the phase transition occurs in the
temperature interval between -140o С and -20o С where both phases coexist. The room temperature
phase transition in PipPbI3 makes it a potential ferroelectric material as suggested elsewhere [6]. No
phase transformations were observed for all bromide perovskitoids regardless organic cation type.
Theoretical crystal and electronic structures
The obtained compounds XRD structures suffered from inaccurately resolved positions of
hydrogen atoms. Therefore, before electronic structure and optical absorption spectra calculations, a
geometry optimization was to be performed. During the optimization spatial symmetries of the
perovskitoids determined by XRD were imposed. The computed crystal lattice vectors are by ~5%
longer than their XRD values, that is presumably due to a neglect of hydrogen bonding. Such
structural changes affect the compounds’ electronic properties. For instance, in the case of
PipPbBr3, the crystal lattice vector elongation leads to a decrease in its bandgap by ~0.2 eV.
However, given structural flexibility of 1D perovskitoids we find such accuracy of theoretical
bandgap estimation acceptable. Figures 3 and 4 depict electronic band structures and density of
states (DOS) for PipPbBr3 and PipPbI3, respectively. The figures show that valence bands (VB) in
both piperidnium perovskitoids contain mostly contributions from halogen anions. In addition to
contributions from halogen anions the perovskitoid conduction bands (CB) have also contributions

Page 9 of 26
Dalton Transactions
from lead cations. The computed energies of indirect VB to CB electronic transitions E/Y to D1 in
View Article Online
DOI: 10.1039/C9DT04543K
PipPbBr3 and PipPbI3 are equal to 3.35 eV and 2.98 eV, respectively. The energies of direct
electronic transitions are predicted to be 3.52 eV and 3.00 eV in PipPbBr3 and PipPbI3,
respectively. Since the energies of both transition types are close but the probability of direct
electronic transitions is significantly higher than that of the indirect ones, the direct transitions are
expected be observed experimentally.
Figure 3.
Electronic band structure (left) and density of states (right) of PipPbBr3.
Figure 4. Electronic band structure (left) and density of states (right) of PipPbI3.
Band structures and DOS of PyPbBr3 and PyPbI3 are presented in Figures 5 and 6. We shall
recall that these data were obtained from our earlier work [34]. An inspection of PyPbBr3 and
PyPbI3 band structures reveals two narrow unoccupied energy levels, which are absent in the band
structures of PipPbBr3 and PipPbI3. According to the DOS plots these energy levels contain mainly

Dalton Transactions
Page 10 of 26
contributions from carbon atoms with some fraction from nitrogen atoms. As was established
View Article Online
DOI: 10.1039/C9DT04543K
earlier these energy levels are formed from antibonding molecular π-orbitals of the pyridinium
cations [35]. Presumably, upon photoexcitation, electrons from VBs can be transferred to the
above-mentioned unoccupied energy levels (UL) of the pyridinium cations. For the PyPbBr3 and
PyPbI3 perovskitoids the predicted energies of VB to UL and VB to CB (in brackets) electronic
transitions Y/Г to R are equal to 2.31 eV (3.18 eV) and 1.57 eV (2.43 eV), respectively.
Figure 5. Electronic band structure (left) and density of states (right) of PyPbBr3.
Figure 6. Electronic band structure (left) and density of states (right) of PyPbI3.
Similar to the band structures of PyPbBr3 and PyPbI3, the band structure of the 3-OH-
PyPbBr3 features two narrow energy levels containing contributions from carbon and nitrogen
atoms as well (Figure 7). The lower unoccupied energy states can be denoted as UL, while the

Page 11 of 26
Dalton Transactions
higher energy carbon states are located deeply inside of the unoccupied Br- and Pb2+ states forming
View Article Online
DOI: 10.1039/C9DT04543K
a conduction band. The DOS plot shows that the two narrow energy levels are comprised of
antibonding molecular π-orbitals of the 3-hydroxypyridinium cation. The computed energies of VB
to UL and VB to CB electronic transitions are equal to 2.54 eV (L/M to X) and 3.02 eV (L/M to
M), respectively. Owing to electron-donating effect of OH group we expect destabilization of the
narrow unoccupied energy levels in the 3-OH-PyPbBr3 as compared to those in PyPbBr3. As a
result, the lower energy narrow unoccupied level (UL) in PyPbBr3 (2.31 eV) becomes blue-shifted
in 3-OH-PyPbBr3 (2.54 eV). Similar blue shift is experienced by the higher energy narrow
unoccupied level in the conduction band rising in energy from 3.18 eV in PyPbBr3 to ~3.80 eV in
3-OH-PyPbBr3.
Therefore, one may conclude that the type of organic cation (aromatic vs. saturated) can
significantly affect electronic structure of 1D perovskitoids due to the presence of aromatic π-
system. Moreover, the change in electronic π-system of an aromatic cation induced by a functional
group (such as -OH) influences the entire electronic structure of 1D perovskitoids. Note that in a
zeroth-order approximation one may relate both VB to UL and VB to CB electronic transitions to
certain bands in absorption and luminescent spectra of the compounds. Accordingly, the alteration
in energies of electronic transitions caused by the presence of π-system and its tuning through
functional groups can lead to the corresponding alterations in absorption and luminescence spectra
of 1D perovskitoids.
Figure 7. Electronic band structure (left) and density of states (right) of 3-OH-PyPbBr3.
Theoretical optical absorption spectra
In this section we present simulated optical absorption spectra of the studied 1D compounds.
Figure 8 demonstrates such spectra for PipPbBr3 and PipPbI3 compounds. It shows that maxima of

