Understanding the origin of broad-band emission in CH3NH3PbBr3

Article abstract of DOI:10.1039/d0tc05003b

Broad-band emissions related to self-trapped excitons (STEs) in the sub-bandgap region in organic-inorganic hybrid perovskites and double perovskite crystals have drawn attention in recent times due to their potential in optoelectronic device applications. In this study, we have shown that the formation of STEs in CH₃NH₃PbBr₃single crystals can be controlled using a suitable sample synthesis procedure. We have observed a broad-band emission (FWHM ～ 80 nm) for crystals whose crystallization is fast, whereas it was absent if we follow slow crystallization procedures. Using UV-Visible absorption spectroscopy, temperature-dependent photoluminescence (PL), time-resolved PL (TRPL), and dc magnetization studies, we concluded that defect-assisted extrinsic self-trapping is dominant here over the intrinsic self-trapping process and excess Pb atoms in interstitial positions are predominantly responsible for the extrinsic self-trapping process.

Full text of DOI:10.1039/d0tc05003b

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Understanding the origin of broad-band emission
in CH3NH3PbBr3
Cite this: J. Mater. Chem. C, 2021,
9, 2793
ab
ab
Pronoy Nandi,
†
Chandan Giri ‡a and Dinesh Topwal
*
Broad-band emissions related to self-trapped excitons (STEs) in the sub-bandgap region in organic–
inorganic hybrid perovskites and double perovskite crystals have drawn attention in recent times due to
their potential in optoelectronic device applications. In this study, we have shown that the formation of
STEs in CH3NH3PbBr3 single crystals can be controlled using a suitable sample synthesis procedure. We
have observed a broad-band emission (FWHM B 80 nm) for crystals whose crystallization is fast,
whereas it was absent if we follow slow crystallization procedures. Using UV-Visible absorption spectro-
scopy, temperature-dependent photoluminescence (PL), time-resolved PL (TRPL), and dc magnetization
studies, we concluded that defect-assisted extrinsic self-trapping is dominant here over the intrinsic
self-trapping process and excess Pb atoms in interstitial positions are predominantly responsible for the
extrinsic self-trapping process.
Received 22nd October 2020,
Accepted 7th January 2021
DOI: 10.1039/d0tc05003b
rsc.li/materials-c
the growth rate,17 pyramidal, rhombohedral or cuboid sized
crystals were obtained. Reduced structural imperfections, and a
Introduction
Organic–inorganic lead halide hybrid perovskites are increasingly variety of morphologies and polymorphic forms are the major
used in optoelectronic devices like photovoltaic cells,1,2 bright advantages of this process. However, the slow growth rate and
light-emitting diodes,3 photodetectors,4 lasers,5 etc. due to the probability of solvent inclusion in the sample are the main
their remarkable optoelectronic properties like high carrier disadvantages.
mobility,6 high photoconversion efficiency,7 tunable bandgap
Another suitable method of crystal growth is the inverse
(with composition),8 long charge carrier diffusion length9 and temperature crystallization (ITC) method wherein the solubility
rare trap density.10 However, defects in these systems can greatly of organic–inorganic metal halides in particular solvents is
influence the luminescence efficiency, electrical conductance, high at room temperature but decreases rapidly with increasing
minority carrier lifetime, and other key aspects of device perfor- temperatures. g-Butyrolactone (GBL), N,N-dimethylformamide
mance. It was observed that vacancies, self-interstitials and (DMF) and dimethylsulphoxide (DMSO) were chosen for ITC growth
antisites have deleterious effects on the performance of a laser, of CH3NH3PbI3/CH(NH2)2PbI3, CH3NH3PbBr3/CH(NH2)2PbBr3
LED and photodetector.11–13 Thus, to enhance the device perfor- and CH3NH3PbCl3 single crystals, respectively, and by setting a high
mance, it is essential to optimize the sample growth process, temperature of the heating bath, single crystals quickly preci-
understand the origins of defects and reduce the same.
The synthesis of organic–inorganic hybrid halide perovskite CH3NH3PbI3 so far (two-inch) using this method.23 Because of a
crystals was first reported by Weber in 1978 by the solution faster growth rate (B1 mm3 ꢀ1), this method can be adopted
pitated.18–22 Liu et al. have grown the largest single crystals of
h
temperature lowering (STL) method where methylammonium for commercial purposes, but structural imperfection and solvent
(MA) and lead-acetate were dissolved in hydrohalide acid, corrosion might arise due to the high growth temperature which
heated at 100 1C and subsequently cooled to 45 1C resulting deteriorates the crystalline quality of these perovskites. However, a
in millimeter-sized crystals.14 Subsequently by judiciously novel space-limited ITC method has recently been developed to
changing the precursors (use methylammonium halide instead fabricate large (120 cm2) single crystal films of CH3NH3PbBr3 on
of methylammonium and hydrohalide acid)15,16 and varying FTO glass for narrow bandgap-photodetectors.24
D. Shi et al. proposed an alternative strategy to synthesize
a Institute of Physics, Sachivalaya Marg, Bhubaneswar 751005, Odisha, India
b Homi Bhabha National Institute, Training School Complex, Anushakti Nagar,
Mumbai 400094, India. E-mail: dinesh.topwal@iopb.res.in
single crystals of metal halide perovskites using the anti-solvent
vapor assisted crystallization (AVC) method.25 This method is
used to synthesize materials which are highly soluble in one
solvent (good solvent) but have poor solubility in other solvents
(bad solvent) and diffusion of a suitable bad solvent vapor into
the precursor solution results in high-quality crystals.26,27
† Present address: Department of Energy Science, Sungkyunkwan University,
Suwon 16419, Republic of Korea.
