Structural and photophysical properties of alkylamino substituted 2-arylidene and 2,5-diarylidene cyclopentanone dyes

Article abstract of DOI:10.1016/j.molstruc.2016.02.009

A series of alkylamino substituted 2-arylidene and 2,5-diarylidene cyclopentanone compounds were synthesized. Spectroscopic and photophysical properties of these compounds have been measured in a variety of solvents. Absorption and fluorescence maxima have been correlated with the Empirical Scale of Solvent Polarity (E_T(30)). Theoretical TD-DFT spectral calculations and Lippert-Mataga analysis support the internal charge transfer (ICT) nature of the S₀ → S₁ excitation for these compounds, with higher degrees of ICT depicted for the alkylamino substituted 2,5-diarylidene cyclopentanones. Photophysical properties consisted of measuring the fluorescence quantum yields (Φ_f) and lifetimes (τ_f) in a variety of solvents. Radiative and nonradiative decay constants have been determined from the Φ_f and τ_f data. Variation with solvent in the nonradiative rate of decay is interpreted in terms of a competition between internal conversion and intersystem crossing. Lastly, two compounds presented have been shown to undergo excited state protonation in glacial acetic acid.

Full text of DOI:10.1016/j.molstruc.2016.02.009

Journal of Molecular Structure 1112 (2016) 97e109
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: http://www.elsevier.com/locate/molstruc
Structural and photophysical properties of alkylamino substituted
2-arylidene and 2,5-diarylidene cyclopentanone dyes
Christopher A. Zoto*, Mine G. Ucak-Astarlioglu, John C. MacDonald, Robert E. Connors
Department of Chemistry and Biochemistry, Worcester Polytechnic Institute, Worcester, MA 01609, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of alkylamino substituted 2-arylidene and 2,5-diarylidene cyclopentanone compounds were
synthesized. Spectroscopic and photophysical properties of these compounds have been measured in a
variety of solvents. Absorption and fluorescence maxima have been correlated with the Empirical Scale of
Solvent Polarity (ET(30)). Theoretical TD-DFT spectral calculations and LipperteMataga analysis support
the internal charge transfer (ICT) nature of the S0 / S1 excitation for these compounds, with higher
degrees of ICT depicted for the alkylamino substituted 2,5-diarylidene cyclopentanones. Photophysical
properties consisted of measuring the fluorescence quantum yields (Ff) and lifetimes (tf) in a variety of
solvents. Radiative and nonradiative decay constants have been determined from the Ff and tf data.
Variation with solvent in the nonradiative rate of decay is interpreted in terms of a competition between
internal conversion and intersystem crossing. Lastly, two compounds presented have been shown to
undergo excited state protonation in glacial acetic acid.
Received 14 November 2015
Received in revised form
3 February 2016
Accepted 3 February 2016
Available online 6 February 2016
Keywords:
2-Arylidene cyclopentanone
2,5-Diarylidene cyclopentanone
Absorption
Fluorescence
Photophysical
Internal charge transfer
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
fluorescence spectra. Additional work in our lab involved studying
the photophysical properties of an asymmetrically substituted 2,5-
The class of organic compounds called diarylidene cyclo-
alkanones are ketocyanine dyes that have received attention in
their use as photosensitizers for various electron energy transfer
processes [1], fluorescent solvent polarity probes [2], fluo-
roionophores [3], and nonlinear optical materials [4], particularly in
their use as chromophores in undergoing two photon absorption.
Previous work has been done in our lab studying the structural
and spectroscopic properties of C2v unsubstituted 2,5-diarylidene
diarylidene
cyclopentanone
dye,
namely
(2E,5E)-2-(4-
cyclo-
cyanobenzylidene)-5-(4-dimethylaminobenzylidene)
pentanone, in a variety of solvents [7]. This asymmetric pushepull
compound exhibits efficient two photon absorption (TPA) when
dissolved in chloroform and is known to have potential for appli-
cations in TPA [4e]. Lastly, recently published work on the photo-
chemistry of compound
I in oxygenated and deoxygenated
solutions has been conducted [8]. The testing of I as a direct
photosensitizer of singlet state oxygen (1Dg), a strong oxidant used
in photodynamic therapy in destroying targeted tumor and
cancerous cells in living organisms, was also carried out in these
studies.
This manuscript provides a presentation and discussion of the
structural, spectroscopic, and photophysical properties of a series of
alkylamino substituted 2-arylidene and 2,5-diarylidene cyclo-
pentanone compounds. These class of ketocyanine compounds are
of interest due to their electron “pushepull” structures. The two
alkylamino substituted 2-arylidene cyclopentanones presented are
examples of DeA compounds and the four alkylamino substituted
2,5-diarylidene cyclopentanones are examples of DeAeD com-
pounds (where D is an electron donor group and A is an electron
acceptor group). The presence of electron donor groups bonded to
the polyene chain should have an effect of the electronic structure
cyclopentanones,
pentanone (1), (2E,5E)-2,5-dicinnamylidene cyclopentanone (2),
and (2E,5E)-2,5-bis(5-phenyl-penta-2,4-dienylidene) cyclo-
namely
(2E,5E)-2,5-dibenzylidene cyclo-
pentanone (3) [5,6]. It was found that fluorescence was not
observed for 1 in any of the solvents studied (protic and aprotic).
Fluorescence was observed for 2 only in protic solvents, and in a
number of protic and aprotic solvents for 3. Solvents which were
able to induce fluorescence in compounds 2 and 3 were believed to
1
do so by inverting the order of 1(n,
p
*) and
(p,
p*) states. It was
also demonstrated that 2 and 3 undergo excited state protonation
in glacial acetic acid, with a signature of dual emission in the
* Corresponding author.
E-mail address: czoto@hotmail.com (C.A. Zoto).
http://dx.doi.org/10.1016/j.molstruc.2016.02.009
0022-2860/© 2016 Elsevier B.V. All rights reserved.

