Phase stability and transformation of the α to ε-phase of Alq3 phosphor after thermal treatment and their photo-physical properties

Article abstract of DOI:10.1016/j.jpcs.2018.05.035

In this study, we analyzed changes in the phases of aluminum tris 8–hydroxyquinoline (Alq₃) after thermal treatment at different temperatures and their photo–physical properties. We prepared α–Alq₃ phosphor via the co–precipitation method. Improvements in the phase purity due to stagnation of the α–phase and the transformation from α to ε–Alq₃ were achieved by thermal treatment in an Ar atmosphere. The initial formation, stagnation, improvement in the phase purity, and conversion to ε–Alq₃ were confirmed by X–ray diffraction (XRD) analysis. The XRD results were also validated by Fourier–transform infrared spectroscopy and Nuclear magnetic resonance spectroscopy. Ultraviolet–visible (UV–Vis) absorption spectroscopy was conducted in the presence of acidic and basic media at concentrations of 10^?6 M to 10^?3 M, respectively. The modifications in the UV–Vis absorption spectra indicated changes in the band gap energy (E_g) after thermal treatment. The variations in E_g for α and ε–Alq₃ supported the stagnation and transformation of the phase. Photoluminescence (PL) analysis of the as–prepared α–Alq₃ determined a λ_emi maximum at 516 nm. A minor blue shift of Δλ = 2 nm was observed as the PL intensity increased for the annealed α–Alq₃. A large blue shift of Δλ = 18 nm as the PL intensity decreased was due to the change in phase from α to ε–Alq₃. PL study of the α and ε–Alq₃ phases in acidic solvent detected blue shifts, whereas red shifts occurred in the basic solvents due to variations in their dielectric constants. The mechanism related to the solvatochromatic effect on the shifts in PL emission was also determined in this study. Thermogravimetric analysis was employed to determine the thermal stability of the prepared phosphors.

Full text of DOI:10.1016/j.jpcs.2018.05.035

Accepted Manuscript
Phase stability and transformation of the α to ε-phase of Alq phosphor after thermal
3
treatment and their photo-physical properties
Deepshikha Painuly, Navin Kumar Mogha, Dhanraj T. Masram, R. Singhal, R.S.
Gedam, I.M. Nagpure
PII:
S0022-3697(17)31897-8
10.1016/j.jpcs.2018.05.035
PCS 8598
DOI:
Reference:
To appear in: Journal of Physics and Chemistry of Solids
Received Date: 7 October 2017
Revised Date: 23 May 2018
Accepted Date: 24 May 2018
Please cite this article as: D. Painuly, N.K. Mogha, D.T. Masram, R. Singhal, R.S. Gedam, I.M. Nagpure,
Phase stability and transformation of the α to ε-phase of Alq phosphor after thermal treatment and
3
their photo-physical properties, Journal of Physics and Chemistry of Solids (2018), doi: 10.1016/
j.jpcs.2018.05.035.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Phase stability and transformation of the to -phase of Alq3 phosphor after
α
ε
thermal treatment and their photo-physical properties
Deepshikha Painuly1, Navin Kumar Mogha2, Dhanraj T. Masram2#, R. Singhal3,
R. S. Gedam4, I. M. Nagpure1*
1Physics, Department of Sciences and Humanities, National Institute of Technology, Uttarakhand–246174, India
2Department of Chemistry, North Campus, University of Delhi, India–110007
3Dept. of Physics and Materials Research Centre, Malaviya National Institute of Technology Jaipur–302017, India
4Dept.of Applied Physics, Visvesvaraya National Institute of Technology, Nagpur–440010, India
Abstract
In this study, we analyzed changes in the phases of aluminum tris 8–hydroxyquinoline (Alq3)
after thermal treatment at different temperatures and their photo–physical properties. We
prepared α–Alq3 phosphor via the co–precipitation method. Improvements in the phase purity
due to stagnation of the α–phase and the transformation from α to ε–Alq3 were achieved by
thermal treatment in an Ar atmosphere. The initial formation, stagnation, improvement in the
phase purity, and conversion to ε–Alq3 were confirmed by X–ray diffraction (XRD) analysis.
The XRD results were also validated by Fourier–transform infrared spectroscopy and nuclear
magnetic resonance spectroscopy. Ultraviolet–visible (UV–Vis) absorption spectroscopy was
conducted in the presence of acidic and basic media at concentrations of 10–6 M to 10–3 M,
respectively. The modifications in the UV–Vis absorption spectra indicated changes in the band
gap energy (Eg) after thermal treatment. The variations in Eg for α and ε–Alq3 supported the
stagnation and transformation of the phase. Photoluminescence (PL) analysis of the as–prepared
α–Alq3 determined a λemi maximum at 516 nm. A minor blue shift of ∆λ = 2 nm was observed as
the PL intensity increased for the annealed α–Alq3. A large blue shift of ∆λ = 18 nm as the PL
intensity decreased was due to the change in phase from α to ε–Alq3. PL study of the α and ε–
Alq3 phases in acidic solvent detected blue shifts, whereas red shifts occurred in the basic
solvents due to variations in their dielectric constants. The mechanism related to the
solvatochromatic effect on the shifts in PL emission was also determined in this study.
Thermogravimetric analysis/differential thermal analysis was employed to determine the thermal
stability of the as–prepared phosphors.
Keywords: Acetic acid (AC), Dichloromethane (DCM), Dimethyl sulfoxide (DMSO), Formic acid (FC),
Meridional aluminum tris 8–hydroxyquinoline, Photoluminescence (PL).
*Corresponding Author: indrajitnagpure@gmail.com (I. M. Nagpure)
#Co–corresponding Author: dhanrajtm@gmail.com (Dhanraj T. Masram)

ACCEPTED MANUSCRIPT
1. Introduction
Materials based on the octahedral type of structure, MN3O3, where M is an trivalent metal and N
and O are the nitrogen and oxygen present in quinoline ligands, are used extensively in
electroluminescence devices [1, 2]. Among the several previously described phosphors,
aluminum tris 8–hydroxyquinoline (Alq3), which was originally reported by Tang and Vanslyke
[3], is still the most efficient low voltage–driven organic phosphor and it is used widely in
organic light emitting devices. The methods used for synthesizing Alq3 have attracted much
attention in order to facilitate the development of efficient electroluminescence devices for
applications in ultra–thin flat and flexible display panels. In addition to its use as an electron
transport layer in hetero–layer devices, Alq3 is employed as a host material for various dyes
where it helps to tune the emission color [4, 5]. Several previous studies have focused on
optimizing the efficiency and long–term stability of Alq3 devices [6–8], but very few have
considered the interactions between Alq3 molecules and solvents, as well as how modifications
of the Alq3 phases by thermal treatment might change the photo–physical properties.
Recently, attempts have been made to understand the structural diversity of Alq3 metal chelate
and its various crystalline phases, which are commonly known by α, β, γ, δ, and ε [9–14]. These
different phases of Alq3 can be obtained using various synthesis techniques and thermal
treatments [15]. Due to this structural diversity, it is essential to understand the correlations
between different phases and their photo–physical properties. The structural properties of Alq3
phases such as α, β, γ, δ, and ε are crucial parameters for defining the intermolecular packing of
molecules, and they are linked to their fluorescence properties. Poly–morphological changes in
Alq3 isomers may occur in the range between ≈130–175°C, whereas different crystalline phases
are formed above ≈185°C [16]. Therefore, the conversion of the Alq3 phase depends on the
synthesis procedure and the thermal treatment employed. Furthermore, the effects of thermal
treatment may determine the crystalline phase purity of the isomeric state immediately prior to
its phase transition temperature. Previous studies also indicate that the δ– and γ–Alq3 phases
exhibit blue–shifted fluorescence in an analogous manner to the yellow–green fluorescence of
the meridional polymorphs of β and ε–Alq3, whereas α–Alq3 exhibits yellow–green fluorescence
under ultraviolet (UV) exposure [11, 17]. The δ–Alq3 phase exhibits blue fluorescence [18],
whereas the α– and β–Alq3 phases exhibits prominent green florescence [10, 11, 18, 19]. Thus,
further analyses are required of the optical properties of ε–Alq3 and the effects of its crystallinity
on the fluorescence of synthesized and thermally treated α–Alq3 phases.