Dalton Transactions
Page 12 of 26
the lowest energy absorptions bands are located at ~3.20 eV (PipPbBr3) and 3.05 eV (PipPbI3).
View Article Online
DOI: 10.1039/C9DT04543K
They match well energies of the VB to CB electronic transitions derived from band structure
computations: 3.35 eV (PipPbBr3) and 2.98 eV (PipPbI3), respectively.
Figure 8. Theoretical optical absorption spectra of PipPbBr3 (left) and PipPbI3 (right).
Simulated optical absorption spectra of PyPbBr3 and PyPbI3 are given in Figure 9. Maxima
of the absorption bands in the figure can not be easily identified. However, in our earlier work we
demonstrated that both compounds feature at least two absorption bands. In the case of PyPbBr3
their maxima are located at ~2.70 eV and ~3.10 eV, while PyPbI3 shows absorption band maxima at
~2.10 eV and 2.40 eV. One can see that an accordance between the absorption band maxima and
computed energies of VB to UL and VB to CB electronic transitions in these compounds is rather
moderate: PipPbBr3 (2.31 eV; 3.18 eV) and PipPbI3 (1.57 eV; 2.43 eV). Notably, this accordance is
significantly better for the higher energy transitions.
Figure 9. Theoretical optical absorption spectra of PyPbBr3 (left) and PyPbI3 (right).

Page 13 of 26
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT04543K
Figure 10 shows simulated optical absorption spectrum of 3-OH-PyPbBr3. The spectrum
features two absorption bands with the maxima at ~2.90 eV and ~3.70 eV whose accordance with
the computed energies of both VB to UL and VB to CB electronic transitions (2.54 eV and 3.02 eV,
respectively) is rather poor.
Figure 10. Theoretical optical absorption spectrum of 3-OH-PyPbBr3.
However, the simulated absorption spectrum of 3-OH-PyPbBr3 does confirm a possibility of two
different electronic transitions corresponding to different mechanisms of light absorption (VB to
UL and VB to CB) yielding two absorption bands in the optical spectrum.
Experimental optical absorption spectra
Spectral characteristics of the powdered samples were studied by diffuse reflectance
spectroscopy. Given the absorbance coefficient A=1-R, where R is a reflection coefficient,
absorption spectra of the compounds can be plotted (Figure 11). According to [25], in PipPbBr3 and
PipPbI3 electronic transitions proceed from occupied p-orbitals of halogen anions to vacant p-
1-
orbitals of lead cations and halogen anions of 1D inorganic chains [PbX3 ]n. Thus, absorption
spectra of the piperidinium compounds are only due to intrachain transitions. The piperidinium
cations do not contribute into the spectra because their intracation transitions are laying in a far UV
region [49]. This conclusion is supported by our DFT results (Figures 3 and 4) and other modeling
studies [25]. They show that molecular orbitals of piperidinium cations neither electronically
interact with the inorganic subsystem nor do they significantly contribute into the perovskitoid band
structures. As a consequence, they do not manifest themselves in the perovskitoid optical spectra. In
other words, piperidinium cations play only a structural role. We also would like to emphasize a

Dalton Transactions
Page 14 of 26
nice agreement between diffuse reflectance spectra of powdered PipPbI3 and 2PipPbI3·DMFA
View Article Online
DOI: 10.1039/C9DT04543K
samples. This match confirms an absence of DMFA molecules in the grinded “solvate” sample.
Both PipPbBr3 and PipPbI3 possess valence and conduction bands typical of semiconductors
(see Figure 3 and Figure 4) that enables the use of diffuse reflectance spectra for their bandgap
estimation. For this purpose, we employed the Tauc plot method and a method based on a spectral
position of the maxima of first derivative of logarithm of the modified Kubelka-Munk function
lnF(E)E as a function of Е (eV) [50].
Figure 11. Optical absorption spectra of perovskitoids. (a): PipPbBr3 (1) and PipPbI3 (2); (b):
PyPbBr3 (1), PyPbI3 (2), and 3-OH-PyPbBr3 (3).
In the case of direct electronic transitions whose probability is higher than that of the
indirect ones, the Tauc method provides bandgaps of 3.51 eV and 2.86 eV for PipPbBr3 and
PipPbI3, respectively (Figures S3a and S3b). The second method of bandgap estimation gives 3.50
eV and 2.68 eV in PipPbBr3 and PipPbI3, respectively (Figure S4). Thus, the bandgaps determined
by both methods agree well with each other and a reasonably good agreement is also observed
between experimental and theoretical direct bandgaps. We shall recall that the predicted DFT
bandgaps equal 3.35 eV and 2.98 eV for the PipPbBr3 and PipPbI3, respectively (Figures 3 and 4).
Unlike piperidinium-based 1D perovskitoids,, their analogs containing aromatic cations
possess unoccupied π-system energy levels. Consequently, the terms either “direct” or “indirect”
transitions are not appropriate for the optical transitions between the valence bands and unoccupied
levels in organic cations. Therefore, neither Tauc plot nor first derivative method can be applied to
estimate band gaps in these 1D compounds. In this case we use a simple method of first derivative
absorption maxima to estimate energies of possible optical transitions.
A comparison of the absorption spectra in Figure 11 shows that given the same halogen
anion, electronic transitions in the perovskitoids with unsaturated organic cations have lower
energies than those in the saturated piperidinium analogs. This difference is better seen on the