98
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
Fig. 1. Reaction schemes for the preparations of (a) IA and IIA, (b) I and II, and (c) III.
and photophysical properties, making this particular set of com-
pounds different than the compounds presented in Ref. [5]. In
solvent environments and their fluorescence is quenched and red
shifted relative to the LE state. A section of this manuscript com-
pares the structural and spectroscopic properties of compounds II
and IV. Compound II contains free-moving dimethylamino groups
substituted on the aryl moieties and can potentially undergo the
formation of a TICT state. On the contrary, compound IV is spatially
restricted on the aryl moieties due to the presence of closed,
saturated ring systems that structurally prevent it from having a
TICT state. Comparisons of the spectroscopic and photophysical
properties of these two compounds will lead to determining
whether or not II undergoes the formation of a TICT state.
Furthermore, the work presented in this manuscript is expected
to be of importance to those interested in studying the structural,
spectroscopic, and photophysical properties of these class of com-
pounds and related compounds. A thorough understanding of the
properties of these compounds is pertinent to examining their
applications.
addition, several of the compounds presented have longer
jugated chains compared to compound 1 in Ref. [5] and compound I
in Ref. [7]. The length of a conjugated chain also has an effect on
p con-
p
electronic structure, spectroscopic and photophysical properties.
A unique feature of these types of electron “pushepull” polyene
compounds are their photoinduced internal charge transfer (ICT)
properties. The ICT state is an excited state characterized by the
transfer of electron density from the electron donor end of the
molecule to the electron acceptor end of the molecule upon
photoexcitation. The nature of ICT excited states gives rise to lower
energies of fluorescence than the locally excited (LE) states. In other
words, the fluorescence of an ICT state is red shifted relative to the
LE state. A special type of ICT state is termed the twisted internal
charge transfer (TICT) state, where the plane of a dimethylamino
group is at a twisted angle with respect to the rest of the molecule
due to rotation about the CeN bond. TICT states form in highly polar

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
99
Fig. 2. Reaction scheme for the preparation of IV.
Fig. 3. Single crystal X-ray structure of I.
2. Experimental and computational methods
a Perkin Elmer® Spectrum One IR spectrometer.
Fig. 1 shows the reaction schemes for the preparations of com-
pounds IA and IIA (1(a)), I and II (1(b)), and III (1(c)). Compound IA
was synthesized by combining an equimolar mixture of cyclo-
pentanone and 4-dimethylaminobenzaldehyde in the presence of
2.1. Compound syntheses and structural characterizations
Presented in this section are general descriptions for the syn-
theses of all target compounds. Please refer to the Supplementary
Information for detailed descriptions.
N,N-dimethylammonium-N',N'-dimethylcarbamate
(DIMCARB)
[9,10]. Compound IIA was synthesized via an intermolecular base-
catalyzed crossed aldol condensation reaction between limiting
(E)-4-dimethylaminocinnamaldehyde and excess cyclopentanone.
Compounds I and II were synthesized via an intermolecular base-
catalyzed crossed aldol condensation reaction between cyclo-
pentanone (1 mole equivalent) and the appropriate aldehyde (2
mole equivalents) (4-dimethylaminobenzaldehyde for I and (E)-4-
dimethylaminocinnamaldehyde for II). Compound III was
Purity of all compounds were confirmed by 1H NMR spectros-
copy (>99% for all target compounds) and thin layer chromatog-
raphy (TLC), which showed one spot upon development. NMR and
IR spectroscopy were the structural characterization methods used.
1H and 13C NMR spectra were obtained using a Bruker® AVANCE
400 MHz NMR spectrometer and a Bruker® AVANCE III 500
(500 MHz) NMR spectrometer. ATR-IR spectra were obtained using

100
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
2.2. X-ray diffraction measurements
The X-ray structure of I was solved using a Bruker-AXS Kappa
APEX CCD Diffractometer, collecting diffraction data at ambient
room temperature using graphite monochromated Mo-K
a
radia-
tion (
l
¼ 0.71073 Å). The unit cells and space groups were deter-
mined using the SAINT þ program [14]. Structures were solved by
direct methods and refined by full matrix least-squares using the
SHELXTL program [15]. Mercury 2.4 software was used to examine
the molecular structure and crystal packing arrangements of the
solved X-ray crystal structures [16]. Single crystal X-ray structures
were properly refined to give overall refinement values of <5%.
2.3. Spectroscopic and photophysical measurements
Absorption and fluorescence emission spectra were measured at
ambient room temperature. Absorption spectra were measured
with both
kineElmer® Lambda 35 UV/VIS spectrometer. Fluorescence emis-
sion spectra were obtained with
PerkineElmer® LS 50B
a a Per-
Shimadzu® UV2100U spectrometer and
a
spectrophotometer equipped with a red sensitive R928 phototube
detector. All solvents used were spectroscopic or HPLC grade.
Fluorescence quantum yields (Ff) were determined by comparing
the corrected integrated fluorescence spectrum of the sample with
that of a standard according to [17]. For Ff measurements of IA and
IIA, coumarin-481 in acetonitrile (Ff ¼ 0.11) was used as the
standard [18]; for all 2,5-diarylidene cyclopentanones, fluorescein
in 0.1 N NaOH (Ff ¼ 0.95) was used [16]. Equation (1) defines Ff
!
ꢀ
ꢁ
ꢀ
ꢁ
Astd
A
n2
n2std
D
Dstd
Ff ¼ Ff ;std
(1)
Fig. 4. Absorption and fluorescence spectra of II in (a) 2,2,2-trifluoroethanol (TFE), (b)
2-propanol, (c) ethyl acetate, (d) acetone, (e) benzene, and (f) carbon disulfide.
where A is the absorbance at the excitation wavelength, n is the
refractive index of the solvent, and D is the integrated area under
the corrected fluorescence spectrum. Solutions of both the stan-
dard and sample were prepared initially with the absorbance at
lmax approximately equal to 0.5, followed by an accurate tenfold
dilution. Fluorescence spectra used to determine Ff were corrected
with sets of correction factors that were obtained by measuring the
spectra of compounds with known emission spectra [17]. Fluores-
cence lifetimes (tf) were obtained with a Photon Technology In-
ternational® TM-3 time resolved spectrofluorometer with pulsed
nitrogen/dye laser excitation. All solutions were degassed with N2
to prevent excited state quenching by molecular oxygen.
synthesized via an intermolecular base-catalyzed crossed aldol
condensation
dimethylaminocinnamaldehyde.
Fig. 2 shows the three step reaction scheme for the preparation
of compound IV. solution of julolidine and N,N-
dimethylformamide (DMF) was added to a solution of phospho-
rous oxychloride (POCl3) in DMF, yielding 9-
reaction
between
IA
and
(E)-4-
A
julolidinylcarbaldehyde [11]. Two carbon homologation of 9-
julolidinylcarbaldehyde using silver perchlorate (AgClO4) cata-
lyzed addition of the zirconocene complex derived from the
hydrozirconation of 1-ethoxyethyne with bis(cyclopentadienyl)
zirconium hydridochloride followed by acid catalyzed dehydration
yielded (E)-3-(9-julolidinyl)-prop-2-en-1-al [12,13]. The final step
consisted of running an intermolecular base-catalyzed crossed
aldol condensation reaction between cyclopentanone (1 mole
equivalent) and (E)-3-(9-julolidinyl)-prop-2-en-1-al (2 mole
equivalents), affording IV.
2.4. Quantum chemical calculations
DFT and TD-DFT quantum chemical calculations were carried
out using the Gaussian09 program [19]. Minimum energy struc-
tures were confirmed by obtaining all real frequencies for the
Fig. 5. Solutions of II in (a) 2,2,2-trifluoroethanol, (b) ethanol, (c) 2-propanol, (d) acetonitrile, (e) acetone, (f) ethyl acetate, (g) benzene, (h) toluene, and (i) cyclohexane.