ACCEPTED MANUSCRIPT
In addition to thermal treatment, solvents have important effects on the absorption and
fluorescence characteristics of optoelectronic organic compounds, as well as on dyes when
tuning the emission color [20–21]. Therefore, the effects of the solution phases on the phosphor
solute have been studied widely. The effects of different solvents and functional group
substitutions on the absorption and florescence spectra have been reported in several studies [22–
23]. Therefore, it is essential to understand the effects of solvent on the complex photo–physics
and photochemistry of metal–organic complexes. The dipole moments of short–lived species are
of considerable interest because they provide information about the electronic and geometrical
structures of these transients. Anomalous shifts occur in the electronic absorption bands of
unsaturated molecules to shorter wavelengths when the solvent is changed from one with a low
dielectric constant to one with a higher dielectric constant, and vice versa [24– 25].
In the present study, we obtained the α–Alq3 complex via the co–precipitation method in an air
atmosphere and we investigated the stagnation of the α–phase and the conversion to ε–Alq3 after
thermal treatment in an Ar atmosphere. We focused primarily on the modification of the
crystalline phases of Alq3 by thermal treatment and its effects on their photo–physical properties,
such as absorption and fluorescence, rather than their isomeric transformation. In addition, we
investigated the solvatochromatic effect due to the presence of acidic (acetic acid (AC) and
formic acid (FC)) and basic (dichloromethane (DCM) and dimethyl sulfoxide (DMSO)) solvents.
Shifts in the ultraviolet–visible (UV–Vis) absorption and fluorescence wavelengths (λmax) were
determined in order to extract useful information about the variations in the ground and excited
states of the Alq3 metal complex. X–ray diffraction (XRD), Fourier transform infrared
spectroscopy (FTIR), and nuclear magnetic resonance (NMR) spectroscopy were utilized for
structural analysis, and thermogravimetric analysis/differential thermal analysis (TGA/DTA) was
conducted to determine the thermal stability of the as–prepared phosphors.
2. Experimental
2.1. Preparation of the materials
The α–Alq3 phosphor was prepared via the co–precipitation technique, before further annealing
at 200°C and then 320°C in an Ar atmosphere. AR/GR grade raw materials with calculated
stoichiometric weights were used to synthesize the phosphor, where 8–hydroxyquinoline, 30 ml
double distilled water, and 30 ml AC were used as the raw materials based on the molar ratio
calculations. The procedure is shown in reaction 1, which indicates that the reaction between 8–
hydroxyl quinoline and dilute AC led to the formation of a transparent dilute oxine solution [26].

ACCEPTED MANUSCRIPT
Aluminum nitrate was dissolved in 30 ml double distilled water and then stirred at 50°C for 2 h
to obtain a transparent aluminum hydroxide solution [26]. This procedure is shown in reaction 2.
The products of reaction 1 and reaction 2 were slowly mixed and a suspension was formed.
Ammonium hydroxide solution (NH4OH) was added in a dropwise manner to this suspension to
increase the yield of the precipitate [27]. This procedure is illustrated in reaction 3.
The precipitate obtained from reaction 3 was filtered and then washed several times with excess
amounts of fresh double distilled water, before rinsing the precipitate and drying at 80°C. The
product obtained in the form of a powder was designated as sample A and it was confirmed as
the α–Alq3 phase by XRD analysis. The procedure is illustrated in reaction 4.
Sample A was subjected to thermal treatment by annealing at ≈ 200°C in an Ar atmosphere for ≈
3 h. Annealing was applied to achieve better phase purity and to remove any unwanted
impurities present in the product. The product obtained after thermal treatment was designated as
sample B. Sample B was again subjected to thermal treatment by annealing at ≈ 320°C in an Ar
atmosphere for ≈ 5 h. Further annealing of sample B at a higher temperature was applied to
obtain changes in the compound. The product obtained after further thermal treatment was

ACCEPTED MANUSCRIPT
designated as sample C. Sample B was confirmed as α–Alq3 and sample C as ε–Alq3 by XRD
analysis. Thus, we estimated that the α–phase was converted into ε–Alq3 at ≈ 320°C.
2.2. Characterization techniques
XRD analysis was performed using a PAN–analytical diffractometer equipped with a Cu–Kα
source (λ = 1.5418 Å). The FTIR spectra of the powder samples were acquired with a Perkin
Elmer FTIR spectrometer using KBr pellets. NMR spectra were obtained using a JEOL NMR
spectrometer at a frequency of 400 MHz with CDCl3 as the solvent. UV–Vis absorption
measurements were performed using a Perkin Elmer LS55λ UV–Vis spectrophotometer.
Photoluminescence (PL) analyses of the powders and in solution were performed using a Perkin
Elmer LS45λ fluorescence spectrometer at room temperature with a mono–chromatized xenon
flash lamp as the excitation source. TGA was performed with a Perkin Elmer Diamond TG/DTA
thermal analyzer. Temperature and enthalpy calibrations were performed with α–alumina
(Al2O3) as the standard at a heating rate of 10°C from room temperature to 600°C in an N2
atmosphere.
3. Results and Discussion
3.1. XRD analyses
The XRD patterns obtained for the Alq3 crystalline phases are shown in Fig. 1(a–c). The dotted
lines in Fig. 1 indicate the differences in the 2θ positions for sample A, sample B, and sample C.
The XRD pattern obtained for sample A is shown in the Fig. 1(a). The data obtained agreed with
those reported previously for α–Alq3 [9, 11, 14]. The additional peak at 5.32° in Fig. 1(a)
indicates the presence of a small amount of intercalated water impurity in sample A and slight
shifts in the other peaks can also be seen. Therefore, in order to improve the phase purity and
obtain possible structural modifications in α–Alq3, sample A was subjected to thermal treatment
in an Ar atmosphere. The purification process generally involved the removal of adsorbed and
absorbed water, other unreacted impurities, and defect levels from the material in the initial stage
of process. Thus, sample A was annealed at 200°C and stagnation of the α–phase with improved
phase purity are shown by the XRD pattern in Fig. 1(b), which indicates the removal of the
impurity peak at 5.32°, modifications of the diffraction peaks at 14.9°, 15.5°, and 15.8°, and the
appearance of additional smaller peaks at 17.4°, 20°, and 32.6°. The stagnation of the α–phase of
Alq3 and the improved phase purity of sample B agreed with previously reported results [9, 11,
14]. The XRD pattern shown in Fig. 1(c) indicates the phase conversion from α to ε–Alq3. The
formation of the ε–Alq3 obtained by thermal treatment was accompanied by the appearance of
additional peaks in the 2θ range between 8.5–11°. The presence of additional peaks at 9.2°, 9.8°,