Page 15 of 26
Dalton Transactions
dependencies of a first derivative of the absorption spectra on photon energy (Figure S5).
View Article Online
DOI: 10.1039/C9DT04543K
According to the data shown in Figure S5, in PyPbBr3, in addition to electronic transitions
characteristic of inorganic chains in PipPbBr3 (~3.50 eV), there is a longer wavelength transition
with a maximum at ~3.10 eV. Similarly, in PyPbI3, in addition to electronic transition present in
PipPbI3 (~2.70 eV) there is also a longer wavelength transition at ~2.40 eV. In the 3-OH-PyPbBr3,
the longer wavelength absorption edge is located at ~2.50 eV. Thus, optical absorption spectra of
the PyPbBr3, PyPbI3, and 3-OH-PyPbBr3 containing unsaturated organic cations along with
absorption bands corresponding to electronic transitions characteristic of inorganic subsystems in
PipPbBr3 and PipPbI3 also demonstrate absorption bands at longer wavelengths.
These longer wavelength absorption bands in spectra measured for the PyPbBr3 (3.10 eV),
PyPbI3 (2.40 eV), and 3-hydroxypyridinium lead bromide (2.50 eV) cannot be assigned to the
1-
electronic transitions in the inorganic chains [PbX3 ]n, which require a higher photon energy
corresponding to the shorter wavelength absorption bands. Neither they can be assigned to intra-
cation electronic transitions since those corresponding absorption lays in a near-UV spectral region
(see Figure S6). The nature of the longer wavelength absorption bands becomes clear based on our
previous study [34]. Therein, we have demonstrated the presence of two flat conduction bands in
the pyridinium perovskitoids. Both bands are formed from pyridinium π-system (Figures 5, 6 and
7). A comparison of our experimental and theoretical data let us conclude that the longer
wavelength absorption bands in the pyridinium perovskitoids correspond to the electronic
transitions from VB comprised of occupied p-orbitals of halogen anions to the states composed of
unoccupied molecular π*-orbitals of the pyridinium cation (UL). This conclusion is based on a
good match between theoretical VB to UL energies of electronic transitions in the pyridinium
perovskitoids (3.18 eV – PyPbBr3; 2.43 eV - PyPbI3) [35] with the positions of band maxima in
their absorption spectra (~3.10 eV – PyPbBr3; and ~2.40 eV PyPbI3).
Finally, we note that in the compounds with unsaturated organic cations absorption band
intensities due to electronic transitions between organic and inorganic subsystems are smaller than
1-
those due to [PbX3 ]n intrachain electronic transitions (Figure 11).
Experimental photoluminescence and excitation spectra
None of the studied 1D synthesized materials exhibit luminescent emission at room
temperature. However, the compounds do demonstrate luminescence at a liquid nitrogen
temperature. Figure 12 shows photoluminescence emission and excitation spectra of single crystal
and powdered PipPbBr3 samples recorded at a temperature of 77 K. It is wise to note that
luminescence measurements in the temperature range from 77 K to 113 K demonstrate no alteration

