New NLO active cyclopalladated chromophores in main-chain polymers

Article abstract of DOI:10.1016/S0020-1693(03)00495-X

Four new OH-functionalized orthopalladated complexes bearing an electron-donor and an acceptor group in a trans-like arrangement with respect to the metal have been prepared and characterized. The ligands are Schiff bases, bound as C,N- or N,O-chelating moieties. Four new polyesters were obtained by polycondensation from the monomeric complexes and pentyloxyterephthaloyl chloride. Two "model" complexes, each one related to one monomer, were also synthetized, in order to obtain further and more detailed characterization data, and in particular to perform measurements of μβ.

Full text of DOI:10.1016/S0020-1693(03)00495-X

Inorganica Chimica Acta 357 (2004) 548–556
www.elsevier.com/locate/ica
New NLO active cyclopalladated chromophores
in main-chain polymers q
a
F. Cariati , U. Caruso , R. Centore , A. De Maria , M. Fusco , B. Panunzi
b
b
b
b
c,
*
,
b
A. Roviello , A. Tuzi
b
a
Universitaꢀ degli Studi di Milano, Dipartimento di Chimica Inorganica, Metallorganica ed Analitica, via Venezian 21, 20133 Milan, Italy
Universitaꢀ degli Studi di Napoli ‘‘Federico II’’, Dipartimento di Chimica, Complesso Universitario di Monte S. Angelo, 80126 Naples, Italy
Universitaꢀ degli Studi di Napoli ‘‘Federico II’’, Dipartimento di Scienza degli Alimenti, via Universitaꢀ 100, 80055 Portici (Naples), Italy
b
c
Received 24 July 2003; accepted 2 August 2003
Abstract
Four new OH-functionalized orthopalladated complexes bearing an electron-donor and an acceptor group in a trans-like ar-
rangement with respect to the metal have been prepared and characterized. The ligands are Schiff bases, bound as C,N- or N,
O-chelating moieties. Four new polyesters were obtained by polycondensation from the monomeric complexes and pentyloxyte-
rephthaloyl chloride. Two ‘‘model’’ complexes, each one related to one monomer, were also synthetized, in order to obtain further
and more detailed characterization data, and in particular to perform measurements of lb.
Ó 2003 Elsevier B.V. All rights reserved.
Keywords: Cyclopalladated complexes; Nonlinear optics; Main-chain polymers
1. Introduction
non-centrosymmetric coordination compounds, with
fair second order NLO response. Variation in molecular
design of cyclopalladated complexes was used to obtain
information on the correlations between structure and
NLO properties. In this work, we describe further de-
velopments in this area.
Four metal complexes with the general formula re-
ported in Scheme 1 have been prepared, characterized
and successively used as monomers in the preparation of
a series of polyesters.
It is to note that the above complexes are organo-
metal derivatives. Organometallics can form polymers
or can be included in polymers, just as simple organic
molecules, and the use of NLO active monomers is ex-
pected to produce NLO active materials. The synthesis
of Pd(II) based organometal side-chain NLO active
polyacrylates has already been described [4]. A variety
of examples of metallic main-chain polymers, with the
second order NLO active units in the polymeric back-
bone, have been reported in the literature, but most of
them contain non-organometal coordinated groups, as
in the case of bis(salicylaldiminate) derivatives of cop-
per, nickel and cobalt [8] or zinc [9]. Very few examples
A significant part of the research in the field of
nonlinear optics is being focused on the design of new
chromophoric coordination compounds [1–3]. Metal
complexes can possess a large NLO response due to the
attainment of low-energy excited states with dipole mo-
ment significantly different from the ground-state dipole
moment. Most of these excited states are involved in
metal-to-ligand and/or ligand-to-metal charge-transfer
and have large oscillator strengths. In these cases, charge
transfer is expected to provide a substantial contribution
to the second order molecular hyperpolarizability [3].
The NLO properties of several types of transition
metal complexes have been previously examined by the
authors [4–6]. In particular, a series of cyclopalladated
derivatives of Schiff bases was prepared, developing a
synthetic strategy to obtain thermally stable and soluble
q Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/S0020-1693(03)00495-X.
* Corresponding author. Tel.: +39081-674170; fax: +39081674367.
E-mail address: bpanunzi@chemistry.unina.it (B. Panunzi).
0020-1693/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0020-1693(03)00495-X

F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
549
content of complexes and polymers and their decom-
position temperature. X-ray diffraction patterns were
recorded at room temperature by a flat film camera
using Ni-filtered CuKa radiation on polymer samples.
Single crystal X-ray data were collected on a Nonius
MACH3 automated single crystal diffractometer using
graphite monochromated MoKa radiation (k ¼ 0:71069
1
ꢁ
A). H NMR spectra were recorded in CDCl3, Me2SO–
d6 (DMSO) or acetone–d6 at 200 MHz using a Varian
Scheme 1. General formulas of the monomeric complexes.