ACCEPTED MANUSCRIPT
10.2°, and 10.5°, the modifications of the peaks at 14.92°, 15.13°, 15.71°, 25.4°, 25.6°, and
25.86°, and the disappearance of the peak at 17.44° confirmed the formation of ε–Alq3 in sample
C. The XRD data in Fig. 1(c) agreed with those obtained for ε–Alq3 by Manju et al. [12].
According to Manju et al. [12], α–Alq3 exists in the p1 space group with the unit cell parameters
of:
89.464 (5),
space group with the unit cell parameters of:
18.7110 (8), (Å) = 95.4940 (19), (Å) = 109.774 (2),
(2), and = 6. The red dotted square in Fig. 1(a–c) in the 2θ range between 8 –11° indicates the
a
(Å)
=
6.2455 (10),
(Å) = 82.52 (2), volume = 1102.1 (4), and
(Å) 13.5190 (5),
(Å) = 114.6270 (16), volume = 3296.8
b
(Å)
=
12.8710 (18),
b
(Å)
=
14.739 (3),
= 2. In addition, ε–Alq3 exists in the p1
(Å) 15.8550 (6), (Å) =
α (Å) = 69.890 (6), β (Å) =
γ
z
a
=
b
=
b
α
β
γ
z
difference between the α and ε phases. It is well known that the physical properties of crystalline
organic metal complexes generally change when they are heated to temperatures far below their
melting point [28–29]. These changes are related to modifications of their morphological
structure and they indicate that the organic metal complex samples are not in a state of
thermodynamic equilibrium in most cases. Therefore, thermal treatment reorganizes the structure
and leads to a state of order with a lower free energy. The whole crystallites or parts of them may
melt and new crystallites can develop from the molten phase. We applied thermal treatment to
our as–prepared α–Alq3 at 200°C and 320°C in an Ar atmosphere, which led to stagnation of the
phase and improvements in the phase purity at 200°C, and modification from α to ε–Alq3 at
320°C. These changes are supported by the XRD data shown in Fig. 1(a–c). Furthermore, the
formation of the phases, stagnation of α–Alq3, and conversion to ε–Alq3 by thermal treatment
were confirmed by FTIR and NMR analyses, as described in the following.
3.2. FTIR analysis
Sample A, sample B, and sample C were studied by vibrational analysis to investigate their
formation and structural modification after thermal treatments. The FTIR spectra are compared
in Fig. 2(a–c). The inset in Fig. 2(a1–c1) shows the FTIR spectrum between 4000–2000 cm–1.
The FTIR peaks between 3500–3000 cm–1 and 1800–1600 cm–1 in Fig. 2(a1) for sample A
indicate the presence of adsorbed/absorbed and/or intercalated water molecules in the sample
[30–31]. The small intensity peaks in Fig. 2 (a1) for sample A between 3000–1800 cm–1 can be
attributed to the unreacted alkyl C–H, nitrile C≡N, N–H, and alkyne C≡C groups [30–31]. The
presence of water molecules and minor unwanted impurities in sample A may be explained by
the use of ammonium hydroxide, the presence of water during the reaction, and the performance
of the procedure in an air atmosphere and atmospheric carbon dioxide. The absence of these
peaks from the FTIR spectrum in Fig. 2(b1–c1) indicate the removal of unreacted alkyl C–H,
nitrile C≡N, and alkyne C≡C groups, and water molecules due to the effect of the thermal

ACCEPTED MANUSCRIPT
treatment in an Ar atmosphere. The small hump at ≈2306 cm–1 is attributable to the N–H group
and the peak at ≈ 3050 cm–1 to the intercalated water molecules, as shown in Fig. 2(b1 and c1).
The observed orientations of the vibrational bands shown in Fig. 2(a–c) were analogous in all the
samples, but there were slight differences in the peak positions for sample C compared with
sample A and sample B. The signature absorption peaks for the metal to ligand bond attributed to
Al–O/Al–N bonding vibrations appeared at 490 and 544 cm–1 in sample A, but these peaks
shifted slightly toward the higher frequency region at 575 and 643 cm–1 in sample B and sample
C [32]. The shifting of the Al–O/Al–N metal ligand peak positions may have been due to the
effects of thermal treatments on the samples. The absorption bands at 1603 and 1579 cm–1 were
attributed to the C=C/C=N plane bending and stretching vibration, respectively, in sample A and
sample B. However, these bands shifted slightly toward the higher frequency region in sample C
[33]. The shifts of the plane bending and stretching vibration bands toward the higher frequency
region may indicate the conversion from the α– to ε–phase of Alq3. The vibrational bands at
1472 and 1498 cm–1 in sample A and sample B correspond to the characteristics peaks for the
pyridine/phenyl groups of Alq3, and a slight shift in the peak position toward the higher
frequency zone was found in sample C. The absorption peaks at 1334 and 1384 cm–1 were
attributed to q–molecule plane/ring deformation. The peaks observed at 746, 790, and 822 cm–1
were more prominent in sample B and sample C, thereby indicating the higher vibrational mode
of C–H wagging due to the absence of adsorbed and/or absorbed water molecules with
irradiation by incident light [34]. Slight shifts in the frequencies of the C–H and C–O/C–N
stretching infrared bands were also observed among sample A, sample B, and sample C. In
addition, the FTIR spectra obtained for sample B and sample C did not contain any additional
absorption bands for the decomposed species. Therefore, the FTIR data confirmed the formation
of Alq3. However, it was quite difficult to predict the structural transformation from the α– to ε–
phase due to the effect of annealing based on the FTIR analysis results.
3.3. NMR analysis
Proton NMR provides information about the number of protons and the environment where they
are located in a particular molecule. The absence or shift of a peak for a particular proton in
NMR can be attributed to a change in its environment due to the formation of new bonds.
Therefore, the absence of the hydroxy proton peak in NMR for sample A, sample B, and sample
C indicated the formation of coordination bonds with the Al atom. The proton (1H) NMR results
for sample A, sample B, and sample C are presented in Figs 3(a), 3(b), and 3(c), respectively.
The absence of a peak for the free hydroxyl group of 8–hydroxyquinoline at around 10 ppm in
Fig. 3 indicates the formation of meridional Alq3 (mer–Alq3). The NMR pattern for sample–A

