Metal induced enhancement of fluorescence and modulation of two-photon absorption cross-section with a donor-acceptor-acceptor-donor receptor

Article abstract of DOI:10.1016/j.jorganchem.2007.07.013

A metal ion sensing fluorophore L that exhibits a large two-photon absorption cross-section has been synthesized in good yields. The influences of different metal ion inputs, on the one- and two-photon spectroscopic properties of L, have been investigated. The ligand itself does not show any fluorescence although in presence of a metal ion like Zn(II), Cd(II), Mg(II) or Ca(II), a ～25 time enhancement of fluorescence is observed. The ligand with symmetrical "donor-acceptor-acceptor-donor" characteristics exhibits a large two-photon absorption cross-section measured by femtosecond open-aperture Z-scan technique at 880 nm. However, presence of any of the above metal ions lowers its two-photon absorption cross-section (δ) to different extents at 880 nm. Theoretical calculation carried out in DFT formalism on the ligand and its Zn(II) complex corroborate experimental results.

Full text of DOI:10.1016/j.jorganchem.2007.07.013

Journal of Organometallic Chemistry 692 (2007) 4969–4977
www.elsevier.com/locate/jorganchem
Metal induced enhancement of fluorescence and modulation
of two-photon absorption cross-section with
a donor–acceptor–acceptor–donor receptor
*
Sanjib Das, Amit Nag, Kalyan K. Sadhu, Debabrata Goswami, Parimal K. Bharadwaj
Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India
Received 17 May 2007; received in revised form 12 July 2007; accepted 13 July 2007
Available online 19 July 2007
Abstract
A metal ion sensing fluorophore L that exhibits a large two-photon absorption cross-section has been synthesized in good yields. The
influences of different metal ion inputs, on the one- and two-photon spectroscopic properties of L, have been investigated. The ligand
itself does not show any fluorescence although in presence of a metal ion like Zn(II), Cd(II), Mg(II) or Ca(II), a ꢀ25 time enhancement
of fluorescence is observed. The ligand with symmetrical ‘‘donor–acceptor–acceptor–donor’’ characteristics exhibits a large two-photon
absorption cross-section measured by femtosecond open-aperture Z-scan technique at 880 nm. However, presence of any of the above
metal ions lowers its two-photon absorption cross-section (d) to different extents at 880 nm. Theoretical calculation carried out in DFT
formalism on the ligand and its Zn(II) complex corroborate experimental results.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Metal ion receptor; OPA; TPA; Fluorescence; DFT
1. Introduction
structure–property relationships for third order nonlinear
optical chromophores and TPA activity are still absent.
However, search for molecular guidelines to have
large TPA cross-section (d) has led to molecules with asym-
metric donor–p-bridge–acceptor (D–p–A) and symmetric
The prospect of developing new materials exhibiting
third-order NLO property measured in terms of two-photon
absorption cross-section (d) has received increasing atten-
tion in recent years, due to their potential applications [1]
in opto-electronics and photonic devices. These molecules
are of relevance in the emerging areas of two-photon laser
scanning microscopy (TPLSM) [1a], optical power limiting
[1b], three-dimensional (3D) optical data storage [1c], two-
photon up-conversion lasing [1d], photodynamic cancer
therapy [1e], and so on. Particularly, fluorescence active
molecules with large two-photon absorption cross-section
donor–p-bridge–donor
(D–p–D),
acceptor–p-bridge–
donor–p-bridge–acceptor (A–p-D–p–A), and donor–
p-bridge–acceptor–p-bridge–donor (D–p–A-p–D) conju-
gated structural motifs. Both theoretical considerations
and experimental results lead to the molecules with high
TPA activity as the ones where donor and acceptor groups
are symmetrically disposed resulting in a substantial sym-
metric intramolecular charge re-distribution upon excita-
tion [2]. These studies also reveal that the effect of
molecular charge transfer symmetry, donor/acceptor group
strength, chromophore number density and length of the
conjugated p backbone are the most important factors
affecting the molecular TPA activity [3]. So far, the TPA
phenomenon has focused mostly on organic dipolar [4]
and less on quadrupolar [5], octupolar [6], and dendritic
d
(i.e., d > 1000 GM; 1 GM = 1 · 10ꢁ50 cm4 s pho-
tonꢁ1 moleculeꢁ1) with the excitation wavelength well
removed from the range, 600–1300 nm are important in
TPLSM applications. However, well-defined molecular
*
Corresponding author. Tel.: +91 512 259 7034; fax: +91 512 259 7436.
E-mail address: pkb@iitk.ac.in (P.K. Bharadwaj).
0022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.07.013

4970
S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
Scheme 1. Plan for modulation of TPA response in a D–p–D0 type receptor system. D, donor; p, ethylenic conjugated moieties; Mn+, Zn(II), Cd(II),
Mg(II) and Ca(II) ions.
[7] molecules and to very few extents with metal complexes
[8]. Very recently, we have shown that simple metal ion rec-
ognition events lead exceptionally large values of TPA
cross-sections known for any coordination metal complexes
in a weakly polarized conjugated Schiff base ligand [9].