Gemini 200 spectrometer. Inherent viscosities (ginh
)
of main-chain polymers with metal–carbon complexes
as chromophores are known, such as metallocene or
metal-containing acetylene polymers for third-order
NLO activities [2]. This could be ascribed to the higher
stability displayed by ‘‘classical’’ coordination mono-
mers with respect to organometallic species in the con-
ditions required for polycondensation. A modification
of a synthetic route leading to stable and soluble mate-
rials previously [10] followed by us in order to get
interfacial polycondensation of cyclopalladated mono-
mers was used in this work to prepare four main-chain
polymers. Direct polycondensation of OH-functional-
ized cyclopalladated NLO active monomers with pent-
yloxyterephthaloyl chloride was achieved. In all the
monomeric metal complexes, one Schiff base acts as
C,N-chelate and a second one as N,O-chelate. These
bases contain an electron-donor group or an electron-
acceptor group and the final complexes contain both an
electron-donor and an acceptor group in a trans-like
arrangement with respect to the metal. More particu-
larly, one group is in para or meta position with respect
to the Pd–C bond, while the other one is in para or meta
position with respect to the Pd–O bond. Two models
(Mod1–2, Scheme 4) of the monomers, in which the OH
functions are protected, have also been synthetized in
order to obtain further and more detailed characteriza-
tion data, especially of structural type, and lb measured
by electric field induced second harmonic (EFISH)
technique. The new models Mod3–4 correspond to the
complexes with the donor and acceptor groups in para
position. Herein their features are compared with those
of Mod1–2 which have been reported previously [7] and
have the donor and acceptor groups in meta position.
were measured in 1,1,2,2,-tetrachloroethane or in di-
chloromethane solutions at 25.0 °C using a Ubbelohde
viscometer. UV–Vis absorption spectra were recorded
at room temperature utilizing a Perkin–Elmer Lambda
7 spectrometer. Infrared spectra were recorded on
a JASCO FT-IR 430 spectrometer in chloroform
solutions.
2.2. Materials
The reagents 4-aminophenol, 4-(dimethylamino)
benzaldehyde, 4-nitrobenzaldheyde, 2-hydroxy-4-(dieth-
ylamino)benzaldehyde, 3-nitrobenzaldheyde, 3-(dimeth-
ylamino)benzaldehyde,
4-nitroaniline,
anisidine,
salicylaldehyde, and 2-hydroxy-5-nitrobenzaldehyde
were commercially available (Aldrich). The compound
4-nitrosalicylaldehyde was obtained by adaptation of
literature methods [11]. The compound 2-hydroxy-5-[(4-
nitrophenyl)azo]benzaldehyde, though commercially
available, was obtained from 4-nitroaniline and salicyl-
aldehyde by standard methods [12]. The compound 4-
aminophenyl benzoate was synthetized according to the
literature [13].
The imines I1, I2, I5, I6 and I6m (see formulas in
Scheme 2), though commercially available, were syn-
thetized from the correspondent amines and aldehydes,
by standard methods [12].
The imine I7m in Scheme 2 was obtained from stoi-
chiometric amounts of 2-hydroxy-5-[(4-nitrophenyl)azo]
benzaldehyde and anisidine in boiling dimethyl sulfox-
ide. Pentyloxyterephthaloyl chloride was synthetized
from hydroxyterephthalic acid by standard methods
[12].
2.3. Synthesis of 3-(dimethylamino)benzaldehyde
2. Experimental
The compound was obtained by a reductive methyl-
ation, adapting a procedure reported in the literature
[14].
To 7.50 g (0.050 mol) of 3-nitrobenzaldehyde dissolved
in 120 ml of ethanol 33 ml of aqueous solution of form-
aldehyde (40% by weight) and 10 ml of Nickel Raney
suspension (containing about 4 g of Nickel) were added.
The suspension was stirred for 5 h under hydrogen pres-
sure (at about 2 atm). After this time, the suspension was
filtered, concentrated in vacuum, and the crude product
2.1. Methods
Optical observations on low molecular weight species
and polymers were performed by use of a Zeiss polar-
izing microscope equipped with a Mettler FP5 micro-
furnace. Phase transition temperatures and enthalpies
were measured with a Perkin–Elmer DSC-7 apparatus
at a scanning rate of 10 K minꢀ1, under a dry nitrogen
flow. Thermogravimetric analysis was performed using a
Mettler TG50 apparatus to evaluate the palladium

550
F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
Scheme 2. Formulas of the organic ligands.
was purified by distillation (T ¼ 100 °C, P ¼ 1:5 mmHg).
Yield ¼ 60%.
1H NMR (CDCl3): d 3.0 (6H, s); d 6.7 (1H, m); d 6.9
(1H, d); d 7.1 (1H, s); d 7.3 (1H, d); d 9.9 (1H, s).
2.4. Synthesis of Schiff bases (see formulas in Scheme 2)
The imines I3 and I3m were obtained in about 10 min
from stoichiometric amounts of 3-(dimethyla-
mino)benzaldehyde and, respectively, 4-aminophenol or
4-aminophenyl benzoate in boiling ethanol. I3m was re-
crystallized from boiling petroleum benzin. M.p. ¼ 184
°C, yield ¼ 80% (I3). M.p. ¼ 95.3 °C; yield ¼ 45% (I3m).