‡ Present address: Department of Chemistry and Biochemistry, The University of
Alabama, Tuscaloosa, Alabama, 35487-0336, USA.
This journal is The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 2793ꢀ2800 | 2793

View Article Online
Paper
Journal of Materials Chemistry C
Single crystals of metal halide perovskites using the AVC
method were synthesized using toluene and dichloromethane
(DCM) as antisolvents.28–30 The proposed room temperature
crystallization technique is important in particular for thermally
sensitive samples, where organic groups are present. Further-
more, the presence of an antisolvent changes the solvent activity
of the solution which has a more profound effect on the
morphology and polymorphic form than that of ITC and STL
methods, whereas, the use of additional raw material (i.e.,
antisolvent) might create problems in the purity of crystals and
may also enhance the capital cost.
Though several groups have synthesized and char-
acterized14–23,25–30 single crystals of metal halide perovskites using
the above-discussed methods, no comparative study is available,
focusing on the crystallographic and spectroscopic nature of the
obtained samples prepared by different methods. In this report, a
comparative study of heat capacity, room temperature powder X-ray
diffraction (XRD), room temperature absorbance and temperature-
dependent photoluminescence (PL) on single crystals of
CH3NH3PbBr3 synthesized by STL, ITC, dichloromethane (DCM)
and ethylacetate (EA) diffused AVC methods has been carried out.
Samples grown by these four methods are found to be crystal-
lographically identical, however, discrepancies are observed in their
optical response. A broad-band emission spectrum between 600 nm
and 750 nm was observed in the crystals grown by rapid crystal-
lization techniques like STL, ITC and DCM diffused AVC methods,
whereas this broad-band emission spectrum is absent in the EA
diffused AVC method (slow crystallization technique). The large
Stokes shift, broadband emitting spectrum with full width at half
maximum (FWHM) larger than 80 nm and high recombination time
(B0.48 ms) indicated that these broad-band emissions are related to
self-trapped excitons (STEs).13 Short-range lattice distortion due to
the localization of free excitons (FEs) is responsible for the formation
of STEs and can be divided into two types: (a) intrinsic and (b)
extrinsic self-trapping.31–34 Transient distortions of the lattice
create intrinsic self-trapping, example: structural distortion in InBr6
octahedra in the lead free indium-based counterparts,35,36 while
lattice distortion due to permanent defects in materials like doping
Fig. 1 Room temperature powder X-ray diffraction patterns (upper panel)
and heat capacity (lower panel) of CH3NH3PbBr3 grown by ITC, STC, DCM
and EA diffused AVC methods.
or alloying causes the extrinsic self-trapping, example: Pb or Br Sharp and well-resolved diffraction peaks indicate that the crystallite
vacancies in interstitial positions.37
size of CH3NH3PbBr3 grown by the above-mentioned techniques is
In this article, we have made a comparison of formation quite large and is free from lattice strain. Temperature-dependent
energy calculations with the experimentally obtained energy heat capacity measurements were performed to identify the phase
levels and suggest that excess Pb atoms in interstitial positions transition temperatures (Fig. 1(b)).39 It was observed that all four
are predominantly responsible for the extrinsic self-trapping samples undergo crystallographic phase transitions at the same
process which is supported by dc-magnetization studies.
temperatures. With the decrease of temperature from 280 K to 70 K,
sharp anomalies were observed at 236.9 K, 154.6 K and 149.3 K for
all the samples indicating a change in internal energies arising from
the crystallographic phase transition of the samples. A comparison
with previous reports suggest that all four samples undergo a
Results and discussion
Powder X-ray diffraction (PXRD) and heat capacity measure- crystallographic phase transition from cubic to tetragonal (I), tetra-
ments were performed on all four CH3NH3PbBr3 samples to gonal (I) to tetragonal (II) and tetragonal (II) to orthorhombic at
assess their structure and phase purity. Fig. 1(a) shows that 236.9 K, 154.6 K and 149.3 K, respectively.40 From these observations
samples are crystalline in nature and diffraction peaks of we might conclude that all the four samples are crystallographically
all four samples appear at the same 2y position, (indexed in identical.
the figure), which match well with those of the CH3NH3PbBr3
The optical properties of all four as-synthesized samples
were investigated using steady-state absorption and PL studies.
30,38
%
perovskite with a cubic phase structure (Pm3m space group).
2794 | J. Mater. Chem. C, 2021, 9, 2793ꢀ2800
This journal is The Royal Society of Chemistry 2021

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 2 depicts the room temperature UV-visible absorption
spectra of CH3NH3PbBr3 synthesized via ITC, DCM diffused
AVC, STL and EA diffused AVC techniques. Each spectrum was
acquired from a pellet (diameter of 10 mm) made by crushing
an equal amount (120 mg) of single crystals of CH3NH3PbBr3
and consists of three broad features; (i) a sub-bandgap absorp-
tion at low energy (600 nm and above), (ii) a strong and sharp
peak at 556 nm (2.23 eV) followed by (iii) band-edge transitions
which start from 506 nm (2.45 eV).39,41 Steeper band edges of
CH3NH3PbBr3 suggest that it is a direct bandgap semiconductor
with ordered nature. Apart from dual transitions at 506 nm and
556 nm (with an error bar of ꢁ2 nm) a broad hump between
600 nm and 750 nm was observed for samples synthesized by
ITC, STL and DCM diffused AVC methods. Furthermore, the
relative absorbance of the broad hump was maximum for
samples grown by the ITC method, followed by DCM diffused
AVC and STL methods, respectively. In contrast, no broad
absorbance peak was observed for the sample grown by the EA
diffused AVC method, suggesting that the optical response can
vary depending on the sample synthesis conditions.