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
101
Fig. 6. Plot of absorption and fluorescence maxima of (a) IA and I and (b) IIA and II in various solvents against the ET(30) scale. Black: IA and IIA; Blue: I and II. Solid symbols
represent aprotic solvents; open symbols represent protic solvents. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)
calculated normal modes of vibration. Solvent effect studies were
computed using the Self-Consistent Reaction Field (SCRF) Polariz-
able Continuum Model (PCM) options. The DFT gas phase geome-
tries were used in the solvent calculations without further
optimization. Molecular Orbital Package for Spectroscopy-Fujitsu
(MOSF) was the molecular orbital program used for carrying out
atomic charge calculations both in the ground and excited states for
excited state protonation studies [20].
In the predicted DFT structures of IA and IIA, approximately 30ꢀ
rotational angles were seen within the cyclopentanone ring for
each compound, attributed to ring folding. For IA, an approximately
11ꢀ rotation of the substituted phenyl ring was also observed. In
addition to the predicted DFT geometry of I, the DFT structures of its
higher homologs, II, III, and IV were essentially planar. The folding
within the cyclopentanone ring is lost with extended
conjugation.
p
Absorption and fluorescence spectra of all alkylamino
substituted 2-arylidene and 2,5-diarylidene cyclopentanones
were examined in solvents of various polarities (please refer to the
Supplementary Information for tabulated data). It has been shown
that all compounds undergo bathochromic shifts in absorption
and fluorescence with respect to an increase in solvent polarity,
some more significant than others. Although minor red shifts are
observed in the absorption spectra, a larger degree of red shifting
is observed in fluorescence, suggesting that the excited states of
these compounds are more polar in nature than the ground state.
Absorption and fluorescence spectra of II in six solvents of
differing polarity and hydrogen bonding strength (two alcohols,
two polar aprotic, and two nonpolar) are shown in Fig. 4. Com-
pound II exhibits a large degree of solvatochromism, as illustrated
in Fig. 5, in that its color changes from yellow / orange / red /
purple in going from nonpolar to polar, aprotic and protic
solvents.
3. Results and discussion
3.1. Structural and spectroscopic properties
Compound I crystallizes in an orthorhombic system, belonging
to the P212121 space group. The single crystal X-ray structure of I, in
two views, is shown in Fig. 3. Compound I is essentially planar, with
approximately 3ꢀe7ꢀ rotations of the phenyl rings on each end
relative to the cyclopentanone ring. Direct comparison between the
X-ray geometry and its predicted ground state geometry at the
B3LYP/6-31G(d) level of theory shows excellent agreement in both
bond lengths and bond angles. Absolute differences between
experimental and calculated bond lengths varies between 0.001 Å
e 0.025 Å and bond angles between 0.01ꢀe 4.4ꢀ. The predicted DFT
geometry of I is planar, with ~1ꢀ rotation of the phenyl rings relative
to the cyclopentanone ring.

102
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
Fig. 7. LipperteMataga plots for (a) IA, (b) IIA, (c) I, (d) III, (e) II, and (f) IV. Solid symbols represent aprotic solvents; open symbols represent protic solvents.
The absorption and fluorescence maxima of (a) IA vs. I and (b)
IIA vs. II in wavenumbers plotted against the ET(30) scale are shown
in Fig. 6. It is seen specifically in these plots that the absorption and
fluorescence maxima are shifted to the red in going from one sided
to two sided conjugation, showing the effect of polyene chain
length on spectroscopic location (quantum mechanical particle in a
box system).
p
Excited state dipole moments (me) of all six compounds have
been determined by treatment of the LipperteMataga method
[21,22], using the data for aprotic solvents. The Stokes shifts for
each compound have been plotted against the solvent's orientation
Table 1
DFT B3LYP/6-31G(d) calculated mg and a values, along with LipperteMataga results
of me and Dm for all compounds.
polarization function (Df). The Stokes shift is the difference, in
Compound a (Ǻ) (B3LYP/6-31G(d)) mg (D) (B3LYP/6-31G(d)) Dm (D) me (D)
wavenumbers, between the absorbance ðꢁn Þ and fluorescence ðꢁnF Þ
A
IA
IIA
I
4.88
5.19
5.77
5.97
5.95
6.34
6.33
7.39
5.22
5.60
5.84
5.76
10.3
10.5
13.7
15.5
15.2
16.0
16.6
17.9
18.9
21.1
21.0
21.8
spectral maxima of a molecule in a solvent. The orientation polar-
ization function (Df) is given by
III
II
IV

