Synthesis, crystal structures, and nonlinear optical (nlo) properties of new schiff-base nickel(ll) complexes. toward a new type of molecular switch?

Article abstract of DOI:10.1021/ic048578n

An H₂L Schiff-base ligand that was obtained from the monocondensation of diaminomaleonitrile and 4-(diethylamino)salicylaldehyde is reported together with four related nickel(ll) complexes formulated as [Ni(L)(L )] (L = MePhCHNH₂, 'PrNH₂, Py, and PPh₃). Crystal structures have been solved for H₂L, [Ni(L)(MePhCHNH ₂)], and [Ni(L)(¹PrNH₂)]. Surprisingly, the complexation process leads to the formation of a rather unusual nickel amido (-NH-Ni) bond by deprotonation of the primary amine of H₂L. A reduction of the quadratic hyperpolarizability (ss) from 38 x 10 ^-30 to 17.5 x 10^-30 cm⁵ esu^-1 is evidenced on H₂L upon metal complexation by the electric-field- induced second-harmonic (EFISH) technique. Qualitative ZINDO/SCI quantum chemical calculations indicate that, in [Ni(L)(MePhCHNH₂)], the ss orientation strongly depends on the laser wavelength. In particular, a ss rotation strictly equal to 90° could be obtained with 1.022 μm incident light on passing from [Ni(L)(MePhCHNH₂] to a hypothetical [Ni(HL)(MePhCHNH₂]+ protonated complex, thus raising the possibility for a new type of molecular switch.

Full text of DOI:10.1021/ic048578n

Inorg. Chem. 2005, 44, 1973−1982
Synthesis, Crystal Structures, and Nonlinear Optical (NLO) Properties of
New Schiff-Base Nickel(II) Complexes. Toward a New Type of Molecular
Switch?
Jean Pierre Costes, Jean Franc¸ois Lame`re, Christine Lepetit, Pascal G. Lacroix,* and
Francoise Dahan
¸
Laboratoire de Chimie de Coordination du CNRS, 205 route de Narbonne,
31077 Toulouse, France
Keitaro Nakatani
PPSM, Ecole Normale Supe´rieure de Cachan, UMR 8531, 61 AVenue du Pdt Wilson,
94235 Cachan, France
Received October 11, 2004
An H2L Schiff-base ligand that was obtained from the monocondensation of diaminomaleonitrile and 4-(diethylamino)-
salicylaldehyde is reported together with four related nickel(II) complexes formulated as [Ni(L)(L )] (L′ ) MePhCHNH2,
′
iPrNH2, Py, and PPh3). Crystal structures have been solved for H2L, [Ni(L)(MePhCHNH2)], and [Ni(L)(iPrNH2)].
Surprisingly, the complexation process leads to the formation of a rather unusual nickel amido (
deprotonation of the primary amine of H2L. A reduction of the quadratic hyperpolarizability (
17.5
(EFISH) technique. Qualitative ZINDO/SCI quantum chemical calculations indicate that, in [Ni(L)(MePhCHNH2)],
the orientation strongly depends on the laser wavelength. In particular, a rotation strictly equal to 90 could
be obtained with 1.022
m incident light on passing from [Ni(L)(MePhCHNH2] to a hypothetical [Ni(HL)(MePhCHNH2]+
protonated complex, thus raising the possibility for a new type of molecular switch.
−
NH
−
NiII) bond by
×
30
â
) from 38
10- to
×
10-30 cm5 esu-1 is evidenced on H2L upon metal complexation by the electric-field-induced second-harmonic
â
â
°
µ
Introduction
electronic behaviors, or the possibility of leading to multi-
function components in relation to the emerging concept of
molecular switches.6,7 This is especially true in nonlinear
optics, for which molecular chemistry has provided molecular
units with greater capabilities than those of the ferroelectric
crystals that are commercially available (e.g., LiNbO3 or
KH2PO4)8 by virtue of their large hyperpolarizabilities (â),9
together with an ultrafast response time, high damage
threshold, and inherent taylorability.10
For about four decades, molecular chemistry has provided
intriguing benchmark units that are used widely to test
various models that describe the electronic behaviors of
solids. Magnetism,1,2 electronic conductivity,3 and nonlinear
optics4,5 have been the most explored properties in these
derivatives. Besides an academic interest, these have also
been expected to possess unique capabilities, such as
enhanced properties with respect to the traditional (non-
molecular) materials in some cases, highly anisotropic
Over the past few years, the possibility of achieving a
switch in the quadratic nonlinear optical (NLO) response has
been considered.11 Basically, most chromophores are built
* Corresponding author. E-mail: pascal@lcc-toulouse.fr.
(1) Kahn, O. Molecular Magnetism; VCH: Weinheim, Germany, 1993.
(2) Miller, J. S.; Epstein, A. J. Angew. Chem., Int. Ed. Engl. 1994, 33,
385.
(6) Lehn, J.-M. Supramolecular Chemistry: Concepts and PerspectiVes;
VCH: Weinheim, Germany, 1995.
(3) See, for example, Special Issue on Molecular Conductors. J. Mater.
Chem. 1995, 5(10).
(7) Ward, M. D. Chem. Soc. ReV. 1995, 24, 121.
(8) Kurtz, S. K. In Laser Handbook; Arecchi, F. T., Schultz-Dubois, E.
O. Eds.; North-Holland: Amsterdam, 1972; Vol. 1, p 923.
(9) Williams, D. J. Angew. Chem., Int. Ed. Engl. 1984, 23, 690.
(10) (a) Dalton, L. R.; Harper, A. W.; Ghosn, R.; Steier, W. H.; Ziari, M.;
Fetterman, H.; Shi, H.; Mustacich, R. V.; Jen, A. K. Y.; Shea, K. J.
Chem. Mater. 1995, 7, 1060. (b) Verbiest, T.; Houbrechts, S.;
Kauranen, M.; Clays, K.; Persoons, A. J. Mater. Chem. 1997, 7, 2175.
(4) Optical Nonlinearity in Chemistry, a special issue of Chem. ReV. 1994,
94 (Jan).
(5) (a) Molecular Nonlinear Optics: Materials, Physics and DeVices; Zyss,
J., Ed.; Academic Press: Boston, 1994. (b) Nonlinear Optics of
Organic Molecules and Polymers; Nalwa, H. S., Miyata, S., Eds; CRC
Press: New York, 1997.
10.1021/ic048578n CCC: $30.25
Published on Web 02/12/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 6, 2005 1973

Costes et al.
Scheme 1
Scheme 2
up from three components: (i) an electron-rich substituent
connected through (ii) a π-electron bridge to (iii) an electron-
acceptor counterpart, leading to charge delocalization by
resonance, as exemplified in Scheme 1. Therefore, switching
the NLO response has been achieved by changing either the
donating12 or accepting13 strength of the substituents or by
changing the conjugated capabilities14 of the bridge. In all
instances, the (on f off) switch arises invariably from a
reduction of the magnitude of the quadratic hyperpolariz-
ability, defined by the molecular polarization as follows:9
µi(E) ) µ0i + RijEj + âijkEjEk + ...