Fig. 3 shows the temperature-dependent (77 K to 300 K)
steady-state PL spectra of CH3NH3PbBr3. The overall features of
the presented PL spectra from samples synthesized using
different processes are somewhat similar and can be broadly
characterized by three peaks marked as 1, 2, and 3 in the
figures, but with different intensities. For the sample grown by
the ITC method three PL emission peaks were observed
(i) peak-1 at 543.5 nm, (ii) peak-2 at 561 nm and (iii) peak-3
at 649 nm at 77 K, which are consistent with the literature.42–44
Also, as the temperature is increased, peak-1 is blue-shifted,
while peak-2 is red-shifted continuously to 529.5 nm and
573.5 nm, respectively, at room temperature (300 K), and peak
Fig. 3 Temperature-dependent photoluminescence spectra of CH3NH3
PbBr3 grown by (a) ITC, (b) STC, (c) DCM diffused and (d) EA diffused AVC
method. Inset shows the variation of first two peaks with temperature.
3 is red-shifted. It may here be noted that peak 3 shows strong
temperature dependence and considerable FWHM as compared
to peak 1 and peak 2. This observation is consistent for samples
grown by all 4 methods. The bandgap energy of most of the
semiconductors tends to decrease (red shift) with increasing
temperature and at higher temperature, interatomic bonds are
weakened and weaker bonds mean less energy required to send
an electron in the conduction band. However, the blue-shifting
behavior of peak-1 in the hybrid perovskite is inconsistent with
that of traditional semiconductors and can be explained using
the Varshni model where the temperature coefficient of band-
gap (dEg/dT) can be expressed as:39,45
ꢀ
ꢁ
ꢀ
ꢁ
ꢀ
ꢁ ꢀ
ꢁ
dEg
dT
dEg
dT
dEg
d lnðvÞ
dT
¼
þ
(1)
d lnðvÞ
p
v
T
p
Here the first term represents the electron–phonon coupling
through the deformation coupling, which represents a decrease
in bandgap with an increase of temperature and the second
term is the lattice dilation term which causes the increase in
the bandgap with temperature. It is suggested that the blue
shifting nature of peak-1 arises because of valence band maxima
stabilization due to lattice dilation with an increase in the
temperature.39
The origin of dual emission in PL spectra (peak-1 and
peak-2) is a widely debated topic in this series of compounds
with no consensus so far; few authors claim it to be surface and
bulk effect,18,19 some refer it to be arising from bound and free
Fig. 2 Room temperature UV-VIS absorbance spectra of four
CH3NH3PbBr3 samples grown by ITC, STC, DCM and EA diffused AVC
methods.
This journal is The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 2793ꢀ2800 | 2795

View Article Online
Paper
Journal of Materials Chemistry C
exciton emission,42 while others believe that tetragonal inclusions excitonic emission. Furthermore, careful observation of the present
in the low temperature are responsible.43 However, based on our data reveals that excitonic peak positions (peak 1 and 2) show a
experiments and band structure calculations,46 we believe that redshift of 41–45 meV for ITC, 32–35 meV for the DCM diffused
methylammonium orientations inside the PbBr6 octahedra can AVC and 27–28 meV for the STL method grown sample compared
result in two closely spaced minima in free energy vs. molecular to the EA diffused AVC sample throughout the temperature region.
orientation plot.47 The electronic structures associated with these This result clearly indicates that the local configurations of trap
two configurations confirm a small shift in the valence band and states and electronic properties are quite sensitive to their charge
no shift in the conduction band and these closely spaced valence states introduced by the permanent lattice defects or e-STEs and
bands might be responsible for dual transitions in hybrid can result in the shifting of valence band maxima (VBM) or
perovskites.39,44
conduction band minima (CBM).54
The results of the temperature-dependent steady-state PL
Time-correlated single-photon counting (TCSPC) of peak-1
spectra of the samples grown by STL and DCM diffused AVC and peak-3 (presented in Fig. 4) could shed some light on the
methods are depicted in Fig. 3(b) and (c), respectively, and are recombination time (of electrons and holes) and help in
similar to the samples grown by the ITC method (Fig. 3a). identifying permanent lattice defects or e-STE results of the
However, the relative intensity of peak-3 (at 77 K) w.r.t. peak-1 is same for samples prepared by the ITC method. The lifetime of
maximum for the sample grown by the ITC method and is the energy levels responsible for peak-1 and peak-2 (not shown
minimum for the sample grown by the STL method. By looking here) is relatively small (ns range), while, that of peak-3 is three
at the position of peak 3 in the energy level diagram, one can orders of magnitude higher in ms range confirming a different
assume it to be originating from some deep level trap states origin of peak-3. The recombination lifetime of CH3NH3PbBr3
which appears at the interband positions and originates due to was determined by measuring PL decay at the emission peak
the rapid growth rate at high temperature and fast diffusion by wavelengths and the curves are fitted with the biexponential
the low boiling point solvent (DCM: 39.6 1C). A similar observa- function of time (t) given by:
tion was made in the UV-visible absorbance study of the
I(t) = I1 expꢀt /t + I2 expꢀt /t
(2)
1
1
2 2
corresponding samples (Fig. 2). Furthermore, it is important
to note that peak-3 is absent at all temperatures for the samples
grown at room temperature by high boiling point antisolvent
diffusion (EA: 77.1 1C) methods, where the growth rate is
relatively lower. Thus, in conclusion, broadband peak-3 was
observed when the sample was synthesized by rapid crystal-
lization methods (ITC, STL, and DCM diffused AVC methods),
while no broad peak was observed for samples grown by the
relatively slow crystallization method (EA diffused AVC
method).