Further, it is important to develop systems that not only
show high d values but also have the ability to respond the
presence of various ionic/neutral analytes through modula-
tion of their TPA responses. Few synthetic approaches [10]
and theoretical calculations are reported [10,11] in the liter-
ature on ways to modulate molecular TPA properties.
Jiang et al. has synthesized different stilbene based symmet-
rically (D–p–D) and asymmetrically (D–p–A) substituted
conjugated dipolar chromophores [10] and showed a com-
parative study of the TPA activity for symmetrically and
asymmetrically substituted chromophores. The TPA effi-
ciency decreases on going from D–p–D to D–p–A mole-
cules. Theoretical calculations carried out by this group
for symmetrically and asymmetrically substituted chro-
mophores point to the fact that the transition dipole
part (p0, as obtained in HOMO of ligand) the p conjuga-
tion is affected a lot. Binding of metal ions such as Zn(II),
Cd(II), Mg(II) or Ca(II) to the imine N as well as sulfur
atoms of L renders them as acceptors i.e. D–p–p0–p–D
changes to a D–p–r–p–D analogue with reduction of
TPA activity (Scheme 1) where the binding of metal ion
with thio ether moiety plays the crucial role by snapping
the conjugation. Interestingly, the ligand does not show
any significant fluorescence but on complexation with
any of the above mentioned metal ions, exhibits ꢀ25-fold
enhancement of fluorescence. Although enhancement is
not very large, it provides a starting point for ligand design.
The metal ions probed in this study are biologically
important. Many enzymatic reactions are mediated by
Mg(II) ions whereas Ca(II) ion acts as an universal second
messenger in cells [12]. Few studies have been reported [13]
in the literature on the TPA properties of chromophores
that respond to different ions such as calcium orange, cal-
cium green-I, and calcium crimson with d values in the
range of 1–30 GM in the unbound state. Upon binding
of calcium ions, d values increase to 50–100 GM. On the
other hand, Mg(II) was shown to lower the d values by
50% while showing fluorescence enhancement in an aza-
crown ether connected to distyrylbenzenes in the D–p–A–
p–D format [14].
0
moment (Mee or Mee Þ function is the most important factor
in modulating the molecular TPA efficiency, which is very
high for symmetrically substituted stilbene chromophores
in comparison to asymmetrically substituted chromoph-
ores. Relevance of transition dipole moment in predicting
TPA efficiency was originally put forth by Marder and
coworkers [2]. However, majority of regular dipolar
compounds show d values that typically do not exceed
2. Experimental
10–300 GM (1 GM = 10ꢁ50 cm4 s photonꢁ1 moleculeꢁ1
measured by femtosecond laser.
)
2.1. Materials
Metal ions are excellent 3D template and can assemble
simple organic NLO-phores around with concomitant tun-
ing of the molecular nonlinear optical property by virtue of
inducing a strong intra-ligand charge-transfer (ILCT) as
well as low-energy metal–ligand charge transfer (MLCT)
transitions. Herein, we report a new highly TPA active acy-
clic receptor (L), synthesized by Schiff base condensation of
4-(dimethylamino)benzaldehyde with 1,2-di(o-aminophe-
nolthio)ethane. The ligand L is a D–p–p0–p–D analogue
since the imine part possesses less electron-donating char-
acter with respect to the dimethylamine moiety and instead
of having a small-saturated carbon chain in the thioether
Reagent grade 4-(dimethylamino)benzaldehyde, 1-flu-
oro-2-nitrobenzene, palladium on carbon 5%, Rhoda-
mine-6G, 1,2-ethanedithiol and perchlorate salts of
different metals used in the study were acquired from
Aldrich and used as received. Reagent grade hydrazine
hydrate, K2CO3 and all the solvents were received from
S.D Fine Chemicals, India. All solvents were freshly dis-
tilled prior to use.
2.2. Synthesis of L
The synthetic route adopted for L is given in Scheme 2.

S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
4971
NH2NH2.H2O
5% Pd / C
F
HS
SH
S
S
S
S
K2CO3
DMF
Reflux for 6 hr
EtOH
NO2
NH2
H2N
NO2
O2N
Reflux for 24 hr
Ln
Lm
S
N
S
N
N
CHO
EtOH
Reflux for 48 hr
L
L
N
N
Scheme 2. Synthetic route for D–p–D0 type highly TPA active cationic receptor.
2.2.1. 1,2-Di(o-nitrophenolthio)ethane (Lm)
(s, 12H), 6.69–6.62 (m, 4H), 7.11–7.07 (m, 4H), 7.31–7.28
(m, 4H), 7.73–7.71 (m, 4H), 9.72 (s, 2H); ES-MS (m/z):
537(20%) [MꢁH]+. Anal. Calc. for C32H34N4S2: C, 71.34;
H, 6.36; N, 10.40. Found: C, 71.42; H, 6.45; N, 10.49%.
