Photoactive azoimine dyes: 4-(2-Pyridylazo)-N,N-diethylaniline and 4-(2-pyridylazo)-N,N-dimethylaniline: Computational and experimental investigation

Article abstract of DOI:10.1016/j.saa.2011.11.007

4-(2-Pyridylazo)-N,N-dimethylaniline and 4-(2-pyridylazo)-N,N- diethylaniline, two photoactive azoimine dyes, were prepared from the reaction of 2-aminopyridine with N,N-dialkyl-1,4-nitrosoaniline at room temperature. Structural characterizations of these dyes using single crystal X-ray diffraction, ¹H NMR, elemental analysis, mass spectroscopy and IR spectroscopy have been carried out. The X-ray structure indicates a trans configuration around the azo group. The photochemical behavior of these compounds differs from that of 2-phenylazopyridine, the non-dialkylamino substituent compound. The synthesized compounds show emission spectra at room temperature while 2-phenylazopyridine does not. The excitation spectra of these compounds differ from their absorption spectra which can be explained on the basis of the trans to cis photoisomerization which is supported by the TD-PBE0/6-31G(d,p) calculations. Both oxidation of the dialkylamino substituents (-NR₂; R = -CH₃ and -C₂H₅) and reduction of -NN-/-NN-^- and -NN-^-/-NN-^2- were observed in the cyclic voltammogram indicating a π-acidity of both dyes.

Full text of DOI:10.1016/j.saa.2011.11.007

Spectrochimica Acta Part A 86 (2012) 538–546
Contents lists available at SciVerse ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Photoactive azoimine dyes: 4-(2-Pyridylazo)-N,N-diethylaniline and
4-(2-pyridylazo)-N,N-dimethylaniline: Computational and
experimental investigation
Suthirat Yoopensuka, Pornthip Tongyingb, Kanidtha Hansongnerna, Chaveng Pakawatchaia,
Saowanit Saithonga, Yuthana Tantirungrotechaib,c,∗∗, Nararak Leesakula,∗
a Department of Chemistry and Center for Innovation in Chemistry, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand
b National Nanotechnology Center, National Science and Technology Development Agency, Thailand Science Park, Klong Luang, Pathumthani 12120, Thailand
c Department of Chemistry, Faculty of Science and Technology, Thammasat University, Pathumthani 12121, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 May 2011
Received in revised form 8 November 2011
Accepted 9 November 2011
4-(2-Pyridylazo)-N,N-dimethylaniline and 4-(2-pyridylazo)-N,N-diethylaniline, two photoactive
azoimine dyes, were prepared from the reaction of 2-aminopyridine with N,N-dialkyl-1,4-nitrosoaniline
at room temperature. Structural characterizations of these dyes using single crystal X-ray diffraction,
1H NMR, elemental analysis, mass spectroscopy and IR spectroscopy have been carried out. The X-ray
structure indicates a trans configuration around the azo group. The photochemical behavior of these
compounds differs from that of 2-phenylazopyridine, the non-dialkylamino substituent compound.
The synthesized compounds show emission spectra at room temperature while 2-phenylazopyridine
does not. The excitation spectra of these compounds differ from their absorption spectra which can
be explained on the basis of the trans to cis photoisomerization which is supported by the TD-PBE0/6-
31G(d,p) calculations. Both oxidation of the dialkylamino substituents (–NR2; R = –CH3 and –C2H5)
and reduction of –N N–/–N N–− and –N N–−/–N N–2− were observed in the cyclic voltammogram
indicating a ␲-acidity of both dyes.
Keywords:
Photoactive dye
Photoisomerization
Azoimine dye
TDDFT
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
There has been much interest in designing new photoactive dyes
based on the azo chromophore [10,11]. Azoimine (–N N–C N–)
Azo (–N N–) compounds are well known photoactive mate-
rials. They are widely used as dyes in the textile industry [1,2],
as substrates for photo aligning for liquid crystals [3], in opti-
cal data storage devices [4] and light sensitizers in dye-sensitized
solar cells [5]. A good example of these compounds is azoben-
zene: a simple but important azo moiety widely used to synthesize
many new optical compounds. Azobenzene is an emissive fluo-
rescence compound. It can also undergo two important reactions:
photoisomerization and photocyclization [6]. The trans-to-cis pho-
toisomerization of azobenzene takes place around the azo moiety.
Many applications of azobenzene and its derivatives have been
explored based on these photoisomerization and emission char-
acteristics [7–9].
has been a popular modification [12,13]. The azoimine dyes can
be used as optical substances on their own owing to their pho-
tochromic properties [13,14]. Moreover, because of their low
lying ␲*-molecular orbitals, azoimine dyes are extensively used
as chelating ligands with transition metal ions such as Ru(II) and
Os(II) [15–17]. Nevertheless, the photochromic characteristics of
azoimine dye and its complexes with transition metal ions have
been less explored [16].
A
derivative of 2-phenylazopyridine (azpy), namely 4-(2-
pyridylazo)-N,N-dimethylaniline or dmazpy, is an azoimine
(–N–C–N N–) dye which has been reported in the literature [18]
with an acronyms like PADA [19] and azpy-NMe2 [20]. As a result
of the high ␲-acidity of azo (–N N–) compounds, dmazpy has been
utilized as a ligand for forming many transition metal complexes
[16,19,21,22] some of which have pharmacological applications
[21,23]. Recently we solved the X-ray structure of dmazpy and
reported it in the literature [24].
By adding the dialkylamino substituents (–NR2; R = –CH3 and
–C2H5), the photo-physical properties of 2-phenylazopyridine
(azpy, see Fig. 1a) can be tuned. We observed experimentally that
azpy is a non-emissive dye at room temperature [25] whereas
∗
Corresponding author.
Tel.: +66 74 28 8421; fax: +66 74 55 8841.
∗∗
Corresponding author at: Department of Chemistry, Faculty of Science and
Technology, Thammasat University, Pathumthani 12121, Thailand. Tel.: +66 2 564
4440x2400.
E-mail addresses: yt203y@gmail.com (Y. Tantirungrotechai),
nararak.le@psu.ac.th (N. Leesakul).
1386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2011.11.007

S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
539
Acetonitrile for electrochemical studies was dried over Fisher
˚
Type 4 A molecular sieves. Tetrabutylammoniumhexafluorophos-
phate (TBAH, puriss. from Fluka) was recrystallized in acetonitrile.
Ferrocene from Fluka was used as an internal standard in the elec-
trochemical measurement.