ACCEPTED MANUSCRIPT
indicated the presence of some additional peaks from around 8.4 to 8.8 ppm and 6.6 to 6.9 ppm,
but the others peak were at their standard positions, as reported by Costa et al. [15]. Additional
impurity peaks were expected in sample A due to the use of ammonium hydroxide, the presence
of water during the reaction, and the performance of the procedure in the air and atmospheric
carbon dioxide. These changes were also supported by the XRD and FTIR data. The NMR
profiles for sample B and sample C were at their standard positions, as reported in a previous
study [11], i.e., the peak at 8.8 ppm could be attributed to the protons in –N–CH– and the peak at
8.2 ppm corresponded to –N–CH–CH–CH– protons. Similarly, the peaks at 7.4–7.6 ppm were
attributed to two other aromatic protons in the quinoline molecule, and the peak at about 7.0 ppm
corresponded to protons in –O–C–CH–. These observations agreed with the XRD and FTIR data.
However, the presence of a small amount of impurity according to FTIR was attributed to the N–
H group and intercalated water molecules in sample B and sample C, but these were not clear in
the NMR spectra.
3.4. UV–Vis absorption analysis
The UV–Vis absorption spectra for sample A, sample B, and sample C in the solution state are
shown in Fig. 4(a–d) and Fig. 5(a–b). The acidic and basic media had different dielectric
constants when the concentrations of the solvents were 10–3 M and 10–6 M. We investigated the
changes in the electronic transition between 250
–
450 nm in the intermolecular structures of
Alq3) by adding different solvents. The
sample A ( Alq3), sample B ( Alq3), and sample C (ε–
α
–
α–
solvatochromatic effect (solute–solvent interaction) was expected to lead to changes in the
positions, intensities, and shapes of the absorption bands determined for the quinoline ligands of
the Alq3 metal complex. The absorption bands normally observed between 300–350 nm are π–π*
transitions according to the vibronic progression. These bands can be attributed to the ring
deformation localized on the aromatic ring of the ligand [35]. The absorption bands that usually
appear at higher wavelengths between ≈350–400 nm correspond to the n–π* transitions, and
these bands can be ascribed to the ligand centered electronic transitions [36]. The bands between
350–400 nm are produced by electronic transitions from the highest occupied molecular orbital
(HOMO), which is mainly situated on the phenoxide ring, to the lowest unoccupied molecular
orbital (LUMO) situated on the pyridyl ring [37].
Figure 4(a) shows the characteristic maximum peak at 350 nm obtained for sample A in AC at a
concentration of 10–3 M, and other peaks appeared at ≈ 320 and 310 nm. When FC was used as
the solvent at a concentration of 10–3 M, the main absorption peak was at 360 nm, with shoulder
peaks at ≈ 310 nm and 317 nm. However, in the basic solvents i.e., DMSO and DCM at a
concentration of 10–3 M, the characteristics peaks were similar, where the characteristic peak

ACCEPTED MANUSCRIPT
positions remained unchanged with the maximum at 385 nm and shoulders at around 320 and
333 nm. It should be noted that as the polarity of the solvents increased from AC to FC and from
DCM to DMSO, there were small blue shifts in the absorption bands. This type of blue shift with
increases in the polarity of the solvent has been reported previously [38]. Furthermore, as the
concentration decreased from 10–3 M to 10–6 M for sample A, the peak positions remained
unchanged in the absorption spectrum. However, major peak intensity enhancements were found
when AC was used as the solvent, as shown in Fig. 4(b). The absorption spectra obtained for
sample B (α–
Alq3) at concentrations of 10–3 M and 10–6 M are shown in Figs 4(c) and 4(d).
Figures 4(c) and 4(d) show that the characteristic peak positions remained unaltered but minor
shifts were found when DCM and DMSO were used as the solvents. A minor shift in the tail of
the absorption band toward higher wavelengths was also found for sample B compared with
sample A. The observed shifts in the tails of the absorption bands were linked to the decrease in
the band gap for sample B, probably due to the improved phase purity of α–Alq3 after thermal
treatment at 200°C. These observations confirmed that sample A and sample B belonged to the
α–phase and the slight change in the band gap for sample B was due to the improved phase
purity. The absorption data collected for sample C (ε–phase) were slightly inconsistent with the
results for sample A (α–Alq3) and sample B (α–Alq3), as shown in Figs 5(a) and 5(b). Small
changes in the absorption spectra were observed with a blue shift of ≈ 3 to 5 nm in the peak
position in the acidic and basic media. The observed shifts in the tails of the absorption bands
toward lower wavelengths in sample C are shown in Fig. 5 (a) and 5(b), which were due to the
increase in the band gap compared with sample A and sample B. The observed inconsistent
behavior of the absorption spectrum obtained for sample C may have been due to the phase
transformation from α– to ε–Alq3 after thermal treatment. Therefore, we may conclude that the
different solvent concentrations did not affect the shifts in the peak positions for sample A and
sample B, but instead they enhanced the peak intensity. However, changes were found in the
spectrum obtained for sample C. The absorption data supported the results obtained by XRD,
FTIR, and NMR analyses.
The UV–Vis results showed that as the polarity of the solvent increased in acidic medium from
AC to FC at 10–3 M, the characteristic transition peaks exhibited minor red shifts in the n–π*
transition, whereas no changes in the π–π* transitions were observed. In the basic solvents, no
changes were observed in the n–π* and π–π* transitions. The results were similar for the basic
solvents, but there were no changes with the acidic solvents at 10–6 M. The minor shifts in the
absorption spectra with acidic solvents as the dielectric constant increased may have been due to
the fewer contacts between the H–atom present in the –COOH group and the solvent molecule
due to its geometrical confirmation. The increase in the absorption intensity with the basic

ACCEPTED MANUSCRIPT
solvents indicated the formation of hydrogen bonding interactions between the solute and
solvent. The shifts determined in the tails of the absorption bands were due to the effects of
annealing [39, 40], where the phase purity improved at 200°C and the phase changed from α– to
ε–Alq3 at 320°C.
3.5. Band gap analysis
Solvents have important roles in tuning the photo–physical properties of luminescent materials,
and thus their effects were investigated in this study. In order to understand the photo–physical
properties of a material, the optical band gap (Eg) is an essential parameter because it is directly
influenced by the solvatochromatic effect due to the interaction between the solute and solvent.
Hence, in order to collect useful information regarding the band gap, we calculated Tauc’s
relationship for sample A, sample B, and sample C [41]. In general, two bands are observed in
the absorption spectrum, where that appearing at the shorter wavelength is represented as
fundamental and the other at the longer wavelength is λonse . λonset was used to calculate the numerical
t
values of Eg for the samples according to Tauc’s relationship given by Equation 6 [41]:
1/n
(α h v
)
=
αo
(h v
–
Eg) ,
(6)
where
h
is Plank’s constant, α is the absorption coefficient, αo is the proportionality constant, hν
is the photon energy, n is the type of transition, and Eg is the band gap energy. The Alq3 metal
complex exhibited a direct band gap transition, as reported previously [42]. Therefore, the value
of n was taken as 1/2 and Tauc’s relationship was modified to Equation 7.
(
α h v)2 = αo
(h v
–
Eg
)
(7)
The Eg values were estimated for sample A, sample, B and sample C by extrapolating the straight
line in the linear fitted region, which met the energy axis where (αhv)2 = 0 and this corresponded
to the direct band gap, as shown in Figs 6–8. The Eg values for sample A and sample B were
between 2.82–3.09 eV at concentrations of 10–3 M and 10–6 M, as shown in Fig. 6(a–h) and Fig.
7 (a–h), respectively. There were small decreases in the band gap from ≈ 0.2 eV to 0.06 eV for
sample B in acidic and basics solvents compared with the results determined for sample A. The
small decrease in the band gap determined for sample B compared with sample A may have been
due to the removal of unwanted impurities and defects after thermal treatment at 200°C.
Furthermore, sample C had an Eg value between 2.93–3.15 eV, as shown in Fig. 8 (a–h). The
observed band gap for sample C was ≈0.06 eV to 0.11 eV larger compared with those
determined for sample A and sample B. The increase in the value of Eg for sample C was clearly
due to the phase transformation from α– to ε–Alq3. These changes also agreed with the observed
blue shifts in both the acidic and basic solvents as their dielectric constants increased [39, 40].