(1)
In this equation, µ0 is the permanent dipole moment, R is
the polarizability, and â is the quadratic hyperpolarizability
responsible for the NLO properties. Until now, the most
convincing NLO switch seemed to be that reported by Coe
et al,12 who have used coordination chemistry to design
ruthenium(II) complexes in which drastic â reduction is
observed upon RuII f RuIII oxidation.
amine, pyridine, and Ni(OAc)2‚4H2O (Aldrich) were used as
purchased. High-grade solvents (propane-2-ol, dichloromethane,
diethyl ether, acetone, ethanol, and methanol) were used for the
syntheses of ligands and complexes. Elemental analyses were
carried out at the Laboratoire de Chimie de Coordination in
Toulouse, France, for C, H, and N. IR spectra were recorded on a
GX system 2000 Perkin-Elmer spectrophotometer. Samples were
run as KBr pellets. UV-visible spectra were recorded on a Hewlett-
Packard 8452A spectrophotometer. 1D 1H NMR spectra were
acquired at 250.13 MHz on a Bruker WM250 spectrometer. 1D
13C spectra using 1H broadband decoupling {1H}13C and gated 1H
decoupling with selective proton irradiation were recorded with a
Bruker WM250 instrument working at 62.89 MHz. Two-dimen-
sional 1H COSY experiments using standard programs and 2D
pulse-field gradient HMQC 1H-13C correlation using a PFG-HMQC
standard program were performed on a Bruker AMX400 spectrom-
eter. Chemical shifts are given in ppm versus TMS (1H and 13C)
using (CD3)2SO as a solvent.
Metal complexes exhibit charge-transfer behaviors of much
greater complexity than those of the first generation of
“push-pull” organic molecules that were investigated in
nonlinear optics,4,5 together with a variety of novel structures
and a diversity of electronic and magnetic properties. The
issue of molecular switches induced by property interplays
arises naturally in metal complexes. For instance, the design
of paramagnetic15 or magnetically coupled16 Schiff-base
metal complexes with NLO responses has been envisioned.
Nevertheless, the possibility for an interaction between both
NLO and magnetic behaviors remains a challenging issue.17
In the present contribution, we will report on the synthesis,
crystal structures, and NLO properties of new nickel(II) metal
complexes. Their molecular structure is illustrated in Scheme
2. The effect of the metal in the environment of the ligand
charge-transfer pathway will be investigated experimentally
by the electric-field-induced second-harmonic (EFISH) method
and computationally within the ZINDO method. In the last
section, the possibility of achieving molecular switches by
a â rotation instead of a â reduction will be evaluated.
Synthesis. (2Z)-2-Amino-3-({(1E)-[4-(diethylamino)-2-hydroxy-
phenyl]methylene}amino)but-2-enedinitrile (H2L): 4-(diethylami-
no)salicylaldehyde (1.93 g, 1 × 10-2 mol) and diaminomaleonitrile
(1.08 g, 2 × 10-3 mol) were mixed in methanol (80 mL) and stirred
at room temperature. The addition of 1 drop of concentrated sulfuric
acid induced a color change along with immediate precipitation.
The solid was filtered off 3 h later, washed with methanol and
diethyl ether, and air-dried. Yield: 2.66 g (93%). Anal. Calcd for
C15H17N5O: C, 63.6; H, 6.1; N, 24.7. Found: C, 63.3; H, 5.9; N,
24.4. 1H NMR (250 MHz, 20 °C, DMSO-d6): δ 1.23 (t, J ) 7 Hz,
6H, CH3), 3.50 (q, J ) 7 Hz, 4H, CH2), 6.21 (d, J ) 2 Hz, 1H,
C(3)H), 6.41 (dd, J ) 2 and 9 Hz, 1H, C(5)H), 7.47 (l, 2H, NH2),
7.72 (d, J ) 9 Hz, 1H, C(6)H), 8.46 (s, 1H, NdCH), 10.70 (l, 1H,
OH). 13C{1H} NMR (62.896 MHz, 20 °C, DMSO-d6): δ 13.4 (s,
CH3), 44.8 (s, CH2), 97.5 (s, ArC(3)H), 105.7 (s, CNH2), 105.4 (s,
ArC(5)H), 109.9 (s, ArC1), 115.2 (s, NCCNH2), 116.1 (s, NCCN),
123.4 (s, NCCN),133.3 (s, ArC(6)H), 152.6 (s, ArC(4)NMe2), 156.2
(s, HCdN), 161.5 (s, ArC(2)OH). Characteristic IR absorptions
(KBr): 3416, 3316, 2232, 2206,1634, 1596, 1575, 1516, 1475,
Experimental Section
Materials and Equipment. 4-(Diethylamino)salicylaldehyde,
diaminomaleonitrile, (S)-(-)-R-methylbenzylamine, isopropyl-
(11) Coe, B. J. Chem.sEur. J. 1999, 5, 2464.
(12) See, for example, Coe, B. J.; Houbrechts, S.; Asselberghs, I.; Persoons,
A. Angew. Chem., Int. Ed. 1999, 38, 366.
(13) See, for example, Hendrickx, E.; Clays, K.; Persoons, A.; Dehu, C.;
Bre´das, J. L. J. Am. Chem. Soc. 1995, 117, 3547.
(14) See, for example, (a) Gilat, S. L.; Kawai, S. H.; Lehn, J. M. Chem.s
Eur. J. 1995, 1, 275. (b) Nakatani, K.; Delaire, J. A. Chem. Mater.
1997, 9, 2682.
(15) Di Bella, S.; Fragala`, I.; Ledoux, I.; Marks, T. J. J. Am. Chem. Soc.
1995, 117, 9481.
(16) (a) Averseng, F.; Lacroix, P. G.; Malfant, I.; Pe´risse´, N.; Lepetit, C.;
Nakatani, K. Inorg. Chem. 2001, 40, 3797. (b) Margeat, O.; Lacroix,
P. G.; Costes, J. P.; Donnadieu, B.; Lepetit, C.; Nakatani, K. Inorg.
Chem. 2004, 43, 4743.
(17) (a) Lacroix, P. G.; Malfant, I.; Be´nard, S.; Yu, P.; Rivie`re, E.; Nakatani,
K. Chem. Mater. 2001, 13, 441. (b) Averseng, F.; Lepetit, C.; Lacroix,
P. G.; Tuchagues, J. P. Chem. Mater. 2000, 12, 2225.
1344, 1257, 1134, 1077, 822, 783, 590 cm-1
.
[Ni(L)(MePhCHNH2)]: H2L (0.28 g, 1 × 10-3 mol), (S)-(-)-
R-methylbenzylamine (0.12 g, 1 × 10-3 mol), and Ni(OAc)2‚4H2O
(0.25 g, 1 × 10-3 mol) were mixed in methanol (50 mL) and stirred.
Heating for 15 min induced precipitation of a green powder that
was filtered off, washed with methanol and diethyl ether, and air
dried. Yield: 0.36 g (78%). Anal. Calcd for C23H26N6NiO: C, 59.9;
1
H, 5.7; N, 18.2. Found: C, 60.0; H, 5.3; N, 18.2. H NMR (250
1974 Inorganic Chemistry, Vol. 44, No. 6, 2005

New Schiff-Base Ni(II) Complexes
MHz, 20 °C, DMSO-d6): δ 1.20 (t, J ) 7 Hz, 6H, CH3), 1.83 (d,
J ) 6 Hz, 3H, CH3), 3.45 (q, J ) 7 Hz, 4H, CH2), 3.74 and 3.86
(m, 1H + 1H, NH2), 4.06 (m, 1H, CHNH2), 5.26 (s, 1H, NH),
6.14 (d, J ) 2 Hz, 1H, C(3)H), 6.36 (dd, J ) 2 and 9 Hz, 1H,
C(5)H), 7.39 (t, J ) 7 Hz, 1H, ArCH(p)Am), 7.41 (d, J ) 9 Hz,
1H, C(6)H), 7.49 (t, J ) 7 Hz, 2H, ArCH(m)Am), 7.58 (d, J ) 7
Hz, 2H, ArCH(o)Am), 7.71 (s, 1H, NdCH). 13C{1H} NMR (62.896
MHz, 20 °C, DMSO-d6): δ 12.9 (s, CH3), 24.4 (s, CH3Am), 44.0
(s, CH2), 51.2 (s, CHAm), 99.0 (s, ArC(3)H), 101.6 (s, CNH), 104.5
(s, ArC(5)H), 111.3 (s, ArC1), 113.2 (s, NCCNH), 116.1 (s, NCCN),
126.7 (s, ArCHAm), 127.4 (s, ArCHAm), 128.5 (s, ArCHAm),
129.9 (s, NCCN),133.6 (s, ArC(6)H), 144.2 (s, ArCAm), 147.4 (s,
HCdN), 151.0 (s, ArC(4)NMe2), 163.7 (s, ArC(2)O). Characteristic
IR absorptions (KBr): 3325, 2222, 2167, 1611, 1583, 1557, 1504,
1403, 1369, 1353, 1267, 1246, 1217, 1141, 1063, 815, 775, 699,
Phos), 134.2 (s, ArCHPhos), 148.0 (s, HCdN), 151.9 (s, ArC(4)-
NEt2). Characteristic IR absorptions (KBr): 3366, 2218, 2178, 1609,
1583, 1565, 1514, 1496, 1435, 1351, 1269, 1242, 1213, 1140, 1097,
826, 752, 695, 528 cm-1
.