where ti is the decay time and Ii is the weight factor of each
decay channel. The peak-1 fitting yielded following parameters
t1 = 17.8 ns with 59.32% weight% and t2 = 3.5 ns with 40.68%
weight%. Here, slow (17.8 ns) and fast (3.5 ns) recombination
times are assigned as bulk and surface recombination.18,22,39 In
contrast, the lifetime of energy levels associated with peak-3
was as large as t1 = 0.48 ms with 57.12% weight% and t2 = 3.32 ms
with 42.88% weight%, suggesting that the transitions respon-
sible for emission of peak-3 originate from defect levels.
The broad emission/absorption in the midgap region for
semiconductors and insulators is best described in terms of
intrinsic self-trapping of electrons and excitons (STEs).31–34
Additionally, it has been suggested that the presence of intrinsic
defects or impurity atoms in the crystal can add further inho-
mogeneity to the excited-state potential surface, leading to
escalation of the broad mid-bandgap emission response.37
Accordingly, STEs as well as carrier trapped permanent lattice
defects are believed to be responsible for broadband white light
emission in zero dimensional perovskites like (C4H14N2)2
In2Br10,36 (Cs2InBr5ꢂH2O),35 two dimensional hybrid perovskites
like ((N-MEDA)[PbBr4] (N-MEDA
=
N1-methyl ethane-1,2-
diammonium),48 (EDBE)[PbX4] (EDBE = 2,20 (ethylenedioxy)bis
(ethylammonium); X = Cl and Br)),49 in double perovskites
like Cs2AgInCl6,50 Cs2NaInCl6,51 Cs4SnBr6, etc.52 and in ion
(Yb3+/Er3+) doped all inorganic hybrid perovskite CsPb
(Br/Cl)3.13 Similar to our steady state photoluminescence obser-
vations, Priante et al.,42 Fu et al.44 and Shi et al.53 also observed
broad-band PL emission between 560 nm and 690 nm at low
temperature (77 K) for CH3NH3PbBr3, which disappeared above
180 K. Using variable fluence and excitation (below band-edge)
dependent steady state and time resolved PL studies, they ascribed
this feature (appearing as peak 3 in our case) to self-trapped
Fig. 4 Time correlated single photon counting (TCSPC) data corres-
ponding to 650 nm and 531 nm (Inset) emission peaks at 77 K of
CH3NH3PbBr3 grown by the ITC method. Y-Axis is normalized at
t = 0 ns and presented in log-scale for better visualization. Hollow dots
represent the experimental data.
2796 | J. Mater. Chem. C, 2021, 9, 2793ꢀ2800
This journal is The Royal Society of Chemistry 2021

View Article Online
Journal of Materials Chemistry C
Paper
Depending on the location of the energy states in the band in metal halide perovskite crystals. These experimental observations
diagram, defect states might be characterized as shallow or suggest that point defects due to self-doping of Pb atoms are
deep levels. Shallow defects have energy levels within few tens responsible for the creation of the midgap defect states
of meV from respective band edges, whereas deep level defects evidenced here. Furthermore, formation energy calculations
reside within the middle third of the bandgap. Alternatively, predict that unipolar self-doping of Pb at interstitial positions
deep levels are associated with highly localized wave functions can create interband defect states,55 while density functional
whereas shallow level wave functions are as extended as the theory (DFT) calculations also predict mid-gap defect states in
far-reaching coulomb potential. Based on these definitions, the calculated absorption spectra of CH3NH3PbI3.58 If the inter-
almost all dominant defects in CH3NH3PbBr3 can be character- stitial Pb atoms are indeed responsible for defect states of
ized from shallow levels except the following cases (deep level methylammonium lead halide samples, then uncompensated
effects):55–57 (i) CH3NH3(MA), Pb, Br vacancies (VMA, VPb, VBr), spins of Pb can align in the applied magnetic field direction like
(ii) MA, Pb, Br interstitial (MAi, Pbi, Bri), (iii) MA on Pb (MAPb
and Pb on MA (PbMA) cation substitution and (iv) MA on Br
)
Mn-doped CH3NH3PbBr3 micro-rods.59
Magnetization measurements (M vs. H) as a function of the
(MABr), Pb on Br (PbBr), Br on MA (BrMA), Br on Pb (BrPb) antisite applied field were performed on all CH3NH3PbBr3 samples at
substitutions. Formation energy calculations reveal that MAi, 100 K and 300 K, which are presented in Fig. 6. It is observed
PbMA, VBr and MABr are shallow donors whereas Bri, MAPb, VMA
,
that the sample synthesized by the EA diffused AVC method is
and VPb are shallow acceptors which are not responsible for the diamagnetic in nature with susceptibility w = ꢀ4.43 ꢃ 10ꢀ9
observed midgap defect states. The remaining four defects Pbi, (inset of Fig. 6) which is consistent with the literature
PbBr, BrMA and BrPb produce deep levels in the bandgap of reports.59,60 However, other three samples synthesized by
CH3NH3PbBr3.55 Formation energy calculations reveal that Pbi STL, DCM diffused AVC and ITC methods show a mixture of
and PbBr levels are situated nearer the conduction band, para and ferromagnetic behavior. It is interesting to observe
whereas, BrMA and BrPb exist closer to the valence band, and
are responsible for the interband transition.