ACCEPTED MANUSCRIPT
The observed blue shift in the absorption spectrum determined for ε–Alq3 compared with those
for sample A and sample B indicated an increase in the band gap for this material. It is well
known that the intermolecular π–π overlapping of quinoline ligands is weaker in the ε–phase
compared with α–Alq3, which may be due to the lack of symmetry present in ε–Alq3 compared
with the molecular structure of the α–phase. Hence, the presence of weak bonding in ε–Alq3
reduces the electron cloud density on phenoxide ring and this weakens the intermolecular
conjugated interaction between adjacent Alq3 molecules. Therefore, the wider optical band gap
determined for ε–Alq3 compared with the α–Alq3 complex mainly contributed to the blue shift in
the absorption band, which was due to the electron transition from HOMO to LUMO.
3.6. PL analysis
The packing and intermolecular interactions inside a molecular structure are important for its
optical properties. In order to understand the effects of intermolecular interactions on the optical
properties of the Alq3 complex, we conducted PL analysis using sample A, sample B, and sample
C in the solid state as well as in solution. The PL results obtained for sample A, sample B, and
sample C in powder form are shown in Fig. 9. The PL results obtained in solution with various
molar concentrations of different solvents are shown in Fig. 10(a–d) and Fig. 11(a–b). The
sample A powder (α–Alq3) exhibited prominent green PL emission with a maximum at 516 nm
at an excitation wavelength of 393 nm. The sample B powder (α–Alq3 annealed at 200°C)
exhibited green PL emission with a maximum at 514 nm and the PL intensity was enhanced
under excitation at 392 nm. The other excitation peaks as shoulders at ≈288 nm and 340 nm, as
shown in Fig. 9, were also present in sample A and sample B. A minor blue shift of ∆λ = 2 nm
was obtained and the PL emission intensity was significantly enhanced due to the better phase
purity of sample B after thermal treatment compared with sample A. The minor blue shifts in the
PL emission spectrum and the excitation spectrum were correlated with the small decrease in the
band gap for sample B compared with sample A, which was consistent with the band gap
analysis. The PL emission peaks for α–Alq3 were ascribed to the favorable π–π* transition inside
the molecular structure. For sample C (ε–Alq3), the maximum PL emission was at 498 nm and
the PL emission intensity decreased at an excitation wavelength of 385 nm, as shown in Fig. 9. A
major fluorescence blue shift of ≈∆λ = 18–16 nm was found in sample C compared with sample
A and sample B. Similarly, a noticeable blue shift of ≈∆λ = 7–8 nm was found in the PL
excitation spectrum. The blue shifts in the PL excitation and emission spectra for sample C may
have been due to the phase transformation from α– to ε–Alq3 after thermal treatment at 320°C. A
reduction in the chelation coordination between metal–nitrogen bonding in the Mq3 type

ACCEPTED MANUSCRIPT
complex probably increased the band gap and contributed to the blue shift in the emission
spectrum [39, 42, 43].
PL analysis was conducted in the solution state in order to understand the effects of different
solvents (i.e., acidic and basic media) on the packing and intermolecular interactions in the
molecular structure of the Alq3 complex. The PL emission profiles obtained for sample A,
sample B, and sample C in different solutions are shown in Fig. 10(a–d) and Fig. 11(a–b). The
fluorescence blue shifts in acidic media and the red shifts in basic media are shown in Fig. 10
and Fig. 11, which were consistent with the polarity and dipole moments of the solvents
employed [44]. The PL emission spectrum obtained for sample A (α–Alq3) in AC solvent (an
acidic medium) at 10–3 M showed that λmax was at 506 nm with a blue shift of ≈ ∆λ = 10 nm
compared with its powder form, as illustrated in Fig. 10(a). Similarly, the fluorescence spectrum
for sample A in FC at 10–3 M showed that the PL emission λmax was at 502 nm with a blue shift
of ≈ ∆λ = 14 nm. The emission spectrum obtained for sample A (α–Alq3) in DCM at 10–3
M
showed that λmax was at 518 nm with a red shift of ≈ ∆λ = 2 nm compared with its powder
sample, as illustrated in Fig. 10(a). Similarly, the fluorescence spectrum for sample A in DMSO
at 10–3 M showed that λmax was at 524 nm with a red shift of ≈ ∆λ = 8 nm. The blue shifts were ≈
∆λ = 12 nm in AC and ≈ ∆λ = 16 nm in FC, with red shifts of ≈ ∆λ = 04 nm in DCM and ≈ ∆λ =
10 nm in DMSO as the PL intensity decreased for sample A at 10–6 M, as shown in Fig. 10(b).
The observed blue shifts in the acidic media and red shifts in the basic media were related to the
dielectric constants for sample A. Furthermore, we conducted PL analysis of sample B in the
solution state. For sample B, the PL emission λmax was at 506 nm in AC with a blue shift of ∆λ =
08 nm. In FC, λmax was at 500 nm with a blue shift of ∆λ = 14 nm at 10–3 M, as shown in Fig.
10(c). The emission spectrum for sample B in DCM showed that λmax was at 518 nm with a red
shift of ≈ ∆λ = 4 nm, and the spectrum for DMSO showed that λmax was at 523 nm with a red
shift of ≈ ∆λ = 9 nm at 10–6 M compared with its powder form, as shown in Fig. 10(c). A similar
blue shift of ≈ ∆λ = 10 nm was found in AC and ≈ ∆λ = 14 nm in FC, with red shifts of ≈ ∆λ = 8
nm in DCM and ≈ ∆λ = 11 nm in DMSO as the PL intensity decreased for sample B at 10–6 M,
as shown in Fig. 10(d).
The variations in the PL properties of sample C were compared with those in sample A and
sample B in the solution state. For sample C, the PL emission λmax was at 494 nm in AC with a
blue shift of ∆λ = 04 nm, and λmax was at 490 nm in FC with a blue shift of ∆λ = 8 nm at 10–3 M,
as shown in Fig. 11(a). In DCM at 10–3 M, λmax was at 509 nm with a red shift of ∆λ = 11 nm,
and in DMSO, λmax was at 516 nm with ∆λ = 16 nm. Similarly, there were minor blue shifts of ≈
∆λ = 2 nm in AC and ≈ ∆λ = 9 nm in FC, and red shifts of ≈ ∆λ = 22 nm in DCM and ≈ ∆λ = 26
nm in DMSO as the PL intensity decreased for sample B at 10–6 M, as shown in Fig. 11(b).