Dalton Transactions
Page 16 of 26
of luminescence spectra of all bromide perovskitoids that is in agreement with XRD data.
View Article Online
DOI: 10.1039/C9DT04543K
Therefore, bromide structures are stable at the temperature of luminescence measurements.
As can be seen, there is a good accordance between spectra registered for both
monocrystalline and powdered samples. The emission spectrum is comprised of a single broad band
with a maximum at ~630 nm (~2.0 eV). Single crystal photoluminescence of PipPbBr3 at a
temperature of 15 K was studied previously [23, 24]. Two bands with the maxima at 415 nm (~3.0
eV) and 650 nm (~1.9 eV) were found in the emission spectra. Positions and intensities of these
bands as a function of temperature were studied in [24]. We found that intensity of the shorter
wavelength band decays rapidly as temperature increases and vanishes completely at a temperature
of 80 K. The longer wavelength band undergoes a blue shift. This is why in PipPbBr3 we have
observed only a longer wavelength luminescence band, which shifts to a shorter wavelength region
at a temperature of 77 K.
The photoluminescence excitation spectra of PipPbBr3 are comprised of several bands in the
region from 250 nm to 400 nm. They correspond to intrachain electronic transitions of the inorganic
subsystem in the absorption spectrum of the compound. The observed large Stokes shift of
photoluminescence suggests a strong electron-phonon interaction in the excited state. Based on the
previous study of PipPbBr3 [23, 24], the photoluminescent band at 630 nm could be assigned to
charge-separated self-trapped excitons localized in 1D inorganic chains. In such exciton, a hole
-
center is reported to consist of several Br- anions with the dimer Br2 being the most probable
candidate. An electron component of exciton, on the other hand, is associated with Pb2+ cation [23].
Figure 12. Photoluminescence (right curves) and photoluminescence excitation (left curves) spectra
of powdered (1, 4) and monocrystalline (2, 3) PipPbBr3 at t = 77 К.

Page 17 of 26
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT04543K
Similar to PipPbBr3, single crystals of PipPbI3 demonstrate only one band (590 nm; 2.1 eV)
in the photoluminescence spectrum (Figure 13). Note, that solvate 2PipPbI3·DMFA sample has a
photoluminescence band almost coinciding with that of the solvent-free sample. Thus, DMFA
molecules in the solvate do not affect its emission spectrum. Photoluminescence of the
monocrystalline PipPbI3 at a liquid helium temperature was studied earlier [21, 22, 26]. Therein,
PipPbI3 synthesis route resembles the one employed in the present study. This let us presume that
the cited studies dealt with PipPbI3 solvates as well. The photoluminescence spectra recorded at t =
37 К and t = 5 К [21, 26] reveal only one band at 590 nm. Our photoluminescence spectra obtained
at t = 77 К feature this band as well. Therefore, we conclude that for this compound the spectral
composition of bands and positions of their maxima do not depend on temperature.
Figure 13. Photoluminescence spectra of monocrystalline 2PipPbI3·DMFA (1), powdered
2PipPbI3·DMFA (2), and monocrystalline PipPbI3 (3) at t = 77 K. Photoluminescence
excitation spectra of monocrystalline PipPbI3 (4), powdered 2PipPbI3·DMFA (5),
monocrystalline 2PipPbI3·DMFA (6) at t = 77 К.
The photoluminescence excitation spectrum of the solvent-free PipPbI3 (Figure 13) has a
complex structure with a longer wavelength edge at 400 nm. This agrees well with the compound’s
absorption spectrum whose bands can be ascribed to intrachain electronic transitions. Akin to
PipPbBr3, the large Stokes shift of the observed PipPbI3 photoluminescent band is presumably due
to charge-separated self-trapped excitons localized in 1D inorganic chains. Analogously, a hole
-
center in such exciton is probably the dimer I2 , while an electron part of exciton is associated with
Pb2+ cation.
In contrast to the photoluminescence excitation spectrum of the solvent-free PipPbI3, the
excitation spectrum of the monocrystalline 2PipPbI3·DMFA demonstrates bands in a visible
spectral range. The shift of spectral bands toward longer wavelength region is presumably due to

Dalton Transactions
Page 18 of 26
the presence of DMFA molecules of the solvate in the crystal structure. One reason for this
View Article Online
DOI: 10.1039/C9DT04543K
difference in excitation could be a difference in crystal structures of the solvate and solvent-free
compound. Figure 13 also shows the photoluminescence excitation spectrum of the grinded
2PipPbI3·DMFA single crystals. The spectrum reveals a significant decrease of the band intensities
due to the solvate that is indicative of a formation of solvent-free powdered PipPbI3 structure
caused by mechanochemical action.
The photoluminescence studies on PyPbBr3 showed that its spectrum depends on the
excitation wavelength. For instance, irradiation of the sample by a light with a wavelength shorter
than 360 nm (Figure 14) gives rise to the broad luminescence band with a maximum at 650 nm. If
the excitation wavelength is longer than 360 mn, an additional narrow band at 525 nm appears in
the compound’s photoluminescence spectrum. An intensity of this band increases with an increase
of the excitation wavelength, while that of the longer wavelength band notably decreases. Thus, the
photoluminescence spectrum of PyPbBr3 at t = 77 К is comprised of two bands generated by a light
of different spectral composition as opposed to a single band in the photoluminescence spectrum of
the piperidinium based species.
This difference is clearly seen in the excitation spectrum of PyPbBr3 (Figure 15) and infers
that the observed bands result from different electronic excitation transitions yielding the emission
from different states in PyPbBr3. Thus, the longer wavelength luminescence with maximum at 650
nm is excited at the edge of the longer wavelength excitation located at 400 nm. As was said above,
the absorption spectrum of PyPbBr3 in this region is comprised of bands formed by electronic
1-
transitions between Br- and Pb2+ ions within the inorganic chains [PbX3 ]n. In general, this
excitation spectrum composed of several bands with clear maxima resembles in both shape and
band positions those in excitation spectra of the piperidinium based materials. Given the significant
Stokes shift of the longer wavelength emission one can presume that it originates from self-trapped
excitons localized in 1D inorganic chains. Thus, this can be considered as a rather general
mechanism in other 1D perovskitoids.