To a solution of 1,2-ethanedithiol (0.94 g, 1 mmol) in
dry DMF (10 mL) was added anhydrous K2CO3 (0.35 g,
2.5 mmol) and the reaction mixture was stirred for 1 h. sub-
sequently, a solution of 1-fluoro-2-nitrobenzene (0.31 g,
2.2 mmol) in DMF (10 mL) was added dropwise for
15 min and the reaction mixture was stirred for 24 h at
RT. The reaction mixture is further heated at 70 ꢁC for
15 h to complete the reaction. After cooling to RT, the
reaction mixture was poured into ice-cold water
(100 mL). The light yellow solid separated was collected
by filtration and washed repeatedly with water
2.3. Methods
The ligand L was characterized by elemental analysis,
1H NMR and ESI-MS spectra. 1H NMR spectra
were recorded on a JEOL JNM-LA400 FT (400 MHz)
instrument in CDCl3 with Me4Si as the internal standard.
UV–Vis spectra were recorded on a JASCO V-570 spec-
trophotometer in CH3CN at 298 K. Fluorescence spectra
in solution phase were recorded on a Perkin Elmer
LS50B, luminescence spectrometer at 298 K. Melting
points were determined with an electrical melting point
apparatus by PERFIT, India and were uncorrected. The
electrospray mass spectra (ES-MS) were recorded on a
MICROMASS QUATTRO Quadruple Mass Spectrome-
ter. Each sample dissolved in acetonitrile and introduced
into the ESI source through a syringe pump at the rate
of 5 ll/min. The ESI capillary was set at 3.5 kV and the
cone voltage was 40 V. The spectra were collected in 6 s
scans and the printouts were averaged of 6–8 scans.
Microanalyses for the complexes were obtained from
CDRI, Lucknow, India.
1
(5 · 100 mL). Yield ꢀ90%; m.p. 175 ꢁC; H NMR: d 2.87
(s, 4H), 7.52–7.41 (m, 8H); ES-MS (m/z): 335(50%)
[MꢁH]+. Anal. Calc. for C14H12N2O4S2: C, 49.99; H,
3.60; N, 8.33. Found: C, 50.08; H, 3.69; N, 8.43%.
2.2.2. 1,2-Di(o-aminophenolthio)ethane (Ln)
In a two-necked round-bottomed flask equipped with a
reflux condenser and a dropping funnel was prepared a sus-
pension of dinitro compound Lm (0.33 g, 1 mmol), palla-
dium on carbon 5% (0.05 g), and absolute ethanol
(15 mL). The mixture was warmed with stirring and subse-
quently hydrazine hydrate 80% (3 mL) in ethanol (10 mL)
was added dropwise over a 1 h period through the drop-
ping funnel with temperature maintained at about 50 ꢁC.
The reaction mixture was refluxed for 6 h and filtered while
hot. On cooling, the filtrate gave the corresponding dia-
mino compound Ln as a white solid after vacuum drying.
Fluorescence quantum yields of all the compounds were
determined by comparing the corrected spectrum with that
of quinine sulfate (/F = 0.54) in 1 N sulfuric acid [15],
1
Yield ꢀ95%; m.p. 154 ꢁC; H NMR: d 2.82 (s, 4H), 7.42–
kex = 350 nm, taking the area under the total emission.
7.35 (m, 8H); ES-MS (m/z): 275 (60%) [MꢁH]+. Anal.
Calc. for C14H16N2S2: C, 60.83; H, 5.83; N, 10.13. Found:
C, 60.89; H, 5.87; N, 10.21%.
2.3.1. Measurement of two photon absorption cross-section
(d)
Two-photon absorption cross-section values are mea-
sured by open aperture Z-scan technique [16]. The femto-
second experimental scheme involves mode-locked
Coherent Mira titanium: sapphire laser (Model 900) which
is pumped by Coherent Verdi frequency doubled Nd:vana-
date laser. The model 900 Mira is tunable from 740 to
900 nm and its repetition rate is 76 MHz. The duration
of the pulse is 150 fs as measured by autocorrelation tech-
nique. We used 880 nm wavelength here for all the mea-
surements. Using a 20 cm focal length lens, the beam is
2.2.3. Synthesis of L
A mixture of 4-(dimethylamino)benzaldehyde (0.37 g,
2.5 mmol) and Ln (0.28 g, 1 mmol) were dissolved in
10 mL absolute ethanol and heated in an oil bath at
80 ꢁC with stirring for 48 h in the dark. Cooling the reac-
tion mixture at room temperature afforded a yellow solid.
The yellow solid was collected by filtration, washed thor-
oughly with absolute ethanol and dried under vacuum.