For a better understanding of the defect states, we substi-
tuted Br ion sites with Cl and synthesized single crystals of
CH3NH3PbCl3 via ITC and EA diffused AVC methods. The
steady-state PL spectra at 77 K of both the Cl replaced samples
are shown in Fig. 5. Here too we observe two peaks at 403 nm
(peak-1) and 410.5 nm (peak-2) for both the samples and an
additional peak at 641.5 nm (peak-3) for the sample synthesized
by the ITC method while peak-3 was absent for the sample
synthesized by the EA diffused AVC method. This result is
similar to the one from CH3NH3PbBr3 thereby confirming that
halogen atoms do not participate in deep level defect formation
Fig. 6 M–H loops of CH3NH3PbBr3 grown by ITC, STC, DCM and EA
diffused AVC methods taken at (a) 100 K and (b) room temperature (300 K).
Fig. 5 Temperature-dependent photoluminescence spectra of CH3NH3
Inset shows the diamagnetic behavior of CH3NH3PbBr3 grown via the EA
PbCl3 grown by ITC and EA diffused AVC methods at 77 K.
diffused AVC method.
This journal is The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 2793ꢀ2800 | 2797

View Article Online
Paper
Journal of Materials Chemistry C
that the magnetic moment is maximum for the sample Experimental section
grown by the ITC method (0.039 emu gꢀ1 and 0.018 emu gꢀ1),
Materials
followed by the DCM diffused AVC method (0.013 emu gꢀ1
All chemicals were purchased from Sigma-Aldrich and used
and 0.009 emu gꢀ1) and the STL method (0.003 emu gꢀ1 and
0.002 emu gꢀ1) at a magnetic field of 60 kOe for 100 K and 300 K,
respectively. The observed trend in the magnitude of magnetiza-
tion is directly proportional to the number of interstitial Pb
atoms (point defects: interstitial and vacancy) and is consistent
with conclusions from PL results. The observed phenomena
(Fig. 6) in CH3NH3PbBr3 can hence be best described as defect
induced magnetism as observed in graphite,61 graphene,62
and ZnO.63 Recently, Sil et al. observed defect-induced ferro-
magnetism at room temperature in the CH3NH3PbI3 system and
suggested that the origin of ferromagnetic ordering in this
as received without further purification: hydrobromic acid
(HBr; 48 wt% in water), methylamine (CH3NH2; 40 wt% in
water), PbBr2 (99.99%), PbI2 (99.99%), and g-butyrolactone
(GBL; 99.9%).
Synthesis of CH3NH3Br
Single crystals of CH3NH3Br were synthesized by reacting
CH3NH2 and HBr in a molar ratio of 1.2 : 1. HBr was added
dropwise into CH3NH2 in a flask under a nitrogen atmosphere
in an ice bath for 3 hours; the resulting solution was evaporated
at 60 1C in a rotary evaporator to remove the unreacted solvent
and white crystals of CH3NH3Br were obtained.
material is due to atomic defects.64 Nafradi et al. also observed
´ ´
ferromagnetic behavior in MAPbI3 after substituting Pb2+ with
Mn2+ and fabricated optically tuned magnetic switches.65 Thus,
we believe that defect states present in the sample are respon-
sible for the deviation from the expected diamagnetic behavior
and that the number of interstitial Pb atoms in the system is the
highest for samples grown by the ITC method and minimum for
the samples grown by the STL method and is almost negligible
for the case of the sample grown by the EA diffused AVC method.
This interesting magnetic behavior for a direct bandgap semi-
conductor like CH3NH3PbBr3 where magnetism can be induced
and controlled by the amount of defects opens up new avenue
for utilizing organic–inorganic halide perovskites for spintronic
applications as well.
Synthesis of CH3NH3PbBr3 single crystals
Method-1: Inverse temperature crystallization (ITC) method.
0.367 g lead(II) bromide (1 mM) and 0.111 g methylammonium
bromide (1 mM) were dissolved in 10 ml N,N-dimethyl-
formamide (DMF) and the precursor solution was kept at
80 1C. After 3 h nice orange-colored single crystals of
CH3NH3PbBr3 were grown.
Method-2: Solution temperature lowering (STL) method. In
this technique, we have reduced the temperature of precursor
solution (same as method-1) from 100 1C to room temperature
over 48 h as reported earlier. This is the most common and
traditional way to synthesize CH3NH3PbBr3 perovskite single
crystals. It is also observed that crystallization increases at
around 80 1C.