Page 19 of 26
Dalton Transactions
View Article Online
DOI: 10.1039/C9DT04543K
Figure 14. Photoluminescence spectra of monocrystalline PyPbBr3 (a) and 3-OH-PyPbBr3 (b) at T
= 77 К; λexc: 390 nm (1), 365 nm (2), 330 nm (3), t = 77 К.
At the same time, when the excitation spectrum is recorded at a shorter wavelength
maximum of emission (520 nm) the majority of excitation bands are located in a visible spectral
region. The absorption spectrum of PyPbBr3 in this region corresponds to electronic transitions
1-
from p-orbitals of Br- anions of the inorganic chains [PbX3 ]n to molecular π*-orbitals of
pyridinium cations (Figure S5). Similar to the piperidinium perovskitoids, we assume that during
excitation of PyPbBr3 a charge-separated exciton may be formed with a hole center being probably
-
a dimer Br2 . The hole’s electronic counterpart could then be a π-system of the pyridinium cation. If
so, the short wavelength band observed in the photoluminescence excitation spectrum of PyPbBr3 is
due to a radiative transition from π-system of the pyridinium cation down to a valence band formed
mainly from p-orbitals of Br- anions. Such transition was demonstrated both experimentally and
theoretically by J.-i. Fujisawa and coworkers [49, 51 - 53]. They showed that in 1D methylviologen
lead iodide electronic excitations proceed from occupied p-orbitals of I- anions and Pb2+ cations
onto vacant molecular orbitals of divalent methylviologen cations.
Figure 15. Excitation spectra of PyPbBr3 photoluminescence recorded at λem = 650 nm (1) and λem
= 520 nm (2), 3-OH-PyPbBr3 at λem = 700 nm (3) and λem = 545 nm (4).
The computed PyPbBr3 band structure in Figure 5 supports the above assumption.
Energetically, the short wavelength maximum of the luminescent band at 520 nm (2.36 eV) is close
to the predicted lower energy level of pyridinium cation (UL) (2.31 eV). Thus, absorption of
PyPbBr3 at 3.1 eV leads to electronic transition from valence band to conduction band which is
followed by a nonradiative transition onto the lower energy level of pyridinium cation (UL) with
the subsequent radiative transition back to the ground state resulting in luminescence.

Dalton Transactions
Page 20 of 26
Therefore, the PyPbBr3 photoluminescence excitation spectra demonstrate that its
View Article Online
DOI: 10.1039/C9DT04543K
pyridinium cation energy levels can be populated not only directly by a light of a suitable
wavelength but also through excitation of inorganic chains followed by electron transition from the
conduction band to the lowest unoccupied levels of pyridinium cations. The second possibility is
indicated by a short wavelength tale in the excitation spectra in the near-UV region. Notably, an
efficiency of such transitions is rather low as demonstrated by low intensities of excitation bands in
the corresponding spectral region.
In contrast to PyPbBr3 photoluminescence spectra that of the monocrystalline PyPbI3
features just one band with a maximum at 625 nm (Figure 16). Its position is independent of an
incident light’s wavelength. The figure shows that the powdered PyPbI3 sample has the same
photoluminescence spectrum as that of the monocrystalline PyPbI3.
Figure 16. Photoluminescence (right curves) and photoluminescence excitation (left curves) spectra
of powdered PyPbI3 (2, 4) and monocrystalline PyPbI3 (1, 3) at t = 77 К.
The origin of this band becomes clear from the photoluminescence excitation spectrum
featuring several bands with easy to distinguish maxima in the range 250 nm – 400 nm. Given the
large Stokes shift we presume that the observed photoluminescence is due to self-trapped excitons
1-
in the inorganic chains [PbX3 ]n.
Thus, unlike PyPbBr3, PyPbI3 does not manifest photoluminescence due to electronic
transitions between energy levels of pyridinium cation and the perovskitoids VB. In fact, its
luminescence behavior is rather similar to that of PipPbI3. This observation of the luminescence
quenching can be an evidence of strong interaction between organic and inorganic electronic