X-ray diffraction data were collected with a Bruker APEX
diffractometer. Structures of single crystals were solved by using
SHLEXL97 [28]. Electronic spectra were recorded with a Shimadzu
UV-160A spectrophotometer. Emission spectra were recorded with
a Perkin Elmer luminescence spectrometer LS55 (FL winlab soft-
ware). Excitation and emission slit widths were 8 nm. Fluorescence
lifetimes were measured by the modulation technique with a dig-
ital storage oscilloscope and fitted by LD APP program [29]. The
modular setup was composed of an LED (370 nm), an excitation
gray filter U340 nm, and a long pass emission filter GG400 nm.
The monitored frequencies were 3–15 MHz. Colloidal silica was
used as a reference for scattered light. FTIR spectra (KBr disk,
4000–400 cm−1) were measured with a BX PerkinElmer FTIR spec-
trophotometer. 1H NMR data were collected by using a CDCl3
solvent with a Bruker 300 MHz NMR spectrometer. Tetramethyl-
silane (TMS) was used as the internal standard. ES-MS data were
recorded on a VG Quattro triple quadrupole mass spectrometer
with an alcohol mobile phase (Wollongong University, Australia).
Elemental analysis was obtained from a Carlo Erba EA 1108 Ele-
mental Analyzer (University of Bristol, England). Electrochemical
measurements were carried out by cyclic voltammetry. The cyclic
voltammograms were recorded with MacLab (4e AD Instruments
with potentiostat) and analyzed by using ECHEM version 1.5.
Three electrodes consisting of a glassy carbon disc electrode, a
platinum-wire auxiliary electrode and a saturated calomel refer-
ence electrode (SCE) were used. A ferrocene–ferrocenium (Fc/Fc+)
reversible couple was observed at 0.44 V at a scan rate of 0.1 V s−1
in acetonitrile.
Fig. 1. Molecular structures of (a) azpy, (b) dmazpy and (c) deazpy with proton
numbering.
the presence of –NR2 substituents pendant to the phenyl ring
causes our proposed dyes to be fluorescence emitters. Further-
more, the photo-physical properties of dmazpy itself have never
been reported.
2.2. Preparation of 4-(2-pyridylazo)-N,N-diethylaniline (deazpy)
We present here
a synthesis and characterization of 4-
2-Aminopyridine (0.50 g, 5 mmol) was dissolved in benzene
(5 mL). The solution was heated at 80 ◦C for 10 min. Then 25 M
NaOH (6 mL) was slowly added into the 2-aminopyridine solu-
tion. After that, N,N-diethyl-1,4-nitrosoaniline (0.80 g, 5 mmol) was
slowly added to the warm solution mixture. During this procedure,
the temperature was kept constant at 80 ◦C and the solution mix-
ture was continuously stirred. More benzene (5 mL) was added into
the solution mixture which was then refluxed for 11 h. A reddish-
orange solution was obtained. The reaction mixture was filtered;
the filtrate was extracted with benzene (25 mL) twice. The mixture
was evaporated to 5 mL. The product was purified by using silica
gel column chromatography. The reddish-orange band was eluted
with hexane:ethylacetate mixtures in ratios of 9:1, 8:2, and 7:3,
respectively. The deazpy solution was evaporated to dryness. Then
the reddish-orange powder of deazpy was collected, washed three
times with diethyl ether, and dried over silica gel in a desiccator
overnight. The yield was 41.2% (0.52 g). m.p. 109–110 ◦C. CHN anal-
ysis: Found: C, 70.06; H, 6.80; N, 21.85%. Calcd for deazpy: C, 70.84;
H, 7.13; N, 22.03%. ES-MS: m/z 255 (M+H)+ 100% and 256 (M+2H)+
35%. The single crystal for the X-ray diffraction study was obtained
(2-pyridylazo)-N,N-diethylaniline (deazpy or azpy-NEt2), a new
azoimine dye, by using the modified azpy synthesis pathway [15].
To our knowledge, deazpy is a novel compound which has not
yet been reported. Only two reports of the bromo-substituted
compound Br-azpy-NEt2 used as a ligand in Os(II) and Pd(II) com-
plexes have appeared in the literature [26,27]. We, therefore, report
structural characterization (single crystal XRD, IR and 1H NMR),
photo-physical properties arising from the photoisomerization and
electrochemical characterization of dmazpy (Fig. 1b) and deazpy
(Fig. 1c). We summarize the X-ray structure of dmazpy here along-
side that of deazpy for completeness as it has already been reported
elsewhere [24]. To gain insights on experimental measurements,
we conduct density functional and time-independent density func-
tional calculations. The PBE0 hybrid functional, a hybrid functional
with a large portion of an exact Hartree–Fock exchange, is chosen
due to its excellent spectroscopic prediction. Our studies reveal that
this class of compound has interesting photo-physical properties
which make them suitable candidates as photoactive dyes or as
ligands for photoharvesting applications.
also from a 3:2 hexane:methanol mixture. IR ␯ (KBr disc)/cm−1
:
2. Experimental methods
2968(m), 1609(s), 1506(s), 1525(s), 1143(s), 824(m), 794(m). 1H
NMR (see from the Supplementary data, CDCl3, ␦, ppm): 8.60 (1H-
py, dd), 7.75 (1H-py, td), 7.70 (1H-py, td), 7.22 (1H-py, dd), 7.92
(2H-Ar, d), 6.66 (2H-Ar, d), 1.17 (4H–CH2, q), 3.42 (6H–CH3, t).
Crystallographic data for the structural analysis of deazpy have
been deposited at the Cambridge Crystallographic Data Centre
(CCDC no. 796323) and can be obtained upon request. A summary
of the crystallographic data and structural information of deazpy
are reported as supplementary data.
2.1. Materials and instrumentation
2-Aminopyridine was obtained from Fluka. N,N-Dimethyl-1,4-
nitrosoaniline and N,N-diethyl-1,4-nitrosoaniline were purchased
from Sigma–Aldrich. All chemicals and solvents for synthesis were
reagent grade and used as received. Silica gel (100 mesh) from
BD Schenker was used for column chromatographic purification.