1H NMR (DMSO) for I3: d 3.0 (6H, s); d 6.7 (2H, d);
d 6.8 (1H, dd); d 7.2 (5H, m); d 8.5 (1H, s); d 9.4 (1H, s).
1H NMR (DMSO) for I3m: d 2.9 (6H, s); d 6.8 (1H,
dd); d 7.3 (6H, m); d 7.6 (2H, m); d 7.7 (1H, m); d 8.1
(2H, d); d 8.6 (1H, s).
The imine I4 was produced in situ in the synthesis of
M4. On a sample obtained for characterization from 2-
hydroxy-4-nitrobenzaldehyde in ethanol, it was found:
m.p. ¼ 251 °C, yield ¼ 68%. 1H NMR (DMSO): d 7.0 (2H,
d); d 7.2 (2H, d); d 7.7 (3H, m); d 9.0 (1H, s); d 9.9 (1H, s); d
13.9 (1H, s);
Scheme 3. Formulas of the dinuclear complexes.
dium(II) acetate with the appropriate ligands following a
previously described general procedure [15].
The complexes D1–3 were obtained, respectively, from
The imine I7 was synthetized by the reaction of stoi-
chiometric amounts of 2-hydroxy-5-[(4-nitrophenyl)azo]
benzaldehyde and 4-aminophenol in boiling N,N-dim-
ethylformamide for about 10 min. M.p. ¼ 297.8 °C;
yield ¼ 91%;. 1H NMR (DMSO) for I7: d 6.9 (2H, d); d 7.1
(1H, d); d 7.4 (2H, d); d 8.0 (4H, d); d 8.3 (1 H, s); d 8.4 (2H,
d); d 9.2 (1H, s); d 9.9 (1H, s);
I1–3, and D3m was obtained from I3m. As an example, the
preparation of D1 is described.
A mixture of 0.53 g (2.18 mmol) of I1 and 0.49 g of
palladium (II) acetate (2.18 mmol) in 20 ml of acetic acid
was gently boiled under nitrogen for ca. 30 min. On
cooling at room temperature a green-yellow microcrys-
talline solid precipitated. The solid was recovered by fil-
tration and washed with ethanol. In the case of D2 the
reaction mixture was kept at reflux for 3 h. Yields:
D1 ¼ 62; D2 ¼ 75; D3 ¼ 42; D3m ¼ 44%.
2.5. Synthesis of the dinuclear complexes
The dinuclear complexes D1–3 and D3m (see Scheme 3
for the formulas) were obtained by reaction of palla-
The dinuclear species were employed as crude products
in the further reactions. Palladium contents calculated by

F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
551
TGA measurements and thermodynamic data are
reported in Table 1 1H NMR data gave evidence of
the fairly high purity of the products.
(1:3) as the solvent and stirring the reaction mixture at
room temperature for 2 h. After this time, the mixture
was filtered and the solution was concentrated in vacuo
until a brown-red solid precipitated.
1H NMR (DMSO): for D1: d 2.1 (3H, s); d 2.9 (6H, s);
d 5.7 (1H, d); d 6.3 (1H, d); d 6.5 (4H, dd); d 7.1 (1H, d);
d 7.5 (1H, s); d 9.5 (1H, s). For D2: d 1.9 (3H, s); d 6.4
(2H, d); d 6.6 (2H, d); d 7.0 (1H, s); d 7.5 (1 H, d); d 7.9
(1H, dd); d 8.2 (1H, s); d 9.7 (1H, s). For D3: d 1.9 (3H,
s); d 2.9 (6H, s); d 6.6 (2H, d); d 6.8 (2H, dd); d 7.2 (2H,
m); d 7.4 (1H, s); d 8.3 (1H, s); d 9.6 (1H, s). For D3m: d
1.8 (3H, s); d 2.9 (6H, s); d 6.4 (1H, d);d 6.5 (1H, d); d 6.7
(1H, d); d 6.9 (1 H, s); d 7.0 (2H, dd); d 7.4 (2H, dd); d
7.6 (2H, m); d 7.7 (1H, d); d 8.1 (2H, d).
Yields: M1 ¼ 67; M2 ¼ 63; M3 ¼ 52; M4 ¼ 45%.
Thermodynamic data and Pd content (%) for the
1
mononuclear species are reported in Table 1. H NMR
data confirm a good purity degree.