Page 21 of 26
Dalton Transactions
subsystems [37] or such transitions might be forbidden (strongly restricted) due to the presence of
View Article Online
DOI: 10.1039/C9DT04543K
heavy iodine anions.
Since PyPbI3 does not demonstrate a specific luminescence, in the following the effect of functional
group (-OH) on luminescence tuning will be demonstrated only on PyPbBr3 and 3-OH-PyPbBr3
compounds.
Figure 14b shows that emission spectra of 3-OH-PyPbBr3 also depend on a wavelength of
the actinic light. At wavelengths shorter than 340 nm the compound’s photoluminescence spectrum
is comprised of a broad longer wavelength band with a maximum at ~700 nm and low intensity
short wavelength peak at 545 nm. In contrast, if the excitation wavelengths are longer than 340 nm
then the spectrum demonstrates only a single intense band at 545 nm, similar to behavior observed
for PyPbBr3.
Figure 15 shows the excitation spectrum of photoluminescence for 3-OH-PyPbBr3. As one
can see, the excitation spectrum recorded for emission wavelength 700 nm corresponding to the
maximum of longer wavelength band, are laying in a region 250-340 nm. This excitation spectral
region is narrower than the corresponding region in the excitation spectrum of PyPbBr3
photoluminescence. The bands in this spectral region can be ascribed to intrachain electronic
transitions of the inorganic subsystem. The longer wavelength luminescence band on the other hand
is ascribable to self-trapped exciton decay.
The excitation spectrum of 3-OH-PyPbBr3 photoluminescence recorded at 545 nm
corresponding to the maximum of shorter wavelength emission band features a superposition of the
excitation bands in a visible spectral region that is similar to PyPbBr3. These excitation spectrum
can be associated with the absorption band in Figure S5 and be assigned to electronic transitions
1-
between the inorganic chains [PbX3 ]n and the 3-OH-Py+ cation. Therefore, just as in the case of
the short wavelength PyPbBr3 emission we attribute this luminescent band at 545 nm to the
radiative transition from energy levels formed from π*-orbitals of the organic cation 3-OH-Py down
to energy levels comprised of p-orbitals of Br- anions (VB).
Noteworthy, this conclusion is supported by the red shift of the luminescent band as
compared to the similar band of PyPbBr3 at 525 nm. This shift is due to the energy lowering of
molecular π*-orbitals of the organic cation 3-OH-Py+ caused by electron-donating -OH group. The
same shift is observed in the case of the largest wavelength absorption band in 3-OH-PyPbBr3 as
compared to its position in PyPbBr3.
The photoluminescence excitation spectrum of 3-OH-PyPbBr3 indicates that energy levels
of 3-OH-Py cation can be populated either directly (see luminescence spectra presented in Figure
S7) or through excitation onto higher energy unoccupied levels of the inorganic subsystem with a
subsequent nonradiative deexcitation. This conclusion is evidenced by an overlap of the excitation

Dalton Transactions
Page 22 of 26
spectra 3 and 4 (see Figure 14) registered for different emission wavelengths, λ = 650 нм and λ
rec
View Article Online
DOI: 10.1039/C9DT04543K
rec = 520 nm respectively.
Conclusions
We reported on comparative studies of 1D organic-inorganic lead trihalide with heterocyclic
cations PipPbBr3, PipPbI3, PyPbBr3, PyPbI3, and 3-OH-PyPbBr3 in order to understand the role of
organic cations in crystal and electronic structures of 1D compounds. In contrast to other studied
1D compounds, 3-OH-PyPbBr3 synthesized for the first time contains doubled inorganic chains
1-
[PbX3 ]n. It is demonstrated that an interaction between organic cations affects the crystal structure
of 1D perovskite and perovskitoid materials. Moreover, the presence of π-system in organic cation
also strongly changes luminescent properties of the 1D perovskitoid materials. The simulated
theoretical as well as recorded experimental optical absorption spectra of PyPbBr3, 3-OH-PyPbBr3,
and PyPbI3 feature bands due to electronic transitions within the inorganic 1D chains as well as
between these chains and the organic cations. Photoluminescence spectra of PyPbBr3 and 3-OH-
PyPbBr3 reveal emission bands due to self-trapped excitons localized in the inorganic chains and
those owing to electronic transitions from molecular π*-orbitals of the organic cations down to p-
orbitals of the halogen anions (VB) (see scheme of the electronic transitions presented in Figure
17).
Figure 17. The scheme of photophysical processes in perovskitoid materials: PipPbX3 (a), PyPbX3
(b), 3-OH-PyPbBr3 (c), here X – Br; I.
In other words, the perovskitoids with saturated organic cations demonstrate radiative transitions
related only to self-trapped excitons. At the same time, in addition to luminescence attributed to
self-trapped excitons in inorganic component, perovskitoids with unsaturated organic cations
1-
feature two optical transitions: 1) inside the inorganic chains [PbX3 ]n whose electronic transitions
form an optical absorption spectrum primarily in the UV region and a luminescence spectrum with
a maximum in the range 580 - 650 nm; and 2) electronic transitions between energy levels of the