Structure Analysis and Refinement. Crystal data for H2L and
[Ni(L)(MePhCHNH2)] were collected on an Enraf-Nonius CAD4
diffractometer using graphite-monochromated Mo KR radiation (λ
) 0.71073 Å). Final unit cell parameters were obtained by means
of least-squares refinement of a set of 25 reflections in both crystal
structures. No significant standard intensity variations ((0.5%) were
observed. Semiempirical absorption corrections from psi scans were
applied to the nickel complex.18 The data for [Ni(L)(iPrNH2)] were
collected on a Stoe imaging plate diffraction system (IPDS)
equipped with an Oxford Cryosystems cooler device using a
graphite-monochromated Mo KR radiation (λ ) 0.71073 Å).
637, 528 cm-1
.
Absorption corrections were applied (Tmin ) 0.5805, Tmax )
0.7691).19
[Ni(L)(iPrNH2)]: This complex was prepared in the same manner
as the previous one with the use of isopropylamine. The solution
was set aside for 24 h, and green crystals appeared. Yield: 0.13 g
(63%). Anal. Calcd for C18H24N6NiO: C, 54.2; H, 6.1; N, 21.1.
The three crystal structures were solved by means of direct
methods using SHELXS-9720 and refined by least-squares proce-
2
dures on Fo using SHELXL-97.21 In H2L, the N(3), C(8), C(9),
1
C(10), and C(11) atoms of the diethylamino substituent were found
to be disordered. Their occupancy factors were first refined and
then kept fixed in the ratio 55/45. H atoms were introduced into
calculations with the rigid model, with Uiso equal to 1.1 times that
of the atom of attachment. Scattering factors were taken from the
International Tables for Crystallography.22 The absolute configu-
ration was determined for [Ni(L)(MePhCHNH2)] with the Flack
parameter23 using 2081 Friedel pairs. Crystallographic data are
summarized in Table 1.
Theoretical Methods. Geometries were fully optimized by DFT,
at the B3PW91/6-31G** level,24 using Gaussian 98.25 The starting
geometries were those of the present X-ray structures for H2L and
[Ni(L)(MePhCHNH2)]. [Ni(L)(MePhCHNH2)] was used as a model
for the starting geometry of [Ni(HL)(MePhCHNH2)]+. Vibrational
analysis was performed at the same level to check the attainment
of a minimum on the potential energy surface and to compute zero-
point vibrational energies. The calculated structures are in good
agreement with the crystallographic data that is available (vide
infra). The calculated structures of H2L, [Ni(L)(MePhCHNH2)], and
[NiH(L)(MePhCHNH2)]+ are given in the Supporting Information.
Found: C, 53.9; H, 5.9; N, 20.9. H NMR (250 MHz, 20 °C,
DMSO-d6): δ 1.19 (t, J ) 7 Hz, 6H, CH3), 1.48 (d, J ) 6.5 Hz,
6H, CH3), 2.98 (sept, 1H, CHNH2), 3.23 and 3.25 (s, 1H + 1H,
NH2), 3.43 (q, J ) 7 Hz, 4H, CH2), 5.25 (s, 1H, NH), 6.07 (d, J )
2 Hz, 1H, C(3)H), 6.36 (dd, J ) 2 and 9 Hz, 1H, C(5)H), 7.41 (d,
J ) 9 Hz, 1H, C(6)H), 7.74 (s, 1H, NdCH). 13C{1H} NMR (62.896
MHz, 20 °C, DMSO-d6): δ 12.9 (s, CH3), 24.1 (s, CH3Am), 43.9
(s, CH2), 43.9 (s, CHAm), 99.0 (s, ArC(3)H), 101.5 (s, CNH), 104.4
(s, ArC(5)H), 111.3 (s, ArC1), 113.1 (s, NCCNH), 115.5 (s, NCCN),
130.1 (s, NCCN),133.6 (s, ArC(6)H), 147.4 (s, HC)N), 151.0 (s,
ArC(4)NMe2), 163.7 (s, ArC(2)O). Characteristic IR absorptions
(KBr): 3300, 3243, 2218, 2168, 1612, 1583, 1558, 1500, 1403,
1373, 1352, 1264, 1240, 1215, 1192, 1138, 1076, 820, 780, 697,
640, 526 cm-1
.
[Ni(L)(Py)]: This complex was prepared in the same manner as
the previous one with the use of pyridine. It precipitated quickly
as a powder that was filtered off, washed with methanol and diethyl
ether, and air dried. Yield: 0.11 g (50%). Anal. Calcd for C20H20N6-
NiO: C, 57.3; H, 4.8; N, 20.1. Found: C, 57.0; H, 4.9; N, 19.9.
1H NMR (250 MHz, 20 °C, DMSO-d6): δ 1.18 (t, J ) 7 Hz, 6H,
CH3), 3.42 (q, J ) 7 Hz, 4H, CH2), 5.96 (l, 1H, NH), 6.02 (s, 1H,
C(3)H), 6.39 (d, J ) 9 Hz, 1H, C(5)H), 7.49 (d, J ) 9 Hz, 1H,
C(6)H), 7.76 (l, 2H, CH(m)Py), 8.14 (t, J ) 7 Hz, 1H, CH(p)Py),
9.13 (l, 2H, CH(o)Py), 10.03 (s, 1H, NdCH). 13C{1H} NMR
(62.896 MHz, 20 °C, DMSO-d6): δ 12.8 (s, CH3), 43.9 (s, CH2),
104.9 (s, ArC(5)H), 112.0 (s, ArC1), 125.9 (l, CH(m)Py), 133.4
(s, ArC(6)H), 138.6 (l, CH(p)Py), 148.4 (s, HC)N), 151.1 (l, CH-
(o)Py), 151.9 (s, ArC(4)NMe2), 164.3 (s, ArC(2)O). Characteristic
IR absorptions (KBr): 3354, 2217, 2178, 1608, 1579, 1566, 1500,
1404, 1362, 1351, 1269, 1238, 1218, 1139, 1127, 1073, 820, 766,
(18) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.,
Sect. A 1968, 24, 351.
(19) Stoe, X. SHAPE: Crystal Optimisation for Numerical Absorption
Correction, revision 1.01; Stoe & Cie: Darmstadt, Germany, 1996.
(20) Sheldrick, G. M. SHELXS-97: Program for Crystal Structure Solution;
University Of Go¨ttingen: Go¨ttingen, Germany, 1990.
(21) Sheldrick, G. M. SHELXL-97: Program for the Refinment of Crystal
Structures from Diffraction Data; University Of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(22) International Tables for Crystallography; Kluwer Academic Publish-
ers: Dordrecht, The Netherlands, 1992; Vol. C, Tables 4.2.6.8 and
6.1.1.4.
698, 600, 526 cm-1
.
(23) Flack, H. D. Acta Crystallogr., Sect. A 1983, 39, 876.
(24) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 564. (b) Perdew, J. P.;
Wang, Y. Phys. ReV. B 1992, 45, 13244.
[Ni(L)(PPh3)]: H2L (0.28 g, 1 × 10-3 mol), PPh3 (0.30 g, 1 ×
10-3 mol), and Ni(OAc)2‚4H2O (0.25 g, 1 × 10-3 mol) were mixed
in methanol (50 mL) and stirred. Heating for 15 min induced
precipitation of a black powder that was filtered off, washed with
methanol and diethyl ether, and air dried. Yield: 0.58 g (95%).
Anal. Calcd for C33H30N5NiOP: C, 65.8; H, 5.0; N, 11.6. Found:
(25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.;
Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.:
Pittsburgh, PA, 1998.
1
C, 65.9; H, 5.0; N, 11.5. H NMR (250 MHz, 20 °C, DMSO-d6):
δ 1.15 (t, J ) 7 Hz, 6H, CH3), 3.36 (q, J ) 7 Hz, 4H, CH2), 5.63
(l, 1H, NH), 6.38 (d, J ) 9 Hz, 1H, C(5)H), 7.52 (d, J ) 9 Hz, 1H,
C(6)H), 7.65 (s, 15H, CH PPh3), 7.73 (s, 1H, NdCH). 13C{1H}
NMR (62.896 MHz, 20 °C, DMSO-d6): 12.7 (s, CH3), 44.1 (s,
CH2), 105.0 (s, ArC(5)H), 128.9 (s, ArCHPhos), 130.8 (s, ArCH-
Inorganic Chemistry, Vol. 44, No. 6, 2005 1975

Costes et al.
Table 1. Crystal Data for H2L, [Ni(L)(MePhCHNH2)], and [Ni(L)(iPrNH2)]
H2L
[Ni(L)(MePhCHNH2)]
[Ni(L)(iPrNH2)]
cryst data
chemical formula
mol wt
C15H17N5O
283.34
C23H26N6NiO
461.21
C18H24N6NiO
399.14
cryst size (mm)
cryst syst
space group
a (Å)
b (Å)
c (Å)
0.5 × 0.2 × 0.1
monoclinic
P21/n
21.080(2)
7.2369(9)
9.9220(11)
91.695(9)
1513.0(3)
1.244
0.5 × 0.15 × 0.1
orthorhombic
P212121
13.9124(13)
25.076(2)
0.5 × 0.5 × 0.4
monoclinic
P21/n
10.5988(11)
14.1297(12)
13.0200(14)
100.790(13)
1915.4(3)
1.384
6.4440(9)
â (deg)
V (Å3)
2248.1(4)
1.363
0.890
F
calcd (Mg/m3)
µ (Mo KR) (mm-1
T (K)
)
0.083
293
1.032
180
293
data collection
radiation (Mo KR) (Å)
scan mode
scan range
2θ range (deg)
no. of reflns
0.71073
ω - 2θ
0.71073
ω - 2θ
0.71073
φ
0 < φ < 250.5°
2.15-26.04
2.95-54
2.95-54
measured
unique
observed [I > 2σ(I)]
3482
3303
1139
5692
4898
3669
18357
3710
3375
refinement
b
b
b
refinement on
no. of variables
H-atom treatment
R [I > 2σ(I)]a
wR2b
Fo
Fo
Fo
235
calcd
0.0320
0.0701
0.109
-0.116
283
calcd
0.0276
0.0541
0.211
-0.156
-0.012(11)
235
calcd
0.0276
0.0539
0.259
-0.324
∆Fmax (e Å-3
∆Fmin (e Å-3
Flack parameter
)
)
a R ) (∑|Fo| - |Fc|)/(∑|Fo|). b wR2 ) {[∑(w(Fo2 - Fc2)2)]/[∑(w(Fo2)2)]}1/2
.
Proton affinities (PA) were calculated as the negative of the
enthalpy changes at 0 K for the addition of a proton to the molecule
M (M(g) + H(g)+ f MH(g)+). The enthalpy changes were obtained
according to the standard definition of proton affinity26
using a nanosecond Nd:YAG pulsed (10-Hz) laser operating at λ
) 1.064 µm. The outcoming Stokes-shifted radiation at λ ) 1.907
µm that was generated by Raman effect in a hydrogen cell (1 m
long, 50 atm) was used as the fundamental beam for second-
harmonic generation (SHG). The compounds were dissolved in
dioxane at various concentrations (0 to 2 × 10-2 mol L-1). The
centrosymmetry of the solution was broken by the dipolar orienta-
tion of the chromophores with a high-voltage pulse (5 kV)
synchronized with the laser pulse. The SHG signal was selected
through a suitable interference filter, detected by a photomultiplier,
and recorded on an ultrafast Tektronic TDS 620 B oscilloscope.
With the NLO response being induced by dipolar orientation of
the chromophores, the EFISH signal is therefore a function of both
the dipole moment (µ) and âvec, the vector component of â along
the dipole moment direction. However, the ZINDO calculation
reveals that µ and âvec are roughly parallel at 1.907 µm with angle
values equal to 8 and 10° for H2L and [Ni(L)(MePhCHNH2)],
respectively. Therefore, âvec and â are assumed to be equivalent
for the present EFISH measurement. The dipole moments were
measured independently by a classic method based on the Guggen-
heim theory.31 Further details of the experimental methodology and
data analysis are reported elsewhere.32
(PA)0 K ) E(M) + ZPE(M) - E(MH+) - ZPE(MH+) (2)
where E stands for the electronic energy and ZPE stands for the
zero-point vibrational energy.
The all-valence INDO (intermediate neglect of differential
overlap) formalism,27 in connection with the sum over state (SOS)
formalism, was employed for the calculation of the electronic
spectra and the molecular hyperpolarizabilities.28 In the present
approach, the monoexcited configuration interaction (CIS) ap-
proximation was employed to describe the excited states. The lowest
100 energy transitions were chosen to undergo CI mixing. All of
the calculations were performed using the INDO/1 Hamiltonian
incorporated into the commercially available software package
ZINDO.29
NLO Measurements. The molecular hyperpolarizabilities were
investigated by the electric-field-induced second-harmonic (EFISH)
technique for H2L and [Ni(L)(MePhCHNH2)]. The principle of the
EFISH technique is reported elsewhere.30 The data were recorded
In addition, the measurement of second-harmonic generation
(SHG) intensity on [Ni(L)(MePhCHNH2)] was carried out by the
Kurtz-Perry powder technique33 using the same laser equipment.
(26) Youjung, S.; Yangsoo, K.; Yongho, K. Chem. Phys. Lett. 2001, 340,
186.
(27) (a) Zerner, M.; Loew, G.; Kirchner, R.; Mueller-Westerhoff, U. J.
Am. Chem. Soc. 1980, 102, 589. (b) Anderson, W. P.; Edwards, D.;
Zerner, M. C. Inorg. Chem. 1986, 25, 2728.
(28) Ward, J. F. ReV. Mod. Phys. 1965, 37, 1.
(29) ZINDO, release 96.0; Molecular Simulations Inc.: Cambridge, U.K.,
1996
(30) (a) Oudar, J. L. J. Chem. Phys. 1977, 67, 446. (b) Levine, B. F.; Betha,
C. G. J. Chem. Phys. 1975, 63, 2666. Levine, B. F.; Betha, C. G. J.
Chem. Phys. 1976, 65, 2429.
(31) Guggenheim, E. A. Trans. Faraday Soc. 1949, 45, 714.
(32) Malthey, I.; Delaire, J. A. Nakatani, K.; Wang, P.; Shi, X.; Wu, S.
AdV. Mater. Opt. Electron. 1996, 6, 233.
1976 Inorganic Chemistry, Vol. 44, No. 6, 2005

New Schiff-Base Ni(II) Complexes
Samples were uncalibrated powders obtained by grinding and were
put between two glass plates.
Results and Discussion
Synthesis and Characterization. H2L is obtained readily
by simple mixing of 4-(diethylamino)salicylaldehyde and
diaminomaleonitrile in alcohol in a 1/1 ratio. Sulfuric acid
has been used previously as a catalyst in the synthesis of
Schiff bases that are built up from the present weakly reacting
diaminomaleonitrile.34,35 1H and 13C NMR and elemental
analysis confirm the composition of the resulting monoimine.
IR spectroscopy provides additional evidence of the
monoimine nature of the compound with a CN stretching
mode split into two components at 2232 and 2206 cm-1, in
contrast to the single band that is observed at 2210 cm-1 in
the symmetric and related diimine ligand that was reported
previously.34
Figure 1. Asymmetric unit and atom labeling system for H2L with
ellipsoids drawn at the 50% probability level. H atoms are omitted for clarity.
The synthesis of the nickel(II) complexes requires the use
of a stoichiometric amount (or a slight excess of a coordinat-
ing amine or phosphine). For instance, no reaction is
observed by heating H2L and Ni(OAc)2‚4H2O in the presence
of a 10-fold excess of NEt3, whereas H2L, NiCl2‚6H2O, and
a primary amine in stoichiometric amounts (1/1) lead to the
desired complex with the deprotonation of H2L. This last
chemical feature is evidenced by IR spectroscopy by the
presence of a narrow peak in the 3350-3300-cm-1 domain
that is attributed to the NH stretching vibration of the
R-NH- function. This is identified clearly in [Ni(L)(py)]
in which no other NH band can interfere. Additional evidence
1
is provided by the H NMR spectra with a signal located
around 6 ppm in each complex (intensity ) 1H). The final
assignment for the 13C NMR signal required the use of 2D
NMR, especially for the diaminomaleonitrile fragments. It
may be interesting to point out that in the case of [Ni(L)-
(py)] no diaminomaleonitrile signal is observed, probably
because of the rotation of the pyridine ligand, which results
Figure 2. Asymmetric unit and atom labeling system for [Ni(L)-
(MePhCHNH2)] with ellipsoids drawn at the 50% probability level. H atoms
are omitted for clarity.
asymmetric unit. Except for the diethylamino substituent, the
molecular structure is planar with its largest deviation of
0.086(1) Å observed at N(5). The disordered nitrogen of the
diethylamino group lies +0.159(7) and -0.472(9) Å from
this mean plan for N(3) and N(3′), respectively. Because of
the quasi-planar overall geometry, a charge transfer upon
electronic transition is expected between the diethylamino
substituent and the dicyano moieties, thus providing a
potential NLO response, as discussed in the next section.
1
in a broadening of the NMR signals. Similarly, the H and
13C pyridine signals are broad, except for the CH in the para
position with respect to the nitrogen atom, for which a
sharper signal is observed.
Deprotonation of the primary amine function of the
tridentate ligand under such mild experimental conditions
is very surprising and has not been observed previously in
the complexation of other tridentate ligands that possess such
an amine function.36 This behavior emphasizes the role of
the nitrile substituents. It has been shown, from several trials
and changes in the experimental conditions, that these two
deprotonations of the H2L ligand are needed for its com-
plexation.
[Ni(L)(MePhCHNH2)] crystallizes in the noncentrosym-
metric P212121 space group with one molecular entity present
in the asymmetric unit. The molecular structure is shown in
Figure 2. Except for the two ethyl groups, the Ni(L)N(6)
fragment is planar with its largest deviation of 0.149(2) Å
observed at N(4). The nickel(II) first coordination sphere is
described in Table 2. The averaged metal-ligand bond
lengths of 1.866(2) Å are in the same range of magnitude as
that of Ni(salen) (1.850(2) Å).37 The metal atom lies in a
nearly square-planar coordination environment and is located
0.0015(3) Å from the O(1), N(1), N(2), N(6) mean plane.
Structural Studies. The X-ray molecular structure of H2L
is presented in Figure 1. The molecule crystallizes in the
P21/n space group, with one H2L entity present in the
(33) (a) Kurtz, S. K.; Perry, T. T. J. Appl. Phys. 1968, 39, 3798. (b)
Dougherty, J. P.; Kurtz, S. K. J. Appl. Crystallogr. 1976, 9, 145.
(34) Lacroix, P. G.; Di Bella, S.; Ledoux, I. Chem. Mater. 1996, 8, 541.
(35) Wo¨hrle, D.; Buttner, P. Polym. Bull. (Berlin) 1985, 13, 57.
(36) (a) Costes, J. P.; Dahan, F.; Dominguez-Vera, J. M.; Laurent, J. P.;
Ruiz, J.; Sotiropoulos, J. Inorg. Chem. 1994, 33, 3908. (b) Costes, J.
P.; Dahan, F.; Laurent, J. P. Inorg. Chem. 1991, 30, 1887.
[Ni(L)(iPrNH2)] crystallizes in the monoclinic P21/n space
group with one molecule present in the asymmetric unit. The
(37) Manfredotti, A. G.; Guastini, C. Acta Crystallogr. 1983, C39, 863.
Inorganic Chemistry, Vol. 44, No. 6, 2005 1977

Costes et al.
Table 2. Selected Bond Lengths (Å) and Angles (deg) in
[Ni(L)(MePhCHNH2)] and [Ni(L)(iPrNH2)] with esd’s in Parentheses
[Ni(L)(MePhCHNH2)]
[Ni(L)(iPrNH2)]
X-ray
X-ray
DFT
Ni-O(1)
Ni-N(1)
Ni-N(2)
Ni-N(6)
1.8295(15)
1.8647(17)
1.835(2)
1.810
1.849
1.828
1.922
1.8290(10)
1.8664(11)
1.8455(12)
1.9259(12)
1.9349(18)
O(1)-Ni-N(1)
N(1)-Ni-N(2)
N(2)-Ni-N(6)
N(6)-Ni-O(1)
O(1)-Ni-N(2)
N(1)-Ni-N(6)
95.88(7)
85.31(8)
92.84(8)
85.98(7)
178.56(8)
177.93(8)
96.1
85.5
94.2
84.1
178.3
179.7
95.73(5)
85.46(5)
93.06(5)
85.97(5)
176.57(5)
175.83(5)
molecular structure is shown in Figure 3. The solid-state
geometry of the two ethyl groups is significantly different
than what was observed in [Ni(L)(MePhCHNH2)]. The
nickel(II) first coordination sphere is described in Table 2.
As observed in [Ni(L)(MePhCHNH2)], the average metal-
ligand bond length is equal to 1.866 Å. The metal atom lies
in a nearly perfect square-planar environment and is located
0.0041(1) Å from the O(1), N(1), N(2), N(6) mean plane.
In contrast to the situation that was observed in [Ni(L)-
(MePhCHNH2)], the Ni(L)N(6) fragment appears to be
slightly bent in the present case with its largest deviation
from a mean plane being 0.323(2) observed at N(5).
However, this situation may likely result from crystal packing
because both [Ni(L)(MePhCHNH2)] and [Ni(L)(iPrNH2)]
exhibit very similar spectroscopic behaviors (vide infra).
In both nickel complexes, the complexation of H2L induces
deprotonation of the phenol function along with deprotona-
tion of the amine function, so the ligand behaves as a
dianionic four-electron donor. The resulting Ni-N amido
bonds (1.835(2) and 1.845(1) Å for [Ni(L)(MePhCHNH2)]
and [Ni(L)(iPrNH2)], respectively) are shorter than the Ni-N
imine bonds (1.865(2) and 1.866(1) Å) and shorter than the
Ni-N amine bonds (1.935(2) and 1.926(1) Å) resulting from
coordination of the primary amine ligand in the fourth
position of the equatorial plane containing the nickel ion.
Gas-phase calculated geometries are in qualitative agree-
ment with the crystal data. In particular, the molecules exhibit
planar salicyleneaminato ethylene (L2-) structures, which
suggests the possibility of long-range electron delocalization
and hence a sizable NLO response. The H2L calculated
geometry is very close to that obtained by X-ray. The largest
difference is observed at the C(12)-C(13) bond length, with
the calculated 1.380 Å value being longer than the experi-
mental value (1.348(2) Å). Similarly, the agreement is
satisfactory for [Ni(L)(MePhCHNH2]. In particular, the
coordination spheres of the nickel atom are compared in
Table 2 and are found to be very similar, with a slight
tendency for bond length shortening (<0.02 Å) in the gas-
phase structure.
Figure 3. Asymmetric unit and atom labeling system for [Ni(L)(iPrNH2)]
with ellipsoids drawn at the 50% probability level. H atoms are omitted
for clarity.
Figure 4. Electronic spectra of [Ni(L)(MePhCHNH2)] in acetone compared
to that of H2L (dotted line).
nm (ꢀ ) 27 000 mol-1 L cm-1) and 456 nm (ꢀ ) 21 800
mol-1 L cm-1). Experimental spectra and ZINDO-calculated
data are gathered in Table 3. There is a significant difference
in λmax between calculation and experiment, but the tendency
for an overestimation of the energy values by means of the
ZINDO approach has been observed previously in related
push-pull nickel(II) Schiff-base complexes.34,38 However,
the red shift and reduced intensity obtained on passing from
H2L to [Ni(L)(MePhCHNH2)] are observed at both experi-
mental and theoretical levels. Similarly, the calculated
relative intensities of the bands (oscillator strength f) are
found to be comparable to the experimental extinction
coefficient (ꢀ) for both the ligand and the metal complex.
On the basis of these observations, it will be assumed that
the electronic properties that are estimated by ZINDO can
be used for the analysis of the NLO response of both H2L
and [Ni(L)(MePhCHNH2)].
All other nickel(II) complexes exhibit similar spectra with
an absorption maximum located at 486 nm and an additional
but less intense band at 454-456 nm (Table 3). This
observation agrees fully with an NLO response dominated
by the push-pull salicyleneaminato ethylene moieties, as
anticipated from our previous investigations.34,38 Because of
Optical Spectroscopy. The experimental spectra of H2L
and [Ni(L)(MePhCHNH2)] that were recorded in acetone are
compared in Figure 4. In H2L, the spectrum is dominated
by an intense transition located at 438 nm (ꢀ ) 45 800 mol-1
L cm-1); a shoulder is present around 420 nm. The complex
exhibits red-shifted but less intense transitions located at 486
(38) Averseng, F.; Lacroix, P. G.; Malfant, I.; Lenoble, G.; Cassoux, P.;
Nakatani, K.; Maltey-Fanton, I.; Delaire, J. A.; Aukauloo, A. Chem.
Mater. 1999, 11, 995.
1978 Inorganic Chemistry, Vol. 44, No. 6, 2005

New Schiff-Base Ni(II) Complexes
Table 3. Experimental and ZINDO-Computed Optical Data for H2L and Its Related Nickel(II) Complexes
λ
max (nm)
composition of CI expansiona
exptl (ꢀ × 10-5
)
calcd (f)
H2L
438 (0.46)
362 (1.29)
279 (0.23)
0.965ø54f55
0.905ø54f6
415-425 (sh)
[Ni(L)(MePhCHNH2)]
486 (0.270)
456 (0.218)
418 (0.78)
302 (0.26)
296 (0.40)
0.966ø82f83
-0.795ø82f84
0.646ø82f83 -0.484ø82f84
[Ni(HL)(MePhCHNH2)]+
[Ni(L)(iPrNH2)]
[Ni(L)(Py)]
392 (1.28)
346 (0.13)
0.923ø82f83
0.557ø80f83 -0.651ø81f83
486 (0.333)
454 (0.255)
486 (0.344)
456 (0.267)
[Ni(L)(PPh3)]
486 (0.328)
454 (0.266)
a Orbital 54 is the HOMO and orbital 55 the LUMO for H2L. 82 is the HOMO and 83 is the LUMO for [Ni(L)(MePhCHNH2)] and [Ni(HL)(MePhCHNH2)]+.
Table 4. Experimental (EFISH) and Calculated (ZINDO) Data (µ in D,
â in 10-30 cm5 esu-1) at 1.907 µm for H2L, [Ni(L)(MePhCHNH2)], and
[Ni(HL)(MePhCHNH2)]+
exptl
calcd
(â2L+ â3L
12 16.4 (23.0 7.6)
12 10.5 (24.4 25.7)
30.3 (39.0 10.2)
µ
â
µ
â
)
H2L
8.6
11
38
17.5
[Ni(L)(MePhCHNH2)]
[Ni(HL)(MePhCHNH2)]+ data not available
these similarities, the NLO properties will be reported and
carefully analyzed for H2L and [Ni(L)(MePhCHNH2)] only.
NLO Properties. The NLO data are reported in Table 4.
Besides a tendency for reduced ZINDO-calculated values
compared to the EFISH data, it appears clearly that the nickel
complex exhibits a lower NLO response (than that of its
related ligand) at both experimental and theoretical levels.
This effect strongly contrasts with the previous report of â
enhancement by nickel complexation that was obtained in
the symmetric and related bis(salicylaldiminato) nickel
complex containing the same nitrile withdrawing units.34 It
also contrasts with the general observation by Di Bella et
al. that metal complexation in salen and salophen symmetric
ligands leads to â enhancement through an additional metal-
to-ligand charge transfer.15 The ZINDO analysis is applied
to investigate the microscopic origin of this unexpected
behavior.
Within the framework of sum-over-state (SOS) perturba-
tion theory, â is related to all excited states of the molecule28
and can be expressed as the sum of two contributions, the
so-called “two-level” (â2L) and “three-level” (â3L) terms. As
observed in most NLO chromophores, â is dominated by
â2L in H2L (Table 4), which allows us to relate the NLO
response to the optical transitions according to the following
relation:30a,39
Figure 5. Frontier orbitals with electron densities involved in the low-
lying charge-transfer transition responsible for the molecular NLO response
of H2L and [Ni(L)(MePhCHNH2)].
(pω being the energy of the incident laser beam). The
calculation reveals that a single low-lying 1 f 2 transition
contributes 54% to the total â2L. It is therefore assumed that
understanding this transition can account for a qualitative
understanding of the NLO response. Ninety-three percent
of the transition is described as the HOMO f LUMO
excitation (Table 3). These orbitals are shown in Figure 5.
A charge transfer is clearly evidenced, with 43.7% of the
electron density located on the dimethylaminophenyl moieties
at the HOMO level and 48.9% of the electron density located
on the maleonitrile counterpart at the LUMO level. This
accounts for the optical nonlinearity of the ligand unambigu-
ously.
In contrast, the NLO response of [Ni(L)(MePhCHNH2)]
appears to be slightly dominated by the â3L term (Table 4).
â3L encompasses summations of contributions in which the
ground state (g) and two excited states (n and n′) are involved
in terms of the products of transition dipole moments
rgn′rn′nrgn (with r12 ) ψ1|r|ψ2 ).28,40 Because direct informa-
tion is not currently available on excited state to excited state
(n f n′) transitions, â analysis is not possible in that case.
In most NLO chromophores, â3L scales to roughly 30-50%
3e2pfi∆µi
2mEi3
Ei4
â2L
)
×
(3)
∑
(Ei2 - (2pω)2)(Ei2 - (pω)2)
i
In this equation, fi, ∆µi, and Ei are the oscillator strength,
the difference between ground and excited-state dipole
moments, and the energy of the ith transition, respectively,
(39) Oudar, J. L.; Chemla, J. J. Chem. Phys. 1977, 66, 2664.
(40) Teng, C. C.; Garito, A. F. Phys. ReV. Lett. 1983, 50, 350.
Inorganic Chemistry, Vol. 44, No. 6, 2005 1979

Costes et al.
1D character â2L and â3L, which result from the contribution
of many i transitions, have orientations that are difficult to
predict precisely. Even if the angle seems hardly predictable,
there is no reason to assume that both vectors are strictly
opposite (angle < 180°). As far as the relative â2L/â3L value
is concerned, it is obvious that larger â2L values at higher
frequencies are related to a resonance effect, according to
the last term in eq 3. To the best of our knowledge, the issue
of how the magnitude of this effect is transposed into â3L
has not been investigated thoroughly, probably because of
the high complexity of the â3L expression.28 This would
probably remain a challenge for theorists.
Additionally, [Ni(L)(MePhCHNH2)] crystallizes in the
noncentrosymmetric P212121 space group, which leads to a
nonvanishing NLO response in the solid state. The compound
therefore exhibits an SHG efficiency roughly equal to 1.25
times that of urea.
Figure 6. â vector and its â2L and â3L components calculated for [Ni-
(L)(MePhCHNH2)] at various wavelengths. At 1.907 µm (top), â is directed
mostly in the direction of â3L (main â component at low frequency). At
1.064 µm (bottom), â is directed mostly in the direction of â2L (main â
component at high frequency), which leads to a â rotation of 45°.
as much as â2L contributions and is of opposite sign (angle
of 180°). The reasons for having an NLO response dominated
by â3L have not been investigated thoroughly, even if it has
been discussed occasionally.41 However, the deprotonation
of the primary amine of H2L leads to the introduction of a
formal but highly donating -NH- fragment. This is evi-
denced by the examination of the frontier orbitals in Figure
4. In contrast, versus H2L, the amido-maleonitrile fragment
is turned from an acceptor to an efficient donor in the metal
complex, a modification that affects both â2L and â3L
necessarily and therefore the overall â. Nevertheless, and
because of the complexity of the â3L expression, any attempt
to relate â to a set of specific transitions would probably be
somewhat irrelevant.
Another unexpected result arises from a careful examina-
tion of the ZINDO calculation. If â3L is dominant at lower
laser frequency, then the calculation reveals that â2L becomes
increasingly important as the frequency increases. Moreover,
the angle between â2L and â3L is equal to 155° (instead of
180°). Consequently, the relative increase of â2L leads to a
rotation of the overall hyperpolarization (â2L + â3L). This
behavior is illustrated in Figure 6. Increasing â values is a
well-known trend at higher frequency. However, this ten-
dency usually leaves the â direction unaffected. For instance,
the calculation carried out on H2L indicates a â rotation of
4° as the laser operating wavelength switches from 1.907 to
1.064 µm. In contrast, â undergoes a rotation of about 45°
in [Ni(L)(MePhCHNH2)] between 1.907 and 1.064 µm, an
effect that arises from the balance between â2L and â3L
(Figure 6).
Critical Evaluation of NLO Switches Obtained with
[Ni(L)(MePhCHNH2)]. Except in the intriguing case of
octupolar geometries (e.g., Td, D3h, D2d),42 the third-rank â
tensor in eq 1 can be restricted to its vectorial component in
most molecules. Therefore, the parameter of interest to
2
account for the NLO response is |â| × |E| × (cos θ)2, with
θ being the angle between the â vector and E (electric field
component of the incident light). The issue of a possible NLO
switch by â rotation is therefore naturally addressed, on the
basis of the idea that the second-harmonic signal, which is
optimized with θ ) 0° (on state), may vanish with θ ) 90°
(off state). The ZINDO analysis suggests that the â orienta-
tion is strongly dependent on the laser wavelength in the
complex and might therefore become perpendicular to that
of the ligand at a suitable wavelength. Of course, the design
of such a device and its practical use would raise many
critical issues, the first one being how to tune reversibly the
metal complexation at the molecular level. Certainly, achiev-
ing molecular motion has become an important issue in
contemporary research.43 Nature has provided many ex-
amples of such devices, each of them occurring in different
types of cells for a particular function.44 However, even if
there is no reason to think that molecular engineering will
never be able to reach this goal, it seems to be hardly
accessible for the present generation of scientists.
A more realistic strategy for a switch induced by â rotation
in molecules such as [Ni(L)(MePhCHNH2)] should imply a
chemical modification that is easier to monitor than a metal
complexation, for instance, a simple protonation. A literature
survey conducted on the Cambridge crystallographic database
(CCDB) revealed that 41 structures contain the -NH-Ni(II)
unit versus 766 based on -NH2-Ni(II). Among the 41
structures, only 20 are built up from a primary amine, but
their synthesis requires the use of a strong base.45 Therefore,
the possible protonation of the actual nickel complex could
be envisioned according to the following equation:
The ultimate understanding of the origin of the â rotation
in the nickel complex is hampered by the complexity of â3L
(see above). It would imply an understanding of (1) the
reason for having an angle of 155 instead of 180° between
â2L and â3L and (2) the reason for having â2L increasing more
rapidly than â3L at higher frequency. Concerning the first
issue, strictly 1D structures possess a unique charge-transfer
axis; therefore, the angle between vectors â2L and â3L is equal
to 180°. In the present nickel complex, because of reduced
[Ni(L)(MePhCHNH2)] + H+ h
(41) See, for example, (a) Lepetit, C.; Lacroix, P. G.; Peyrou, V.; Saccavini,
C.; Chauvin, R. J. Comput. Methods Sci. Eng. 2004, in press. (b) Di
Bella, S.; Fragala`, I. Eur. J. Inorg. Chem. 2003, 2606. (c) Di Bella,
S.; Fragala`, I. New J. Chem. 2002, 26, 285. (d) Kanis, D. R.; Lacroix,
P. G.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10089.
[Ni(HL)(MePhCHNH2)]+ (4)
The corresponding proton affinity (PA) is compared to those
1980 Inorganic Chemistry, Vol. 44, No. 6, 2005

New Schiff-Base Ni(II) Complexes
Scheme 3
Figure 7. Frontier orbitals with electron densities involved in the low-
lying charge-transfer transition responsible for the molecular NLO response
of [Ni(HL)(MePhCHNH2)]+.
of related amines in Scheme 3 to provide computational
support for significant basicity in the -NH-NiII unit. In
particular, the protonation at -NH-NiII appears to be more
favorable than that achieved on the primary MePhCHNH2
amine.
Until now, the intermolecular proton transfer-NLO prop-
erties relationship has received very little attention.46-48
Computationally, [Ni(HL)(MePhCHNH2)]+ is found to be
roughly similar to the parent [Ni(L)(MePhCHNH2)] chro-
mophore, the most noticeable modification being the elonga-
tion of the N(2)-C(13) bond upon protonation from 1.345
to 1.457 Å, consistent with the vanishing donating effect that
was observed previously at N(2). Additionally, the conju-
gated pathway between N(3) (diethylamino) and N(5) (nitrile)
is slightly modified ((0.025 Å) by a shortening of the single
bonds and an elongation of the double bonds. These
modifications lead to restoration of an important electron
delocalization between the diethylamino substituent and the
maleonitrile moieties. Therefore, the optical properties of [Ni-
(HL)(MePhCHNH2)]+ appear to be reminiscent of those of
the parent H2L ligand with an intense charge-transfer
transition located at 392 nm (f ) 1.28) and a weak shoulder
at higher energy. This leads to an enhanced â value equal to
30.3 × 10-30 cm5 esu-1 (Table 4). As observed for H2L, the
ZINDO data indicate that â2L dominates the nonlinearity
(Table 4) with an intense HOMO f LUMO based transition
Figure 8. â rotation obtained upon protonation of [Ni(L)(MePhCHNH2)]
into [Ni(HL)(MePhCHNH2)]+, expressed as a function of the laser
wavelength.
being responsible for 44% of the effect. These similarities
are illustrated further by the description of the frontier orbitals
in Figure 7. As evidenced in H2L, push-pull character arises
from electron density localized mainly on the dimethylami-
nophenyl at the HOMO level and on the maleonitrile
counterpart at the LUMO level. Therefore, â is only weakly
subjected to rotation with the laser frequency in H2L and
[Ni(HL)(MePhCHNH2)]+, in contrast to the situation that is
observed in [Ni(L)(MePhCHNH2)]. The â rotation resulting
from the protonation of [Ni(L)(MePhCHNH2)] is illustrated
in Figure 8. The angle value is equal to 23.4° at zero
frequency (λ f ∞), rises to 29.7° at λ ) 1.907 µm, and
then becomes strongly dependent on the laser wavelength.
Interestingly, the calculation reveals that the value is strictly
equal to 90° at λ ) 1.022 µm, thus providing a potential
NLO switch. This intriguing situation, which results from
the assumption of a pure vectorial â tensor, has to be
discussed with caution. Nevertheless, it suggests that, in this
unusual situation, a new type of NLO switch may be
envisioned theoretically.
(42) For a general introduction to octupolar molecules, see Zyss, J.; Ledoux,
I. Chem. ReV. 1994, 94, 77.
(43) Molecular Machines, special issue of Acc. Chem. Res. 2001, 34(6).
(44) Vale, R. D.; Milligan, R. A. Science 2000, 288, 88.
(45) See, for example, (a) Wilkes, E. N.; Hambley, T. W.; Lawrance, G.
A.; Maeder, M. Aust. J. Chem. 2000, 53, 517. (b) Meyer, F.; Hyla-
Kryspin, I.; Kaifer, E.; Kircher, P. Eur. J. Inorg. Chem. 2000, 771.
(c) Arion, V.; Wieghardt, K.; Weyhermueller, T.; Bill, E.; Leovac, V.
L.; Rufinska, A. Inorg. Chem. 1997, 36, 661. (d) Richter, R.; Hartung,
J.; Beyer, L.; Langer, V. Z. Anorg. Allg. Chem. 1993, 619, 1295.
(46) Lacroix, P. G.; Lepetit, C.; Daran, J. C. New J. Chem. 2001, 25, 451.
(47) Evans, C. C.; Bagieu-Beucher, M.; Masse, R.; Nicoud, J.-F. Chem.
Mater. 1998, 10, 847.
Of course, there would necessarily be many issues to take
into account before considering such complexes in a
perspective of real application in material chemistry, the first
one being the stability of the species in acidic media. It is
important to point out that it was not possible to isolate [Ni-
(48) Pan, F.; Wong, M. S.; Gramlich, V.; Bosshard, C.; Gu¨nther, P. J. Am.
Chem. Soc. 1996, 118, 6315.
Inorganic Chemistry, Vol. 44, No. 6, 2005 1981

Costes et al.
(HL)(MePhCHNH2)]+ by the addition of acid to the solution.
Instead, a lowering of the UV-visible band at 486 nm was
observed after a few minutes, with the appearance of NMR
signals ascribed to H2L. Therefore, [Ni(L)(MePhCHNH2)]
will certainly not be the ultimate candidate for this type of
NLO switch. At a more theoretical level, additional com-
putational approaches could be applied tentatively to verify
the ZINDO trends and predictions, for instance, within the
newly emerging time-dependent density functional theory
(TD-DFT). This method, until now, has shown limited
accuracy for describing molecular hyperpolarizabilities of
extended π systems.49,50 Nevertheless, a recent report has
pointed out that TD-DFT â values of stilbazolium-type
chromophores can agree quite closely with the experimental
data.51 Therefore, this approach will probably be more fruitful
in the future. Finally, the experimental issues of how to
achieve the proton transfer and how to measure the expected
effect cannot be avoided. Protonic conductors are well-known
transparent structures that are fully suitable for NLO ap-
plications.52,53 Some of them (e.g., those based on phosphates
or phosphonates)54 could lead to hybrid organic-inorganic
networks with potential NLO and proton-transfer behaviors.
Conclusions
Several new nickel(II) complexes that were obtained from
a tridentate (H2L) Schiff-base ligand have been reported. The
synthetic process leads to the formation of the rather unusual
nickel-amido (-NH-NiII) bond by deprotonation of the
primary amine located on H2L. The proton affinities esti-
mated by DFT indicate that the (-NH-NiII) group is more
basic than a primary amine, although the chemical stability
of the resulting (-NH2-NiII)+-containing species is very
poor. ZINDO calculations suggest that the â orientation in
[Ni(L)(MePhCHNH2)] is surprisingly strongly dependent on
the laser frequency, which is in strong contrast to that in
[Ni(HL)(MePhCHNH2)]+. A careful examination reveals that
the â rotation that is achieved upon protonation of [Ni(L)-
(MePhCHNH2)] is precisely equal to 90° when the laser is
operating at 1.022 µm.
Until now, switching the NLO response of a molecule has
been achieved in only a few instances. The present investiga-
tion points out for the first time that a switch obtained by â
rotation of 90° could be envisioned within a single molecule.
We are now considering the possibility of carrying out
frequency-dependent NLO measurements on chiral single
crystals to test the real potential of these metal complexes
as molecular materials with switching NLO capabilities. The
study of the frequency dependence of the EFISH signal could
also be envisioned for a better understanding of the NLO
response in these molecules.
(49) For critical reviews on â calculations by TD-DFT, see (a) Money, V.
Khim. Fakultet 2001, 91, 69 and 85. (b) Matsuzawa, N. N.; Dixon,
D. A. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. B 2000, 26, 17. (c)
Monev, V. Mol. Eng. 1999, 8, 217.
(50) For examples of limitation of the NLO response estimation by means
of TD-DFT, see (a) Cai, Z. L.; Sendt, K.; Reimers, J. R. J. Chem.
Phys. 2002, 117, 5543. (b) Van Gisbergen, S. J. A.; Schipper, P. R.
T.; Gritensko, O. V.; Baerends, E. J.; Snijders, J. D.; Champagne, B.;
Kirtman, B. Phys. ReV. Lett. 1999, 83, 694. (c) De Boeij, P. L.;
Kootstra, F.; Berger, J. A.; Van Leeuwen, R.; Snijders, J. G. J. Chem.
Phys. 2001, 115, 1995.
(51) Coe, B. J.; Harris, J. A.; Brunschwig, B. S.; Garin, J.; Orduna, J.;
Coles, S. J.; Hursthouse, M. B. J. Am. Chem. Soc. 2004, 126, 10418.
(52) (a) Protonic Conductors. Special issue of Solid State Ionics 2001, 145.
(b) Solid State Protonic Conductors. Special issue of Solid State Ionics
1997, 97.
Acknowledgment. We thank CALMIP (CALcul en Midi
Pyre´ne´es - Toulouse) for parallel computing facilities.
Supporting Information Available: Gas-phase structures of
H2L, [Ni(L)(MePhCHNH2)], and [NiH(L)(MePhCHNH2)]+. X-ray
crystallographic files including the structural data for H2L, [Ni(L)-
(MePhCHNH2)], and [Ni(L)(iPrNH2)] in CIF format have been
deposited with the Cambridge Crystallographic Data Centre (num-
bers CCDC 243546, CCDC 243547, and CCDC 243548, respec-
tively). This material is available free of charge via the Internet at
http://pubs.acs.org.
(53) Protonic Conductors; Colomban, Ph., Ed.; Cambridge University
Press: Cambridge, U.K. 1992.
(54) For a review on PO43-- and RP(O)(OMe)2-based protonic conductors,
see, for example, Alberti, G.; Casciola, M. Chem. Solid State Mater.
1992, 2, 238.
IC048578N
1982 Inorganic Chemistry, Vol. 44, No. 6, 2005