540
S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
The preparation of dmazpy followed the same procedure as
that of deazpy. The N,N-diethyl-1,4-nitrosoaniline was replaced
by an equivalent amount of N,N-dimethyl-1,4-nitrosoaniline. The
reaction mixture was refluxed for 9 h. Finally, a red crystal was
obtained with a yield of 46% (0.52 g). The melting point of dmazpy is
104–105 ◦C. CHN analysis: Found: C, 69.85; H, 6.34; N, 23.81%. Calcd
for dmazpy: C, 69.00; H, 6.24; N, 24.76%. ES-MS: m/z 227 (MH+,
100%). IR ␯ (KBr disc)/cm−1: 2896(m), 1610(s), 1509(s), 1528(s),
1147(s), 815(m). 1H NMR (see from the Supplementary data, CDCl3,
␦, ppm): 8.60 (1H-py, dd), 7.76 (1H-py, td), 7.70 (1H-py, td), 7.22
(1H-py, dd), 7.95 (2H-Ar, d), 6.69 (2H-Ar, d), 3.04 (6H–CH3, s).
2.3. Computational study
All compounds were geometry optimized in gas phase by using
the PBE0:6-31G(d,p) method. A Hessian calculation was performed
to verify the stationary structure and to obtain the vibrational
frequencies. The scaling factor of 0.96 was applied to the calcu-
Fig. 2. The experimental and scaled PBE0/6-31G** (shown as vertical lines) IR spec-
tra of (a) dmazpy and (b) deazpy.
lated frequencies as recommended in the literature [30,31]. The 1
H
NMR spectrum was calculated by using B3LYP/EPR-II and WP04/6-
31G(d,p) at the optimized geometry. The EPR-II basis set is rather
flexible in the inner core region, and is therefore suitable for mag-
netic properties predictions [32]. Moreover, we considered the
WP04 functional which is a modified B3LYP functional specifi-
cally reparameterized for proton NMR prediction [35]. Jain et al.
assessed the performance of several functionals and basis sets [36].
They recommended the WP04/6-31G(d,p) model for practical 1H
NMR calculation. The TMS compound was used as the reference
for the chemical shift calculation. Absorption and excitation spec-
tra were predicted by calculating the electronic excitation energy at
the TD-PBE0/6-31++G(d,p) level with the effect of methanol solvent
represented by the CPCM model. All calculations were performed
by using Gaussian 03 [33]. To simulate the IR spectrum and to visu-
alize the molecular orbital involved in the spectroscopic transition,
the Avogadro program was used [34].
of N N bond order in dmazpy and deazpy is due to the elec-
tron delocalization from dialkylamino electron donating group.
The delocalized ␲-electrons are partially filled in the ␲* orbital of
the azo chromophore, hence decreasing the N N bond order. The
dialkylamino substituent, therefore, increases the electron delocal-
ization causing the lowering of ␯(N N) frequencies.
Fig. 2 shows an overall agreement between the experimental
and simulated vibrational spectra of both compounds. Although,
they yield the correct qualitative picture, the calculated N N fre-
quencies from the gas phase structure are greater than those
obtained from the experimental data. The scaled N N stretch-
ing frequency of trans-azpy appeared at 1609 cm−1 while that of
dmazpy (and deazpy) appeared at 1509 (1506) and 1528 (1525)
cm−1. We expect a better agreement if the solvent effect is incor-
porated in the frequency calculation.
Table 1 compares the experimental chemical shift of dmazpy
and deazpy with those of the B3LYP/EPR-II and WP04/6-31G(d,p)
models. Overall agreement with the experiment can be observed
in the WP04/6-31G(d,p) model. The WP04/6-31G(d,p) model
performs much better than the B3LYP/EPR-II model; the mean
unsigned error of B3LYP/EPR-II is 0.21 kcal/mol whereas that of
WP04/6-31G(d,p) is 0.16–0.17 kcal/mol. This confirms the recom-
mendation by Jain et al. [34] that the WP04/6-31G(d,p) model is
a best practical model for predicting the proton chemical shift.
The maximum error of 1H B3LYP/EPR-II shift occurs at the 4-H
proton attached to the pyridine ring (see Fig. 1). The B3LYP/EPR-
II overshoots the proton shift by about 1.0 ppm whereas the
WP04/6-31G(d,p) correctly predicts this chemical shift. In general,
it appears that the B3LYP/EPR-II model overestimates the proton
shift whereas the WP04/6-31G(d,p) underestimates it. Conversely,
the WP04/6-31G(d,p) result is in agreement with the experimental
obtained proton chemical shift.
3. Results and discussion
3.1. Synthesis
Dmazpy and deazpy compounds were synthesized from the
reactions between 2-aminopyridine, and either N,N-dimethyl-1,4-
nitrosoaniline or N,N-diethyl-1,4-nitrosoaniline in a strong basic
solution according to Scheme 1.
Under the experimental conditions, the yields of dmazpy and
deazpy were both moderate. The nitrosoaniline could not react
completely with 2-aminopyridine. The green band of nitrosoani-
line in the column chromatography purification was still present
even if we doubled the amount of 2-aminopyridine. Other attempts
such as carrying out the reaction in the absence of 25 M NaOH or
using toluene as solvent could not drive the reaction to completion.
The dmazpy and deazpy compounds obtained are stable to air and
moisture. They are completely soluble in all organic solvents and
slightly soluble in water.
3.3. X-ray structure determination
Tables 2 and 3 summarize the crystallographic data of both com-
pounds (see Fig. 3 for atom numbering). Both compounds adopt
the trans configuration around the N N azo group. Both pyridine
and phenyl rings are almost coplanar with the dihedral angles
around the azo moiety, ∠C(5)N(2)N(3)C(6), being −179.68(10)◦ and
−179.2(2)◦ for dmazpy and deazpy, respectively. This coplanarity
allows good electron transfer from the dialkylamino substituents to
the azo group. The corresponding bond angles of both compounds
are almost the same. The N N bond lengths of dmazpy and deazpy
3.2. Infrared and 1H NMR spectra
Some characteristic peaks of dmazpy and deazpy are in the
1600–1400 cm−1 region (see Fig. 2). They are the stretching modes
of C C and C N in pyridine and phenyl rings. The most impor-
tant intense peak was ␯(N N) which signifies the ␲-acidity of the
azo dye. The ␯(N N) of dmazpy and deazpy appeared at 1402 and
1394 cm−1, respectively, whereas that of azpy was at 1424 cm−1
[15]. The red shift of ␯(N N) indicates that the N N bond of the
dmazpy and deazpy is weaker than that of free azpy. The decrease
˚
are 1.2566(16) and 1.211(3) A. Structural comparison with azpy is
impossible because no X-ray data of trans-azpy is available [37]. We

S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
541
Scheme 1.
Table 1
Comparison of experimental and calculated 1H NMR of dmazpy and deazpy (see Fig. 1 for proton numbering).
H-positions
ı, ppm
Dmazpy
Expt.
Deazpy
Expt.