Page 23 of 26
Dalton Transactions
organic cations and VBs giving rise to the longer wavelength bands in the absorption spectra and to
View Article Online
DOI: 10.1039/C9DT04543K
the shorter wavelength luminescence (520–550 nm). A possibility of tuning the position of this
luminescence band by introducing the electron-donating group -OH into pyridinium ring is
demonstrated both theoretically and experimentally.
Acknowledgment
This work was financially supported by Russian Foundation for Basic Research (19-03-
00836-a) and performed within the project by Saint-Petersburg State University (Pure ID
51124539) to create the laboratory “Photoactive nanocomposite materials”. The authors are grateful
to the Resource Center for X-ray Diffraction Studies, the Resource Center “Nanophotonics”, and
the Computational Resource Center of Research Park at St-Petersburg State University.
References
1.
A. Lemmerer, A thesis submitted to the Faculty of Science, University of the Witwatersrand,
Johannesburg, in fulfillment of the requirements for the degree of Doctor of Philosophy,
Johannesburg, 2007.
2.
3.
4.
H. Lin, C. Zho, Y. Tian, T. Siegrist, and B. Ma, ACS Energy Lett. 2018, 3, 54-62.
M.I. Saidaminov, O.F. Mohammed, and O.M. Bakr, ACS Energy Lett. 2017, 2, 889-896.
I.B. Koutselasy, L. Ducassez, and G.C. Papavassiliou, J. Phys.: Condens. Matter 1996, 8,
1217-1227.
5. C. C. Stoumpos, L. Mao, C. D. Malliakas, and M. G. Kanatzidis, Inorg. Chem. 2017, 56, 56−73
6.
H.-Y. Ye, Y. Zhang, D.-W. Fu, and R.-G. Xiong, Angew. Chem. Int. Ed. 2014, 53, 11242-
11247.
7.
H.-B. Duan, H.-R. Zhao, X.-M. Ren, H. Zhou, Z.-F. Tiana, and W.-Q. Jina, Dalton Trans.
2011, 40, 1672-1683.
8.
9.
S.-P. Zhao and X.-M. Ren, Dalton Trans. 2011, 40, 8261-8272.
Z. Yuan, C. Zhou, Y. Tian, Y. Shu, J. Messier, J.C. Wang, L.J. van de Burgt, K.
Kountouriotis, Y. Xin, E. Holt, K. Schanze, R. Clark, T. Siegrist, and B. Ma, Nat. Commun.
2017, 8, 14051.
10. S. Elleuch, Y. Abid, A. Mlayah and H. Boughzala, J. Raman Spectrosc. 39 (2008) 786-792.
11. G.-E. Wang, G. Xu, M.-S. Wang, J. Sun, Z.-N. Xu, G.-C. Guo and J.-S. Huang, J. Mater.
Chem. 2012, 22, 16742-16744.
12. D.G. Billing and A. Lemmerer, Acta Cryst. 2004, C60, m224-m226.
13. G.-E. Wang, X.-M. Jiang, M.-J. Zhang, H.-F. Chen, B.-W. Liu, M.-S. Wang and G.-C. Guo,
CrystEngComm 2013, 15, 10399-10404.

Dalton Transactions
Page 24 of 26
14. K. Xiong, W. Liu, S.J. Teat, L. An, H. Wang, T.J. Emge, J. Li, J. Solid State Chem. 2015,
View Article Online
DOI: 10.1039/C9DT04543K
230, 143-148.
15. A. Triguia, H. Abida, A. Mlayahb, Y. Abida, Synth. Met. 2012, 162, 1731-1736.
16. D.G. Billing and A. Lemmerer, CrystEngComm 2006, 8, 686-695.
17. J.-H. Im, J. Chung, S.-J. Kim and N.-G. Park, Nanoscale Res. Lett. 2012, 7, 353.
18. A. Mancini, P. Quadrelli, G. Amoroso, C. Milanese, M. Boiocchi, A. Sironi, M. Patrini, G.
Guizzetti, L. Malavasi, J. Solid State Chem. 2016, 240, 55-60.
19. G.V. Gridnova, E.A. Ziger, V.M. Koshkin, S.V. Lindemann, Yu.T. Struchkov and V.E.
Shklover, Doklady Akademii Nauk, 1984, 278, 414.
20. A.B. Corradi, S. Bruni, F. Cariati, A.M. Ferrari, A. Saccani, F. Sandrolini, P. Sgarabotto,
Inorganica Chim. Acta, 1997, 254, 137-143.
21. A. Nagami, K. Okamura, T. Ishihara, Physica B 1996, 227, 346-348.
22. T. Fukumoto, M. Hirasawa, T. Ishihara, J. Lumin. 2000, 87-89, 497-499.
23. J. Azuma, K. Tanaka, Int. J. Mod. Phys. B 2001, 15, 3646-3650.
24. J. Azuma, K. Tanaka and K. Kan’no, J. Phys. Soc. Jpn. 2002, 71, 971-977.
25. J. Azuma, K. Tanaka, M. Kamada and K. Kan’no, J. Phys. Soc. Jpn. 2002, 71, 2730-2735.
26. I. Akimoto, M. Sakai, K. Kan’no, J. Lumin. 2004, 108, 31-35.
27. T. Kawasaki, M. Takahashi, T. Ohhara, I. Tanaka, K. Kusaka, T. Hosoya, T. Yamada, and K.
Kurihara, J. Phys. Soc. Jpn. 2012, 81, 094602.
28. K. Ramalingam, C. Rizzoli, T. Rajaraman and M. Prabu, Main Group Met. Chem. 2016, 39,
175-181.
29. M. Geselle and H. Fuess, Z. Kristallographie -New Crystal Structures 1997, 212, 235.
30. J. Jóźków, R. Jakubas, G. Bator, and A. Pietraszko, J. Chem. Phys. 2000, 16, 7249-7246.
31. K. Yamada, T. Ohtani, S. Shirakawa, H. Ohki, T. Okuda, T. Kamiyama, and K. Oikawa, Z.
Naturforsch., 1996, 51a, 739-744.
32. T. Okuda, Y. Aihara, N. Tanaka, K. Yamada, S. Ichiba, J. Chem. Soc. Dalton Trans. 1989,
631-634.
33. K. Yamada, T. Tsuda, C. Holst, T. Okuda, H. Ehrenberg, I. Svoboda, H. Krane, H. Fuess,
Bull. Chem. Soc. Jpn. 2001, 74, 77-83.
34. T.-M. Huong Duong, S. Nobusue, and H. Tada, Appl. Phys. Express 2018, 11, 115502.
35. N.I. Selivanov, A.A. Murashkina, R. Kevorkyants, A.V. Emeline and D.W. Bahnemann,
Dalton Trans. 2018, 74, 16313-16319.
36. A.E. Maughan, J.A. Kurzman, and J.R. Neilson, Inorg. Chem. 2015, 54, 370-378.
37. B. Saparov and D. B. Mitzi, Chem. Rev. 2016, 116, 4558−4596.