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
103
Table 2
Onsager cavity radii for all compounds were calculated at the DFT
B3LYP/6-31G(d) level of theory. The positive slopes observed in the
LipperteMataga plots are attributed to the internal charge transfer
properties portrayed by these compounds and the overall general
trend is that both Dm and me become greater in magnitude from
least conjugated (IA) to most conjugated (II and IV) chain lengths.
Table 1 presents the computed ground state dipole moments and
Onsager cavity radii, along with the LipperteMataga results for me
Absorbance and fluorescence spectral maxima of I, II, III, and IV in toluene and
methanol compared to literature cited compounds.
Compound Toluene
Methanol
Dl (methanol-
toluene)
labs (nm) lf (nm) labs (nm) lf (nm) Dlabs (nm) Dlf (nm)
I
II
III
IV
2 [5]*
3 [5]
I [7]
452
495
475
505
e
482
553
530
588
e
479
523
500
523
408
451
473
635
735
737
750
522
625
693
27
28
25
18
e
153
182
207
162
e
and Dm
.
A comparative study between the absorbance and fluorescence
maxima locations of I, II, III, and IV and cited compounds 2 and 3
(in Ref. [5]) and I (in Ref. [7]) in toluene and methanol was inves-
tigated. Toluene and methanol are two solvents with significant
differences in polarity. It is seen in Table 2 that I, II, III, and IV
presented in this manuscript exhibit a higher magnitude in the
spectral shifts in absorbance and fluorescence in going from
toluene to methanol (with more pronounced shifts in fluorescence)
when compared to the cited compounds. This data shows that the
430
455
520
558
21
18
105
135
*No observable fluorescence in aprotic solvents.
ꢂ
ꢃ
n2 ꢂ 1
ðε ꢂ 1Þ
ꢂ
ꢃ
Df ¼
ꢂ
(2)
2n2 þ 1
ð2ε þ 1Þ
presence of electron donor functional groups onto the
p conjugated
polyene chain has an effect on the electronic structural and spec-
troscopic properties for these class of compounds.
where ε is the dielectric constant and n is the index of refraction of
the solvent. LipperteMataga plots have been constructed for each
compound (compiled set shown in Fig. 7), relating the Stokes shift
against the orientation polarization function in various solvents.
The LipperteMataga equation is defined as
TD-DFT spectral calculations were carried out at the B3LYP/6-
31G(d) level of theory from the optimized ground state struc-
tures. For IA, in the gas phase, S1 is predicted to be a weak transition
localized at 3.37 eV (368 nm, f ¼ 0.014) with HOMO-1 (n) / LUMO
(p*) as the orbital configuration. Excitation to the S2 state is pre-
2Dm2
hca3
ꢁ
ꢁ
dicted to be a strong transition occurring at 3.60 eV (344 nm,
nA ꢂ nF
¼
Df þ constant
(3)
f ¼ 0.80) with major CI configuration of HOMO / LUMO, corre-
sponding to a (p, p*) transition. In going from the gas phase to
solvent environment, the energy of the 1(
p
,
p
*) state falls below the
*) state, regardless of solvent polarity. In the gas phase, the
*) state, predicted to be S2, is localized at 3.60 eV (344 nm). In
*) and the
energy of this state continuously decreases with respect to an in-
crease in solvent polarity. The energy of the S1( *) state de-
where Dm is the difference in the electronic dipole moment be-
tween S0 and S1, h is Planck's constant, c is the speed of light in a
vacuum, and a is the Onsager cavity radius for the spherical inter-
action of the dipole in a solvent. Ground state dipole moments and
1(n,
p
(p, p
1
solvent, S1 is assigned an orbital configuration of (p, p
p, p
0.8
0.6
creases in the following order: 3.40 eV (365 nm, n-hexane), 3.36 eV
(369 nm, toluene), 3.34 eV (372 nm, chloroform), and 3.32 eV
(374 nm, methanol). This is characteristic of the typical bath-
ochromic shifting observed with respect to an increase in solvent
(a)
0.4
polarity. The inversion of the 1(
p, p*) state from S2 in the gas phase
0.2
to S1 in solvent is consistent with the observation of fluorescence in
0
the experimental spectra for all solvents studied.
0.8
0.6
The 1(n,
p
*) state inverts from being S1 in the gas phase to S2 in
solvent. In the gas phase, the 1(n,
*) state is computed to be
located at 3.37 eV (368 nm). In solvent, the 1(n,
*) state undergoes
characteristic hypsochromic (blue) shift, increasing in the
p
(b)
0.4
p
0.2
a
following order: 3.46 eV (358 nm, n-hexane), 3.47 eV (358 nm,
toluene), 3.50 eV (354 nm, chloroform), and 3.54 eV (350 nm,
0
0.8
0.6
(c)
0.4
Table 3
TD-DFT computed energy gaps of the lowest lying 1
(p
,
p
*) and 1(n,
p*) states and f
0.2
values for the 1(n,
p*) states of IA and IIA.
0
E[1(n, *)-1(
p
p
,
p
*)]j (cmꢂ1
)
f[1(n,
p*)]
0.6
Solvent
jD
IA
Gas
0.4
1896
536
833
1367
1833
0.014
0.32
0.091
0.030
0.015
(d)
n-Hexane
Toluene
Chloroform
Methanol
IIA
0.2
0
200
300
400
500
600
Gas
326
0.23
Wavelength (nm)
n-Hexane
Toluene
Chloroform
Methanol
2004
2606
3177
3663
0.0042
0.0026
0.0014
0.0008
Fig. 8. Experimental absorption spectra of IA in (a) methanol, (b) chloroform, (c)
toluene, and (d) n-hexane and their corresponding TD-DFT oscillator strengths at the
B3LYP/6-31G(d) level of theory.

104
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
methanol). As depicted in Fig. 8, good agreement is established
between the experimental absorption spectra of IA and the TD-DFT
calculations in each of the four solvents with excitation to S1(p, p*)
being the strongest electronic transition. It is also shown that from
the TD-DFT data, with respect to an increase in solvent polarity, (i)
(394 nm, f ¼ 0.23) with HOMO-1 (n) / LUMO (
p*) as the primary
CI configuration. Excitation to the S2 state is predicted to be a strong
electronic transition at 3.19 eV (389 nm, f ¼ 0.97) with a major CI
configuration of HOMO (p) / LUMO (p*). As with IA, the unusually
large oscillator strength computed for the n / p* transition in the
the S2(n,
of the oscillator strengths of S2(n,
The increase in the S2eS1 energy gap is attributed to the hyp-
sochromic shift of (n, *) and the bathochromic shift of ( *).
The unusually large oscillator strength predicted for the lowest
n /
manifestation of a well-known limitation with the TD-DFT method,
where nearly degenerate excited states undergo strong configura-
tion mixing [23]. This mixing causes a blending of the properties of
p
*)-S1(
p
,
p
*) energy gap becomes larger, and (ii) the ratio
gas phase is a result of configuration mixing of nearly degenerate
p
*) to S1( *) becomes smaller.
p
,
p
states (
D
E ¼ 326 cmꢂ1) and the inversion of the S1 state to (
p, p*) in
solvent is consistent with the observation of fluorescence seen in
the experimental spectra.
The three alkylamino substituted 2,5-diarylidene cyclo-
pentanone compounds, I, II, and IV are all members of the C2v point
group and for purposes of classification, a C2v symmetry is
p
p, p
p
* transition of IA in n-hexane (f ¼ 0.32) appears to be a
assumed. Table 4 presents the TD-DFT spectral calculations of the
1
lowest lying
(p, p p*) states of I, II, and IV in gas and
*) and 1(n,
the nearly degenerate excited states. The energy separation of S1(
p
,
solvent environments (arranged in order of increasing polarity). In
the gas phase, TD-DFT computes S1 as a strong symmetry allowed
1
p
*) and S2(n,
and 1(n,
*) properties, including the oscillator strength. As seen
from Table 3, the larger the 1(n, *)e1(
*) energy gap, the smaller
the 1(n,
*) oscillator strength. For large gaps, e.g. methanol, the
p (p, p*)
*) is only 536 cmꢂ1, hence the blending of
p
B2 ) A1 transition of the (
excitation a2( ) / b1( *), where a2(
the LUMO. The lowest lying (n, *) state was computed to be a
symmetry forbidden A2 ) A1 transition, arising from the orbital
excitation b2(n) / b1( *), where b2 is the nonbonding molecular
orbital on the oxygen atom. The S2 state was predicted to be (n, *)
for I and II; S3 for IV. Fig. 9 shows the computed molecular orbitals
p
,
p
*) type, arising from the orbital
p
p,
p
p
p
p
) is the HOMO and b1( *) is
p
p
p
low oscillator strength expected to be computed for an n /
transition is observed (f ¼ 0.015).
p*
p
Similar trends were found in the TD-DFT calculations for IIA. S1
p
is predicted to be (n, p*) in the gas phase and (p, p*) in solvent. In
the gas phase, the S1 state is computed to be a transition at 3.15 eV
of I, II, and IV. The HOMO was computed to be a
p
bonding orbital
Table 4
1
TD-DFT spectral calculations of the lowest lying
(
p,
p
*) and 1(n,
p*) states of I, II, and IV in gas and solvents.
Solvent
State
Energy
cmꢂ1
f
nm
I
Gas
S1
2 (n,
S1
2 (n,
S1
2 (n,
S1
2 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
(
p
,
p
*) (B2)
23641
24510
22124
25253
21834
25381
22026
25316
21598
25907
21505
25906
21277
26178
21413
26178
423
408
452
396
458
394
454
395
463
386
465
386
470
382
467
382
1.45
0.00
1.66
0.00
1.70
0.02
1.67
0.00
1.66
0.00
1.67
0.00
1.67
0.00
1.64
0.00
S
p
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
Carbon Tetrachloride
Carbon Disulfide
Toluene
(p, p
S
p
(p, p
S
p
(p, p
S
p
Tetrahydrofuran
Dichloromethane
Dimethylformamide
Methanol
(p, p
S
p
(p, p
S
p
(p, p
S
p
(p, p
S
p
II
Gas
S1
2 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
(
p
,
p
p
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
20492
22523
19194
23256
18692
23529
18868
23474
18382
23697
18349
23641
488
444
521
430
535
425
530
426
544
422
545
423
2.02
0.00
2.22
0.00
2.29
0.00
2.26
0.00
2.26
0.00
2.23
0.00
S
n-Hexane
(p, p
S
p
Carbon Disulfide
Toluene
(p, p
S
p
(p, p
S
p
Dichloromethane
Methanol
(p, p
S
p
(p, p
S
p
IV
Gas
S1(
3 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
S1
3 (n,
p
,
p
p
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
*) (B2)
*) (A2)
19802
22676
18519
23041
18215
23041
17668
22936
17606
22883
505
441
540
434
549
434
566
436
568
437
2.03
0.01
2.22
0.00
2.26
0.00
2.25
0.00
2.23
0.00
S
n-Hexane
(p, p
S
p
Toluene
(p, p
S
p
Dichloromethane
Methanol
(p, p
S
p
(p, p
S
p