B3LYP/EPR-II
WP04/6-31G(d,p)
B3LYP/EPR-II
WP04/6-31G(d,p)
1-H
2-H
3-H
4-H
5-H
6-H
7-H
8-H
8.60
7.76
7.70
7.22
7.95
6.69
3.04
8.76
7.50
8.11
8.26
8.25
6.90
3.14
8.58
7.08
7.53
7.76
7.79
6.53
3.00
8.60
7.75
7.70
7.22
7.92
6.66
3.42
1.17
8.74
7.47
8.10
8.24
8.19
7.05
3.53
1.16
8.55
7.05
7.53
7.75
7.76
6.59
3.26
1.13
MUE (MAX)a
0.21 (0.50)
0.17 (0.30)
0.21 (0.49)
0.16 (0.29)
a
MUE and MAX denote the mean and the maximum unsigned error respectively.
therefore resort to the DFT computation for the geometry of these
compounds.
difference between deazpy and dmazpy. Comparing with the cal-
culated N N bond length of azpy, the –NR2 substituent clearly
weaken the N N bond of deazpy and dmazpy. The electron delo-
calization from the dialkylamino substituent on the phenyl ring
to the N N chromophore affects the C(phenyl)-N(azo) bond to a
greater degree than the C(pyridine)-N(azo) bond. The C(phenyl)-
The overall agreement between selected experimental and
PBE0/6-31G(d,p) geometry parameters is excellent (see Table 4).
For example, the calculated N N bond length of trans-dmazpy is
˚
only 0.003 A greater than the experimental bond length. The result
˚
supports the experimental observation that there is no structural
N(azo) bond length is 1.396 A while the C(pyridine)-N(azo)
Table 2
Summary of crystallographic information for dmazpy [24] and deazpy.
Crystal parameters
Dmazpy
Deazpy
Identification code
Empirical formula
Formula weight
Temperature
Dmazp
C13H14N4
226.28
293(2) K
0.71073 A
C15H18N4
254.33
293(2) K
˚
˚
Wavelength
0.71073 A
Crystal system
Space group
Unit cell dimensions
Monoclinic
P21/n (No. 14)
Orthorhombic
P212121 (No. 19)
◦
◦
◦
◦
◦
˚
˚
˚
˚
a = 6.2322(4) A, ˛ = 90 b = 19.9353(11) A, ˇ = 96.003(1) c = 9.6404(6) A, ꢀ = 90
a = 7.4904(7) A ˛ = 90
˚
b = 9.1010(8) A, ˇ = 90
◦
˚
c = 21.0145(18) A, ꢀ = 90
3
3
˚
˚
Volume
1191.16 (13) A
1432.6 (2) A
Z
4
4
Density (calculated)
Absorption coefficient
F(000)
1.262 mg/m3
1.179 mg/m3
0.073 mm−1
544
0.079 mm−1
480
Crystal size
Theta range for data collection
Index ranges
0.279 mm × 0.256 mm × 0.061 mm
0.312 mm × 0.165 mm × 0.032 mm
2.04–25.00◦
1.94–24.99◦
−7 ≤ h ≤ 7, −23 ≤ k ≤ 23, −11 ≤ l ≤ 11
−8 ≤ h ≤ 8, −10 ≤ k ≤ 10,
−24 ≤ l ≤ 24
Reflections collected
12,811
13,573
Independent reflections
Completeness to theta = 25.00◦
Absorption correction
Refinement method
2101 [R(int) = 0.0231]
100.0%
Semi-empirical from equivalents
Full-matrix least-squares on F2
2101/0/156
2511 [R(int) = 0.0455]
100.0%
Semi-empirical from equivalents
Full-matrix least-squares on F2
2511/0/175
Data/restraints/parameters
Goodness-of-fit on F2
1.042
1.056
Final R indices [I > 2ꢁ(I)]
R indices (all data)
Largest diff. peak and hole
R1 = 0.0379, wR2 = 0.1014
R1 = 0.0531, wR2 = 0.1056
R1 = 0.0801, wR2 = 0.1167
0.237, −0.141 e A
R1 = 0.0457, wR2 = 0.1077
−3
−3
˚
˚
0.116, −0.185 e A

542
S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
Fig. 3. X-ray structures of (a) dmazpy and (b) deazpy.
˚
possible through a photochemical process [9] as we will discuss
later.
Analysis of molecular packing in the deazpy crystal structure
found weak intermolecular C–H· · ·N hydrogen bonds between the
N(pyridine) atom and the H(pyridine) on the adjacent molecule,
bond length is 1.416 A. Calculation also shows that phenyl and
pyridine rings are almost coplanar. The ∠C(4)C(5)C(9)C(10) dihe-
dral angle is 11.7◦ which is comparable to the experimental
result.
Although both dmazpy and deazpy are determined to be in the
trans configuration in the ground state, another possible configu-
ration is the cis configuration. The cis form has a shorter N N bond
than the trans form. The coplanarity between phenyl and pyridine
rings is lost in the cis form due to the steric effect. Using the PBE0/6-
31G(d,p) calculation, the cis form is about 14.5 kcal/mol greater in
energy than the trans form. The trans to cis conversion is usually
i
˚
C(2)–H(2)· · ·N(1) , with the C· · ·N distance of 3.345(4) A (where
the symmetry code i = −x + 1, y − 1/2, −z + 5/2, see Supplementary
data). Intermolecular ␲-␲ stacking interactions occur between the
pyridine ring and the adjacent phenyl ring from edge-to-face ori-
entation. The centroid–centroid distances are found to alternate
˚
between 3.734 and 3.761 A as illustrated in supplementary data.
Table 3
˚
Selected bond lengths (in A) and bond angles (in degree) for dmazpy and deazpy
derived from X-ray data (see Fig. 3 for atom numbering).
Bond parameters
Dmazpy
Deazpy
˚
Bond lengths (A)
N(1)–C(5)
N(1)–C(1)
C(4)–C(5)
N(2)–C(5)
N(2)–N(3)
N(3)–C(6)
N(4)–C(9)
1.3311(16)
1.3316(18)
1.3863(19)
1.4337(16)
1.2566(16)
1.4035(16)
1.3586(17)
1.313(3)
1.326(3)
1.356(4)
1.495(3)
1.211(3)
1.439(3)
1.361(3)
Bond angles (◦)
N(1)–C(5)–N(2)
C(4)–C(5)–N(2)
C(5)–N(2)–N(3)
N(2)–N(3)–C(6)
N(3)–C(6)–C(7)
N(3)–C(6)–C(11)
C(12)–N(4)–C(9)
C(13)–N(4)–C(9)
C(8)–C(9)–N(4)
C(10)–C(9)–N(4)
112.72(11)
123.06(12)
112.11(11)
116.04(11)
115.89(12)
126.11(12)
121.01(11)
121.16(11)
121.36(12)
121.68(12)
110.6(2)
125.2(2)
109.1(2)
113.6(2)
114.2(2)
127.0(2)
122.3(2)
122.8(2)
121.2(2)
122.0(2)
Fig. 4. The UV–vis absorption spectra of azpy (solid line), dmazpy (dot line) and
deazpy (dashed line). The TD-PBE0/6-31++G** excitation energy and corresponding
oscillator strength is shown here as a set of vertical lines.