Page 25 of 26
Dalton Transactions
38. Wouter T.M. Van Gompel, Roald Herckens, Kristof Van Hecke, Bart Ruttens, Jan D’Haen,
View Article Online
DOI: 10.1039/C9DT04543K
Laurence Lutsen, and Dirk Vanderzande, СhemComm 2019, 55, 2481-2484
39. María C. Gélvez-Rueda, Wouter T. M. Van Gompel, Roald Herckens, Laurence Lutsen, Dirk
Vanderzande, and Ferdinand C. Grozema, J. Phys. Chem. Lett. 2020, 11, 824-830.
40. James V. Passarelli, Daniel J. Fairfield, Nicholas A. Sather, Mark P. Hendricks, Hiroaki Sai,
Charlotte L. Stern, and Samuel I. Stupp, J. Am. Chem. Soc. 2018, 140, 23, 7313-7323
41. H.K. Henisch, J. Dennis and J.I. Hanoka, J. Phys. Chem. Solids 1964, 26, 493-500.
42. J. Dennis and H.K. Henisch, J. Electrochem. Soc. 1967, 114, 263-266.
43. J.P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865-3868.
44. J.P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett. 1997, 78, 1396-1396 (erratum).
45. X. Gonze, B. Amadon, P.M. Anglade, J.-M. Beuken, F. Bottin, P. Boulanger, F. Bruneval, D.
Caliste, R. Caracas, M. Cote, T. Deutsch, L. Genovese, Ph. Ghosez, M. Giantomassi, S.
Goedecker, D. Hamann, P. Hermet, F. Jollet, G. Jomard, S. Leroux, M. Mancini, S. Mazevet,
M.J.T. Oliveira, G. Onida, Y. Pouillon, T. Rangel, G.-M. Rignanese, D. Sangalli, R. Shaltaf,
M. Torrent, M.J. Verstraete, G. Zérah, J.W. Zwanziger, Comput. Phys. Commun., 2009, 180,
2582-2615.
46. N. Troullier, J.L. Martins, Phys. Rev. B 1991, 43, 1993-2006.
47. H.J. Monkhorst and J.D. Pack, Phys. Rev. B 1976, 13, 5188–5192.
48. E.E. Salpeter and H.A. Bethe, Phys. Rev. 1951, 84, 1232-1242.
49. J. Fujisawa, N. Tajima, K. Tamaki, M. Shimomura, and T. Ishihara, J. Phys. Chem. C, 2007,
111, 1146-1149.
50. M.V. Kharlamova, N.A. Sapoletova, A.A. Eliseeev, A.V. Lukashin, Tech. Phys. Lett. 2008,
34, 288-291.
51. J. Fujisawa and T. Ishihara, Phys. Rev. B. 2004, 70, 113203.
52. J. Fujisawa and N. Tajima, Phys. Rev. B. 2005, 72, 125201.
53. J. Fujisawa and G. Giorgi, Phys. Chem. Chem. Phys. 2014, 16, 17955-17959.

Dalton Transactions
Page 26 of 26
View Article Online
DOI: 10.1039/C9DT04543K