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
105
Fig. 9. Computed molecular orbitals of I, II, and IV.
with an even distribution of electron density along the conjugated
framework. The LUMO was computed to be a * antibonding
orbital. The HOMO-2 was computed to be a nonbonding orbital
localized on the carbonyl oxygen atom. The same orbital configu-
rations were predicted for II and IV. Examination of the HOMO and
LUMO reveals the internal charge transfer (ICT) nature of the S0 /
The plots in Fig. 10 show the fluorescence quantum yields
plotted against the maximum frequency of fluorescence of I, II, and
IV in the solvents presented in the tabulated data (see
Supplementary Information). The appearance of these plots is
reminiscent of two compounds from our previous work [5,7],
where similar plots were shown for an unsubstituted 2,5-
diarylidenecyclopentanone ((2E,5E)-2,5-bis-(5-phenyl-penta-2,4-
dienylidene) cyclopentanone) [5] and an asymmetrically
p
S1 excitation, in that
p electron density is transferred from the
electron donor ends of the molecule in the HOMO to the electron
acceptor end of the molecule in the LUMO. In solvent, TD-DFT
substituted
2,5-diarylidene
cyclopentanone
((2E,5E)-2-(4-
cyclo-
predicts S1 as (
Table 4, the calculated energy of the S1 state decreases with respect
to increasing solvent polarity, and that the energy of the 1(n,
*)
p, p*) of B2 symmetry for I, II, and IV. As seen in
cyanobenzylidene)-5-(4-dimethylaminobenzylidene)
pentanone) [7]. In all three plots, it is seen that Ff reaches a
maximum approximately in the middle of the frequency range
plotted, then falls off towards higher and lower frequencies.
Related to these results are the plots shown in Fig. 11 of knr
against the maximum frequency of fluorescence. As shown in the
knr plot for I (Fig. 11(a)), in carbon tetrachloride, knr is high
p
state increases, consistent with the trends observed in the experi-
mental data.
3.2. Photophysical properties
(2.59 ꢃ 109
s
ꢂ1), reaches a minimum at 17378 cmꢂ1 in N,N-
Photophysical properties consisted of measuring the fluores-
cence quantum yields (Ff) and fluorescence lifetimes (tf) in a va-
riety of solvents. Fluorescence quantum yields were very low for IA
and IIA, with values ranging between 0.0006 (in carbon tetra-
chloride, carbon disulfide, and diethyl ether) to 0.01 (in N,N-
dimethylformamide) for IA, and between 0.001 (in n-hexane) to
0.01 (in N,N-dimethylformamide, pyridine, and methanol) for IIA.
Fluorescence lifetimes for IA and IIA were not measured due to
weak fluorescence. The low Ff values observed for IA and IIA are
attributed to efficient singlet / triplet intersystem crossing, in
accordance to El-Sayed's rule [24].
dimethylformamide (8.80 ꢃ 108
s
ꢂ1), then rises to 3.36 ꢃ 109 sꢂ1
in methanol (nf ¼ 15758 cmꢂ1). Fig. 11(a) can be subdivided into
two regions, the region from the minimum to the low frequency
side (region 1) and the region from the minimum to the high fre-
quency side (region 2). In region 1, knr increases from
8.80 ꢃ 108 sꢂ1 (N,N-dimethylformamide, nf ¼ 17378 cmꢂ1) to
3.36 ꢃ 109 sꢂ1 (methanol, nf ¼ 15758 cmꢂ1). In region 2, knr de-
creases from 2.59 ꢃ 109 sꢂ1 (carbon tetrachloride, nf ¼ 20907 cmꢂ1
to 8.80 ꢃ 108 sꢂ1 (N,N-dimethylformamide, nf ¼ 17378 cmꢂ1).
)
Noting that knr ¼ kic þ kisc, where kic is the rate of internal
conversion from S1 to S0 and kisc is the rate of intersystem crossing
from the singlet to the triplet manifold of states, we believe that the
variation in knr seen in Fig. 11(a) can be attributed to opposing
behavior of these two rates with respect to solvent polarity. In
shifting from nonpolar (higher nf) to polar solvents (lower nf), kic
increases while kisc decreases. In region 1, where knr increases with
respect to a decrease in nf, the increase in kic dominates the decrease
in kisc; whereas in region 2, where knr decreases with respect to a
decrease in nf, the decrease in kisc dominates the increase in kic. The
order of magnitude increase in knr found in region 1 is attributed to
the energy gap law for internal conversion, which predicts an
For compounds I, II, III, and IV, further photophysical mea-
surements involved calculating the radiative (kf) and nonradiative
(knr) decay constants from the quantum yield and lifetime data
using the following equations
Ff
tf
kf ¼
(4)
(5)
!
1
Ff
knr
¼
ꢂ 1 kf
exponential dependence of kic on the S0eS1 energy gap (DE).