S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
543
Fig. 5. Molecular orbital energy diagrams (in eV) of trans-azpy, trans- and cis-forms of dmazpy and deazpy. Selected molecular orbitals of trans- and cis-dmazpy are displayed
with the isovalue of 0.025.
3.4. Optical properties
The calculated vertical excitation energies agree with the exper-
imental absorption spectra. The lowest allowed transitions of azpy,
dmazpy and deazpy, all in trans-form, appear at 340.4, 426.7 and
430.3 nm which can be assigned as mainly HOMO → LUMO transi-
tion (see Table 5). This is a very strong transition judging from the
value of the oscillator strength. Taking trans-dmazpy as an exam-
ple, the analysis of the atomic contribution to molecular orbitals
involved in such a transition (see Supplementary data) shows that
the HOMO is mainly at the phenyl ring (41.57%) and N(CH3)2
(39.70%) whereas the LUMO is mainly at N N (40.52%), pyridine
(27.48%) and phenyl (22.94%). Similar conclusion can be drawn in
the case of trans-deazpy. The excitation therefore moves an elec-
tron from –N(CH3)2 towards N N. Hence it might be possible to
consider these compounds as donating ligands in transition metal
complexes for light-harvesting applications.
3.4.1. Absorption
The studied azoimine compounds show two absorption bands
corresponding to n-␲* and ␲-␲* transitions in the 200–500 nm
region (see Fig. 4). The UV/Vis spectrum of deazpy has the absorp-
tion maximum of the most intense band at a longer wavelength
than that of dmazpy and azpy. The absorption spectrum of azpy
exhibits the most intense band of ␲-␲* transition at 320 nm and the
weak band of n-␲* transition at 450 nm. In contrast to azpy, both
dmazpy and deazpy show an intense ␲-␲* band at lower energy
around 430–440 nm. Moreover, the n-␲* band shifts to a higher
energy of around 274 nm. This indicates the reordering of n and ␲
orbitals in these compounds.
Fig. 4 also compares experimental absorption spectra and calcu-
lated TD-PBE0/6-31++G** vertical excitation energies of trans-azpy,
dmazpy and deazpy. The height of each vertical line corresponds
to the oscillator strength for each allowed electronic transition.
From the TD-PBE0 calculation, the next three allowed exci-
tations are weak transitions appearing in the 200–300 nm
region. They are characterized as mainly HOMO → LUMO + 1,
Table 4
˚
Selected PBE0/6-31G(d,p) bond lengths (in A), bond angles and dihedral angles (in degree). X-ray data are shown here for comparison.
Optimized geometry
Calc. PBE0/6-31G**
Exp. X-ray
Trans-azpy
Trans-dmazpy
Trans-deazpy
Cis-dmazpy
Cis-deazpy
Trans-dmazpy
Trans-deazpy
N(2)–N(3)
N(3)–C(6)
N(2)–C(5)
N(1)–C(5)
C(4)–C(5)
C(6)–C(11)
N(1)–C(1)
N(4)–C(9)
N(1)–C(5)–N(2)
C(4)–C(5)–N(2)
C(5)–N(2)–N(3)
N(2)–N(3)–C(6)
N(3)–C(6)–C(7)
N(3)–C(6)–C(11)
C(12)–N(4)–C(9)
C(13)–N(4)–C(9)
C(8)–C(9)–N(4)
C(10)–C(9)–N(4)
C(4)–C(5)–C(6)–C(7)
C(4)–C(5)–N(2)–N(3)
N(2)–N(3)–C(6)–C(11)
C(5)–N(2)–N(3)–C(6)
C(10)–C(9)–N(4)–C(13)
1.252
1.412
1.420
1.333
1.400
1.401
1.331
1.258
1.396
1.416
1.335
1.402
1.404
1.331
1.369
113.0
123.7
113.5
115.1
116.4
125.0
120.0
120.2
121.2
121.0
0.0
1.258
1.396
1.416
1.335
1.402
1.403
1.331
1.371
113.0
123.8
113.6
115.0
116.6
125.1
121.7
121.9
121.5
121.2
0.1
1.246
1.414
1.425
1.332
1.401
1.403
1.334
1.372
116.2
119.3
123.7
123.7
115.1
126.3
119.8
119.9
121.3
121.2
37.2
1.246
1.413
1.424
1.332
1.401
1.403
1.335
1.373
116.3
119.2
123.7
123.9
115.0
126.7
121.7
121.8
121.6
121.4
38.2
1.257(2)
1.404(2)
1.434(2)
1.331(2)
1.386(2)
1.395(2)
1.332(2)
1.359(2)
112.7(1)
123.1(1)
112.1(1)
116.0(1)
115.9(1)
126.1(1)
121.0(1)
121.2(1)
121.4(1)
121.7(1)
0.7
1.211(3)
1.439(3)
1.495(3)
1.313(3)
1.356(4)
1.393(4)
1.326(3)
1.361(3)
110.6(2)
125.2(2)
109.1(2)
113.6(2)
114.2(2)
127.0(2)
122.3(2)
122.8(2)
121.2(2)
122.0(2)
0.6
112.8
123.6
113.8
114.5
115.4
124.3
0.0
−0.1
0.0
0.0
0.9
−69.1
−34.3
−10.0
−4.9
−69.4
−32.9
−10.1
−5.7
10.4(2)
2.1(2)
6.1(3)
0.1
−0.9
−3.2(4)
−179.2(2)
−5.2(4)
180.0
–
−180.0
−179.9
3.9
−179.7(1)
−0.1
−7.5(2)

544
S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
Table 5
TD-PBE0/6-31++G** vertical excitation energy and the corresponding experimental ꢂmax of deazpy and dmazpy in both trans- and cis-forms. Assignments of major electronic
transitions for each excitation are reported.