1H NMR (acetone): for M1: d 2.6 (6H, s); d 5.4 (1H,
d); d 6.2 (1H, dd); d 6.9 (4H, m); d 7.2 (3H, m); d 7.3
(2H, d); d 7.6 (3H, m); d 8.0 (1H, s); d 8.4 (1H, s);d 8.5
(1H, s); d 8.7 (1H, d). For M2 (DMSO): d 1.2 (3H, t);d
3.4 (4H, dd); d 5.8 (1H, d); d 6.1 (1H, dd); d 6.8 (3H, m);
d 6.9 (2H, d); d 7.1 (2H, d); d 7.4 (5H, m); d 7.5 (1H, d); d
7.7 (1H, d); d 7.8 (1H, s); d 8.6 (2H, s). For M3 (DMSO):
d 2.7 (6H, s); d 5.5 (1H, d); d 6.1 (1H, d); d 6.3 (1H, d); d
6.8 (4H, m); d 7.2 (3H, m); d 7.5 (2H, d); d 8.0 (1H, dd);
d 8.2 (1H, s);d 8.4 (1H, s); d 8.5 (1H, d); d 9.7 (2H, s).
For M4 (DMSO): d 2.9 (6H, s); d 5.6 (1H, d); d 6.1 (1H,
d); d 6.5 (1H, d); d 6.9 (5H, m); d 7.3 (2H, d); d 7.6 (2H,
d); d 7.9 (3H, m); d 8.3 (5H, m); d 9.7 (2H, s).
2.6. Synthesis of the mononuclear complexes
The preparation of the four mononuclear complexes
M1–4 (formulas in Scheme 4) was carried out by the
same general procedure.
The monomers M1 and M2 were obtained, respec-
tively, from the dinuclear complex D1 by reaction with
the O,N-ligand I4 and from D2 by reaction with I5. The
monomers M3 and M4 were obtained from the same
dinuclear complex D3 by reaction, respectively, with the
O,N-ligands I6 and I7.
In a typical preparation of M3 1.10 g (4.0 mmol) of I6
was dissolved in 20 ml of N,N-dimethylformamide (only
in this case of M1 the O,N-ligand was synthetized in
situ). To this solution were added 1.5 g of sodium ace-
tate, 1.0 g of potassium carbonate and finally 2.00 g of
D3 (2.0 mmol). The suspension was stirred for about 30
min at 40 °C, while the color faded. After this time the
mixture was filtered and poured in about 100 ml of
ethanol/water (1:3) with 0.5 g of sodium acetate. A red
precipitate was obtained, which was washed with water
and dried at 120 °C. For M1 the reaction time was about
1 h at room temperature.
2.7. Synthesis of the model complexes
The model complexes Mod3–4 (see formulas in Scheme
4) were synthetized by the same general procedure. Here is
given a typical preparation of Mod3.
To a solution of I6m, 0.11 g (0.4 mmol) in 8 ml of N; N-
dimethylformamide were added 0.15 g of sodium acetate,
0.10 g of potassium carbonate and finally 0.20 g of D3m
(0.2 mmol). The suspension was stirred at room temper-
ature for about 20 min, while the color faded. After this
time the product was precipitated by addition of ethanol
(2–3 ml). It was recrystallized from dichloromethane/pe-
troleum benzin, obtaining red crystals. Yield 50%.
Mod4 was obtained by reaction of D3m with I7m and
crystallized from dichloromethane/petroleum benzin,
obtaining dark red crystals with 52% yield.
The monomer M2 was obtained by the same general
procedure, but employing ethanol/dichloromethane
Thermodynamic data and Pd content (%) for the
model complexes are reported in Table 1.
1H NMR data confirm a very good purity degree.
1H NMR For Mod3 (CDCl3): d 2.8 (6H, s);d 3.9 (3H,
s); d 5.5 (1H, d); d 6.2 (1H, dd); d 6.5 (1H, d); d 6.8 (1H,
d); d 6.9 (2H, d); d 7.2 (1H, s); d 7.4 (2H, d); d 7.5 (4H,
m); d 7.7 (2H, d); d 8.0 (1H, d); d 8.1 (2H, t); d 8.2 (1H,
d); d 8.3 (2H, m). For Mod4: d 2.9 (6H, s); d 3.9 (3H, s); d
5.5 (1H, d); d 6.2 (1H, dd); d 6.6 (1H, d); d 6.8 (1H, d); d
7.0 (4H, m); d 7.2 (4H, m); d 7.3 (2H, d); d 7.5 (2H, d); d
7.9 (5H, m); d 8.0 (1H, s); d 8.1 (1H, s); d 8.3 (2H, d).
Table 1
Thermodynamic data and Pd content (%) of dinuclear, monomeric and
model complexes
Td (°C)a Tm (°C)
DHm (J/g)
%Pdcalc
%Pdexp
b
c
D1
D2
275
312
207
210
283
275
220
226
26.29
26.14
26.20
20.92
17.60
16.68
17.65
15.05
14.80
12.90
28.19
25.78
26.79
20.84
17.18
16.48
17.83
15.17
15.06
12.75
D3
D3m
M1
M2
M3
2.8. Synthesis of the polymers
M4
Mod3
Mod4
285
290
241
142
32
42
The synthesis of the four polymers P1–4 (formulas in
Scheme 5) was performed by a modification of the al-
ready used procedure [10] from M1–4 and pentyloxyte-
rephthaloyl chloride.
a Decomposition temperature, calculated at 5% by weight loss.
b Calculated palladium content.
c Experimental palladium content.