ACCEPTED MANUSCRIPT
There were large blue shifts in the fluorescence spectra obtained for sample A and sample B, and
small red shifts for sample C (ε–Alq3) compared with their powder samples.
The fluorescence properties of
ε–Alq3 in the solution state were more susceptible to the basic
solvents according to the large red shifts, but less affected by the acidic solvents with small blue
shifts. The observed shifts in the PL emission spectra obtained for sample A, sample B, and
sample C were consistent with the variations in the dielectric constants of the acidic and basic
solvents [45]. Further details of the PL spectra measured at concentrations of 10–3 M and 10–6 M,
are shown in Fig. 10 and Fig. 11, respectively, which indicate that there were decreases in the PL
intensity as the solubility of the solute in the solvent increased [46]. Fluorescence analysis of
sample C (ε–Alq3) in the solution state showed that the PL emission peaks shifted toward lower
wavelengths in both the acidic and basic solvents at concentrations of 10–3 M and 10–6
M
compared with sample A and sample B. The shift in the PL emission peak toward lower
wavelengths was also supported by the increase in the band gap due to the phase transition from
α– to ε–Alq3. Thus, we may conclude that the acidic and basic solvents had solvatochromatic
effects on the PL emissions according to their dielectric constant values. The solvatochromatic
effect was more effective in acidic solvents for α–Alq3, but in basic solvents for ε–Alq3.
3.7. TGA/DTA analysis
The TGA/DTA results obtained for sample A, sample B, and sample C are shown in Fig. 12(a),
12(b), and 12(c), respectively. Sample A and sample B exhibited similar thermal behavior
(TGA), where they typically remained unchanged because both were in the α–Alq3 phase.
Different initial weight losses were found for sample A compared with sample B, as shown in
Fig. 12(a–b). An initial weight loss of ≈ 3% to 5% was found for sample A, which was
associated with desorption as well as the release of adsorbed/absorbed moisture (H2O) and
unreacted impurities present in the sample. The removal of the unreacted impurities present in
sample A was correlated with the release of energy in the DTA curve around 62°C. However, the
corresponding initial weight loss was reduced to ≈ 1 to 2% according to the TGA curve for
sample B, as shown in Fig. 12(b), which was probably due to the improved phase purity and the
removal of unwanted moisture, intercalated H2O, gases, and impurities during thermal treatment
in an Ar atmosphere [27]. Furthermore, the weight loss was less than ≈ 2% to 3% according to
the TGA curve for sample A from 110°C to the onset point at 360°C, and the weight loss was
similar for sample B up to its onset point at 364°C according to the TGA curve. The TGA curve
indicated that sample A was thermally stable up to ≈ 360°C, but up to to ≈ 364°C for sample B
due to its improved phase purity. The DTA curves obtained for sample A and sample B were
dissimilar. The TGA curve obtained for sample C (Fig. 12(c)) indicated that the initial weight

ACCEPTED MANUSCRIPT
loss of ≈ 3% to 5% at 120°C was similar to that shown by the TDA curve. However, this change
was inconsistent with the TGA curve for sample C due to the absence of intercalated H2O and
unwanted impurities in the complex after thermal treatment in an Ar atmosphere. The initial
variations in the DTA curve at 50°C and 120°C for sample C were similar to those determined
for sample B. The DTA curve obtained for sample C was also inconsistent with those produced
for sample A and sample B. The TGA analysis for sample C indicated that it was thermally
stable up to ≈ 350°C and it appeared to be less stable compared with sample B and sample A,
probably due to the change in phase and the presence of weak bonding in the complex. Similarly,
the TGA and DTA curves obtained for sample A, sample B, and sample C indicated that the
decomposition temperatures were different due to the initiation of decomposition by coordinated
8–hydroxyquinoline in the Alq3 complexes [47]. The final pyrolysis analyses of the Alq3
complexes showed that overall decomposition occurred at more similar temperatures for sample
A and sample C compared with sample B. These results support the presence of more stable α–
Alq3 in sample A and sample B, whereas ε–Alq3 was weaker and it broke down more readily.
Therefore, the thermal behavior of sample C was quite different from that of sample A and
sample B due to the phase transformation from α– to ε–Alq3 after thermal treatment.
3.7. Solvent effect mechanism
Variations in the energy of the electronic states of molecules in the condensed phase, i.e., solid or
liquid, and the intermolecular solute–media interactions alter the photo–physical properties of
molecules. The transitions in the photo–physical properties are influenced by the nature of
phosphor and the solvent properties, such as the charge distribution or the dipole moment for the
state of the solvent. The state will be dissimilar for the ground and excited molecules, and this
will contribute to the spectral shift. This phenomenon is usually assumed to be a
solvatochromatic shift [45]. In this study, UV–Vis absorption and PL measurements were
obtained in different acidic and basic solvents, and the solvatochromatic effects were determined.
The quinoline molecule can act as a proton acceptor in an acidic medium but as a proton donor in
a basic medium, and the mechanism of the solvatochromatic shift is illustrated in Fig. 13(a–d).
In this study, we used powder Alq3 as a solute and added it to AC and FC as acidic solvents. The
electron pair present on the quinoline ring made the Alq3 molecule a proton acceptor due to the
solvation effect. The methyl group present in AC was a weak electron donor and it could pull
electrons toward the oxygen molecule, where the oxygen molecule connected to the carbon
molecule could act as an electron lone pair and support the proton donation process, as shown in
Fig. 13(a). The properties of the FC solvent were similar as a proton acceptor when it was added
to Alq3 in solution, as shown in Fig. 13(b). The large blue shift in the PL spectrum with AC was

ACCEPTED MANUSCRIPT
expected, but it was reduced greatly with FC because the absence of a methyl group in FC led to
an asymmetric electron transition in contrast to the symmetric transition with AC. This result
also supports the increase in the value of the dielectric constant from AC to FC. Therefore, the
observed blue shift in the PL spectrum was due to the specific chemical effect of the solvent on
one or both electronic states (i.e., LUMO and HOMO of quinoline) [47]. The hydrogen bonds
present in the acidic AC and FC solvents were related to the quinoline group, where the phenol
in the quinoline ring was a proton acceptor and the charge was transferred from phenol to the
solvent molecule. This process was responsible for the blue shift of the PL emission spectrum
due to destabilization of the excited state [47]. In addition, when Alq3 powder was added to basic
solvents, i.e., DCM and DMSO, the quinoline group of the Alq3 complex could act as a proton
donor and contribute to the red shift in the PL emission spectrum. The DCM solvent helped to
deprotonate the hydrogen molecule and converted the quinoline ring into a proton donor, as
shown in Fig. 13(c). Similarly, when DMSO was used as the solvent, the quinoline ring could act
as a strong electron acceptor due to the presence of the more electronegatively charged sulfur
and the greater number of alkyl groups, as shown in Fig. 13(d). The sulfur and alkyl group in
DMSO converted quinoline into a strong proton donor and this led to the large red shift in the PL
spectrum compared with the DCM solvent. DCM contains no additional alkyl groups and thus
the quinoline group could act as a weak proton donor. The electronic charge was relocated from
the solvent molecule to the phenolic oxygen in the quinoline ring in the DCM and DMSO
solvents (i.e., basic solvents) to act as a proton donor. This process simultaneously stabilized the
ground and excited state, where the latter greatly cancelled some of the negative charge flow
from the phenolic oxygen to the ring after excitation. Thus, the red shift occurred in the PL
spectrum. Geometrical rearrangement also occurred after excitation and the fluorescence was red
shifted. For the basic solvents, the mechanism first involved the ionization of both DCM and
DMSO solvents. Furthermore, the negative ions were surrounded by Alq3 rings and they pulled
the ring to act as possible electron donors to the quinoline group.
According to the mechanism described above, the PL spectrum for sample A in a basic medium
of DMSO in the solution state had a red shift of ≈ ∆λ = 6 nm compared with DCM. Similarly, for
sample B, the red shift was ≈ ∆λ = 4 nm, and that for sample C was ≈ ∆λ = 5 nm in DMSO
compared with that in DCM at 10–3 M. The red shift with DMSO was not much larger although
quinoline could act as a strong proton donor because the Alq3 complex did not support this
process. For sample A in an acidic medium at a concentration at 10–3 M, the PL spectrum
exhibited a blue shift of ≈ ∆λ = 4 nm in AC compared with FC. For sample B in an acidic
medium at a concentration of 10–3 M, there was a blue shift of ∆λ = 6 nm, and the blue shift was
∆λ = 4 nm for sample C. The observed ∆λ was similar in both the acidic and basic media. The