Wavelength (nm)
Oscillator strength
Major assignments
Exp. ꢂmax (nm)
Trans-dmazpy
426.7
289.0
283.9
271.3
Cis-dmazpy
454.6
359.6
Trans-deazpy
430.3
291.0
283.5
274.0
Cis-deazpy
453.2
361.7
1.039
0.063
0.046
0.125
HOMO → LUMO (99%)
423.6
268.9
HOMO-2 → LUMO (11%), HOMO → LUMO + 1 (76%)
HOMO-2 → LUMO (83%), HOMO → LUMO + 1 (−10%)
HOMO-3 → LUMO (−24%), HOMO → LUMO + 2 (73%)
0.142
0.373
HOMO-1 → LUMO (−40%), HOMO → LUMO (56%)
HOMO-1 → LUMO (51%), HOMO → LUMO (42%)
1.100
0.049
0.050
0.114
HOMO → LUMO (99%)
432.6
272.0
HOMO-3 → LUMO (−13%), HOMO → LUMO + 1 (74%)
HOMO-2 → LUMO (88%)
HOMO-3 → LUMO (29%), HOMO → LUMO + 2 (67%)
0.159
0.406
HOMO-1 → LUMO (41%), HOMO → LUMO (56%)
HOMO-1 → LUMO (50%), HOMO → LUMO (−43%)
HOMO-2 → LUMO and HOMO → LUMO + 2 transitions. The
LUMO + 1 orbital is essentially localized at the pyridine while
LUMO + 2 is localized at the phenyl and –NR2 groups. Therefore,
these weak transitions can be described as electron transfer from
–NR2 to the pyridine ring, the ␲-electron of the pyridine to N N,
and the ␲-␲* transition of the substituted phenyl ring.
Fig. 7 also displays the excitation spectrum of deazpy along
with its absorption and emission spectra. The excitation spec-
trum at the collected emission wavelengths is not identical to the
absorption spectra. The maximum wavelength of the excitation
spectrum appears at 350 nm while that of the absorption band
occurs at 430 nm. This discrepancy implies that deazpy undergoes
photochemical change. Upon irradiation, there must be an unsta-
ble molecular configuration of the excited state. The dmazpy and
deazpy compounds should exist in more than one structural form
in the ground and excited states. We believe that the photoisomer-
ization behavior of these compounds is similar to azobenzene. The
X-ray data tells us that the trans-form is a stable form in the ground
state. The excited dmazpy and deazpy molecules might convert
themselves from trans- to cis-form.
To prove this hypothesis, the theoretical absorption spectrum of
cis-form was calculated. The absorption spectrum of the cis-form
agrees perfectly with the excitation spectrum from the exper-
iment. This suggests that the photoisomerization occurs after
the dmazpy and deazpy dyes are irradiated and the cis-form is
generated. The cis-form is the conformer responsible for the result-
ing fluorescence. We observe that the excited states of dmazpy
and deazpy decay monoexponentially independent of the fre-
quency. The excited state lifetimes of dmazpy and deazpy in
dichloromethane range from 7.0 to 8.3 ns (see Table 6). Fluores-
cence quantum yields were determined using a deaerated solution
of coumarin 6 in ethanol (˚R = 0.78) [38] as a reference standard.
3.4.2. Fluorescence emission, lifetime and fluorescence quantum
yield
The steady-state fluorescence of azpy, dmazpy and deazpy are
studied at room temperature (300 K) in dichloromethane, acetoni-
trile and ethanol (see Fig. 6). In these conditions, azpy does not
fluoresce in any of the solvents studied. In contrast, both dmazpy
and deazpy are photo-active, especially in dichloromethane;
dmazpy and deazpy show an intense broad band of emission at 510
and 517 nm respectively. However, the characteristic fluorescence
bands are observed in acetonitrile. Fig. 6 reveals that both dyes
exhibit weak emission at 510 nm with a small shoulder at 610 nm.
The lowest singlet S0 → S1 transition of both compounds corre-
sponds to the ␲-␲* transition. Unlike dmazpy, deazpy exhibits a red
shifted emission wavelength which correlates with the shift of ꢂmax
observed in the absorption spectrum (see Table 5). This observation
indicates that a diethylamino (–N(C2H5)2) substituent can donate
an electron through the ␲-delocalized system of the azpy moiety
more efficiently than a dimethylamino (–N(CH3)2) substituent.
Fig. 6. Normalised emission spectra of deazpy in dichloromethane, acetonitrile and
ethanol (ꢂex = 350 nm, ꢂem = 510 nm).
Fig. 7. Normalised absorption, excitation and emission spectra of deazpy in ace-
tonitrile (ꢂex = 350 nm, ꢂem = 510 nm).

S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
545
Table 6
Summary of photo-physical measurement and cyclic voltammetric data for studied compounds.
a
b
d
Compound
ꢂabs (nm),
ꢂem (nm)
ꢃc (ns) 3␴
ꢄF
E1/2 e (ꢅEp f)reduction, V
E1/2 e (ꢅEp f) oxidation, V
ε × 10−4 (M−1 cm−1
)
Dmazpy
Deazpy
Azpy
424(2.86)
433 (3.20)
314(1.18)
510
517
–
7.0 0.1
8.3 0.3
–
0.08
0.18
–
−1.35 (0.22)−0.96 (0.05)
−1.33 (0.19)−0.95 (0.04)
1.24 (0.12)
1.00 (0.08)
–
0.98 (0.10)
1.26 (0.08)
–
–
−1.11 (0.20)
a
Measured in CH2Cl2 solution.
b
c
Excited at 350 nm. Phase angles were in between 10◦ and 50◦.
Measured in CH2Cl2 solution dearated by saturation with argon.
d
Determined in CH2Cl2 solution at 25 ◦C with diluted concentrations of 5 ␮M for all samples (Abs < 0.05, preventing self quenching reaction). The 5 ␮M coumarin 6 (ꢄ = 0.78)
[38] in ethanol solution was the standard. All substances were deaerated by saturation with argon. Excited at 350 nm, slid width 8 nm for all samples.
e
Measured at 10−4 M in CH3CN containing 0.1 M n-Et4NPF6 as a supporting electrolyte at a scan rate 50 mV/s under argon. Potentials were in V versus SCE. Redox couple
of ferrocence/ferrocenium (Fc/Fc+) at 0.44 V was the internal standard.
f
ꢅEp = (Epa − Epc).
The following equation recommended by Kotelevskiy [39] was
used [40]
ꢀ
ꢁ
IS(1 − 10−OD
)
)
nS
nR
2
R
˚
S
=
˚
R
(1)
IR(1 − 10−OD
S
where ˚S and ˚R are the fluorescence quantum yields of the sam-
ples (dmazpy and deazpy) and reference (coumarin 6). IS and IR
are the fluorescence integrated areas of corrected emission spectra
of the samples and reference. ODR and ODS are the optical den-
sities of the reference and samples at the excitation wavelength
(350 nm). nS and nR are the refractive indices of the solvents of the
sample (CH2Cl2) and of the reference (ethanol) at 25 ◦C. The fluores-
cence quantum yields for dmazpy and deazpy in dichloromethane
are 0.08 and 0.18, respectively. The fluorescence quantum yields of
dmazpy and deazpy increase with the number of alkyl carbons in
the –NR2 substituent. This observation correlates with the fact that
–N(C2H5)2 is a better electron donating group than –N(CH3)2. The
increasing fluorescence quantum yields result from the LUMO of
the ␲* orbital of deazpy being lower in energy than that of dmazpy.
As stated earlier, the calculated vertical excitation energy of the
cis-form matches well with the excitation energy (see Fig. 7). More-
over, its excitation spectrum appears to be close to a mirror image
of its emission spectrum. We conclude that the excitation spectrum
of deazpy is due to the cis-form generated from the photoiso-
merization of trans-deazpy. The calculation also provides spectral
assignment for a strong peak around 350 nm and a broad band
as shoulder around 450 nm observed in the excitation spectrum.
For deazpy, the first allowed excitation at 453 nm is composed
mainly of HOMO → LUMO (56%) and HOMO-1 → LUMO (41%) tran-
sitions. The oscillator strength of this excitation is less than that
of the second allowed excitation at 361.7 nm. The second allowed
excitation is also composed mainly of HOMO → LUMO (43%) and
HOMO-1 → LUMO (50%) transitions but with different proportions
(see Fig. 5 and Table 5).
3.5. Electrochemical studies
Fig. 8 displays the cyclic voltammograms of dmazpy and deazpy
in acetonitrile. The cyclic voltammetric data of dmazpy, deazpy
and azpy are summarized in Table 6. Electrochemical behavior of
dmazpy and deazpy are similar. At negative potential scanning (0 to
−2 V relative to SCE), dmazpy shows two quasi-reversible couples
at E1/2 = −0.96 V and −1.35 V while deazpy shows a reversible cou-
ple at −0.95 V and one quasi-reversible process at −1.33 V. These
two couples at negative potentials are due to the reduction of
the azo moiety (–N N–). Dmazpy and deazpy can accommodate
two electrons since the LUMO is dominated by the azo group as
reported in the literature [41,42] and found in our calculation (see
Supplementary data). This can be represented by
+e
[–N N–]ꢀ[–NN· · ·N]−[–NN–]2−
(2)
Fig. 8. Cyclic voltammograms of (a) dmazpy (b) deazpy and (c) N,N-diethyl-1,4-
nitrosoaniline.
−e

546
S. Yoopensuk et al. / Spectrochimica Acta Part A 86 (2012) 538–546
During a positive scan (0–1.8 V), two signals are obtained. The
Appendix A. Supplementary data
first signal, an obvious quasi-reversible couple, is observed at
1.24 V and 1.00 V for dmazpy and deazpy, respectively. The sec-
ond one, an anodic peak at 1.33 V, corresponds to an irreversible
electrochemical reaction. The reversal scan does not give an obvi-
ous cathodic peak potential. This redox couple is assigned to the
dialkylamino substituent. Furthermore, comparing with the cyclic
voltammograms of nitrosoaniline (Fig. 8c), a starting material,
this quasi-reversible couple is due to –NR2/[–NR2]+oxidation. The
quasi-reversible couple of nitrosoaniline appears at a similar poten-
tial. This behavior is different from azpy. On scanning in the same
positive potential range, we observe no couple in the voltammo-
gram of azpy.
Because the ␲-accepting property of a compound can be judged
from a high reduction potential, it is worth noting that dmazpy
and deazpy are better ␲-acceptors than azpy. Although the LUMO
energy of azpy is lower than the two substituent compounds, the
HOMO energy of dmazpy and deazpy are higher than that of azpy.
The calculated molecular orbital energy diagram (see Fig. 5) also
confirms this trend. We therefore conclude that the ␲-acidity of
dmazpy and deazpy is greater than that of azpy which leads to
them being stronger ␲-backbonding ligands in transition metal
complexes.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.saa.2011.11.007.
References
[1] T. Nakamura, T. Shibusawa, Textile Res. J. 70 (2000) 801–809.
[2] R. Bianchini, G. Catelani, R. Cecconi, F. D’Andrea, E. Frino, J. Isaad, M. Rolla,
Carbohydr. Res. 343 (2008) 2067–2074.
[3] W.M. Gibbons, P.J. Shannon, S.-T. Sun, B.J. Swetlin, Nature 351 (1991) 49–50.
[4] M.S. Ho, A. Natansohn, P. Rochon, Macromolecules 28 (1995) 6124–6127.
[5] K.R. Millington, K.W. Fincher, A.L. King, Solar Energy Mater. Solar Cells 91 (2007)
1618–1630.
[6] Z. Lei, A. Vaidyalingam, P.K. Dutta, J. Phys. Chem. B 102 (1998) 8557–8562.
[7] H. Rau, Photoreactive Organic Thin Films, Academic Press, San Diego, 2002, pp.
3–47.
[8] H. Rau, S. Yu-Quan, J. Photochem. Photobiol. A: Chem. 42 (1988) 321–327.
[9] H. Satzger, S. Spörlein, C. Root, J. Wachtveitl, W. Zinth, P. Gilch, Chem. Phys.
Lett. 372 (2003) 216–223.
[10] H. Dinc¸ alp, S. Yavuz, Ö. HaklI, C. Zafer, C. Özsoy, I. Durucasu, Si. Ic¸ li, J. Photochem.
Photobiol. A: Chem. 210 (2010) 8–16.
[11] H. Joshi, F.S. Kamounah, C. Gooijer, G. van der Zwan, L. Antonov, J. Photochem.
Photobiol. A: Chem. 152 (2002) 183–191.
[12] L. Angiolini, T. Benelli, L. Giorgini, F.M. Raymo, Polymer 50 (2009) 5638–5646.
[13] M.Z. Al-Noaimi, R.J. Abdel-Jalil, S.F. Haddad, R.H. Al-Far, R.J. Crutchley, Inorg.
Chim. Acta 359 (2006) 2395–2399.
[14] P. Pratihar, T.K. Mondal, A.K. Patra, C. Sinha, Inorg. Chem. 48 (2009) 2760–
2769.
[15] R.A. Krause, K. Krause, Inorg. Chem. 19 (1980) 2600–2603.
[16] P. Byabartta, P.K. Santra, T.K. Misra, C. Sinha, C.H.L. Kennard, Polyhedron 20
(2001) 905–913.
4. Conclusions
[17] T.K. Mondal, J. Dinda, A.M.Z. Slawin, J. Derek Woollins, C. Sinha, Polyhedron 26
(2007) 600–606.
[18] T.A. Liss, J. Org. Chem. 32 (1967) 1146–1150.
[19] H.A. Gazzaz, E. Ember, A. Zahl, R. van Eldik, Dalton Trans. (2009) 9486–9495.
[20] A.F.A. Peacock, A. Habtemariam, S.A. Moggach, A. Prescimone, S. Parsons, P.J.
Sadler, Inorg. Chem. 46 (2007) 4049–4059.
[21] S.J. Dougan, M. Melchart, A. Habtemariam, S. Parsons, P.J. Sadler, Inorg. Chem.
45 (2006) 10882–10894.
[22] Y. Onoe, S. Tsukahara, H. Watarai, Bull. Chem. Soc. Jpn. 71 (1998) 603–608.
[23] S.J. Dougan, A. Habtemariam, S.E. McHale, S. Parsons, P.J. Sadler, Proc. Natl. Acad.
Sci. U. S. A. 105 (2008) 11628–11633.
[24] N. Leesakul, S. Yoopensuk, C. Pakawatchai, S. Saithong, K. Hansongnern, Acta
Crystallogr. E: Struct. Rep. Online 66 (2010) o1923.
4-(2-Pyridylazo)-N,N-dimethylaniline (dmazpy) and 4-(2-
pyridylazo)-N,N-diethylaniline (deazpy) were synthesized and
characterized. The dmazpy and deazpy are stable at room tem-
perature (300 K). The compound structures have been confirmed
by single X-ray diffraction, spectroscopic techniques and electro-
chemical studies. The trans configuration around N N is found to
be a stable form of both compounds in the ground state. However,
the trans-form can be converted to cis-form upon irradiation. This
is evidenced by the difference of excitation and absorption spec-
trum and confirmed by the calculated vertical excitation energy
of the cis-form. The HOMO is mainly at the phenyl and –NR2
substituents. Owing to the planarity of pyridine and phenyl rings,
the electrons can delocalize to the LUMO which mainly has an azo
(–N N–) character. The HOMO → LUMO transition is characterized
as a phenyl → azoimine transition. Cyclic voltammograms indicate
the greater reduction ability of deazpy and dmazpy than that
of azpy which can stabilize the transition metal ion. From their
electrochemical and photo-physical properties, we believe that
dmazpy and deazpy should be considered as potential candidates
for use as photo-active dyes and as ligands for transition metal
complexes in photoharvesting applications.
[25] N. Leesakul, unpublished result.
[26] Y. Li, Z.-Y. Lin, W.-T. Wong, Eur. J. Inorg. Chem. 2001 (2001) 3163–3173.
[27] A. Ohashi, T. Tsukuda, H. Watari, Anal. Sci. 19 (2003) 1085–1086.
[28] G. Sheldrick, Appl Catal. A 64 (2008) 112–122.
[29] S. Landgraf, Spectrochim. Acta A 57 (2001) 2029–2048.
[30] A.P. Scott, L. Radom, J. Phys. Chem. 100 (1996) 16502–16513.
[31] J.P. Merrick, D. Moran, L. Radom, J. Phys. Chem. A 111 (2007) 11683–11700.
[32] N. Rega, M. Cossi, V. Barone, J. Chem. Phys. 105 (1996) 11060.
[33] K.W. Wiitala, T.R. Hoye, C.J. Cramer, J. Chem. Theory Comput.
1085–1092.
2 (2006)
[34] R. Jain, T. Bally, P.R. Rablen, J. Org. Chem. 74 (2009) 4017–4023.
[35] M. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, J. Mont-
gomery, T. Vreven, K. Kudin, J. Burant, J. Millam, S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. Petersson, H. Nakatsuji, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. Knox, H. Hratchian, J. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. Stratmann, O. Yazyev, A. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Ayala, K. Morokuma, G. Voth, P. Salvador, J. Dannen-
berg, V. Zakrzewski, S. Dapprich, A. Daniels, M. Strain, O. Farkas, D. Malick, A.
Rabuck, K. Raghavachari, J. Foresman, J. Ortiz, Q. Cui, A. Baboul, S. Clifford, J.
Cioslowski, B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Martin,
D. Fox, T. Keith, A. Laham, C. Peng, A. Nanayakkara, M. Challacombe, P. Gill, B.
Johnson, W. Chen, M. Wong, C. Gonzalez, J. Pople, Gaussian 03, Revision C.02,
Gaussian Inc., 2003.
[36] Avogadro program, http://avogadro.openmolecules.net.
[37] K. Panneerselvam, K. Hansongnern, N. Rattanawit, F.-L. Liao, T.-H. Lu, Anal. Sci.
16 (2000) 1107–1108.
[38] G.A. Reynolds, K.H. Drexhage, Opt. Commun. 13 (1975) 222–225.
[39] S.I. Kotelevskiy, J. Lumin. 79 (1998) 211–214.
[40] A. Rosspeintner, G. Angulo, M. Weiglhofer, S. Landgraf, G. Grampp, J. Pho-
tochem. Photobiol. A: Chem. 183 (2006) 225–235.
Acknowledgements
N.L. acknowledges financial support from the Center for Inno-
vation in Chemistry (PERCH–CIC), the Commission on Higher
Education, and the Ministry of Education. Ms. Sriwipha Onganu-
sorn is acknowledged for supplying compound in ES-MS and CHN
analysis. Y.T. is grateful to the National Nanotechnology Center
(to NANOSIM) and the Thailand Research Fund (RSA-5180010) for
financial support and HPC services, NECTEC for computing facilities.
S.Y. is grateful for a TGIST scholarship from the National Science
and Technology Development Agency. We thank the Department
of Chemistry, Wollongong University, Australia for ES-MS mea-
surements and the Department of Chemistry, University of Bristol,
England for elemental analysis. Finally, we thank Michael A. Allen
for his proofreading.
[41] S. Goswami, R. Mukherjee, A. Chakravorty, Inorg. Chem. 22 (1983) 2825–2832.
[42] T. Mathur, J. Dinda, P. Datta, G. Mostafa, T.-H. Lu, C. Sinha, Polyhedron 25 (2006)
2503–2512.