552
F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
Scheme 4. Formulas of the monomeric and model complexes.
The synthesis of P1 is reported as an example. In 60
and washed first with water, then in the order with:
ethanol; twice with chloroform/heptane (1:2); ethanol
containing a small amount of sodium acetate, water,
and ethanol.
Finally, the polymer was dissolved in 1,1,2,2-tetra-
chloroethane at 70 °C (about 30% by weight of the crude
material did not dissolve and was filtered off) and pre-
cipitated in heptane. It was recovered by filtration, wa-
shed with hexane, and dried at 80 °C. Yield ¼ 69%.
For P2–4 yields are, respectively, 61, 60, and 65%.
Thermodynamic data and Pd content (%) of the poly-
mers are reported in Table 2.
ml of water solution containing 0.292 g of KOH and 0.2
g of tetrabutylammonium hydrogen sulfate 1.017 g (1.69
mmol) of M1 were dispersed. To this mixture was added
30 ml of chloroform (dried by treatment with basic
alumina) containing 0.488 g (1.69 mmol) of pent-
yloxyterephthaloyl chloride, under vigorous stirring.
The reaction proceeded under stirring for about 8 min.
About 100 ml of heptane were added to the mixture to
precipitate the polymer. This was recovered by filtration
Table 2
Thermodynamic data and Pd content (%) of the polymers
Td (°C)a
ginh (dL/g)b
%Pdcalc
%Pdsper
c
d
Tg (°C)
P1
P2
P3
P4
178
180
161
187
291
287
260
266
0.17ꢁ
13.01
12.56
13.00
11.50
13.60
13.34
13.71
11.79
0.20ꢁ
0.10ꢁꢁ
0.17ꢁꢁꢁ
a Calculated at 5% by weight loss.
b
ꢁ
Measured at 25.0 °C at 0.50 g/dl concentration using as solvent:
:
1,1,2,2-tetrachloroethane; ꢁꢁ: N; N-dimethylformamide; ꢁꢁꢁ: N-methyl-
pyrrolidone.
c Calculated palladium content.
d Experimental palladium content estimated by TGA measurements
in air.
Scheme 5. Formulas of the polymers.

F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
553
Table 4
2.9. X-ray analysis of the model complexes
ꢁ
Selected bond lengths (A) and angles (°) for Mod3 and Mod4
Mod3
Mod4
Single crystals of Mod3 and Mod4, suitable for single
crystal analysis, were obtained by evaporation from a
chloroform and chloroform/toluene solution, respec-
tively, both in the form of well-developed elongated
prisms. Absorption effects were corrected using DI-
FABS [16]; the structures were solved by direct methods
(SHELXS program of SHELX-97 package [17]), com-
pleted by difference Fourier methods and refined by the
full matrix method (SHELXL program of the same
package). Refinement was on F 2 against all independent
measured reflections; sigma weights were introduced in
the last refinement cycles. Largest peak and hole in the
Pd1–N2
Pd1–N3
Pd1–O4
Pd1–C6
N1–C3
N2–C9
2.04(1)
2.036(9)
2.059(8)
2.02(1)
1.40(2)
1.29(2)
2.02(1)
2.05(1)
2.04(1)
2.03(1)
1.38(2)
1.26(2)
N2–Pd1–C6
N2–Pd1–O4
N2–Pd1–N3
N3–Pd1–C6
N3–Pd1–O4
O4–Pd1–C6
C1–N1–C3
C2–N1–C3
C1–N1–C2
C9–N2–C10
C7–C9–N2
C27–N3–C30
81.6(5)
87.5(4)
174.7(5)
101.7(4)
89.3(4)
16.7(4)
120(1)
118(1)
113(1)
119(1)
11(1)
80.0(6)
87.8(5)
176.9(5)
103.1(6)
89.1(5)
167.2(7)
118(2)
120(2)
114(2)
121(1)
118(2)
114(1)
ꢁꢀ3
last difference Fourier were 0.85, )0.50 e A for Mod3
ꢁꢀ3
and 0.78, )0.50 e A for Mod4. In both the structures
C, N, O, Pd atoms were given anisotropic displacement
parameters. H atoms were placed in calculated positions
and refined by the riding model (Uiso ¼ Ueq of the carrier
atom). Solvent is present in both structures, a chloro-
form molecule in Mod3 and a toluene molecule in Mod4,
and this occurrence seems to be correlated to the pres-
ence of voids in the crystal packing owing to the peculiar
shape of the molecules. Some crystal, collection and
refinement data are given in Tables 3 and 4.