106
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
kic ¼ aexpðꢂbDEÞ
(6)
According to the energy gap law, kic is expected to increase as
the S0eS1 energy gap decreases due to greater vibrational overlap
(Frank-Condon factor) between the S0 and S1 states [25]. An order
of magnitude change in knr in this region was also found for II
(Fig. 11(b)) and IV (Fig. 11(c)).
In region 2, where intersystem crossing from the singlet to
triplet state manifolds is the major nonradiative decay channel
from S1 to S0, we believe that the solvent modulated location of the
(n,
the rate of singlet to triplet intersystem crossing. The positions of
(n, *) states and ( *) states behave differently under the in-
fluence of a change in solvent polarity. Whereas (n, *) states un-
dergo a hypsochromic shift with increased solvent polarity, ( *)
states undergo a bathochromic shift [26]. A mechanism involving
thermally activated intersystem crossing from S1( *) to a higher
lying 3(n,
*) state has been offered to explain solvent effects on
p*) states relative to the S1(p, p*) and T1(p, p*) states influence
p
p, p
p
p, p
p, p
p
intersystem crossing for I, II, and IV and other related molecules
[27]. We consider vibronic spineorbit coupling [28] to be an
alternative mechanism that is less restrictive than the thermally
activated intersystem crossing mechanism in its requirement for
the magnitude of the S1(
p,
p
*)-T(n,
p*) energy gap relative to kBT. In
this mechanism, spineorbit coupling between (
p
,
p
*) and (n, *) in
p
different spin manifolds and vibronic coupling within the same
spin manifold are operative in promoting intersystem crossing.
Recall that I, II, and IV are compounds belonging to the C2v point
group. The C2v symmetries of these three compounds can be used
to make the discussion of vibronic spineorbit coupling more spe-
cific. Group theory demonstrates that states 1B2(
and 3B2( *)/1A2(n,
coupling and that b1 corresponds to the irreducible representation
of the vibration that induces the vibronic coupling between S1(
*) and Sm(n, *) and T1( *) and Tm(n, *) states for these
p,
p
*)/3A2(n,
p*)
p
,
p
p*) mix through first order spineorbit
p,
p
p
p,
p
p
compounds. The scheme below shows the vibronic-spin orbit
coupling mechanism applied to these three C2v compounds.
1
3
ðS1Þ B2ðp; pꢄÞ)VCðb1Þ/1A2ðn;pꢄÞ)SOC/ðT1Þ B2ðp;pꢄÞ
1
3
ðS1Þ B2ðp; pꢄÞ)SOC/3A2ðn;pꢄÞ)VCðb1Þ/ðT1Þ B2ðp;pꢄÞ
A solvent induced increase in the spacing between S1/T1 and the
appropriate intermediate (n,
p*) state(s) attenuates the degree of
state mixing which in turn reduces the rate of S / T intersystem
crossing [29]. Both the thermally activated mechanism and the
vibronic spineorbit coupling mechanism are favored over direct
spineorbit coupling between S1(p, p*) and T1(p, p*) because of the
well-established understanding that the matrix element for
spineorbit coupling between electronic states of different orbital
configuration is greater than the matrix element for spineorbit
coupling between electronic states of the same orbital configura-
tion [24]. The decrease in the knr values shown in region 2 in the
plots in Fig.11 is consistent with a gradual, solvent induced increase
in the spacing between S1/T1 and higher energy (n,
p*) states.
Table 5 provides theoretical support for this interpretation with the
results of TD-DFT spectral calculations modeled in different sol-
vents. From the TD-DFT calculations of I, II, and IV, the decrease in
knr is consistent with a gradual solvent induced increase in the
spacing between S1/T1 (
resulting in an attenuation in the degree of mixing between (n,
and ( *) states of either the singlet or triplet manifolds.
p
,
p
*) and the higher energy (n,
p
*) states,
p*)
p, p
Fig. 10. Fluorescence quantum yields (Ff) of (a) I, (b) II, and (c) IV plotted against the
fluorescence spectral maxima in various solvents. Circles represent aprotic solvents;
diamonds represent protic solvents.

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
107
Direct comparison of the TD-DFT computed energy gaps and
experimental knr values of I to II in toluene, carbon disulfide, and
dichloromethane led to another interesting observation. A larger
change in the knr value of I than II was observed between toluene
and dichloromethane, which were directly correlated to the mag-
nitudes of the changes in the energy spacings between S1/T1(p, p*)
and the appropriate intermediate (n, p*) states. To exemplify, for I,
the knr value is reduced from 19.2 ꢃ 108 sꢂ1 to 10.9 ꢃ 108 sꢂ1
(
p
D
¼ 8.3 ꢃ 108
s p, p*)-(n,
ꢂ1). The differences in the lowest lying (
*) energy gaps were calculated to be 1168 cmꢂ1 (triplet) and
1112 cmꢂ1 (singlet). Similarly, for II, the knr value is reduced from
6.98 ꢃ 108 sꢂ1 to 4.81 ꢃ 108 sꢂ1
(D
¼ 2.17 ꢃ 108
s
ꢂ1) and the cor-
responding changes in the lowest lying (p, p*)-(n, p*) energy gaps
were calculated to be 1014 cmꢂ1 (triplet) and 709 cmꢂ1 (singlet).
From a conceptual perspective, the larger change in knr observed
for I from toluene to dichloromethane is attributed to a larger
change in the (
in energy gaps implies a smaller degree of vibronic coupling be-
tween (n, *) and ( *) states of the same spin manifold, which in
p, p*)-(n, p*) energy spacings. The larger difference
p
p, p
turn results in a larger change in kisc reduction. Conversely, the
smaller change in knr seen for II is attributed to a smaller difference
in the (n,
energy gaps means a higher degree of vibronic coupling between
(n, *) and ( *) states of the same manifold, which in turn results
p*)-(p, p*) energy spacings. The smaller difference in
p
p, p
in a smaller change in kisc reduction.
Experimental spectroscopic and photophysical results and TD-
DFT calculations of II and IV make it evident that these two com-
pounds behave similarly. Therefore, the tendency for II to undergo
fluorescence quenching in polar solvents cannot be attributed to
twisting of the dimethylamino group to form a polar TICT state
since its spatially restricted julolidine analogue, IV, behaves in a
similar manner.
Compound III exhibited anomalous fluorescence behavior in
highly polar solvents, particularly in alcohols, with Ff values
ranging between 0.11 (in methanol) to 0.36 (in 1-butanol). To
explain the relatively high quantum yield values observed in alco-
hols, TD-DFTcalculations of III in several alcohols were investigated
and compared to I and II, in which fluorescence quenching in al-
cohols was noticeable. Table 6 lists the computed TD-DFT energy
gaps for the lowest lying singlet and triplet (n,
II, and III in methanol, ethanol, and 2-propanol with experimen-
tally observed Ff values. TD-DFT predicted the 3(n,
*) state located
*) state, suggesting that direct spineorbit coupling
*) and T1( *) is weak (low kisc). Vibronic
p*)-(p, p*) states of I,
p
above the S1(p, p
between S1(
p,
p
p, p
spineorbit coupling can be offered as a mechanism for explaining
the high Ff values observed for III in alcohols. As shown in Table 6,
even though the lowest lying triplet (n,
III fall in between I and II, the corresponding singlet (n,
gaps of III are larger than for I and II. The larger energy gap between
p*)-(
p,
p*) energy gaps of
p*)-(p, p*)
1
(p, p p*) states suggests an attenuation in the degree of
*) and 1(n,
vibronic coupling between these two states, resulting in a slower
rate of singlet / triplet intersystem crossing. The higher popula-
tion of molecules in the S1(p, p*) state remains more populated
than the T1( *) state, thereby inducing fluorescence.
p, p
3.3. Excited state protonation studies
Both I and III were found to undergo excited state protonation
when dissolved in glacial acetic acid (see Fig. 12). In knowing the
energies of fluorescence of both the unprotonated (A) and
Fig. 11. Nonradiative decay constant (knr) of (a) I, (b) II, and (c) IV plotted against the
fluorescence spectral maxima in various solvents. Circles represent aprotic solvents;
diamonds represent protic solvents.