1
Yield ꢀ 90%; m.p. 165 ꢁC; H NMR: d 2.83 (s, 4H), 3.06

4972
S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
focused into a 1 cm long cell filled with sample, where it
easily produces GW-level intensity at the focal point of
the lens. The sample is scanned through the focal point
using a motorized translation stage (model ESP 300),
which can step with a minimum resolution of 0.1 lm. This
allows a smooth intensity scan for the samples in this wave-
length. The transmitted beam, after passing through the
sample is focused into the aperture of a UV-enhanced
amplified silicon photo detector (Thorlabs DET 210) by
using a 7.5 cm focal length lens. The signal is measured
in an oscilloscope (Tektronix TDS 224), which is finally
interfaced with the computer using GPIB card (National
Instruments). The data are acquired using LabVIEW pro-
gramming. The nonlinear absorption coefficient b is
obtained [17], by fitting our measured transmittance values
to the following formula:
where A0 and A are the absorbance of the metal-free li-
gand and the ML complex, respectively and [M] is the
concentration of various metal ions added for complexa-
tion. When A0/(A ꢁ A0) is plotted against the metal ion
concentration [M]ꢁ1, Ks is directly obtained from the
intercept/slop ratio. The details of the theoretical analysis
of the stoichiometries and corresponding binding constant
determination are available [20] in the literature. Here the
reported values are consistent with good correlation coef-
ficients (P0.98).
3. Results and discussion
In ligand L, two imine moieties along with two sulfur
groups (N2S2 unit) form a 3-D cavity with moderately large
residual electron density capable of binding metal ions
(Scheme 1). This makes L an attractive candidate for mod-
ulation of TPA property through metal ion recognition
events.
TðzÞ ¼ 1 ꢁ bI0L=ð2ð1 þ z2=z02ÞÞ;
where b = nonlinear absorption coefficient, I0 = on-axis
electric field intensity at the focal point in absence of the
sample, L = sample thickness, z0 ¼ Rayleigh range ¼
pw20=k; w0 is the minimum spot size at the focal point.
The b values are obtained by curve fitting the measured
open-aperture traces with the above equation. After getting
the value of b, the TPA cross-section d in GM unit of one
solute molecule is given by the following expression:
3.1. UV–Vis absorption spectroscopy and binding constant
determination in metal complexes
Compound L is highly soluble in DCM but sparingly sol-
uble in CH3CN solvent. Thus, all the photophysical proper-
ties of L were studied in mixed DCM/CH3CN solvents. It
exhibits a broad absorption band at kmax = 358 nm in
90:10 (v/v) mixture of DCM and CH3CN solution with
d ¼ bhm=N ꢂ c ꢂ 10ꢁ3
;
e
max = 7.5 · 104 mol Lꢁ1 cmꢁ1 (Table 1), which is attribut-
where m is the frequency of the incident laser beam, N is
Avogadro constant, c is the concentration of the com-
pound in 90:10 (v/v) mixture of DCM and CH3CN solvent.
Rhodamine 6G is taken as the reference to calibrate the
measurement technique for which the d value is known in
the literature [18].
able [21] to the combined locally excited (LE) and intra-
ligand charge transfer (ILCT) from the dimethylamine moi-
ety to the imine moiety. Upon addition of metal ions, two
well-resolved broad peaks appear where the kmax of higher
energy band is located in the range of 337–355 nm and the
lower energy band is located at kmax = 450 nm. The emax val-
ues are highly sensitive to the metal ion inputs. Upon com-
plexation, there is a considerable change in the emax value of
the lower energy band in case of Zn(II) and Mg(II) ion input
compare to Ca(II) and Cd(II) ion input. The reverse situa-
tion appears in case of the higher energy band (Table 1).
To explain this phenomenon, the complex stability constant
values (logKs) are calculated with 1:1 metal–ligand ratio as
confirmed by the isobestic point obtained at 400 nm (Fig. 1).
For the complexation of L with Zn(II) and Mg(II), the
logKs values are found to be 4.6 and 4.4 respectively,
whereas the logKs values for Cd(II) and Ca(II) are 3.6 and
3.4, respectively. Based on the stability constant data, it is
2.3.2. Binding constant determination
The binding constants (Ks) for the complexes between L
and metal ions were determined by UV–Vis spectroscopic
titration of dilute solution of L against metal ion solution
in 90:10 (v/v) mixture of DCM and CH3CN solvent. The
concentration of metal ions was in the range 1 · 10ꢁ5
–
1 · 10ꢁ3M. The ligand concentration was 1 · 10ꢁ5 M.
The linear fit of the absorption spectral data at a particular
wavelength for 1:1 complexation was obtained by applying
the following equation [19]
A0=ðA ꢁ A0Þ ¼ ½a=ðb ꢁ aÞꢃ½ð1=Ks½MꢃÞ þ 1ꢃ;
Table 1
Liner and nonlinear spectroscopic data for L
Compound
k
abs/nm (e · 104/mol Lꢁ1 cmꢁ1
)
kmax-em (nm)
d (GM)
logKs
/ · 10ꢁ3
Dd (%)
L
358 (7.5)
422
384
409
386
407
2300
1580
1960
1650
1930
0.44
12.8
4.2
10.2
2.7
L–Zn2+
L–Cd2+
L–Mg2+
L–Ca2+
338 (4.4), 450 (7.6)
347 (5.9), 463 (0.6)
338 (5.2), 450 (5.1)
355 (6.7), 465 (0.7)
4.6
3.6
4.4
3.4
31
14
28
16

S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
4973
0.4
0.3
0.2
0.1
0.0
Y = 0.00449 + 1.11909 x 10-7
0.060
X
R = 0.99854
0.045
0
A
0.030
/-A
0
A
0.015
300
350
400
450
500
0
200000
400000
600000
[M]-1
Wavelength (nm)
Fig. 1. (a) UV–Vis spectra of L upon addition of Zn(II) ion. (b) Linear regression plot of A0/(A ꢁ A0) as a function of inverse concentration of Zn(II) ion
added to L.
speculated that due to small size both Zn(II) and Mg(II) ions
can easily fit into the cavity of L and localize the electrons of
the imine moieties to greater extent compared to Ca(II) and
Cd(II) ions which, in turn, becomes a strong acceptor. As a
result, Zn(II) and Mg(II) ions induce more asymmetric D–
p–A character to L in compare to Ca(II) and Cd(II) ions.
to the metal-ion free receptor L (Table 1). The isomeriza-
tion of the imine bond in the excited state prevail the fluo-
rescence property of the free ligand [23]. However, in
presence of metal ion the isomerization of the imine bond
is restricted due to the formation of metal–nitrogen bond,
which reflects the increment of quantum yield of the
metal–ligand complex.
3.2. Emission studies
3.3. Nonlinear optical properties
Compound L shows very low broad emission in a mix-
ture of DCM and CH3CN solvent. The quantum yield
(/F = 0.00044) is very low due to weak ICT from the termi-
nal dimethylamino to the imine (referenced to quinine sul-
fate [15] in 1 N sulfuric acid, kex = 350 nm, /F = 0.54). In
presence of metal ions, L exhibits enhanced fluorescence
intensity with two well-resolved peaks assignable to the
locally excited (LE) transition and the intramolecular
charge transfer (ICT) transition from the donor dimethyla-
mino to the metal bound imine moieties, respectively
(Fig. 2) [22]. In case of Zn(II) and Mg(II) ions, the fluores-
cence quantum yield increases up to ꢀ25 times compared
The TPA cross-section (d) value of L was measured in
90:10 (v/v) mixture of DCM/CH3CN solvent in absence
and presence of metal ions at 880 nm. The Z-scan traces
are shown in Fig. 3 while the d values for free and com-
plexed L are given in Table 1. The high d value of 2300
GM for L can be attributed to the presence of two D–p–
p0–p–D chromophoric units with extended conjugation
through the two aromatic rings for each unit. Several
reports are available in the literature [24] showing incre-
ment of the d values to different extents with the increase
of the chromophore number density.
0.9
Ca2+
L
Zn2+
Zn2+
Cd2+
0.6
Cd2+
Mg2+
Ca2+
L
Mg2+
0.3
0.0
400
500
600
300
400
500
600
Wavelength (nm)
Wavelength (nm)
Fig. 2. (a) UV–Vis spectra of L alone and in presence of different metal ion inputs in 90:10 (v/v) mixture of DCM/CH3CN (1 · 10ꢁ5 M). (b) Emission
spectra of L alone and in presence of Zn(II), Cd(II), Mg(II) and Ca(II) in 90:10 (v/v) mixture of DCM/CH3CN (1 · 10ꢁ5 M).

4974
S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
L-Ca2+
1.0
L-Mg2+
L
1.0
0.8
0.6
0.4
1.0
0.9
0.8
0.7
0.8
0.6
0.6
Experimental data
Experimental data
Experimental data
Theoritical fit
Theoritical plot
0.4
Theoritical fit
0.5
-8
-4
0
4
8
-8
-4
0
4
8
-8
-4
0
4
8
Z Position (cm)
Z Position (cm)
Z Position (cm)
L-Zn2+
L-Cd2+
1.0
0.8
0.6
0.4
1.0
0.8
0.6
Experimental data
Theoritical fit
Experimental data
Theoritical plot
0.4
-8
-4
0
4
8
-8
-4
0
4
8
Z Position (cm)
Z Position (cm)
Fig. 3. Open-aperture Z-scan traces of L (conc. 5 · 10ꢁ4 M) and in presence of Ca(II), Mg(II), Zn(II) and Cd(II) ions input. Solid lines are the best fitted
of the experimental data.
When a metal ion is added to L, it binds at the N2S2
donor site as shown in Scheme 1. The metal coordination
induces considerable localization of the electrons causing
lowering of the d values compared to the metal-free L.
Both Zn(II) and Mg(II) ions decrease the TPA cross-sec-
tional value more in comparison to Ca(II) and Cd(II) con-
sistent with higher binding constant values for the former
two ions making the molecular charge transfer character
of L more asymmetric in nature.
erties of the chromophores, quantum chemical calculations
were performed for three different models of the ligand as
well as three different metal–ligand complex models exclud-
ing the anions for simplicity. The model compounds are
based upon cis/trans isomerism of imine bonds and the
models are trans–trans, trans–cis and cis–cis geometry
along the two strands of the ligand.
The global energy-minimized geometry is obtained for
the trans–trans orientation of the imine bonds for the ligand
(Fig. 4). But in case of metal complex the nature of two
imine bonds become different and the lowest energy struc-
ture is obtained for trans–cis orientation. This change in
geometry results in lower TPA values for the metal–ligand
complex. The trans–trans geometry in the ligand is more
favorable than the other two models mainly due to less ste-
ric hindrance between the p-arms of the ligand; hence p con-
jugation is quite effective through the chromophoric length
of the ligand. Any one of the model complexes for ligand
does not provide a favorable molecular geometry for metal
3.4. Theoretical studies
The molecular geometries of the ligand and its Zn(II)
complex were optimized at the B3LYP/6-31G* level in
the density functional theory (DFT) formalism by using
the GAUSSIAN 03 program package [25] to study the effect
of the nature of ligand in the metal complex on two photon
absorption properties. The geometries corresponding to
global minima obtained with the semi-empirical AM1
method (gradient <0.001, Hyperchem version 7.0; Hyper-
cube Inc.) were similar to the DFT results for all the cases.
The transition energies of the compounds were then calcu-
lated by using time-dependent DFT (TD-DFT) method
using the same functional (B3LYP) with the same basis
set (6-31G*). The design of the pushing donor groups
allows assessing the effect of donor strengths. To gain bet-
ter insight into the effects of Zn+2 on the two-photon prop-
˚
complexation as the N(1)–N(2) distances are more than 9 A
i.e., too long to bind a metal ion. As a result, the geometry
of the ligand is severely modified via rotation of the central
C–C bond between two sulfur atoms. The Zn–S bond dis-
tance is the guiding factor in stabilization of this particular
geometry over the other two models. In all the cases, the sul-
fur atoms in the ligand play the role of weak coordinating
atoms to bind with metal. The calculated Zn–S bond

S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
4975
Fig. 4. The global energy-minimized geometry of (a) the ligand and (b) its Zn(II) complex.
Fig. 5. Contour surfaces of HOMOꢁ1, HOMO, LUMO and LUMO+1 for (a) the ligand and (b) its Zn(II) complex.
˚
distances (2.45 A) are similar to the values reported for
other Zn(II) complexes with thioether ligands [26]. In the
present case, metal–sulfur interaction snaps the conjugation
of the p electron cloud that was present throughout the side
arms in the free ligand (Fig. 5). This reduces charge transfer
and in turn reduces the two-photon absorption cross-

4976
S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
section. In some case, this decrease of the transition dipole
moment can be explained by analyzing the transition nature
of the higher-lying (TPA target) excited state as well as the
lower-lying (intermediate) state. There is a delocalization of
p electron from the donor dimethylamino group to the
nitrogen atom of imine bond. The planar geometry in the
two arms (torsion angle between dimethylamino group
and imine group is less than 1ꢁ) in the favorable geometry
of the ligand is distorted to a greater extent when complexed
with Zn(II) (torsion angle is grater than 14ꢁ). These distor-
tions from planarity lead to diminished conjugation and in
turn to low TPA cross-section in the complex. The summa-
tion of the angle around the imine nitrogen atom after com-
plexation suggests sp2 geometry of the nitrogen in the
metal-bound state. This is consistent with the imine nitro-
gen atoms being strongly, but not completely, decoupled
from the p system when Zn+2 ion is complexed by the
ligand.
Comparing the energy level diagrams of the ligand and
its complex with Zn(II) ion, it is found that complexation
decreases the HOMO–LUMO energy gap. This is corrobo-
rated by single photon absorption spectra that show red
shift upon metal complexation. Table 2 summarizes the
calculated one-photon absorption (OPA) properties (the
maximum OPA wavelengths kmax, transition natures and
oscillator strengths f) of studied molecules and the avail-
able experimental values. As shown in Table 2, our calcu-
lated results deviate slightly from the experimental values.
This is understandable, because the experimental values
were measured in the polar solvent and were affected by
the counter ions in solution, while our calculations were
carried out in vacuum. Upon coordination, the acceptor
strength of the imine moiety is enhanced, resulting in a
red shift of the maximum OPA wavelength.
LUMO are mainly positioned on the other strand. By close
look at the contour diagrams we can say that the HOMO–
LUMO transition is nothing but the p ! p* transition of
the chromophore. In case of Zn complex, the HOMOꢁ1
and LUMO are distributed in the strand where trans geom-
etry is observed around the imine bond. Here, HOMO and
LUMO+1 are mainly positioned in the cis imine based
strand. Thus, the HOMO–LUMO transition can be said
to be the reflection of intra-ligand charge transfer (ILCT).
From the viewpoint of the HOMO of ligand as well as Zn
complex it can be concluded that the p conjugation from
one strand of the ligand to the other strand is sufficiently
reduced in the metal complex. This is due to the involve-
ment of p orbitals of the sulfur atoms in the ligand but this
p cloud around the sulfur atoms are almost negligible in the
Zn complex as it forms a weak r bond with the metal.
The two-photon absorption properties of conjugated sys-
tems can be described by looking at the transition dipole
moments [2]. For this, a simple model can be invoked con-
taining three electronic states: the ground state (g), the low-
est excited state (e), and a higher lying two-photon state (e0).
The two-photon absorption cross-section depends on the
transition dipole moments (M) between states g and e, and
between states e and e0. After determining the transition
moment Mge for the g ! e transition through TD-DFT
0
method, and assuming Cge ꢄ 0:1 eV (a value consistent with
the bandwidth of the two-photon spectra for many chro-
0
mophores), the excited-state transition moment Mee can
be estimated, based on experimental data, via the available
literature method [14]. In this context, the changes in the
magnitude of two-photon absorption cross-section of the
ligand on cation binding can be attributed to a moderate
reduction of the excited-state transition dipole moment
0
ðMee Þ from 12.80 to 5.46 D (Table 2) while Mge is calculated
Generally the frontier molecular orbitals, especially
HOMOꢁ1, HOMO, LUMO and LUMO+1, have impor-
tant contributions to electronic transition during two-pho-
ton excitation. In Fig. 5, the contours of the non-
degenerate HOMOꢁ1, HOMO, LUMO and LUMO+1
for the ligand and its Zn complex are shown. In the case
of free ligand HOMOꢁ1 and LUMO+1 are mainly distrib-
uted in the same strand of the ligand while HOMO and
to be almost opposite for the unbound and bound forms of
the chromophore (6.17 and 10.95 D, respectively). These
changes in the transitional dipole moment values reflect in
the reducing d value of the Zn complex.
4. Conclusion
In conclusion, we have reported the synthesis of a new
ion responsive molecule in the format donor–acceptor–
acceptor–donor that exhibits large TPA cross-section mea-
sured in femtosecond regime. Its TPA activity reduces in
presence of selective metal ions to different extents in con-
sistent with their stability constant values. In case of Zn(II)
and Mg(II) ions, the d value decreases, respectively, by 31%
and 28%. Interestingly, this ligand shows enhancement of
fluorescence upon metal binding where the quantum yield
increases up to 25 times compared to metal-free L. Fluoro-
phores with large two-photon absorption cross-section are
potentially important in TPLSM applications. Theoretical
calculations carried out at the B3LYP/6-31G* level in den-
sity functional theory (DFT) formalism show that upon
metal binding, the p conjugation diminishes significantly
as also the transition moment leading to lower TPA
Table 2
Theoretical OPA properties and transition dipole moments for studied
compounds
Compound
L
L–Zn2+
kabs (nm) (experiment)
kmax (nm) (theoretical)
Transition nature
358
450
464
362, 354, 348
HOMOꢁ2 ! LUMO+1
HOMOꢁ1 ! LUMO+1
HOMO ! LUMO+1
HOMOꢁ1 ! LUMO
HOMO ! LUMO
0.2583, 0.0376
3.26
HOMO ! LUMO
Oscillator strength (f)
Ege (eV)
0.0025
3.11
Mge (D)
6.17
12.80
10.95
5.46
0
Mee (D)
d (GM)
2300
1580

S. Das et al. / Journal of Organometallic Chemistry 692 (2007) 4969–4977
4977
[6] W.H. Lee, H. Lee, J.A. Kim, J.-H. Choi, M. Cho, S.-J. Jeon, B.R.
Cho, J. Am. Chem. Soc. 123 (2001) 10658.
cross-section in the metal complexes. Further studies on
similar systems are in progress in our laboratory.
[7] M. Drobizhev, A. Karotki, A. Rebane, Opt. Lett. 26 (2001) 1081.
[8] (a) Q. Zheng, G.S. He, P.N. Prasad, J. Mater. Chem. 15 (2005) 579;
(b) S.K. Hurst, M.G. Humphrey, T. Isoshima, K. Wostyn, I.
Asselberghs, K. Clays, A. Persoons, M. Samoc, B. Luther-Davices,
Organometallics 21 (2002) 2024.
Acknowledgments
Financial support received from DRDO & DST, New
Delhi, India (to P.K.B.) is gratefully acknowledged. D.G.
thanks DST, MCIT (India) and international SRF program
of Welcome Trust (UK) for the financial grant. K.K.S.
thanks CSIR for SRF and A.N. thanks UGC for SRF.
[9] S. Das, A. Nag, D. Goswami, P.K. Bharadwaj, J. Am. Chem. Soc.
128 (2006) 402.
[10] X.M. Wang, D. Wang, G.Y. Zhou, W.T. Yu, Y.F. Zhou, Q. Fang,
M.H. Jiang, J. Mater. Chem. 11 (2001) 1600.
˚
[11] L. Frediani, Z. Rinkevicius, H. Agren, J. Chem. Phys. 122 (2005)
244104.
[12] (a) Y. Suzuki, H. Komatsu, T. Ikeda, N. Saito, S. Araki, D. Citterio,
H. Hisamoto, Y. Kitamura, T. Kubota, J. Nakagawa, K. Oka, K.
Suzuki, Anal. Chem. 74 (2002) 1423;
(b) A. Takahashi, P. Camacho, J.D. Lechleiter, B. Herman, Physiol.
Rev. 79 (1999) 1089.
[13] (a) C. Xu, R.M. Williams, W. Zipfel, W.W. Webb, Bioimaging 4
(1996) 198;
(b) C. Xu, W.W. Webb, J. Opt. Soc. Am. B 13 (1996) 481.
[14] S.J.K. Pond, O. Tsutsumi, M. Rumi, O. Kwon, E. Zojer, J.L. Bredas,
S.R. Marder, J.W. Perry, J. Am. Chem. Soc. 126 (2004) 9291.
[15] S.R. Meech, D. Phillips, J. Photochem. 23 (1983) 193.
[16] M. Sheik-Bahaei, A.A. Said, T. Wei, D.J. Hagan, E.W. Van Stryland,
IEEE J. Quantum Electron. 26 (1990) 760.
Appendix A. Supplementary material
1
Characterization of the ligands L: ESI-MS, H NMR,
and Open-aperture Z-scan traces of reference Rhodamine
6G. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
2007.07.013.
References
[1] (a) R.H. Kohler, J. Cao, W.R. Zipfel, W.W. Webb, M.R. Hansen,
Science 276 (1997) 2039;
[17] (a) D. Kim, A. Osuka, M. Shigeiwa, J. Chem. Phys. A 109 (2005) 2996;
(b) H. Rath, J. Sankar, V. Prabhuraja, T.K. Chandrasekhar, A. Nag,
D. Goswami, J. Am. Chem. Soc. 127 (2005) 11608.
[18] P. Sengupta, J. Balaji, S. Banerjee, R. Philip, G. Ravindra Kumar, S.
Maiti, J. Chem. Phys. 112 (2000) 9201.
[19] (a) J. Bourson, J. Pouget, B. Valeur, J. Phys. Chem. 97 (1993) 4552;
(b) S. Banthia, A. Samanta, J. Phys. Chem. B 106 (2002) 5572.
[20] S. Fery-Forgues, M.T. Le Bries, J.P. Guette´, B. Valeur, J. Phys.
Chem. 92 (1988) 6233.
(b) J.D. Bhawalkar, G.S. He, P.N. Prasad, Opt. Commun. 119 (1995)
587;
(c) J.H. Strickler, W.W. Webb, Adv. Mater. 5 (1993) 479;
(d) Q. Zheng, G.S. He, T.-C. Lin, P.N. Prasad, J. Mater. Chem. 13
(2003) 2499;
(e) J.D. Bhawalkar, N.D. Kumar, C.F. Zhao, P.N. Prasad, J. Clin.
Laser Med. Surg. 15 (1997) 201.
[2] M. Albota, D. Beljonne, J.-L. Bredas, J.E. Ehrlich, J.-Y. Fu, A.A.
Heikal, S.E. Hess, T. Kogej, M.D. Levin, S.R. Marder, D. McCord-
Maughon, J.W. Perry, H. Ro¨ckel, M. Rumi, G. Subramaniam,
W.W. Webb, X.-L. Wu, C. Xu, Science 281 (1998) 1653.
[3] (a) W.J. Yang, D.Y. Kim, M.-Y. Jeong, H.M. Kim, Y.K. Lee, X.
Fang, S.-J. Jeon, B.R. Cho, Chem. Eur. J. 11 (2005) 4191;
(b) S.J. Chung, K.-S. Kim, T.H. Lin, G.S. He, J. Swiatkiewicz, P.N.
Prasad, J. Phys. Chem. B 103 (1999) 10741;
[21] P.R. Bangal, S. Chakravorti, J. Photochem. Photobiol. A 116 (1998)
191.
[22] (a) C.H. Zhang, Z.B. Chen, Y.B. Jiang, Spectrochim. Acta A 60
(2004) 2729;
(b) W. Rettig, Angew. Chem., Int. Ed. Engl. 25 (1986) 971.
[23] J.-S. Wu, W.-M. Liu, X.-Q. Zhuang, F. Wang, P.-F. Wang, S.L.
Tao, X.H. Zhang, S.K. Wu, S.T. Lee, Org. Lett. 9 (2007) 33.
[24] (a) S.J. Chung, K.-S. Kim, T.H. Lin, G.S. He, J. Swiatkiewicz, P.N.
Prasad, J. Phys. Chem. B 103 (1999) 10741;
(b) J. Yoo, S.K. Yang, M.Y. Jeong, H.C. Ahn, S.J. Jeon, B.R. Cho,
Org. Lett. 5 (2003) 645.
[25] M.J. Frish et al., Gaussian 03 Program (Revision C. 02), Gaussian
Inc., Wallingford, CT, 2004.
[26] (a) B. Kersting, Angew. Chem. Int. Ed. 40 (2001) 3987;
(b) D.C. Haagenson, D.F. Moser, L. Stahl, Inorg. Chem. 41 (2002)
1245.
(c) J. Yoo, S.K. Yang, M.-Y. Jeong, H.C. Ahn, S.-J. Jeon, B.R.
Cho, Org. Lett. 5 (2003) 645;
(d) L. Antonov, K. Kamada, K. Ohta, F.S. Kamounah, Phys. Chem.
Chem. Phys. 5 (2003) 1193.
[4] K.D. Belfield, K.J. Schafer, W. Mourad, B.A. Reinhardt, J. Org.
Chem. 65 (2000) 4475.
[5] O. Mongin, L. Porres, L. Moreaux, J. Mertz, M. Blanchard-Desce,
Org. Lett. 4 (2002) 719.