115(1)
N2–Pd1–O4–C32
C6–Pd1–N3–C30
Pd1–N3–C30–C31
Pd1–O4–C32–C31
C9–N2–C10–C11
C30–N3–C27–C28
C12–C13–O2–C16
C36–C35–N4–N5
N4–N5–C42–C38
)179(1)
171(1)
3(2)
162(1)
)166(1)
)9(3)
7(3)
)2(2)
)53(2)
)67(1)
)70(1)
)66(2)
65(2)
66(2)
176(2)
175(2)
2.10. NLO measurements
The molecular quadratic optical nonlinearities were
experimentally determinated on a sample of Mod3 and
Mod4 by using the EFISH technique. The set-up allows
for the determination of the scalar lb product where l is
the dipole moment and b the vector part of the qua-
dratic hyperpolarizability tensor. The measurements
Table 5
EFISH measurements data of the model complexes
lb (esu)
Mod1
Mod2
Mod3
Mod4
120 ꢂ 10ꢀ48
75 ꢂ 10ꢀ48
35 ꢂ 10ꢀ48
790 ꢂ 10ꢀ48
Table 3
Crystal, collection and refinement data for Mod3 and Mod4
Mod3
Mod4
were carried out in chloroform solutions at a funda-
mental wavelength of 1.907 lm, using a Q-switched,
mode locked Nd3þ-YAG laser with pulse durations of
15 ns at 10 Hz repetition rate, whose 1.064 lm initial
wavelength was shifted by stimulated Raman scattering
in a high-pressure hydrogen cell. EFISH measurements
data are summarized in Table 5 that also reports data
for Mod1–2 taken from [10] for better comparison.
Chemical formula
C36H30N4O6Pd ꢂ
CHCl3
840.41
C42H34N6O6Pd ꢂ
C7H8
Formula weight
Temperature (K)
Crystal system
Space group
917.29
293
monoclinic
293
triclinic
P1
P21=c
ꢁ
a (A)
9.418(4)
13.933(8)
14.832(8)
68.20(4)
80.27(4)
83.37(5)
1778(2)
2, 1.570
0.800
21.74(2)
6.407(7)
31.26(3)
90
ꢁ
b (A)
ꢁ
c (A)
a (°)
b (°)
c (°)
98.6(1)
90
3. Results and discussion
3
ꢁ
V (A )
4306(7)
4, 1.415
0.489
Z, Dx (g/cm3)
The synthetic route adopted for the preparation of
the polymers in summarized in Scheme 6, where the
groups involved in coordination, those deputed to
electron donation or withdrawal, and the hydroxylic
functions are schematically shown.
l (mmꢀ1
Data/parameters
)
6242/460
7557/524
0.0659
R1 on F (I > 3rðIÞ) 0.0680
P
jjFojꢀjFcjj
jFoj
P
R1 ¼
.

554
F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
and the acceptor in the C,N-chelate (both in meta po-
sitions), while this arrangement is reversed in M2.
In the four complexes the two polarizing groups are
in trans-like position with reference to the square planar
environment of the metal. In absence of NLO suited
features for the ligands, this arrangement should be re-
sponsible of the chromophoric activity, as expected both
for the complexes and for the polymers because of the
active role of the metal linking the electron-donor to the
acceptor group.
Within the above general arrangement some variation
has been introduced, in order to get some information
on the corresponding changes in NLO activity. First,
in M3–4 one of the polarizing groups is in para position
with respect to the metal-bonded carbon and the other
in para to the coordinated oxygen, at variance with the
analogous complexes M1–2, where the corresponding
meta positions are involved. A different NLO behavior is
expected for the two arrangements. In fact, a more ex-
tended and therefore more active [18] dipolar system
should be involved in the first case. Another variation
has been introduced by increasing the extension of the
electron delocalization in M4 with respect to M3, where
a shorter conjugated withdrawing system is present.
The molecular structure of the two new model com-
pounds Mod3 and Mod4 is reported in Figs. 1 and 2.
Scheme 6. The synthetic route adopted for the preparation of the
polymers.
3.1. Ligands and dinuclear complexes
Seven ligands (I1–7) (see Scheme 2) bearing a phenolic
group apt to afford copolymerization by condensation
were synthetized. I1–3 were suited for C,N-chelation and
I4–7 for N,O-chelation. On the ground of the two-level
model [18] these ligands are not expected to display very
significant NLO activity. In fact, no one of them shows
the typical structure of the nonlinear optically active
compounds, where both an electron-donor and an ac-
ceptor group are conjugated by a delocalized p-electron
system.
For the ligands (as well as for the dinuclear and
mononuclear complexes, see below) some derivatives
were prepared, in which the phenolic function was
substituted by an ether group. The reason for the
preparation of these ‘‘model’’ complexes is their easier
purification and characterization, which in turn com-
pletes and makes easier the characterization of the
phenolic species. Characterization data for the Schiff
base ligands are reported in Table 1. The 1H NMR data
are in good agreement with the expected patterns.
The dinuclear complexes reported in Scheme 3, are
intermediates in the preparation of the complexes used
as monomers in the successive polymerizations. They
include two phenolic functions and two equal groups
capable of strong electron inductive effect (two donor
groups in D1–2 and two withdrawing groups in D3 and
D3m) and are centrosymmetric. All are microcrystalline
solids which do not show melting before decomposition,
as shown in Table 1.
Fig. 1. Ortep view of Mod3. Anisotropic displacement ellipsoids are
drawn at 30% probability level.
3.2. Monomeric and model complexes
All the mononuclear complexes M1–4 include two
phenolic groups and bear one electron-donor and one
withdrawing group: one on the C,N-chelate, the other
one on the N,O-chelate. On comparing M1 and M2 it is
observed that the two compounds are substantially
isomers, making allowance that in this context N(CH3)2
can be considered equivalent to N(CH2CH3)2, and that
in M1 the electron donor group is in the N,O-chelate
Fig. 2. Ortep view of Mod4. Anisotropic displacement ellipsoids are
drawn at 30% probability level.

F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
555
Selected bond lengths and angles are reported in Table
3. In both compounds, the coordination around Pd atom
is substantially square planar, with standard values
for the distances from metal to coordinated atoms [19].
The Pd atom is shared between a cyclo-palladated (five
membered) planar ring and a salicylaldiminate (six
membered) ring, the latter being substantially planar, at
variance with the analogous compounds we have already
studied [7]. The whole molecular part comprised between
the dimethylamino donor group and the nitro acceptor
takes a full planar conformation which is a required
condition [18] for a significant charge transfer and,
therefore, for relevant second order NLO activity. The
geometry around the iminic N atom is planar trigonal in
both compounds, indicating an sp2 hybridization that
should favor electron donation to the adjacent phenyl
ring. In both compounds, the phenyl rings of the imine
groups are tilted with respect to the planar push–pull core
of the molecule by rotation around N3–C27 and N2–C10
and O2–C13, reasonably in order to release intramolec-
ular steric interactions.
As for NLO activity, it can be noted from Table 5
that the lb value of Mod4 is the largest one. It must be
noted that the lb value of Mod4 is comparable or even
greater in some cases than those reported for some
bis(salicylaldiminato) Ni(II) or Cu(II) coordination
complexes [20–22]. Moreover, its lb value is also higher
than the value reported for Disperse Red (lb is
580 ꢂ 10ꢀ48 esu [18]), the typical organic NLO chromo-
phore based on the 4-nitroazobenzene moiety. The
higher lb value for Mod4 could be explained on the
basis of a larger extension of the p-electron delocaliza-
tion. If so, the result gives evidence of the role of co-
ordinated metal in the charge transfer. The lb value for
Mod1 and Mod2 is a little higher than for Mod3. Mod1
and Mod3 display the same push–pull grouping, but in
different positions in the metal environment. Substan-
tially, the same observation holds if Mod2 is compared
to Mod3.
can ensure solubility and amorphism to the polymers.
NLO active chromophores are directly linked to the ri-
gid rod main-chain, while the alkoxy groups are pen-
dants of the macromolecular backbone. So, we expect
that the poling process of the polymers should involve
rotation of the entire polymer backbone, with the mo-
tion of the chromophores substantially decoupled from
those of the alkoxy side chain. Long thermal stability of
the polar alignment can be expected [24] due to the re-
stricted mobility of the chromophores below the glass
transition.
The 1H NMR spectra are in good agreement with the
expected patterns. In particular, the comparison of the
intensity of the signals related to the alkoxy protons (due
only to the comonomer) with the intensity of the aromatic
protons (due to both the comonomer and the organo-
metallic monomer) confirms the stoichiometry. The IR
spectra show the strong band characteristic for ester
group centered at 1730 cmꢀ1. The polymers are soluble,
and stable for several days at room temperature in sol-
vents such as N; N-dimethylformamide, N-methylpirr-
olidone and 1,1,2,2-tetrachloroethane. Some relevant
thermodynamic data are presented in Table 3. The pal-
ladium content of all the polymers is always close to the
content calculated on the basis of stoichiometry. The UV–
Vis transmittance spectrum was measured on thin film
samples (obtained by evaporation of N-methylpirroli-
done polymer solutions on glass slides, at about 80 °C). In
Fig. 3, the transmittance spectra of the four polymers are
shown.
As expected, the values of the maximum of transmit-
tance of the polymers match very well the values of ab-
sorbance of the model complexes. Only the polymer
containing an azobenzene group, P4, shows absorption
above 500 nm. The other ones show a large range ab-
sorption free centered around 532 nm, where it can be
recorded as the second harmonic by employing a Nd3þ
YAG laser.
-
All the model compounds show an appreciable and
comparable solvatochromic effect, measured on chlo-
roform and o-dichlorobenzene solutions. A hypsochro-
mic (blue) effect, with shift values of about 10 nm for a
difference of 1.5 D between the dipole moment of the
solvents used was also observed.
3.3. Polymers
Four polyesters P1–4 were obtained from the NLO
active monomeric metal complexes. The use of NLO
active monomers seems to be a convenient strategy to
produce NLO active main-chain organometallic poly-
mers. In fact, rigid rod-like polymers substituted with
flexible side-chains can represent a suitable approach to
processable NLO materials [23]. Thus, insertion of alk-
oxy side chains in the rigid structure of the main chain
Fig. 3. UV–Vis transmittance spectra of thin film samples of the
polymers.

556
F. Cariati et al. / Inorganica Chimica Acta 357 (2004) 548–556
Acknowledgements
The research has been supported by MIUR of Italy.
NMR spectra and X-ray data were recorded at Centro
Interdipartimentale di Metodologie Chimico-Fisiche of
ꢀ
the Universita di Napoli ‘‘Federico II’’.
References
Fig. 4. DSC curve for a sample of P4.
[1] I.R. Whittall, A.M. McDonagh, M.G. Humprey, M. Samoc,
Organometallic Complexes in Nonlinear Optics I: Second-Order
Nonlinearities, Advances in Organometallic Chemistry, vol. 42,
1998.
Thin films of the polymers, obtained by spin-coating of
[2] N.J. Long, Angew. Chem. Int. Ed. Engl. 34 (1995) 21.
[3] S. Di Bella, Chem. Soc. Rev. 30 (6) (2001) 355.
[4] U. Caruso, A. Di Matola, B. Panunzi, A. Roviello, A. Sirigu,
Polymer 42 (2001) 3973.
1,1,2,2-tetrachloroethane polymer solutions, are amor-
phous, also after annealing for 30 min at 10 °C below the
glass transition. Optical observation, RX spectra and
DSC analysis (the only signal detected is the one related to
the glass transition, as can be seen from Fig. 4) confirmed
the absence of crystallinity.
Moreover, preliminary experiments show that second
harmonic intensity measured 10 °C above Tg on thin films
of P1 and P4 is higher than that for P2 and P3, as suggested
[5] U. Caruso, A. De Maria, B. Panunzi, A. Roviello, J. Pol. Sci., Part
A: 40 (2001) 2987.
[6] F. Cariati, U. Caruso, R. Centore, W. Marcolli, A. De Maria, B.
Panunzi, A. Roviello, A. Tuzi, Inorg. Chem. 41 (25) (2002) 6597.
[7] R. Centore, A. Roviello, A. Tuzi, B. Panunzi, Acta Cryst. C 58
(2002) m26.
ꢀ
[8] S. Di Bella, I. Fragala, Synth. Met. 115 (2000) 191.
[9] X.T. Tao, H. Suzuki, T. Watanabe, S.H. Lee, S. Miyata, H.
Sasabe, Appl. Phys. Lett. 70 (12) (1997) 1503.
by the NLO responses of the model compounds Mod1–4
.
The good thermochemical stability of all the polymers is
remarkable: as indicated by TGA measurements in ni-
trogen, there is a large temperature range, in which the
materials can be poled, from the glass transition until
decomposition.
[10] M. Ghedini, S. Morrone, U. Caruso, A. Roviello, Inorg. Chim.
Acta 292 (1999) 163.
[11] A.Roviello and co-workers, sent to Polymer.
[12] Vogel, Chimica Organica Pratica, Furniss, Hannaford, Rogers,
Smith, Tatchell, editors. II edizione CEA, 1988.
[13] K. Penttinen, Suom. Kemistilehti B 44 (1) (1971) 34.
[14] B.R. Samant, A.B. Kulkarni, Indian J. Chem. 5 (12) (1967) 652.
[15] J. Barbera, P. Espinet, E. Lalinde, M. Marcos, J.L. Serrano, Liq.
Cryst. 2 (1987) 833.
4. Conclusions
[16] N. Walker, D. Stuart, Acta Cryst. A 39 (1983) 158.
[17] G.M. Sheldrick, SHELX-97 and Shelxs97, University of Gottin-
gen, Germany, 1997.
€
The synthesis of four main-chain polyesters, contain-
ing Schiff bases cyclopalladated mononuclear complexes,
has been achieved. The polymers are amorphous, display
good chemical and thermal stability, and are fairly soluble
in standard solvents. On the basis of the NLO responses
performed on model complexes we can expect NLO ac-
tivity for the polymers [24]. In particular, the large value
of lb (comparable with those recorded for some bis-
salicylaldiminato complexes [22]) measured for Mod4 es-
tablishes the potentiality of the correspondent polymer,
with the only requirement of a laser operating at wave-
length over 1200 nm for the poling process.
[18] P.N. Prasad, D.J. Williams, Introduction to Nonlinear Optical
Effects in Molecules and Polymers, Wiley, New York, 1991.
[19] B.J. OÕKeefe, P.J. Steel, Acta Cryst. C 56 (2000) 1440s.
ꢀ
[20] D. Vitalini, P. Mineo, S. Di Bella, I. Fragala, P. Maravigna, E.
Scamporrino, Macromolecules 29 (1996) 4478.
[21] F. Aversang, P.G. Lacroix, I. Malfant, G. Lenoble, P. Cassoux,
K. Nakatani, I. Maltey-Fanton, J.A. Delaire, A. Aukauloo,
Chem. Mater. 11 (1999) 995.
[22] P.G. Lacroix, Eur. J. Inorg. Chem. (2001) 339.
[23] C. Heldmann, M. Schulze, G. Wegner, Macromolecules 29 (1996)
4686.
[24] C. Heldmann, D. Neher, H.J. Winkelhahn, G. Wegner,
Macromolecules 29 (1996) 4697.