ACCEPTED MANUSCRIPT
influence of the dipole moment of the solvents on the blue or red shift in the PL spectrum should
not be neglected.
Conclusion
In this study, we prepared α–Alq3 using a co–precipitation method. Stagnation of the α–Alq3
phase and improved phase purity were achieved by thermal treatment in an Ar atmosphere at
200°C. Furthermore, the phase changed from α– to ε–Alq3 after thermal treatment at 320°C in an
Ar atmosphere. The formation of α–Alq3, its improved phase purity, and the change of phase
from α– to ε–Alq3 were confirmed by the XRD data, as well as the FTIR and NMR results. The
UV–Vis absorption spectra obtained with acidic (AC and FC) and basic (DCM and DMSO)
media indicated that blue or red shifts occurred in the PL spectra with different solvents. The
increased or decreased absorption shifts were attributed to the effects of the dielectric constants
(i.e., polarity) of the solvents employed. The band gap energy value was lower for sample B
compared with sample A, but higher for sample C and the ε–phase had a “wobbly” structure. The
PL emission spectrum obtained for the powder sample A showed that λmax was at 516 nm as well
as at 514 nm for sample B and at 498 nm for sample C. The large fluorescence blue shift was due
to the conversion from α– to ε–Alq3. Furthermore, the blue shifts in the PL spectra with acidic
media and the red shifts with basic media at concentrations of 10–3 and 10–6 M were due to the
different dielectric constants of the solvents. The fluorescence shift of ε–Alq3 in the solution state
was more susceptible in the basic solvents but less susceptible in the acidic solvents compared
with α–Alq3. TGA/DTA analysis also supported the presence of the α–Alq3 phase in sample A
and sample B and it was more stable, whereas ε–Alq3 was weaker and it broke down more
readily. The solvatochromatic effect based on the electron acceptor and proton donor nature of
the solvent was responsible for the blue and red shifts in the PL spectra obtained for the Alq3
samples.
Acknowledgments: The authors are thankful to DST (SERB) (Sanction Project Ref. No. SB/
FTP/PS–006/2014, dt. 17/03/2015) for financial assistance and NIT Uttarakhand for constant
encouragement. The authors are also thankful to the MRC laboratory, MNIT Jaipur for UV–Vis
absorption and PL measurements.
References
[1] R. Martin, J. Kress, I. Campbell, D. Smith, Phys. Rev. B, 61 (2000) 15804.
[2] M. Amati, F. Lelj, Chem. Phys. Lett., 358 (2002) 144.
[3] C.W. Tang, S.A. VanSlyke, Appl. Phys. Lett., 51 (1987) 913.

ACCEPTED MANUSCRIPT
[4] J.A. Gómez, F.A. Castro, F. Nüesch, L. Zuppiroli, C.F.O. Graeff, Appl. Phys. A,
108(3) (2012) 727.
[5] M. Janghouri, E. Mohajeranin, A. Khabazi, Z. Abedi, H. Razavi, J. Lumin., 140 (2013) 7.
[6] T. Abhijith, T.V. Arun Kumar, V.S. Reddy, Nanotechnology, 28 (2017) 9.
[7] K.P. Dhakal, H. Leea, J. Kima, Synth. Metals, 190 (2014) 44.
[8] B.–R. Huang, C.–C. Liao, W.–T. Fan, J.–H. Wu, C.–C. Chen, Y.–P. Lin, J.–Y. Li, S.–P.
Chen, W.–C. Ke, N.–C. Chen, Appl. Surface Sci., 303 (2014) 319.
[9] M. Brinkmann, G. Gadret, M. Muccini, C. Taliani, N. Masciocchi, A. Sironi, J. Am. Chem.
Soc., 122 (2000) 5147.
[10] M. Cölle, J. Gmeiner, W. Milius, H. Hillebrecht, W. Brütting, Adv. Mater., 13 (2003) 108.
[11] M. Rajeswaran, T.N. Blanton, C.W. Tang, W.C. Lenhart, S.C. Switalski, D.J. Giesen, B.J.
Antalek, T.D. Pawlik, D.Y. Kondakov, N. Zumbulyadis, R.H. Young, Polyhedron, 28
(2009) 835.
[12] M. Rajeswaran, T.N. Blanton, J. Chem. Crystallography, 35 (1) (2005)71.
[13] F. Suzuki, T. Fukushima, M. Fukuchi, H. Kaji, J. Phys. Chem. C, 117 (2013) 18809.
[14] T. Fukushima, H. Kaji, Org. Electron.,13(2012)2985.
[15] J.C.S. Costa, C.F.R.A.C. Lima, L.M. N.B.F. Santos, J. Phys. Chem. C, 118 (2014) 21762.
[16] M. Cölle, W. Brűtting, Phys. Stat. Sol., 201 (2004) 1095.
[17] M. Braun, J. Gmeiner, M. Tzolov, M. Cölle, F.D. Meyer, W. Milius, H. Hillebrecht, O.
Wendland, J.U. von Schutz, W. Brütting, J. Chem. Phys., 114 (2001) 9625.
[18] B.S. Xua, H. Wanga, Y. Haob, Z. Gaob, H. Zhou, J. Lumin., 122–123 (2007) 663.
[19] M. Cölle, J. Gmeiner, W. Milius, H. Hillebrecht, W. Brütting, Proceedings of the EL2002,
Ghent (2002) 133.
[20] X. Liu, J. M. Cole, K. Sing Low, J. Phys. Chem. C 117 (2013) 14731.
[21] J.–J. Aaron, M. Buna, C. Parkanyi, M.S. Antonious, A. Tine, L. Cisse, J. Fluoresc., 5 (1995)
337.
[22] C. Parkanyi, M.R. Stem–Beren, O.R. Martinez, J.J. Aaron, M. Bulaceanu–Macnair, A.F.
Arrieta, Spectrochim. Acta 60A (2004) 1805.
[23] A.B. Mabire, M.P. Robin, W.–D. Quan, H. Willcock, V. G. Stavros, R. K. O'Reilly, Chem
Commun (Camb.) 51(47) (2015) 9733.
[24] D. Singh, S. K. Gupta, Y. Sonvanea, I. Lukacevic, J. Mater. Chem. C, 4 (2016) 6386.
[25] S.A. –Hatab, V.A. Spata, S. Matsika, S.A. –Hatab, V.A. Spata, S. Matsika, J. Phys. Chem.
A, 121 (6) (2017) 1213.
[26] I.M. Nagpure, M.M. Duvenhage, S.S. Pitale, O.M. Ntwaeaborwa, J.J. Terblans, H.C. Swart,
J. Fluore., 22 (2012) 1271.