108
C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
Table 5
TD-DFT computed energy gaps of the lowest lying (n, p*) and (p, p*) states and experimental knr values for II (in comparison to I for the same solvents) and IV.
Toluene
Carbon disulfide
Dichloromethane
D(DCM/Toluene)
II
T3(n,
S3(n,
p
p
*)-T1(
*)-S1(p,p
p
,
p
*)
*)
)
8346 cmꢂ1
4606 cmꢂ1
6.98
8464 cmꢂ1
4837 cmꢂ1
6.87
9360 cmꢂ1
5315 cmꢂ1
4.81
1014 cmꢂ1
709 cmꢂ1
2.17
knr ꢃ 10ꢂ8 (sꢂ1
I
T3(n,
Sn(n,
p
p
*)-T1(
p,
p
*)
*)y
)
6547 cmꢂ1
3290 cmꢂ1
19.2
6676 cmꢂ1
3547 cmꢂ1
17.6
7715 cmꢂ1
4402 cmꢂ1
10.9
1168 cmꢂ1
1112 cmꢂ1
8.3
*)-S1(p,p
knr ꢃ 10ꢂ8 (sꢂ1
IV
n-Hexane
8449 cmꢂ1
4522 cmꢂ1
10.0
Toluene
8741 cmꢂ1
4826 cmꢂ1
7.53
Dichloromethane
9801 cmꢂ1
5268 cmꢂ1
7.39
T
S3(n,
m(n,p*)-T1(p,p
*)z
p*)-S1(
p
,
p*)
knr ꢃ 10ꢂ8 (sꢂ1
)
yIn toluene and carbon disulfide, m ¼ 2; in dichloromethane, m ¼ 3.
zIn n-hexane and toluene, m ¼ 3; in dichloromethane, m ¼ 4.
protonated (HAþ) forms, the difference in pKa between the excited
solvent polarity, some more significant than others. The observa-
tion of bathochromic shifts in absorbance and fluorescence, TD-DFT
spectral calculations, and LipperteMataga analysis are indicative of
the internal charge transfer (ICT) natures of the S0 / S1 excitation
portrayed by these compounds. Photophysical properties also
showed solvent dependence. The behavior of knr with respect to the
fluorescence spectral maxima and solvent polarity can be divided
into two regions. In region 1, the increase in the rate of internal
conversion allows it to dominate over intersystem crossing;
whereas in region 2, intersystem crossing is the major nonradiative
decay channel from S1. The variation in the rates of internal con-
version and intersystem crossing are interpreted to be related to
solvent induced changes in energy gaps on the singlet and triplet
spin manifolds.
state and ground state (
DpKa) of each compound were calculated
from the Fӧrster cycle [30,31].
ꢄ
ꢅ
fl
fl
~
~
NAhc yA ꢂ yHAþ
2:303RT
DpKa ¼ pKa ꢄ ꢂ pKa ¼
(7)
where NA is Avogadro's number, h is Planck's constant, c is the speed
of light in a vacuum, R is the universal gas constant, and T is absolute
temperature of the solution. Experimental observations show that
the unprotonated and protonated forms of I fluoresce at 16750 cmꢂ1
and 13569 cm-1, and those for III at 15576 cm-1 and 12225 cm-1. From
the emission data, the
6.7 and 7.1. Positive pKa values are indicative that the compounds
are more basic in the singly excited ( *) state than in the ground
DpKa values of both I and III were found to be
D
The tendency for compound II to undergo fluorescence
quenching in polar aprotic and protic solvents cannot be attributed
to twisting of the dimethylamino group to form a polar TICT state
since is spatially restricted julolidine analogue, IV, behaves in a
similar manner and is incapable of twisting to form a TICT state.
p, p
singlet state (pKa* > pKa). Additional quantitative support on the
excited state protonation of I and III in glacial acetic acid involved
calculating atomic charges (Q) in both the ground singlet and lowest
lying singly excited (p, p*) states, looking specifically at the carbonyl
oxygen atom. For I, QO(S0)
D
¼
ꢂ0.55, Q(1(
p
p, p
,
p
*))
¼ ꢂ0.66,
QO ¼ ꢂ0.11; for III, QO(S0) ¼ ꢂ0.55, Q(1(
*)) ¼ ꢂ0.63,
D
QO ¼ ꢂ0.077. The charge of the carbonyl oxygen atom becomes
1
more negative in going from the S0 to
(p, p*) state, which is sup-
portive that the compounds are stronger bases in the excited state
and have a higher affinity of protonation.
4. Conclusion
A series of alkylamino substituted 2-arylidene and 2,5-diary-
lidene cyclopentanone compounds have been synthesized and
structurally characterized. The spectroscopic and photophysical
properties of all compounds studied have been found to vary with
Table 6
TD-DFT computed energy gaps of S1/T1
(p, p*) and the appropriate intermediate (n,
p*) states of I, II, and III in methanol, ethanol, and 2-propanol with experimentally
observed Ff values.
Compound
D
E [cmꢂ1] and Ff
Methanol
Ethanol
2-Propanol
I
D
D
ES ¼ 4765
ET ¼ 8081
D
D
ES ¼ 4697
ET ¼ 8001
D
D
ES ¼ 4697
ET ¼ 7977
(
D
D
Ff ¼ 0.025)
(
D
D
(
D
D
(
Ff ¼ 0.11)
(
D
D
(
D
D
(
Ff ¼ 0.18)
II
III
ES ¼ 5292
ET ¼ 9702
ES ¼ 5326
ET ¼ 9624
ES ¼ 5326
ET ¼ 9639
(
D
D
Ff ¼ 0.022)
Ff ¼ 0.076)
Ff ¼ 0.15)
ES ¼ 5543
ET ¼ 9672
ES ¼ 5480
ET ¼ 9622
ES ¼ 5480
ET ¼ 9603
Fig. 12. Room temperature absorption and fluorescence spectra of (a) III and (b) I in
glacial acetic acid.
(
Ff ¼ 0.11)
Ff ¼ 0.23)
Ff ¼ 0.33)

C.A. Zoto et al. / Journal of Molecular Structure 1112 (2016) 97e109
109
[10] A.E. Rosamilia, M.A. Giarrusso, J.L. Scott, C.R. Strauss, Green. Chem. 8 (2006)
1042.
[11] G. Cai, N. Bozhkova, J. Odingo, N. Berova, K. Nakanishi, J. Am. Chem. Soc. 115
(1993) 7192.
Direct comparison of the spectroscopic and photophysical proper-
ties of II and IV clearly support this argument. Vibronic spineorbit
coupling and comparative TD-DFT spectral calculations of III with I
and II were used to justify the anomalous behavior of fluorescence
and high Ff values observed for III in alcohols. A supportive argu-
ment of the effect of structural asymmetry on the anomalous
fluorescence behavior for III in alcohols is still inconclusive.
Lastly, compounds I and III have been found to undergo excited
state protonation in glacial acetic acid, consistent with the observed
fluorescence spectra and theoretically calculated partial charges of
[12] H. Maeta, K. Suzuki, Tet. Lett. 34 (1993) 341.
[13] M.G. Ucak-Astarlioglu, The Electronic Structure and Spectroscopy of Diary-
lidene Cycloalkanones and Their Protonated Cations, Ph. D. dissertation,
Worcester Polytechnic Institute, 2003.
[14] Bruker, SAINT (Version 6.14).
[15] SHELXTL (Version 6.14) for WNT/2003, Bruker AXS Inc., Madison, WI, 1999.
[16] CCDC. Mercury 2.4, CCDC, Cambridge, U. K., 2009.
[17] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, second ed., Springer,
New York, 2004.
atoms both in the S0 and lowest lying 1(
p, p*) states.
[18] J.R. Rechthaler, G. Kohler, Chem. Phys. Lett. 189 (1994) 99.
[19] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino,
G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers,
K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam,
M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski,
R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz,
J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford
CT, 2009.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.molstruc.2016.02.009.
References
[1] (a) M.V. Barnabus, A. Liu, A.D. Trifunac, V.V. Krongauz, C.T. Chang, J. Phys.
Chem. 96 (1992) 212;
(b) W.J. Chambers, D.F. Eaton, J. Imaging Sci. 30 (1986) 230;
(c) T.J. Dougherty, C.J. Gomer, B.W. Henderson, G. Jori, O. Kessel, M. Korbelik,
J. Moan, Q. Peng, J. Natl. Cancer. Inst. 90 (1998) 889.
[20] MOS-F version 4.2B, Microsoft® Windows 95/98/NT, A. Matsuura, Fujitsu Labs
Ltd.
[21] E.Z. Lippert, Naturforschung 10a (1955) 541.
[22] N. Mataga, Y. Kaifu, M. Koizumi, Bull. Chem. Soc. Jpn. 29 (1956) 465.
[23] F. Furche, D. Rappoport, Density functional methods for excited states: equi-
librium structure and electronic spectra, in: M. Olivucci (Ed.), Computational
Photochemistry, Elsevier, Amsterdam, 2005, pp. 93e128.
[24] M.A. El-Sayed, J. Chem. Phys. 38 (1963) 2834.
[2] (a) V.G. Pivovarenko, A.V. Klueva, A.O. Doroshenko, A.P. Demchenko, Chem.
Phys. Lett. 325 (2000) 389;
(b) P.K. Das, R. Pramanik, D. Banerjee, S. Bagchi, Spectrochim. Acta A 56 (2000)
2763.
[3] A.O. Doroshenko, A.V. Grigorovich, E.A. Posokhov, V.G. Pivovarenko,
A.P. Demchenko, Mol. Eng. 8 (1999) 199.
[4] (a) J. Kawamata, K. Inoue, T. Inabe, Bull. Chem. Soc. Jpn. 71 (1998) 2777;
(b) J. Kawamata, K. Inoue, T. Inabe, M. Kiguchi, M. Kato, Y. Taniguchi, Chem.
Phys. Lett. 249 (1996) 29;
[25] R. Englman, J. Jortner, Mol. Phys. 18 (1970) 145.
[26] P. Suppan, N. Ghonheim, Solvatochromism, The Royal Society of Chemistry,
Cambridge, United Kingdom, 1997.
(c) P. Kaatz, D.P. Shelton, J. Chem. Phys. 105 (1996) 3918;
(d) J. Kawamata, K. Inoue, T. Inabe, Appl. Phys. Lett. 66 (1995) 3102;
(e) W. Tao, W. Fei-peng, S. Men-quan, Chem. Res. Chin. Univ. 19 (2003) 470.
[5] R.E. Connors, M.G. Ucak-Astarlioglu, J. Phys. Chem. A 107 (2003) 7684.
[6] M.G. Ucak-Astarlioglu, R.E. Connors, J. Phys. Chem. A 109 (2005) 8275.
[7] C.A. Zoto, R.E. Connors, J. Mol. Struct. 982 (2010) 121.
[8] C.A. Zoto, M.G. Ucak-Astarlioglu, R.E. Connors, J. Mol. Struct. 1105 (2016) 396.
[9] U.P. Kreher, A.E. Rosamilia, C.L. Raston, J.L. Scott, C.R. Strauss, Org. Lett. 5
(2003) 3107.
[27] A.O. Doroshenko, V.G. Pivovarenko, J. Photochem Photobiol. A Chem. 156
(2003) 55.
[28] J. Tatchen, N. Gilka, C.M. Marian, Phys. Chem. Chem. Phys. 9 (2007) 5209 (and
references therein).
[29] R.H. Clarke, H.A. Frank, J. Chem. Phys. 65 (1976) 39.
[30] I.F. Pierola, N.J. Turro, P. Kuo, Macromolecules 18 (1985) 508.
[31] A.S. Sharma, S.G. Schulman, Introduction to Fluorescence Spectroscopy,
Wiley-Interscience Publication, Canada, 1999.