Conclusions
Anti-solvent vapor assisted crystallization (AVC) method
In summary, we have carried out an in-depth comparative study
of the structural, optical and magnetic properties of
CH3NH3PbBr3 grown by ITC, STL, DCM and EA diffused AVC
methods through powder XRD, heat capacity, room tempera-
ture UV-VIS and temperature-dependent PL spectroscopy as
well as M–H studies. Identical PXRD patterns and heat capacity
confirm the phase purity and the presence of three crystal-
lographic phase transitions in all four samples synthesized by
different techniques. However, the presence of a broad peak
between 600 nm and 700 nm was observed in the sub-bandgap
region (bandgap of CH3NH3PbBr3 is approx. 506 nm) in the
room temperature absorption and temperature-dependent PL
studies in case of samples grown by ITC, STL and DCM diffused
AVC methods, which was absent in the sample grown by the EA
diffused AVC method. The time-resolved PL study confirms that
the lifetime associated with this broad peak is higher (Bms)
than that of band-edge transitions (Bns). A comparison with
the formation energy calculations confirmed that excess Pb
atoms in interstitial positions create defect levels for the
samples whose growth rate is high. This observation was also
confirmed by dc magnetization studies. Thus, a novel approach
to synthesize defect-free crystals is put forth which will be
We know that the solubility of a compound in one solvent (good
solvent) decreases due to the slow incorporation of another
solvent (bad solvent) in which the solute is not soluble (an anti-
solvent). This can be done in two ways; one is by direct contact
between the saturated solution and the anti-solvent (‘‘layering’’)
and other by allowing the vapors of the anti-solvent to slowly
diffuse into a saturated solution. In this method, ethyl acetate
(EA) and dichloromethane (DCM) vapors are used to reduce the
solubility of the solute.
Method-3: Dichloromethane (DCM) diffused AVC method.
In this method, the precursor solution (same as method-1)
filled vial was kept in a DCM environment and orange crystals
were grown within 1–2 days.
Method-4: Ethyl acetate (EA) diffused AVC method. In order
to achieve very slow crystallization, we have taken a high boiling
point solvent (ethyl acetate: 77.1 1C). Then, the vial containing
the CH3NH3PbBr3 precursor solution (same as method-1) was
kept in an EA environment. With slow diffusion of EA into the
solution, nice orange crystals were grown within 8–10 days.
Characterization techniques
useful to fabricate optoelectronic devices using metal halide X-Ray powder diffraction of all samples was performed by
perovskite single crystals.
using a Bruker D8 Advance diffractometer using Cu Ka
2798 | J. Mater. Chem. C, 2021, 9, 2793ꢀ2800
This journal is The Royal Society of Chemistry 2021

View Article Online
Journal of Materials Chemistry C
Paper
11 S. Mahajan, W. D. Johnston, M. A. Pollack and R. E. Nahory,
Appl. Phys. Lett., 1979, 34, 717–719.
12 Y. Rakita, I. Lubomirsky and D. Cahen, Mater. Horiz., 2019,
6, 1297–1305.
13 F. Jiang, X. Jiang, W. Zheng, Y. Ouyang, Y. Zhang, L. Li,
P. Fan, H. Zhao, Y. Li, Y. Jiang, X. Zhu, X. Zhuang, Y.-X. Feng
and A. Pan, J. Mater. Chem. C, 2020, 8, 13976–13981.
radiation (l = 1.34 Å) to confirm the phase purity of all four
samples. The temperature dependence of specific heat of all
four samples was recorded using a Physical Property Measure-
ment System (PPMS) Evercool II, from Quantum Design, USA.
Heat capacity measurements were performed in the tempera-
ture range between 70 K and 300 K. The optical absorption
spectra were recorded using an ultraviolet-visible-near-infrared
spectrophotometer (Shimadzu, 3101PC) in the wave-
length range of 400–800 nm. Temperature-dependent photo-
luminescence (PL) spectra of all the samples were recorded in
the temperature range of 77 K to 300 K using a FL920 spectro-
meter from Edinburg instruments with an excitation energy of
325 nm. Time-resolved PL experiments were performed on the
same FL 920 spectrometer by means of time-correlated single-
photon counting (TCSPC). For both the measurements, sam-
ples were mounted on a copper plate in a cryostat and cooled by
liquid nitrogen. The dc-magnetization measurements of all
four samples were performed at 300 K using a PPMS system
equipped with VSM head from Quantum Design, USA.
¨
14 D. Weber, Zeitschrift fur Naturforsch. B, 1978, 33, 1443–1445.
15 Y. Dang, Y. Liu, Y. Sun, D. Yuan, X. Liu, W. Lu, G. Liu, H. Xia
and X. Tao, CrystEngComm, 2015, 17, 665–670.
16 P. Nandi, C. Giri, B. Joseph, S. Rath, U. Manju and
D. Topwal, J. Phys. Chem. A, 2016, 120, 9732–9739.
17 Z. Lian, Q. Yan, Q. Lv, Y. Wang, L. Liu, L. Zhang, S. Pan,
Q. Li, L. Wang and J.-L. Sun, Sci. Rep., 2015, 5, 16563.
18 M. I. Saidaminov, A. L. Abdelhady, B. Murali, E. Alarousu,
V. M. Burlakov, W. Peng, I. Dursun, L. Wang, Y. He,
G. Maculan, A. Goriely, T. Wu, O. F. Mohammed and
O. M. Bakr, Nat. Commun., 2015, 6, 7586.
19 A. A. Zhumekenov, M. I. Saidaminov, M. A. Haque,
E. Alarousu, S. P. Sarmah, B. Murali, I. Dursun,
X.-H. Miao, A. L. Abdelhady, T. Wu, O. F. Mohammed and
O. M. Bakr, ACS Energy Lett., 2016, 1, 32–37.
Conflicts of interest
20 G. Maculan, A. D. Sheikh, A. L. Abdelhady,
M. I. Saidaminov, M. A. Haque, B. Murali, E. Alarousu,
O. F. Mohammed, T. Wu and O. M. Bakr, J. Phys. Chem. Lett.,
2015, 6, 3781–3786.
21 Y. Zhang, Y. Liu, Y. Li, Z. Yang and S. (Frank) Liu, J. Mater.
Chem. C, 2016, 4, 9172–9178.
22 P. Nandi, C. Giri, D. Swain, U. Manju, S. D. Mahanti and
D. Topwal, ACS Appl. Energy Mater., 2018, 1, 3807–3814.
23 Y. Liu, Z. Yang, D. Cui, X. Ren, J. Sun, X. Liu, J. Zhang,
Q. Wei, H. Fan, F. Yu, X. Zhang, C. Zhao and S. (Frank) Liu,
Adv. Mater., 2015, 27, 5176–5183.
There are no conflicts to declare.
Acknowledgements
DT would like to acknowledge SERB, DST, India (funding under
project file no. CRG/2020/003108) for financial support.
References
1 J. Y. Kim, J.-W. Lee, H. S. Jung, H. Shin and N.-G. Park,
Chem. Rev., 2020, 120, 7867–7918.
2 P. Yadav, D. Prochowicz, E. A. Alharbi, S. M. Zakeeruddin
24 H.-S. Rao, W.-G. Li, B.-X. Chen, D.-B. Kuang and C.-Y. Su,
Adv. Mater., 2017, 29, 1602639.
¨
25 D. Shi, V. Adinolfi, R. Comin, M. Yuan, E. Alarousu, A. Buin,
Y. Chen, S. Hoogland, A. Rothenberger, K. Katsiev, Y. Losovyj,
X. Zhang, P. A. Dowben, O. F. Mohammed, E. H. Sargent and
O. M. Bakr, Science, 2015, 347, 519 LP–522.
26 N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. Il
Seok, Nat. Mater., 2014, 13, 897–903.
and M. Gratzel, J. Mater. Chem. C, 2017, 5, 7799–7805.
3 Q.-W. Liu, S. Yuan, S.-Q. Sun, W. Luo, Y.-J. Zhang, L.-S. Liao
and M.-K. Fung, J. Mater. Chem. C, 2019, 7, 4344–4349.
4 J. Miao and F. Zhang, J. Mater. Chem. C, 2019, 7, 1741–1791.
5 C. M. Raghavan, T.-P. Chen, S.-S. Li, W.-L. Chen, C.-Y. Lo,
Y.-M. Liao, G. Haider, C.-C. Lin, C.-C. Chen, R. Sankar,
Y.-M. Chang, F.-C. Chou and C.-W. Chen, Nano Lett., 2018,
18, 3221–3228.
27 P. Nandi, C. Giri, U. Bansode and D. Topwal, AIP Conf. Proc.,
2017, 1832, 80065.
6 Y. Wang, Y. Zhang, P. Zhang and W. Zhang, Phys. Chem. 28 D. Niesner, O. Schuster, M. Wilhelm, I. Levchuk, A. Osvet,
Chem. Phys., 2015, 17, 11516–11520.
7 S. De Wolf, J. Holovsky, S.-J. Moon, P. Loper, B. Niesen,
S. Shrestha, M. Batentschuk, C. Brabec and T. Fauster, Phys.
Rev. B, 2017, 95, 75207.
¨
M. Ledinsky, F.-J. Haug, J.-H. Yum and C. Ballif, J. Phys. 29 C. Wang, B. R. Ecker, H. Wei, J. Huang, J.-Q. Meng and
Chem. Lett., 2014, 5, 1035–1039.
Y. Gao, Phys. Chem. Chem. Phys., 2017, 19, 5361–5365.
8 C. M. Sutter-Fella, Y. Li, M. Amani, J. W. Ager, F. M. Toma, 30 P. Nandi, S. K. Pandey, C. Giri, V. Singh, L. Petaccia,
E. Yablonovitch, I. D. Sharp and A. Javey, Nano Lett., 2016,
16, 800–806.
U. Manju, S. D. Mahanti and D. Topwal, J. Phys. Chem. Lett.,
2020, 11, 5719–5727.
9 G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, 31 M. D. Smith, A. Jaffe, E. R. Dohner, A. M. Lindenberg and
¨
M. Gratzel, S. Mhaisalkar and T. C. Sum, Science, 2013,
H. I. Karunadasa, Chem. Sci., 2017, 8, 4497–4504.
32 Q. Li, Z. Chen, B. Yang, L. Tan, B. Xu, J. Han, Y. Zhao,
J. Tang and Z. Quan, J. Am. Chem. Soc., 2020, 142,
1786–1791.
342, 344–347.
10 K. X. Steirer, P. Schulz, G. Teeter, V. Stevanovic, M. Yang,
K. Zhu and J. J. Berry, ACS Energy Lett., 2016, 1, 360–366.
This journal is The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 2793ꢀ2800 | 2799

View Article Online
Paper
Journal of Materials Chemistry C
33 Z. Yang, X. Wang, Y. Chen, Z. Zheng, Z. Chen, W. Xu, W. Liu, 50 M. Cong, B. Yang, F. Hong, T. Zheng, Y. Sang, J. Guo,
Y. (Michael) Yang, J. Zhao, T. Chen and H. Zhu,
Nat. Commun., 2019, 10, 4540.
34 S. Kahmann, E. K. Tekelenburg, H. Duim, M. E. Kamminga
and M. A. Loi, Nat. Commun., 2020, 11, 2344.
S. Yang and K. Han, Sci. Bull., 2020, 65, 1078–1084.
51 R. Zeng, L. Zhang, Y. Xue, B. Ke, Z. Zhao, D. Huang, Q. Wei,
W. Zhou and B. Zou, J. Phys. Chem. Lett., 2020, 11,
2053–2061.
35 L. Zhou, J.-F. Liao, Z.-G. Huang, J.-H. Wei, X.-D. Wang, 52 S. Li, J. Luo, J. Liu and J. Tang, J. Phys. Chem. Lett., 2019, 10,
W.-G. Li, H.-Y. Chen, D.-B. Kuang and C.-Y. Su, Angew.
Chem., Int. Ed., 2019, 58, 5277–5281.
36 L. Zhou, J.-F. Liao, Z.-G. Huang, J.-H. Wei, X.-D. Wang, H.-Y.
1999–2007.
53 J. Shi, Y. Li, J. Wu, H. Wu, Y. Luo, D. Li, J. J. Jasieniak and
Q. Meng, Adv. Opt. Mater., 2020, 8, 1902026.
Chen and D.-B. Kuang, Angew. Chem., Int. Ed., 2019, 58, 54 R. Wang, J. Yang, X. Wu and S. Wang, Nanoscale, 2016, 8,
15435–15440. 8210–8219.
37 J. Zhou, M. Li, L. Ning, R. Zhang, M. S. Molokeev, J. Zhao, 55 T. Shi, W.-J. Yin, F. Hong, K. Zhu and Y. Yan, Appl. Phys.
S. Yang, K. Han and Z. Xia, J. Phys. Chem. Lett., 2019, 10,
1337–1341.
38 P. Nandi, D. Topwal, N.-G. Park and H. Shin, J. Phys. D: Appl.
Phys., 2020, 53, 493002.
39 P. Nandi, C. Giri, D. Swain, U. Manju and D. Topwal,
CrystEngComm, 2019, 21, 656–661.
40 H. Mashiyama, Y. Kurihara and T. Azetsu, J. Korean Phy.
Soc., 1998, 32, S156–S158.
Lett., 2015, 106, 103902.
56 Y. Liu, K. Palotas, X. Yuan, T. Hou, H. Lin, Y. Li and
S.-T. Lee, ACS Nano, 2017, 11, 2060–2065.
57 D. J. Vogel, T. M. Inerbaev and D. S. Kilin, J. Phys. Chem. C,
2018, 122, 5216–5226.
58 X. Zhu, H. Su, R. A. Marcus and M. E. Michel-Beyerle, J. Phys.
Chem. Lett., 2014, 5, 3061–3065.
59 S. Zou, Z. Gong, B. Liang, L. Hou, R. Liu, H. Zhong, B. Zou
and A. V. Kavokin, J. Phys. Commun., 2017, 1, 55018.
60 M. Hirasawa, T. Ishihara, T. Goto, K. Uchida and N. Miura,
Phys. B, 1994, 201, 427–430.
41 Y. Yang, Y. Yan, M. Yang, S. Choi, K. Zhu, J. M. Luther and
M. C. Beard, Nat. Commun., 2015, 6, 7961.
42 D. Priante, I. Dursun, M. S. Alias, D. Shi, V. A. Melnikov,
T. K. Ng, O. F. Mohammed, O. M. Bakr and B. S. Ooi, 61 A. Milev, D. M. A. S. Dissanayake, G. S. K. Kannangara and
Appl. Phys. Lett., 2015, 106, 081902.
A. R. Kumarasinghe, Phys. Chem. Chem. Phys., 2013, 15,
43 M. I. Dar, G. Jacopin, S. Meloni, A. Mattoni, N. Arora,
16294–16302.
A. Boziki, S. M. Zakeeruddin, U. Rothlisberger and 62 R. R. Nair, M. Sepioni, I.-L. Tsai, O. Lehtinen, J. Keinonen,
¨
M. Gratzel, Sci. Adv., 2016, 2, e1601156.
A. V. Krasheninnikov, T. Thomson, A. K. Geim and
44 J. Fu, N. F. Jamaludin, B. Wu, M. Li, A. Solanki, Y. F. Ng,
I. V. Grigorieva, Nat. Phys., 2012, 8, 199–202.
S. Mhaisalkar, C. H. A. Huan and T. C. Sum, Adv. Energy 63 G. Z. Xing, Y. H. Lu, Y. F. Tian, J. B. Yi, C. C. Lim, Y. F. Li,
Mater., 2019, 9, 1803119.
G. P. Li, D. D. Wang, B. Yao, J. Ding, Y. P. Feng and T. Wu,
AIP Adv., 2011, 1, 22152.
64 S. Sil, H. Luitel, J. Dhar, M. Chakrabarti, P. P. Ray,
B. Bandyopadhyay and D. Sanyal, Phys. Lett. A, 2020,
384, 126278.
45 Y. P. Varshni, Physica, 1967, 34, 149–154.
46 P. Nandi, P. Madadevan and D. Topwal, Unpubl. result.
47 S. Sarkar and P. Mahadevan, Phys. Rev. B, 2017, 95, 214118.
48 E. R. Dohner, E. T. Hoke and H. I. Karunadasa, J. Am. Chem.
Soc., 2014, 136, 1718–1721.
´ ´
65 B. Nafradi, P. Szirmai, M. Spina, H. Lee, O. V. Yazyev,
´
A. Arakcheeva, D. Chernyshov, M. Gibert, L. Forro and
49 E. R. Dohner, A. Jaffe, L. R. Bradshaw and H. I. Karunadasa,
J. Am. Chem. Soc., 2014, 136, 13154–13157.
´
E. Horvath, Nat. Commun., 2016, 7, 13406.
2800 | J. Mater. Chem. C, 2021, 9, 2793ꢀ2800
This journal is The Royal Society of Chemistry 2021