ACCEPTED MANUSCRIPT
[27] J.G. Mahakhode, S.J. Dhoble, C.P. Joshi, S.V. Moharil, Bull. Mater.Sci., 34(7) (2011) 1649.
[28] X. Bing–She, H. Yu–Ying, W. Huaa, Z. He–Fenga, L. Xu–Guangc, C. Ming–Weid, Solid
State Comm. 136 (2005) 318.
[29] G. Baldacchini, P. Chiacchiaretta, R.B. Pode, M. A. Vincenti, Q.–M. Wang, Low Temp.
Phys., 38 (2012) 786.
[30] S.M. Sawde, R.R. Patil, S.V. Moharil, Inter. J. of Lumin. and Appls., 6(3) (2016) 152.
[31] F.F. Muhammad, A.I.A. Hapip, K. Sulaiman, J. of Organomet. Chem., 695 (2010) 2526.
[32] Y. Akurai, Y. Hosoi, H. Ishii, Y. Ouchi, G. Galvan, A. Kobitski, K. Seki, J. of Appl. Phys.,
96 (2004) 10.
[33] A.D. Esposti, M. Brinkmann, G. Ruani, J. Chem. Phys., 116 (2002) 798.
[34] N. Khaorapaponga, M, Ogawa, J. Phy. and Chem. of Sol., 69 (2008) 941.
[35] S.S.S. Abuthahir, A.J.A. Nasser, S. Rajendran, G. Brindha, Chem. Sci. Transactions, 3 (1)
(2014) 303.
[36] M.M. Duvenhage, H.G. Visser, O.M. Ntwaeaborwa, H.C. Swart, Physica B, 439 (2014) 46.
[37] J. C.S. Costa, R.J.S. Taveira, C.F.R.A.C. Lima, A. Mendes, L.M.N.B.F. Santos, Opt.
Materials, 58 (2016) 51.
[38] A. Belay, E. Libnedengel, H.K. Kima, Y.–H. Hwanga, Lumin,, 31 (2016) 118.
[39] B. Xu, H. Wang, Y. Hao, Z. Gao, H. Zhou, J. Lumin., 122–123 (2007) 663.
[40] W. Xie, Z. Pang, Y. Zhao, F. Jiang, H. Yuan, H. Song, S. Han, J. Cryst. Growth, 404 (2014)
164.
[41] J. Tauc, R. Grigorovic, A. Vancu, Phy. Sta. Solidi, 15 (1966) 627.
[42] T.A. Hopkins, K. Meerholz, S. Shaheen, M.L. Anderson, A. Schmidt, B. Kippelen, A.B.
Padias, H.K. Hall Jr., N. Peyghambarian, N. R. Armstrong, Chem. Mater., 8(2) (1996) 344.
[43] C.H. Chen, J. Shi, Coord. Chem. Rev., 171 (1998) 161.
[44] X. Liu, J.M. Cole, K.S. Low, J. Phys. Chem. C., 117 (2013) 14731.
[45] M. Homocianu, A. Airinei, D.O. Dorohoi, J. of Adv. Res. in Phys., 2(1) (2011) 011105.
[46] K. Gottfried, K. Rechthaler, Pure & App. Chem., 65 (8) (1993) 1647.
[47] T. Lee, M.S. Lin, Crystal Growth & Design, 7 (9) (2007) 1803.

ACCEPTED MANUSCRIPT
Figure and table captions
Fig. 1: X–ray diffraction patterns obtained for: (a) sample A (i.e., α–Alq3 phase), (b) sample B
(i.e., α–Alq3 phase), and (c) sample C (i.e., ε–Alq3 phase).
Fig. 2: FTIR analysis for: (a) sample A, (b) sample B, and (c) sample C (inset shows the FTIR
spectrum for Alq3 in the range from 2000 to 4000 cm–1).
Fig. 3: NMR analysis of: (a) sample A, (b) sample B, and (c) sample C.
Fig. 4: Absorption spectra obtained for: (a) sample A at 10–3 M, (b) sample A at 10–6 M, (c)
sample B at 10–3 M, and (d) sample B at 10–6 M.
Fig. 5: Absorption spectra obtained for: (a) sample C at 10–3 M and (b) sample C at 10–6 M.
Fig. 6: Band gap analysis results for: sample A (a) in AC at 10–3 M, (b) in AC at 10–6 M, (c) in
FC at 10–3 M, (d) in FC at 10–6 M, (e) in DCM at 10–3 M, (e) in DCM at 10–6 M, (f) in
DMSO at 10–3 M, and (h) in DMSO at 10–6 M.
Fig. 7: Band gap analysis results for: sample B (a) in AC at 10–3 M, (b) in AC at 10–6 M, (c) in
FC at 10–3 M, (d) in FC at 10–6 M, (e) in DCM at 10–3 M, (e) in DCM at 10–6 M, (f) in
DMSO at 10–3 M, and (h) in DMSO at 10–6 M.
Fig. 8: Band gap analysis results for: sample C (a) in AC at 10–3 M, (b) in AC at 10–6 M, (c) in
FC at 10–3 M, (d) in FC at 10–6 M, (e) in DCM at 10–3 M, (e) in DCM at 10–6 M, (f) in
DMSO at 10–3 M, and (h) in DMSO at 10–6 M.
Fig. 9: Photoluminescence analysis results for: (a) sample A (i.e., α–Alq3 phase), (b) sample B
(i.e., α–Alq3 phase), and (c) sample C (i.e., ε–Alq3 phase) in powder form.
Fig. 10: Photoluminescence analysis results for powder samples in solution form in different
acidic and basic media: (a) sample A at 10–3 M, (b) sample A at 10–6 M, (c) sample B at
10–3 M, and (d) sample B at 10–6 M.

ACCEPTED MANUSCRIPT
Fig. 11: Photoluminescence analysis results for powder samples in solution form in different
acidic and basic media: (a) sample C at 10–3 M and (b) sample C at 10–6 M.
Fig. 12: Thermal analysis (TGA/DTA) results for: (a) sample A (i.e., α–Alq3 phase), (b) sample
B (i.e., α–Alq3 phase), and (c) sample C (i.e., ε–Alq3 phase) in powder form.
Fig. 13: Mechanism responsible for the solvatochromatic shift by aluminum quinolates in acidic
and basic media.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT