Effect of the Coordination to M(II) Metal Centers (M) Zn, Cd, Pt) on the Quadratic Hyperpolarizability of Various Substituted 5-X-1,10-phenanthrolines (X) Donor Group) and of trans-4-(Dimethylamino)-4′-stilbazole

Article abstract of DOI:10.1021/om010470h

Coordination to a Zn(II) center of chelating π-delocalized nitrogen donor ligands such as 5-X-1,10-phenanthrolines (X = OMe, NMe₂, trans-CH=CHC₆H₄-4′-NMe₂, trans, trans-(CH= CH)₂C₆H₄-4′-NMe₂) produces a significant enhancement of their quadratic hyperpolarizability measured either by the EFISH technique (β_1.34) or by absorption and emission solvatochromic investigations (β_CT) working with an incident wavelength of 1.34 μm. However, the enhancement factor by metal coordination of the quadratic hyperpolarizability of planar 5-X-1,10-phenanthrolines is lower than that of the nonplanar and flexible ligand 4-(p -(dibutylamino)styryl)-4′-methyl-2,2′-bipyridine because this latter ligand becomes planar and rigid by coordination. Coordination of 5-(trans-CH=CHC₆H₄-4′-NMe₂)-1,10-phenanthroline to the softer Cd(CH₃CO₂)₂ does not produce an enhancement of the quadratic hyperpolarizability. This is due to a significant decrease of δμ_eg (difference of the dipole moment in the excited and ground states of the intraligand charge transfer (ILCT)) upon coordination to the Cd(II) center. Strangely enough, the second-order NLO response of Zn(II) and Pt(II) complexes carrying two 4-trans-NC₅H₄CH=CHC₆H₄-4′-NMe2 stilbazole ligands in a tetra-hedral or planar coordination indicates a significant increase, upon coordination, of the quadratic hyperpolarizability β_CT, along the ILCT transition of the ligand, but an irrelevant increase of β_1.34, which is the vectorial component of the tensor β along the molecular dipole moment axis, although in the free ligand both β_CT and β_1.34 are comparable. The lack of coincidence of the molecular dipole moment axis with the direction of the ILCT transition of the ligand, a major origin of the second-order NLO response, could explain why β_CT and β_1.34 values are not comparable. In fact, a detailed solvatochromic analysis shows that the increase of β_CT upon coordination to Zn(II) or Pt(II) of the single stilbazole ligand is not significant, in agreement with EFISH measurements and the irrelevant red shift of the ILCT transition.

Full text of DOI:10.1021/om010470h

Organometallics 2002, 21, 161-170
161
Effect of th e Coor d in a tion to M(II) Meta l Cen ter s
(M ) Zn , Cd , P t) on th e Qu a d r a tic Hyp er p ola r iza bility of
Va r iou s Su bstitu ted 5-X-1,10-p h en a n th r olin es (X ) Don or
Gr ou p ) a n d of tr a n s-4-(Dim eth yla m in o)-4′-stilba zole
Dominique Roberto,* Renato Ugo, Francesca Tessore, and Elena Lucenti
Dipartimento di Chimica Inorganica, Metallorganica e Analitica and Centro CNR
“CSSSCMTBSO”, Universita` di Milano, Via G. Venezian, 21, I-20133 Milano, Italy
Silvio Quici and Stefano Vezza
Centro CNR Sintesi e Stereochimica di Speciali Sistemi Organici and Dipartimento di
Chimica Organica e Industriale, Via C. Golgi, 19, I-20133 Milano, Italy
PierCarlo Fantucci and Ivana Invernizzi
Dipartimento di Biotecnologie e Bioscienze, Universita` di Milano Bicocca,
P.zza della Scienza 2, I-20126 Milano, Italy
Silvia Bruni
Dipartimento di Scienze Chimiche, Fisiche e Matematiche, Universita` degli Studi
dell’Insubria, Via Valleggio, 11, I-22100 Como, Italy
Isabelle Ledoux-Rak and J oseph Zyss
LPQM, Ecole Normale Supe´rieure de Cachan, 61 avenue du Pre´sident Wilson,
F-94230 Cachan, France
Received J une 4, 2001
Coordination to a Zn(II) center of chelating π-delocalized nitrogen donor ligands such as
5-X-1,10-phenanthrolines (X ) OMe, NMe2, trans-CHdCHC6H4-4′-NMe2, trans,trans-(CHd
CH)2C6H4-4′-NMe2) produces a significant enhancement of their quadratic hyperpolarizability
measured either by the EFISH technique (â1.34) or by absorption and emission solvatochromic
investigations (âCT) working with an incident wavelength of 1.34 µm. However, the
enhancement factor by metal coordination of the quadratic hyperpolarizability of planar
5-X-1,10-phenanthrolines is lower than that of the nonplanar and flexible ligand 4-(p-
(dibutylamino)styryl)-4′-methyl-2,2′-bipyridine because this latter ligand becomes planar and
rigid by coordination. Coordination of 5-(trans-CHdCHC6H4-4′-NMe2)-1,10-phenanthroline
to the softer Cd(CH3CO2)2 does not produce an enhancement of the quadratic hyperpolar-
izability. This is due to a significant decrease of ∆µeg (difference of the dipole moment in the
excited and ground states of the intraligand charge transfer (ILCT)) upon coordination to
the Cd(II) center. Strangely enough, the second-order NLO response of Zn(II) and Pt(II)
complexes carrying two 4-trans-NC5H4CHdCHC6H4-4′-NMe2 stilbazole ligands in a tetra-
hedral or planar coordination indicates a significant increase, upon coordination, of the
quadratic hyperpolarizability âCT, along the ILCT transition of the ligand, but an irrelevant
increase of â1.34, which is the vectorial component of the tensor â along the molecular dipole
moment axis, although in the free ligand both âCT and â1.34 are comparable. The lack of
coincidence of the molecular dipole moment axis with the direction of the ILCT transition of
the ligand, a major origin of the second-order NLO response, could explain why âCT and â1.34
values are not comparable. In fact, a detailed solvatochromic analysis shows that the increase
of âCT upon coordination to Zn(II) or Pt(II) of the single stilbazole ligand is not significant,
in agreement with EFISH measurements and the irrelevant red shift of the ILCT transition.
In tr od u ction
tions, because they offer a great diversity of tunable
electronic properties by virtue of the metal center.1,2 In
particular with pyridines,3,4 stilbazoles,3-5 or bipy-
ridines2,6 bearing an electron-donor substituent, âvec, the
vectorial projection of the quadratic hyperpolarizability
In the past few years, organometallic and coordination
complexes have emerged as interesting molecular ma-
terials for second-order nonlinear optical (NLO) applica-
10.1021/om010470h CCC: $22.00 © 2002 American Chemical Society
Publication on Web 12/11/2001

162 Organometallics, Vol. 21, No. 1, 2002
Roberto et al.
tensor along the ground-state molecular dipole moment
(µ) direction, measured by the solution-phase dc electric-
field-induced second-harmonic (EFISH) generation
method,7 increases upon coordination to a metal center.
As confirmed by a solvatochromic investigation,8,9 this
enhancement is mainly due to a red shift of the
intraligand charge transfer (ILCT) transition. With a
simple ligand such as 4-(dimethylamino)pyridine, the
enhancement is as high as about 102 times3 and is
modulated by the different donor-acceptor properties
of the metal center. Coordination of 4-(dimethylamino)-
pyridine to relatively soft acceptor centers such as “cis-
M(CO)2Cl” (M ) Rh(I) 4d8, Ir(I) 5d8)3 and “fac-Os-
(CO)3Cl2” (Os(II) 5d6)3 produces an increase of âvec about
10 times higher than coordination to the less accepting
and softer center “Cr0(CO)5” (3d6).4 Strangely enough,
a similar increase of âvec has been reported to occur by
coordination to both the relatively soft acid “cis-Rh-
(CO)2Cl”3 and the hard Lewis acid BF3.10 With more
π-delocalized para-substituted pyridines such as stil-
bazoles, the increase of âvec upon coordination not only
is less relevant (enhancement of about 1.5-2 times with
respect to the free ligand) but decreases with an increase
in the length of the π-delocalized bridge between the
donor group and the pyridine ring.3,4 The remarkable
increase of âvec of the flexible chelating ligand 4-(p-
(dibutylamino)styryl)-4′-methyl-2,2′-bipyridine upon co-
ordination to the Zn(II) Lewis acids6 (enhancement with
respect to the free ligand of about 10 times) prompted
us to investigate the effect on âvec of coordination of the
η1 pseudolinear stilbazole 4-trans-NC5H4CHdCHC6H4-
4′-NMe2 and of the rigid chelating ligands 5-X-1,10-
phenanthrolines (X being a donor group such as OMe,
NMe2, trans-CHdCHC6H4-4′-NMe2, and trans,trans-
(CHdCH)2C6H4-4′-NMe2) to M(II) ions of increasing
hardness (from Pt(II) to Cd(II) to Zn(II)). The choice of
position 5 of the 1,10-phenanthroline ring was based on
a preliminary MNDO-CPHF theoretical investigation
(see Table 1 in the Experimental Section) of the iden-
tification of the best position (2, 3, 4, or 5) of the
substituent “trans-CHdCHC6H4-4′-NMe2” (Table 1).
Comparable values of both dipole moment µ and static
quadratic hyperpolarizability â0 were obtained when the
substituent was located at positions 4 and 5, while
substitution at position 3 afforded a slightly higher
value of â0 but a smaller value of µ, therefore giving
comparable µâ0 values. Only in position 2 were values
of both µ and â0 smaller. Therefore, due to the easier
synthesis of the ligands, we investigated the second-
order NLO properties of 1,10-phenanthrolines substi-
tuted at position 5 and their Zn(II) and, in one case,
Cd(II) complexes.
Exp er im en ta l Section
Gen er a l Com m en ts. ZnCl2, Zn(CF3CO2)2‚2H2O, M(CH3-
CO2)2‚2H2O (M ) Zn, Cd), cis-[PtCl2(CH3CN)2)], 4-trans-NC5H4-
CHdCHC6H4-4′-NMe2, 1,10-phenanthroline, 5-methyl-1,10-
phenanthroline, and 4-(dimethylamino)cinnamaldehyde were
purchased from Sigma-Aldrich and were used without further
purification. Some 5-X-1,10-phenanthrolines (X ) OMe (1),
NMe2 (2); see Scheme 1) were prepared according to a reported
synthesis,11 whereas other 5-X-1,10-phenanthrolines (X )
trans-CHdCHC6H4-4′-NMe2 (3), trans,trans-(CHdCH)2C6H4-
4′-NMe2 (4)) and all the M(II) (M ) Zn, Cd, Pt) complexes were
prepared as described below. All solvents were dried over
molecular sieves (4 Å) prior to use. Products were characterized
by 1H NMR (Bruker AC-200 and a Bruker Avance DRX 300
spectrometers) and UV-visible (J asco V-570 spectrophotom-
eter) spectroscopy and by elemental analysis and in few cases
also by infrared spectroscopy (Nicolet MX-1FT and a Bruker
Vector 22 spectrometers) and mass spectrometry (Varian VG-
9090 spectrometer). Dipole moments, µ, were measured in
CHCl3 by using a WTW-DM01 dipole meter (dielectric con-
stant) coupled with a Pulfrich Zeiss PR2 refractometer (refrac-
tive index) and calculated by using the Guggenheim method.12
Elemental analyses were carried out at the Dipartimento di
Chimica Inorganica, Metallorganica e Analitica of the Uni-
versita` di Milano.
Deter m in a tion of th e Secon d -Or d er NLO Resp on ses.
EFISH measurements7 of âvec, the vectorial component of the
quadratic hyperpolarizability tensor â along the molecular
dipole moment axis, were carried out at the EÄ cole Normale
Supe´rieure de Cachan in CHCl3 solutions of different concen-
trations (10-3-10-4 M) working at a fundamental incident
wavelength of 1.34 µm and, in a few cases, of 1.06 µm, using
a Q-switched, mode-locked Nd:YAG laser with pulse duration
of 15 or 90 ns at a 10 Hz repetition rate. All experimental
EFISH âλ values are defined according to the “phenomenologi-
cal” convention.13
The computational evaluation of dipole moments and of the
quadratic hyperpolarizability (vectorial projection of the qua-
dratic hyperpolarizability tensor along the dipole moment
vector, which corresponds to EFISH âλ values) of both 5-X-1,-
10-phenanthrolines and some of their Zn(II) complexes was
carried out using the MNDO14 semiempirical method and the
coupled perturbed Hartree-Fock (CPHF) approach,15 as previ-
ously described for various para-substituted pyridines.3 We
also evaluated the best position (2, 3, 4, or 5) for the substitu-
ent trans-CHdCHC6H4-4′-NMe2 on the 1,10-phenanthroline
system in order to optimize both the ground-state dipole
moment µ and the static quadratic hyperpolarizability (â0 or
(1) (a) Nalwa, H. S. Appl. Organomet. Chem. 1991, 5, 349. (b) Long,
N. J . Angew. Chem., Int. Ed. Engl. 1995, 34, 21. (c) Optoelectronic
Properties of Inorganic Compounds; Roundhill, D. M., Fackler, J . P.,
J r., Eds.; Plenum Press: New York, 1999.
(2) Le Bozec, H.; Renouard, T. Eur. J . Inorg. Chem. 2000, 229.
(3) Roberto, D.; Ugo, R.; Bruni, S.; Cariati, E.; Cariati, F.; Fantucci,
P. C.; Invernizzi, I.; Quici, S.; Ledoux, I.; Zyss, J . Organometallics 2000,
19, 1775.
(4) Kanis, D. R.; Lacroix, P. G.; Ratner, M. A.; Marks, T. J . J . Am.
Chem. Soc. 1994, 116, 10089.
(5) Bruce, D. W.; Thornton, A. Mol. Cryst. Liq. Cryst. 1993, 231,
253.
(6) (a) Bourgault, M.; Mountassir, C.; Le Bozec, H.; Ledoux, I.;
Pucetti, G.; Zyss, J . J . Chem. Soc., Chem. Commun. 1993, 1623. (b)
Bourgault, M.; Baum, K.; Le Bozec, H.; Pucetti, G.; Ledoux, I.; Zyss,
J . New J . Chem. 1998, 517.
(7) (a) Levine, B. F.; Bethea, C. G. Appl. Phys. Lett. 1974, 24, 445.
(b) Singer, K. D.; Garito, A. F. J . Chem. Phys. 1981, 75, 3572. (c)
Ledoux, I.; Zyss, J . Chem. Phys. 1982, 73, 203.
(8) Bruni, S.; Cariati, F.; Cariati, E.; Porta, F. A.; Quici, S.; Roberto,
D. Spectrochim. Acta, Part A 2001, 57, 1417.
(9) Bruni, S.; Cariati, F.; Cariati, E.; Roberto, D. To be submitted
for publication.
â
0,TOT; see Table 1).
The quadratic hyperpolarizability along the charge-transfer
direction, âCT, of Zn(II), Pt(II), or Cd(II) complexes of 4-trans-
NC5H4CHdCHC6H4-4′-NMe2 and 5-X-1,10-phenanthrolines (X
) NMe2, trans-(CHdCH)nC6H4-4′-NMe2, where n ) 1, 2) and
the free ligand 5-(trans-CHdCHC6H4-4′-NMe2)-1,10-phenan-
(11) Shan, Y.; Sullivan, B. P. Inorg. Chem. 1995, 34, 6235.
(12) (a) Guggenheim, E. A. Trans. Faraday Soc. 1949, 45, 203. (b)
Smith, W. Electric Dipole Moments; Butterworth Scientific: London,
1965. (c) Thompson, B. J . Chem. Educ. 1996, 43, 66.
(13) Willetts, A.; Rice, J . E.; Burland, D. M.; Shelton, D. P. J . Chem.
Phys. 1992, 97, 7590.
(10) Lesley, M. J . G.; Woodward, A.; Taylor, N. J .; Marder, T. B.;
Cazenobe, I.; Ledoux, I.; Zyss, J .; Thornton, A.; Bruce, D. W.; Kakkar,
A. K. Chem. Mater. 1998, 10, 1355.
(14) Dewar, M. J . S.; Thiel, W. J . Am. Chem. Soc. 1977, 99, 4899.
(15) Gerrat, J .; Mills, I. M. J . Chem. Phys. 1968, 49, 1719.

Effect of Coordination on Quadratic Hyperpolarizability
Organometallics, Vol. 21, No. 1, 2002 163
Ta ble 1. In vestiga tion , by th e MNDO-CP HF Sem iem p ir ica l Meth od , of th e Best P osition for th e
Su bstitu en t “tr a n s-CHdCHC6H4-4′-NMe2” on th e 1,10-P h en a n th r olin e (P h en ) Rin g in Or d er To Op tim ize
b
both th e Va lu es of Dip ole Mom en t µ a n d of â0 Obta in ed fr om EF ISH Mea su r em en ts
a
molecule
µ, D
â0,TOT
,
10-30 cm5 esu-1
â0,b 10-30 cm5 esu-1
µâ0, 10-30 D cm5 esu-1
5-(trans-CHdCHC6H4-4′-NMe2)-Phen
4-(trans-CHdCHC6H4-4′-NMe2)-Phen
3-(trans-CHdCHC6H4-4′-NMe2)-Phen
2-(trans-CHdCHC6H4-4′-NMe2)-Phen
4.92
4.96
3.67
1.33
6.64
6.98
14.16
14.42
6.37
6.56
9.20
2.35
31.3
32.5
33.8
3.12
a
b
â0,TOT is the total intrinsic quadratic hyperpolarizability at zero frequency (see ref 3). â0 is the vectorial projection of the quadratic
hyperpolarizability tensor (âvec of eq 2) at zero frequency.
throline was obtained by solvatochromic measurements, using
both absorption and emission spectra in various solvents
(cyclohexane, carbon tetrachloride, toluene, chloroform, ethyl
acetate, dichloromethane, 1,2-dichloroethane, acetone, aceto-
nitrile), following the methodology described in the case of
trans-NC5H4CHdCHC6H4-4′-NMe2 and 5-NMe2-1,10-phenan-
throline.8 The quadratic hyperpolarizability tensor âCT along
the charge-transfer axis of the major charge-transfer absorp-
tion involved in the NLO response is calculated according to
the Oudar two-level equation:16
2.93 (s, 6H, CH3), 2.20 (br s, 1H, D2O exchange); MS-FAB+:
m/e 344 (M + 1)+, calcd for C22H21N3O m/e 343. Anal. Calcd
(found): C, 76.94 (76.62); H, 6.16 (6.02); N, 12.24 (12.40).
5-[4-(Dim eth ylam in o)-r-styr yl]-1,10-ph en an th r olin e (3).
A 0.27 g portion of 5 (0.79 mmol) in 8 mL of aqueous 10% H2-
SO4 was maintained at reflux under magnetic stirring for 1
h. The mixture was cooled to room temperature, made alkaline
with aqueous 20% NaOH, and extracted with CH2Cl2 (2 × 60
mL). The organic phase was washed with 50 mL of brine, dried
over Na2CO3, and evaporated to dryness in vacuo. The residue
was purified by column chromatography (SiO2, CH2Cl2/CH3-
OH/NH4OH ) 90/9.8/0.2) to give 0.3 g of a dark orange solid,
crystallized from CH2Cl2/Et2O to afford 0.18 g (72%) of pure 3
νa2reg2∆µeg
3
âCT
)
(1)
2
2
2
2h2c2 (νa - νL2)(νa - 4νL
)
1
as an orange solid. H NMR (CDCl3): δ (ppm) 9.20 (dd, 1H, J
) 1.6 Hz, J ) 4.3 Hz, H2), 9.12 (dd, 1H, J ) 1.7 Hz, J ) 4.3
Hz, H9), 8.62 (dd, 1H, J ) 1.7 Hz, J ) 8.4 Hz, H4), 8.22 (dd,
1H, J ) 1.7 Hz, J ) 8.1 Hz, H7), 7.94 (s, 1H, H6), 7.66 (dd, 1H,
J ) 4.3 Hz, J ) 8.4 Hz, H3), 7.60 (dd, 1H, J ) 4.3 Hz, J ) 8.4
Hz, H8), 7.56 (d, 2H, J ) 16.1 Hz, H11), 7.52 (d, 2H, J ) 8.9
Hz, C6H4 meta), 7.20 (d, 1H, J ) 15.8 Hz, H12), 6.76 (d, 2H, J
) 8 8 Hz, C6H4 ortho), 3.02 (s, 6H, CH3). MS-FAB+: m/e 326
(M + 1)+, calcd for C22H19N3 m/e 325. Anal. Calcd (found): C,
81.20 (80.90); H, 5.88 (6.02); N, 12.91 (12.70).
where reg is the transition dipole moment related to the
integrated intensity f of the absorption band, νa is the
frequency of the charge-transfer absorption band, νL is the
frequency of the incident radiation, and ∆µeg is the variation
of the dipole moment upon excitation.
Syn th esis of 5-X-1,10-P h en a n th r olin es. For the number-
ing used in the attribution of the 1H NMR signals, see the
following structure:
5-{2-Hydr oxy-2-[4-(dim eth yla m in o)p h en yl]-3-bu ten yl}-
1,10-p h en a n th r olin e (6). A 3 mL portion of 2 M lithium
diisopropylamide (LDA) in THF was added dropwise under
N2 at 0 °C to a solution of 0.5 g (2.6 mmol) of 5-methyl-1,10-
phenanthroline in 5 mL of anhydrous THF. The resulting
mixture was stirred at this temperature for 75 min, and then
a solution of 0.91 g (5.2 mmol) of p-(dimethylamino)benzalde-
hyde in 5 mL of THF was added via syringe. The mixture was
left at 0 °C for 75 min and then cooled to room temperature
and reacted overnight. The excess LDA was destroyed with
MeOH, and the solvent was removed under reduced pressure.
The resulting residue was dissolved in 50 mL of CH2Cl2 and
washed with a small amount of H2O. The organic layer was
separated, washed with 10 mL of brine, dried over Na2CO3,
and evaporated, affording a residue which was purified by
column chromatography (SiO2, CH2Cl2/CH3OH/NH4OH ) 95/
4.5/0.5) to give 0.32 g of a brown solid. This product was
purified by crystallization from CH2Cl2/Et2O to afford 0.27 g
5-{2-Hyd r oxy-2-[4-(d im eth yla m in o)p h en yl]eth yl}-1,10-
p h en a n th r olin e (5). A 3 mL portion of 2 M lithium diiso-
propylamide (LDA) in THF was added dropwise under N2 at
0 °C to a solution of 0.5 g (2.6 mmol) of 5-methyl-1,10-
phenanthroline in 5 mL of anhydrous THF. The mixture was
stirred at this temperature for 75 min, and then a solution of
0.8 g (5.2 mmol) of p-(dimethylamino)benzaldehyde in 5 mL
of THF was added with a syringe. The mixture was left at 0
°C for 75 mi and then cooled to room temperature and left
overnight. The excess LDA was destroyed with MeOH, the
solvent was removed under reduced pressure, the residue was
dissolved in 50 mL of CH2Cl2, and this solution was washed
with a small amount of H2O. The organic layer was separated,
washed with 10 mL of brine, dried over Na2CO3, and evapo-
rated, affording a residue which was purified by column
chromatography (SiO2, CH2Cl2/CH3OH/NH4OH ) 95/4.5/0.5)
to give 0.46 g of an orange solid, crystallized from methanol
to afford 0.27 g (30%) of pure 5 as a yellow solid. 1H NMR
(CDCl3): δ (ppm) 9.14 (dd, 1H, J ) 1.6 Hz, J ) 4.3 Hz, H2),
9.12 (dd, 1H, J ) 1.7 Hz, J ) 4.3 Hz, H9), 8.44 (dd, 1H, J )
1.6 Hz, J ) 8.4 Hz, H4), 8.12 (dd, 1H, J ) 1.7 Hz, J ) 8.1 Hz,
H7), 7.62 (s, 1H, H6), 7.60 (dd, 1H, J ) 4.3 Hz, J ) 8.0 Hz, H3),
7.57 (dd, 1H, J ) 4.4 Hz, J ) 7.7 Hz, H8), 7.26 (d, 2H, J ) 8.7
Hz, C6H4 meta), 6.70 (d, 2H, J ) 8.7 Hz, C6H4 ortho), 5.04 (t,
1H, J ) 7.0 Hz, J ) 5.9 Hz, CHOH), 3.52 (m, 2H, CH2OH,
1
(28%) of pure 6 as a yellow solid. H NMR (CDCl3): δ (ppm)
9.14 (dd, 1H, J ) 1.3 Hz, J ) 4.2 Hz, H2), 9.12 (dd, 1H, J )
1.6 Hz, J ) 4.4 Hz, H9), 8.49 (dd, 1H, J ) 1.4 Hz, J ) 8.4 Hz,
H4), 8.14 (dd, 1H, J ) 1.6 Hz, J ) 8.1 Hz, H7), 7.70 (s, 1H, H6),
7.61 (dd, 1H, J ) 4.2 Hz, J ) 8.3 Hz, H3), 7.57 (dd, 1H, J )
4.4 Hz, J ) 8.1 Hz, H8), 7.23 (d, 2H, J ) 8.7 Hz, C6H4 meta),
6.65 (d, 2H, J ) 8.8 Hz, C6H4 ortho), 6.54 (d, 1H, J ) 15.9 Hz,
H
14), 6.14 (dd, 1H, J ) 6.9 Hz, J ) 15 8 Hz, H13), 4.75(q, 1H,
CHOH), 3.40 (m, 2H, CH2OH), 2.90 (s, 6H, CH3), 2.10 (br. s,
1H, D2O exchange). MS-EI: m/e 351 (M - H2O), calcd for
C
24H21N3 m/e 351. Anal. Calcd (found): C, 82.02 (82.23); H,
6.02 (6.05),; N, 11.96 (12.01).
5-{4-[4-(Dim et h yla m in o)p h en yl]-3-b u t a d ien yl}-1,10-
p h en a n th r olin e (4). A 0.22 g portion of 6 (0.60 mmol) in 15
mL of aqueous 10% H2SO4 was maintained at reflux under
magnetic stirring for 1 h. The mixture was cooled to room
temperature, made alkaline with aqueous 20% NaOH, and
extracted with CH2Cl2 (2 × 60 mL). The organic phase was
(16) (a) Oudar, J . L.; Chemla, D. S. J . Chem. Phys. 1977, 66, 2664.
(b) Oudar, J . L. J . Chem. Phys. 1977, 67, 446. (c) Oudar, J . L.; Le
Person, H. Opt. Commun. 1975, 15, 258.

164 Organometallics, Vol. 21, No. 1, 2002
Roberto et al.
washed with 50 mL of brine, dried over Na2CO3, and evapo-
rated to dryness in vacuo. The residue was purified by column
chromatography (SiO2, CH2Cl2/CH3OH/NH4OH ) 90/9.8/0.2)
to give 0.3 g of a dark solid, crystallized from CH2Cl2/Et2O to
afford 0.19 g (72%) of pure 4 as an orange solid. 1H NMR
(CDCl3): δ (ppm) 9.20 (dd, 1H, J ) 1.6 Hz, J ) 4.3 Hz, H2),
9.12 (dd, 1H, J ) 1.7 Hz, J ) 4.4 Hz, H9), 8.60 (dd, 1H, J )
1.6 Hz, J ) 8.4 Hz, H4), 8.22 (dd, 1H, J ) 1.7 Hz, J ) 8.1 Hz,
H7), 7.94 (s, 1H, H6), 7.67 (dd, 1H, J ) 4.3 Hz, J ) 8.4 Hz, H3),
7.60 (dd, 1H, J ) 4.3 Hz, J ) 8.1 Hz, H8), 7.40 (d, 2H, J ) 8.8
Hz, C6H4 meta), 7.25 (d, 1H, J ) 15.0 Hz, H11), 7.14 (dd, 1H,
J ) 14.9 Hz, J ) 15.0 Hz, H12), 6.95 (dd, 1H, J ) 15.3 Hz, J
) 15.3 Hz, H13), 6.74 (d, 1H, J ) 15.3 Hz, H14), 6.71 (d, 2H, J
) 8.8 Hz, C6H4 ortho), 3.00 (s, 6H, CH3); MS-FAB+: m/e 352
(M + 1)+, calcd for C24H21N3 m/e 351. Anal. Calcd (found): C,
82.02 (81.83); H, 6.02 (6.15); N, 11.90 (11.94).
(s, 6H, CH3CO2); MS-FAB+ m/e 498 (M - CH3CO2)+, calcd for
C
26H25N3O4Cd m/e 557. Anal. Calcd (found): C,56.17 (55.72);
H, 4.53 (4.65); N, 7.56 (7.64).
Gen er a l P r oced u r e for th e Syn th esis of [Zn Y2L2]
Com p lexes (L ) 4-tr a n s-NC5H4CHdCHC6H4-4′-NMe2; Y )
Cl, CH3CO2, CF 3CO2). The Zn(II) salt (0.814 mmol) was kept
under vacuum (10-2 Torr) for 3 h at 100 °C in a two-necked
flask (250 mL). Then anhydrous CH2Cl2 (150 mL) and 4-trans-
NC5H4CHdCHC6H4-4′-NMe2 (0.364 g; 1.628 mmol) were added
under N2; the mixture was stirred overnight in the dark (the
flask was covered with aluminum foil), affording a solution
evaporated to dryness to give, in quantitative yield, [ZnY2(4-
trans-NC5H4CHdCHC6H4-4′-NMe2)2] recrystallized from dichlo-
romethane/pentane. For the numbering used in the attribution
of the 1H NMR signals, see the following structure:
Gen er a l P r oced u r e for th e Syn th esis of M(CH 3CO2)2
(M ) Zn , Cd ) Com p lexes of Liga n d s 1-4. In a 50 mL flask,
the 5-X-1,10-phenanthroline (150 mg) was dissolved in CH2-
Cl2 (15 mL) and, then, 1 equiv of the solid salt (M(CH3CO2)2‚
2H2O where M ) Zn, Cd) was added. The mixture was stirred
at room temperature overnight and evaporated to dryness,
affording, in quantitative yield, pure [M(CH3CO2)2(5-X-1,10-
phenanthroline)], recrystallized from dichloromethane/pentane
at room temperature.
1‚Zn (CH3CO2)2: yellow solid; 1H NMR (CDCl3) δ (ppm) 9.29
(dd, 1H, J ) 1.4 Hz, J ) 4.7 Hz, H2), 9.11 (dd, 1H, J ) 1.4 Hz,
J ) 4.8 Hz, H9), 8.80 (dd, 1H, J ) 1.4 Hz, J ) 8.4 Hz, H4),
8.34 (dd, 1H, J ) 1.4 Hz, J ) 8.2 Hz, H7), 7.90 (dd, 1H, J )
4.7 Hz, J ) 8.4 Hz, H3), 7.81 (dd, 1H, J ) 4.7 Hz, J ) 8.2 Hz,
H8), 7.28 (s, 1H, H6), 3.00 (s, 6H, (CH3)2N), 2.10 (s, 6H, CH3-
CO2); MS-FAB+ m/e 347 (M - CH3CO2)+, calcd for C18H19N3O4-
Zn m/e 406. Anal. Calcd (found): C, 53.15 (53.04); H, 4.71
(4.82); N, 10.30 (10.19).
1
[Zn Cl2L2]: red solid; H NMR (CDCl3) δ (ppm) 8.63 (4 H,
2H2 and 2H6), 7.55 (4 H, 2H3 and 2H5), 7.49 (d, 4 H, 2H2′ and
2H6′, J ) 8.6 Hz), 7.41 (d, 2 H, 2CHPh, J trans ) 16.0 Hz), 6.84
(d, 2 H, 2PyCH, J trans ) 16.0 Hz), 6.73 (d, 4 H, 2H3′ and 2H5′,
J ) 8.6 Hz), 3.07 (s, 12 H, 2NMe2); MS (FAB+) m/e 549
(molecular ion peak - Cl). Anal. Calcd (found): C, 61.55
(61.29); H, 5.47 (5.69); N, 9.57 (9.33).
1
[Zn (CH3CO2)2L2]: yellow solid; H NMR (CDCl3) δ (ppm)
8.64 (d, 4 H, 2H2 and 2H6, J ) 5.7 Hz), 7.47 (d, 4 H, 2H2′ and
2H6′, J ) 8.7 Hz), 7.42 (d, 4 H, 2H3 and 2H5, J ) 5.7 Hz), 7.31
(d, 2 H, 2CHPh, J trans ) 16.0 Hz), 6.80 (d, 2 H, 2PyCH, J trans
) 16.0 Hz), 6.72 (d, 4 H, 2H3′ and 2H5′, J ) 8.7 Hz), 3.03 (s, 12
H, 2NMe2), 2.13 (s, 6 H, 2O2CCH3); IR (as KBr) νasym(CO2) and
2‚Zn (CH3CO2)2: white solid; 1H NMR (CDCl3) δ (ppm) 9.30
(dd, 1H, J ) 1.3 Hz, J ) 4.8 Hz, H2), 9.10 (dd, 1H, J ) 1.1 Hz,
J ) 4.8 Hz, H9), 8.88 (dd, 1H, J ) 1.4 Hz, J ) 8.4 Hz, H4),
8.36 (dd, 1H, J ) 1.1 Hz, J ) 8.2 Hz, H7), 7.91 (dd, 1H, J )
4.8 Hz, J ) 8.4 Hz, H3), 7.81 (dd, 1H, J ) 4.7 Hz, J ) 8.2 Hz,
H8), 7.09 (s, 1H, H6), 4.1 (s, 3H, CH3O), 2.05 (s, 6H, CH3CO2);
MS-FAB+ m/e 333 (M - CH3CO2)+, calcd for C17H16N2O5Zn
m/e 392. Anal. Calcd (found): C, 51.86 (51.66); H, 4.09 (4.11);
N, 7.11 (7.14).
ν
sym(CO2) 1642 and 1430 cm-1, respectively, in agreement with
unidentate acetates.17 Anal. Calcd (found): C, 64.56 (64.36);
H, 6.01 (6.13); N, 8.86 (8.99).
[Zn (CF 3CO2)2L2]: yellow solid; 1H NMR (CDCl3) δ (ppm)
8.61 (d, 4 H, 2H2 and 2H6, J ) 6.3 Hz), 7.53 (d, 4 H, 2H3 and
2H5, J ) 6.3 Hz), 7.49 (d, 4 H, 2H2′ and 2H6′, J ) 8.8 Hz), 7.39
(d, 2 H, 2CHPh, J trans ) 16.1 Hz), 6.82 (d, 2 H, 2PyCH, J trans
) 16.1 Hz), 6.73 (d, 4 H, 2H3′ and 2H5′, J ) 8.8 Hz), 3.06 (s, 12
H, 2NMe2); IR (as KBr) νasym(CO2) and νsym(CO2) 1700 and 1433
cm-1, respectively, in agreement with unidentate trifluoroac-
etates;17 MS (FAB+) m/e 625 (molecular ion peak - O2CCF3).
Anal. Calcd (found): C, 55.14 (55.35); H, 4.32 (4.50); N, 7.56
(7.39).
1
3‚Zn (CH3CO2)2: red solid; H NMR (CDCl3) δ (ppm) 9.34
(d, 1H, J ) 4.7 Hz, H2), 9.23 (d, 1H, J ) 4.7 Hz, H9), 8.79 (dd,
1H, J ) 1.1 Hz, J ) 8.5 Hz, H4), 8.47 (dd, 1H, J ) 1.3 Hz, J )
8.3 Hz, H7), 8.06 (s, 1H, H6), 7.94 (dd, 1H, J ) 4.7 Hz, J ) 8.5
Hz, H3), 7.87 (dd, 1H, J ) 4.7 Hz, J ) 8.5 Hz, H8), 7.52 (d, 2H,
J ) 8.8 Hz, H ortho), 7.50 (d, 1H, J ) 15.4 Hz, H11), 7.25 (d,
1H, J ) 15.4 Hz, H12), 6.75 (d, 2H, J ) 8.8 Hz, H meta), 3.03
(s, 6H, (CH3)2N), 2.05 (s, 6H, CH3CO2); MS-FAB+ m/e 450 (M
- CH3CO2)+, calcd for C26H25N3O4Zn m/e 509. Anal. Calcd
(found): C, 61.34 (61.09); H, 4.95 (5.06); N, 8.26 (8.42).
Syn th esis of cis-[P tCl2L2]. In a two-necked flask (250 mL)
under N2, 4-trans-NC5H4CHdCHC6H4-4′-NMe2 (0.160 g; 0.715
mmol) was added to a suspension of cis-[PtCl2(CH3CN)2] (0.124
g; 0.357 mmol) in anhydrous CH2Cl2 (150 mL). After it was
stirred for 1 h at room temperature, the final yellow solution
was evaporated to dryness to give, in quantitative yield, cis-
[PtCl2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2], recrystallized
from dichloromethane/pentane, as a yellow solid. 1H NMR
(CDCl3): δ (ppm) 8.50 (d, 4 H, 2H2 and 2H6, J ) 6.3 Hz), 7.49
1
4‚Zn (CH3CO2)2: red solid; H NMR (CDCl3) δ (ppm) 9.34
(d, 1H, J ) 4.5 Hz, H2), 9.22 (d, 1H, J ) 4.5 Hz, H9), 8.84 (d,
1H, J ) 8.1 Hz, H4), 8.47 (d, 1H, J ) 7.3 Hz, H7), 8.05 (s, 1H,
H6), 7.95 (dd, 1H, J ) 4.7 Hz, J ) 8.5 Hz, H3), 7.87 (dd, 1H, J
) 4.8 Hz, J ) 8.2 Hz, H8), 7.40 (d, 2H, J ) 8.8 Hz, H ortho),
7.20-7.18 (m, 2H, H11, H12), 6.94 (ddd, 1H, J ) 15.4 Hz, J )
3.2 Hz, J ) 2.8 Hz, H13), 6.79 (d, 1H, J ) 15.4 Hz, H14), 6.71
(d, 2H, J ) 8.8 Hz, meta), 3.00 (s, 6H, (CH3)2N), 2.05 (s, 6H,
CH3CO2); MS-FAB+ m/e 474 (M - CH3CO2)+, calcd for
(8 H, 2H3, 2H5, 2H2′ and 2H6′), 7.39 (d, 2 H, 2CHPh, J trans
)
16.1 Hz), 6.84 (d, 2 H, 2PyCH, J trans ) 16.1 Hz), 6.73 (d, 4 H,
2H3′ and 2H5′, J ) 8.6 Hz), 3.06 (s, 12 H, 2NMe2). IR (as
polyethylene pellet): ν(Pt-Cl) ) 390 and 349 cm-1, slightly
lower than those of cis-[PtCl2(CH3CN)2] (397 and 351 cm-1).
Anal. Calcd (found): C, 50.37 (50.19); H, 4.48 (4.60); N, 7.84
(8.01).
C
28H27N3O4Zn m/e 533. Anal. Calcd (found): C, 62.87 (62.64);
H, 5.09 (5.21); N, 7.85 (7.68).
3‚Cd (CH3CO2)2: red solid; 1H NMR (CD3CN) δ (ppm) 8.84
(d, 1H, J ) 8.9 Hz, H4), 8.81 (d, 1H, J ) 4.7 Hz, H2), 8.75 (d,
1H, J ) 4.6 Hz, H9), 8.43 (d, 1H, J ) 8.0 Hz, H7), 8.08 (s, 1H,
H6), 7.78 (dd, 1H, J ) 4.7 Hz, J ) 8.0 Hz, H3), 7.75 (dd, 1H, J
) 4.7 Hz, J ) 8.0 Hz, H8), 7.54 (d, 1H, J ) 15.9 Hz, H11), 7.50
(d, 2H, J ) 8.8 Hz, H ortho), 7.28 (d, 1H, J ) 15.8 Hz, H12),
6.73 (d, 2H, J ) 8.8 Hz, H meta), 2.95 (s, 6H, (CH3)2N), 1.90
Resu lts a n d Discu ssion
1. Syn th esis of Liga n d s a n d of Th eir Com p lexes
a n d Deter m in a tion of Som e Gr ou n d -Sta te P r op er -
(17) Deacan, G. B.; Phillips, R. J . Coord. Chem. Rev. 1980, 33, 227.

Effect of Coordination on Quadratic Hyperpolarizability
Organometallics, Vol. 21, No. 1, 2002 165
Sch em e 1. Syn th esis of Liga n d s 3 a n d 4a
1.2. Com p lexes of Zn (II) a n d P t(II) w ith 4-tr a n s-
NC5H4CHdCHC6H4-4′-NMe2. The complexes [ZnY2(4-
trans-NC5H4CHdCHC6H4-4′-NMe2)2] (Y ) Cl, CH3CO2,
CF3CO2) and cis-[PtCl2(4-trans-NC5H4CHdCHC6H4-4′-
NMe2)2] were prepared by room-temperature reaction
in CH2Cl2 of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 with
the stoichiometric amount of ZnY2 or cis-[PtCl2(CH3-
CN)2]. They were characterized by UV-visible spec-
1
troscopy (Table 3), elemental analysis, infrared and H
NMR spectroscopy, and, in some cases, mass spectrom-
etry (see Experimental Section). Dipole moments were
determined in CHCl3 by the standard Guggenheim
method12 (Table 3).
1
1.2.1. H NMR Sp ectr a a n d Dip ole Mom en ts. In
1
the H NMR spectra (see Experimental Section), coor-
dination to Zn(II) causes a shift to lower field of the
signals of the hydrogens in a position R with respect to
the pyridine nitrogen atom. In accord with the increased
accepting character of Zn(II), δ in CHCl3 goes from 8.52
to 8.64, 8.63, and 8.61 ppm when Y ) CH3CO2, Cl, CF3-
CO2, respectively. However, such a shift is not observed
upon coordination to the softer center “cis-PtCl2” (δ 8.52
ppm), which behaves as “cis-M(CO)2Cl” (M ) Rh, Ir).3
The chemical shift of the NMe2 protons (δ 3.03 ppm) is
not too affected by coordination (δ 3.03-3.07 ppm).3 The
coupling constant between the trans olefinic hydrogens
of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 (J H-H ) 16.2
Hz) does not change upon coordination. Therefore, a fair
comparison of the electronic properties between the free
and metal-coordinated ligands is possible.3 Dipole mo-
ments of Zn(II) and Pt(II) complexes are reported in
Table 3, along with those of related Rh(I),3 Ir(I),3 and
a
Legend: (i) p-(dimethylamino)benzaldehyde, 2 M LDA in
anhydrous THF, 0 °C, 2.5 h, room temperature, 16 h; (ii) 10%
H2SO4 in H2O, reflux (103 °C), 1 h; (iii) p-(dimethylamino)cin-
namylaldehyde, 2 M LDA in anhydrous THF, 0 °C, for 2.5 h,
room temperature for 16 h.
ties. 1.1. 5-X-1,10-p h en a n th r olin es (5-X-P h en ) a n d
Th eir Com p lexes w ith Zn (II) a n d Cd (II). Ligands
1 and 2 (X ) OMe, NMe2, respectively; see Scheme
1) were synthesized according to the literature,11
whereas ligands 3 and 4 (X ) trans-CHdCHC6H4-4′-
NMe2, trans,trans-(CHdCH)2C6H4-4′-NMe2, respec-
tively) were prepared as shown in Scheme 1, starting
from commercially available 5-Me-1,10-phenanthroline.
Complexes were prepared by reaction at room temp-
erature of the ligands dissolved in CH2Cl2 with a
stoichiometric amount of ZnCl2, Zn(CH3CO2)2, or Cd-
(CH3CO2)2 (see Experimental Section). Both phenan-
throline ligands and their complexes were characterized
by UV-visible spectroscopy (Table 2), elemental analy-
10
BF3 complexes. The highest increase of the dipole
moment of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 occurs
upon coordination to the strong Lewis acid BF3 (µ
enhancement factor 3);10 the molecular dipole moment
of the square-planar cis-[PtCl2(4-trans-NC5H4CHd
CHC6H4-4′-NMe2)2] complex is higher than that of
the pseudolinear arrangement in square-planar cis-
[M(CO)2Cl(4-trans-NC5H4CHdCHC6H4-4′-NMe2)] (M )
Rh, Ir) complexes3 and smaller than that of tetrahedral
[ZnY2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2] (Y ) Cl,
CF3CO2) complexes, which increases in the order ex-
pected for the increased electron-accepting strength of
the ancillary ligand Y. While in Rh(I) and Ir(I) com-
plexes and in the adduct with BF3 we have a limited
orthogonal contribution to the dipole moment which
roughly lies along the ligand axis,3 in tetrahedral Zn-
(II) and Cd(II) complexes and in the square-planar Pt-
(II) complex the molecular dipole moment is the vecto-
rial sum of the dipole moments of two pseudolinear
arrangements; therefore, it does not lie along the ligand
axis. For instance, for the Pt(II) complex (angles Cl-
Pt-Cl and N-Pt-N of about 90°), if we assume a value
of about 2.5 D for the dipole moment of each Pt-Cl
bond18 and of 3.9 D (corresponding to the value of the
unperturbed free ligand)3 for the dipole moment of each
Pt-(4-trans-NC5H4CHdCHC6H4-4′-NMe2) segment, by
addition of the two vectorial contributions we obtain a
dipole moment of 9 D, quite close to the experimental
value of 8.0 D. This would suggest that the dipole
moment of the ligand is not enhanced very much by
1
sis, H NMR spectroscopy, and mass spectrometry (see
Experimental Section). Dipole moments were deter-
mined in CHCl3 by the standard Guggenheim method12
(Table 2).
1
1.1.1. H NMR Sp ectr a a n d Dip ole Mom en ts. In
the 1H NMR spectra (see Experimental Section) of
phenanthrolines coordinated to Zn(CH3CO2)2, the sig-
nals of hydrogens in a position R to the phenanthroline
nitrogen atoms are shifted to lower fields (∆δ in CDCl3
0.11-0.14 ppm) due to some electron transfer to Zn-
(II), while the relatively small shift to lower field of the
signals of the NMe2 protons becomes negligible as the
bridge length between the donor group and the metal
increases (∆δ in CDCl3 0.1, 0.01, and 0 ppm for ligands
2-4, respectively). Dipole moments (Table 2) of ligands
1 and 2 are comparable and slightly lower than those
of the more π-delocalized ligands 3 and 4. The dipole
moment enhancement occurring upon coordination to
M(CH3CO2)2 (M ) Zn, Cd) is higher for phenanthrolines
1 and 2 (µ enhancement factor 1.9-2) than for the more
π-delocalized 3 and 4 (µ enhancement factor 1.6, com-
parable to that reported upon coordination of 4-(trans-
CHdCHC6H4-p-NBu2)-4′-CH3-2,2′-bipyridine to Zn(CH3-
CO2)2).6
(18) (a) Chatt, J .; Williams, A. A. J . Chem. Soc. 1951, 3061. (b)
Chatt, J .; Williams, A. A. J . Chem. Soc. 1960, 1378.

166 Organometallics, Vol. 21, No. 1, 2002
Roberto et al.
Ta ble 2. Electr on ic Sp ectr a , Dip ole Mom en ts, a n d EF ISH â1.34 Va lu es in CHCl3 of Su bstitu ted
5-X-1,10-p h en a n th r olin es (5-X-P h en ) a n d Th eir Com p lexes w ith “Zn (CH3CO2)2” a n d “Cd (CH3CO2)2”
a,b
a
λmax
,
µâ0,c 10-30
â1.34,
10-30
molecule
nm
µ,a
D
D cm5 esu-1
cm5 esu-1
EFd
5-MeO-Phen
[Zn(CH3CO2)2(5-MeO-Phen)]
5-NMe2-Phen
[Zn(CH3CO2)2(5-NMe2-Phen)]
5-(trans-CHdCHC6H4-4′-NMe2)-Phen
[Zn(CH3CO2)2(5-(trans-CHdCHC6H4-4′-NMe2)-Phen)]
[Cd(CH3CO2)2(5-(trans-CHdCHC6H4-4′-NMe2)-Phen)]
5-(trans,trans-(CHdCH)2C6H4-4′-NMe2)-Phen
[Zn(CH3CO2)2(5-(trans,trans-(CHdCH)2C6H4-4′-NMe2)-Phen)]
4-(trans-CHdCHC6H4-4′-NBu2)-4′-CH3-2,2′-bipyridinef
[Zn(CH3CO2)2(4-(trans-CHdCHC6H4-4′-NBu2)-4′-CH3-2,2′-bipyridine)]f
272
284
328
344
371
419
411
399
432
388
447
4.0
9.6
74
20
175
127
338
202
215
450
41
4.0e
13
7.2
33
41
77
46
75
112
14
116
7.6
3.8
7.7
4.9
8.0
7.7
4.9
7.7
5.1
7.9
3.2
4.6
1.9
1.1
1.5
450
8.3
a
b
Experimental values in CHCl3.
λ
max (ILCT) mainly involved in the NLO response.8,9 c â measured by EFISH at 1.34 µm and converted
to â0 with the two-level expression (eq 3).16 EF is the â1.34 enhancement factor: â1.34(complex)/â1.34(free phenanthroline). e A similar
â1.34 value (3.7 × 10-30 cm5 esu-1) is obtained by carrying out the EFISH measurement at 1.06 µm and converting it to â1.34 with the
two-level expression (eq 3).16 f Reference 6.
d
Ta ble 3. Electr on ic Sp ectr a , Dip ole Mom en ts, a n d EF ISH â1.34 Va lu es in CHCl3 of Zn (II), P t(II), Rh (I), Ir (I),
a n d BF 3 Com p lexes w ith 4-tr a n s-NC5H4CHdCHC6H4-4′-NMe2
a,b
molecule
λmax
,
nm
µ,a
D
µâ1.34,c 10-30 D cm5 esu-1
â
1.34, 10-30 cm5 esu-1
EFd
4-trans-NC5H4CHdCHC6H4-4′-NMe2 (L)e
[Zn(CH3CO2)2L2]
[ZnCl2L2]
[Zn(CF3CO2)2L2]
cis-[PtCl2L2]
cis-[Ir(CO)2ClL]e
cis-[Ir(COT)2ClL]e,g
cis-[Rh(CO)2ClL]e
BF3Lh
374
376f
410
420
375
431
413
421
438
3.9
8.0
8.4
10.5
8.0
6.0
8.1
7.0
11.8
215
384
706
1019
328
768
664
777
1522
55
48
84
97
41
128
82
111
129
0.9
1.5
1.8
0.8
2.3
1.5
2.0
2.3
a
b
d
Experimental values in CHCl3. ILCT transition. c â measured by EFISH at 1.34 µm. EF is the â1.34 enhancement factor:
g
h
â
1.34(complex)/â1.34(free stilbazole). e Data from ref 3. f For[Cd(CH3CO2)2L2], λmax 375 nm. COT ) cyclooctene. Data from ref 10.
Surprisingly, both µ and EFISH â1.34
values of the ligand 4-trans-NC5H4CHdCHC6H4-4′-NMe2 measured in CHCl3 (µ ) 10 D; â1.34
µm
µm
) 45 × 10-30 D cm5 esu-1) are quite different from the values measured by some of us under similar conditions.3
coordination to a soft metal center such as Pt(II). The
same conclusion can be reached for the tetrahedral
[ZnCl2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2] complex;
assuming that the values are the same as above for the
various contributions and that the angles Cl-Zn-Cl
and N-Zn-N are roughly 109°, the calculated value of
the dipole moment is 7.4 D, slightly lower but not too
far from the experimental value of 8.4 D. In this latter
case MNDO calculations show that the electronic den-
sity distribution on the various atoms of 4-trans-NC5H4-
CHdCHC6H4-4′-NMe2 is not significantly affected by
coordination to “ZnCl2” or “Zn(CF3CO2)2”, in agreement
with a relatively small perturbation.
∆λmax (64 nm) was reported with the strong Lewis acid
BF3.10
The UV-visible spectra of complexes of Zn(II) and Pt-
(II) with 4-trans-NC5H4CHdCHC6H4-4′-NMe2 (Table 3)
are also characterized by such an intense ILCT transi-
tion (log ꢀ ) ca. 4.5); however, the red shift does not
follow a trend which can be easily related to the Lewis
acidity. It is very small upon coordination to “Zn(CH3-
CO2)2” (∆λmax ) 2 nm) and “PtCl2” (∆λmax ) 1 nm), but
it is significant upon coordination to “ZnCl2” (∆λmax
)
36 nm) and particularly to “Zn(CF3CO2)2” (∆λmax ) 46
nm). Therefore, differently from pseudolinear molecules
with one stilbazole ligand linked to a Lewis metal
center,3,10 the ∆λmax of the ILCT transition cannot be
used as a direct measure of the relative Lewis acidity
when two stilbazole ligands are linked to a metal center
in a tetrahedral or square-planar arrangement. The
UV-visible spectrum of free phenanthroline, when X
2. Ch a r ge-Tr a n sfer Ba n d s in UV-Visible Sp ec-
t r a . UV-visible spectra in CHCl3 of the complexes
cis-[M(CO)2Cl(4-trans-NC5H4CHdCHC6H4-4′-NMe2)] (M
) Rh, Ir), cis-[Ir(cyclooctene)2Cl(4-trans-NC5H4CHd
CHC6H4-4′-NMe2)], and fac-[Os(CO)3Cl2(4-trans-NC5H4-
CHdCHC6H4-4′-NMe2)] have one major band (log ꢀ )
ca. 4.5) in the region 413-435 nm, attributed to an
intraligand charge transfer (ILCT) transition, red-
shifted when compared to the same band (λmax 374 nm)
in the free ligand.3,9,10 This transition emanates from
the NMe2 donor end of the stilbazole ligand and is
usually sensitive to the Lewis acidity of the metal
moiety, which increases the π* acceptor properties of
the pyridine ring, thus producing a red shift of the ILCT.
Therefore, ∆λmax was used by some of us as a way to
estimate the relative Lewis acidity of an acceptor metal
center.3 In agreement with this assumption, the higher
) NMe2, shows two bands (Table 2). The first (λmax
)
281 nm) can be attributed to a π f π* transition, by
analogy with the π f π* transition of the unsubstituted
phenanthroline (λmax ) 265 nm),19 whereas the second
(λmax ) 328 nm) can be attributed to the intraligand
charge transfer (ILCT) from the NMe2 group to the π*
system of the phenanthroline ring.8 When X ) 4-trans-
(CHdCH)nC6H4-4′-NMe2 (n ) 1, 2) as in ligands 3 and
4, this ILCT transition is shifted, as expected, to lower
energy (λmax ) 371 and 399 nm for n ) 1 and 2,
(19) (a) Zone, K.; Krumholz, P.; Stammreich, H. J . Am. Chem. Soc.
1955, 77, 777. (b) Ito, T.; Tanaka, N.; Hanazaki, I.; Nagakura, S. Bull.
Chem. Soc. J pn. 1969, 42, 702.

Effect of Coordination on Quadratic Hyperpolarizability
Organometallics, Vol. 21, No. 1, 2002 167
respectively). When X ) MeO, the UV-visible spectrum
shows only one band at ca. 272 nm, attributed to the
overlap of the π f π* transition and the ILCT from the
MeO group to the π* system of the phenanthroline ring.9
The higher energy of the ILCT is reasonable because a
blue shift (∆λ ) 59 nm) of a similar ILCT band was
reported to occur, going from ((dialkylamino)styryl)-
bipyridine to (alkoxystyryl)bipyridine.6 Upon coordina-
tion to “Zn(CH3CO2)2” all these ILCT transitions are
red-shifted: ∆λmax is 12, 16, 33, and 48 nm for X ) OMe,
NMe2, 4-trans,trans-(CHdCH)2C6H4-4′-NMe2, 4-trans-
CHdCHC6H4-4′-NMe2, respectively. Coordination of
ligand 4 (X ) 4-trans-CHdCHC6H4-4′-NMe2) to “Cd(CH3-
CO2)2” also induces a significant red shift (∆λmax ) 40
nm) of the ILCT transition.
3. EF ISH a n d Th eor etica l In vestiga tion of Sec-
on d -Or d er NLO P r op er ties. The molecular quadratic
hyperpolarizabilities (â) of the ligands and their com-
plexes (Tables 2 and 3) were measured in CHCl3 by the
solution-phase dc electric-field-induced second-harmonic
(EFISH) generation method,7 which provides informa-
tion on the quadratic hyperpolarizability through the
equation
the two-level model could not be applied safely in these
latter cases.21
3.1. P h en a n th r olin e Liga n d s. It was shown experi-
mentally that the two-level model applies well to the
ligand 4-trans-NC5H4CHdCHC6H4-4′-NMe2 and its
pseudolinear complexes with “cis-M(CO)2Cl” (M ) Rh-
(I), Ir(I)) or “fac-Os(CO)3Cl2”.3 Even if the molecular
geometries are not pseudolinear, this model appears to
be valid also for 5-X-1,10-phenanthrolines. For instance,
the EFISH âλ value of 5-methoxy-1,10-phenanthroline
was measured both at 1.06 µm and at 1.34 µm incident
wavelengths (Table 2) and it was found that â1.34
,
calculated with the two-level expression (eq 3) from
experimental EFISH â1.06, was in good agreement with
experimental EFISH â1.34 (3.7 × 10-30 compared to 4.0
× 10-30 cm5 esu-1, the difference being within experi-
mental error). Such an agreement would suggest that,
for this kind of ligand, other tensorial components of
the quadratic hyperpolarizability inaccessible to EFISH
measurements are probably not so relevant. In agree-
ment with this hypothesis, values of â0,TOT (the total
static quadratic hyperpolarizability at zero frequency
which takes into account all the âi tensor components)
and static EFISH â0 of the 5-X-1,10-phenanthrolines
investigated (X ) OMe, NMe2, trans-CHdCHC6H4-4′-
NMe2, trans,trans-(CHdCH)2C6H4-4′-NMe2; Table
4), calculated using the semiempirical MNDO-CPHF
method,3 are quite comparable. As expected, experi-
mental EFISH â1.34 values of these ligands increase with
the electron-donating strength of the X substituent and,
more significantly, with an increase of the π-conjugated
electron network linking the NMe2 group to the phenan-
throline ring (Table 2). This trend, occurring also with
4-X-pyridines,3 is in agreement with the red shift of λmax
of the ILCT transition (Table 2) major origin of the
second-order NLO response.8,9 Interestingly, while both
EFISH â1.34 and µâ0 of linear 4-trans-NC5H4CHd
CHC6H4-4′-NMe2 and of rigid 5-(trans-CHdCHC6H4-4′-
NMe2)-1,10-phenanthroline ligands are of the same
order of magnitude, the flexible 4-(trans-CHdCHC6H4-
4′-NBu2)-4′-CH3-2,2′-bipyridine ligand was reported to
have much lower values of EFISH â1.34 and µâ0 (Tables
2 and 3).6
Dipole moments and static EFISH â0 and in some
cases EFISH â1.34 of the various phenanthroline ligands
were calculated by the theoretical MNDO-CPHF method
(Table 4), which proved to be convenient to predict the
experimental dipole moments and the trend of the
quadratic hyperpolarizability of various 4-X-pyridines.3
The calculated dipole moments of the more π-delocalized
5-X-1,10-phenanthrolines (X ) trans-CHdCHC6H4-4′-
NMe2, trans,trans-(CHdCH)2C6H4-4′-NMe2) are in fair
agreement with the experimental values as reported for
4-trans-NC5H4CHdCHC6H4-4′-NMe2.3 However, when
X ) OMe, NMe2, the calculated values are lower than
the experimental ones, as in the case of 4-(trans-CHd
CHC6H4-4′-NBu2)-4′-CH3-2,2′-bipyridine (Table 4).6 MN-
DO-CPHF calculated EFISH â0 values of 5-X-1,10-
phenanthrolines (X ) NMe2, trans-CHdCHC6H4-4′-
NMe2, trans,trans-(CHdCH)2C6H4-4′-NMe2) and of the
flexible chelating ligand 4-(trans-CHdCHC6H4-4′-NBu2)-
4′-CH3-2,2′-bipyridine are underestimated (Table 4) by
γEFISH ) (µâvec/5kT) + γ0(-2ω; ω, ω, 0)
(2)
where µâvec/5kT is the dipolar orientational contribution
and γ0(-2ω; ω, ω, 0), a third-order term at frequency ω
of the incident radiation, is the electronic contribution
which is always considered negligible in these kinds of
ligands and their metal complexes.2-6 In the following,
âvec, which is the vectorial projection of the quadratic
hyperpolarizability tensor along the dipole moment
vector µ, will be reported as EFISH â1.34 or simply â1.34
when 1.34 µm is the fundamental incident wavelength.
Its second harmonic (λ/2 ) 670 nm) occurs quite far from
any significant absorption band (see Tables 2 and 3),
so that dispersive enhancements of the EFISH response
are minimized.20 We have shown in section 2 that both
the ligands and their metal complexes show one major
ILCT absorption band; therefore, the zero-frequency
quadratic hyperpolarizability â0 can be approximatively
calculated from experimental EFISH â1.34 using the two-
level model,16 which assumes that only a one-dimen-
sional charge-transfer process to one excited state is
involved in the nonlinear response. This model leads to
the expression
â0 ) âλ(1- (2λmax/λ)2)(1 - (λmax/λ)2)
(3)
where λ is the fundamental incident wavelength and
max is the value of the maximum of the absorption band
λ
of the ILCT transition (Tables 2 and 3), which, according
to solvatochromic investigations,8,9 is the major origin
of the second-order NLO response. However, the two-
level model applies quite well only to pseudolinear
molecular arrangements3,9,10 characterized by a charge
transfer process which occurs nearly along the dipole
moment axis. This arrangement does not occur in
tetrahedral Zn(II) or square-planar Pt(II) complexes
with two stilbazole ligands, where the dipole moment
axis is in a direction quite different from the ILCT
transition, located along the stilbazole ligand. Therefore,
(21) Brasselet, S.; Zyss, J . J . Nonlinear Opt. Phys. Mater. 1996, 5,
671.
(20) Orr, B. J .; Ward, J . F. Mol. Phys. 1971, 20, 513.

168 Organometallics, Vol. 21, No. 1, 2002
Roberto et al.
Ta ble 4. Com p a r ison of Exp er im en ta l a n d Th eor etica l (MNDO-CP HF Sem iem p ir ica l Meth od ) Va lu es of
Dip ole Mom en ts a n d EF ISH â0 Va lu es of 4-(tr a n s-CHdCHC6H4-4′-NBu 2)-4′-CH3-2,2′-bip yr id in e,
4-tr a n s-NC5H4CHdCHC6H4-4′-NMe2, a n d Va r iou s 5-X-1,10-p h en a n th r olin e (5-X-P h en ) Liga n d s a n d of
Selected Zn (II) Com p lexes
µ, D
MNDO-CPHF
â0 (â0,TOT),a 10-30 cm5 esu-1
molecule
exptlb
exptlb
MNDO-CPHF
4-(trans-CHdCHC6H4-4′-NBu2)-4′-CH3-2,2′-bipyridine
4-trans-NC5H4CHdCHC6H4-4′-NMe2
ZnCl2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2
Zn(CF3CO2)2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2
5-MeO-Phen
5.1c
3.9
8.4
10.5
4.0
3.8
2.8
4.6
16.7
20.5
2.7
2.7
4.9
5.0
8c
2.8d
10.4 (10.4)
15.0
e
35
48f
53f
17.0
2.4
5.7
26
44
2.0 (2.4)
0.57 (0.59)
6.4h (6.6)
8.7 (9.0)
5-Me2N-Phen
5-(trans-CHdCHC6H4-4′-NMe2)-Pheng
5-(trans,trans-(CHdCH)2C6H4-4′-NMe2)-Phen
4.9
4.9
a
b
â
0,TOT is the intrinsic quadratic hyperpolarizability at zero frequency.3 Experimental value in CHCl3; the value of â0 was extrapolated
d
from experimental EFISH â1.34 by the two-level model.16 c Reference 6. The calculated â1.34 value is 3.6 × 10-30 cm5 esu-1; for 4-(trans-
CHdCHC6H4-4′-NMe2)-4′-CH3-2,2′-bipyridine, calculated µ and â0 values are 4.0 D and 10.1 × 10-30 cm5 esu-1, respectively. e Reference
3. f Approximate value because the two-level model16 cannot be applied safely to this tetrahedral Zn(II) complex. By coordination to
g
h
“Zn(CH3CO2)2”, experimental and MNDO calculated dipole moments become 8 and 10.9 D, respectively. The calculated â1.34 value is 8.3
× 10-30 cm5 esu-1
.
a factor of 0.2-0.3, as in the case of 4-X-pyridines,3 when
compared to EFISH â0 extrapolated from experimental
EFISH â1.34 by the two-level approximation (eq 3).
However, the general trend is that expected, confirming
that the MNDO-CPHF method can be used only as a
preliminary way to define the trends of the second-order
NLO response of this kind of molecules.3 Strangely
enough the calculated EFISH â0 value of 5-OMe-1,10-
phenanthroline is quite in agreement with that extrapo-
lated from experimental EFISH â1.34 by the two-level
approximation (eq 3).
factor of EFISH â1.34 upon coordination of 5-(trans-CHd
CHC6H4-4′-NMe2)-1,10-phenanthroline to “Zn(CH3CO2)2”
(EF ) 1.9) is much smaller than that reported to occur
upon coordination of 4-(trans-CHdCHC6H4-p-NBu2)-4′-
CH3-2,2′-bipyridine (EF ) 8.2).6 This latter large effect
is related to the particularly low quadratic hyperpolar-
izability of the flexible bipyridine ligand (Table 2).
MNDO-CPHF calculations (Table 4) show that such a
low value must be attributed to the nonplanarity of both
the bipyridine and the NBu2C6H4 moieties due to steric
repulsions of the butyl groups. Upon cordination to Zn-
(II), the chelated bipyridine becomes rigid and more
planar; therefore, the ligand increases its π* acceptor
properties with a parallel significant increase of its
quadratic hyperpolarizability. In agreement with this
suggestion, MNDO-CPHF calculations show that by
substitution of butyl groups with methyl groups we have
an increased planarity of the ligand and an increase of
EFISH â0 by a factor of 3 (Table 4).
3.3. Com p lexes of Zn (II) a n d P t(II) w ith 4-tr a n s-
NC5H4CHdCHC6H4-4′-NMe2. Experimental EFISH
â1.34 values of complexes [ZnY2(4-trans-NC5H4CHd
CHC6H4-4′-NMe2)2] (Y ) Cl, CH3CO2, CF3CO2) and cis-
[PtCl2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2] are re-
ported in Table 3 along, for comparison, with those of
the ligand3 and its complexes with BF3,10 “cis-M(CO)2Cl”
(M ) Rh, Ir), and “cis-Ir(cyclooctene)2Cl”.3 As suggested
in section 3, the two-level model16 cannot be safely
applied to tetrahedral Zn(II) or square-planar Pt(II)
complexes: EFISH â1.34 of [Zn(CH3CO2)2(4-trans-NC5H4-
CHdCHC6H4-4′-NMe2)2], calculated by the two-level
equation (eq 3) from experimental EFISH â1.91, is quite
different from the experimental value (29 × 10-30
compared to 48 × 10-30 cm5 esu-1). Therefore, µâ0 values
are not reported in Table 3. Although usually EFISH
â1.34 of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 increases
by coordination to a metal center,3,10 strangely enough
EFISH â1.34 is almost unaffected by coordination to the
relatively soft “cis-PtCl2” Lewis acid whereas it increases
significantly upon coordination to the harder “ZnY2”
Lewis acids, the enhancement factor EF being a function
of the ancillary Y ligands (for Y ) Cl, EF ) 1.5; for Y )
CF3CO2, EF ) 1.8). However, EFISH â1.34 does not
increase upon coordination to the less hard “Zn(CH3-
CO2)2” Lewis acid, in agreement with an irrelevant red
shift of the ILCT transition of only 2 nm (when Y ) Cl,
3.2. Com p lexes of Zn (II) a n d Cd (II) w ith 5-X-1,-
10-p h en a n th r olin es. The second-order NLO response
of phenanthrolines with electron donor groups is domi-
nated by the ILCT transition, with the phenanthroline
π* system acting as an electron acceptor, as in para-
substituted pyridines.3 By interaction with a Lewis acid,
a red shift of λmax of the ILCT occurs, which is related
to an increase of the second-order NLO response.3,4
However, although ∆λmax of the ILCT of 5-(trans-CHd
CHC6H4-4′-NMe2)-1,10-phenanthroline is comparable
upon coordination to those of “Cd(CH3CO2)2” or “Zn(CH3-
CO2)2”, this is not reflected in the EFISH â1.34 value,
which is almost unaffected by coordination to the
relatively soft “Cd(CH3CO2)2” acid, whereas it is sig-
nificantly increased by coordination to the harder “Zn-
(CH3CO2)2” acid (Table 2). The enhancement factor (EF)
of EFISH â1.34 upon coordination to “Zn(CH3CO2)2” of
5-X-1,10-phenanthrolines is higher for the better donor
group NMe2 (EF ) 4.6) than for OMe (EF ) 3.2). It
becomes less and less relevant by increasing the length
of the π-delocalized bridge between the donor group
NMe2 and the phenanthroline moiety (Table 2), a trend
already reported to occur with 4-X-pyridines.3 When X
) NMe2, the enhancement factor of EFISH â1.34 (4.6) is
much lower than that (ca. 102) of 4-NMe2-C5H4N when
interacting with BF310 or even with Ir(I), Rh(I), and Os-
(II) metal centers.3 This difference can be explained by
taking into account that the EFISH â1.34 value of
5-NMe2-1,10-phenanthroline is much higher (by a factor
of 102) than that of 4-NMe2-C5H4N,3 as expected for the
larger π-conjugated electron network of the phenan-
throline system, which acts as a better π* acceptor of
the ILCT transition than the pyridine ring. Conse-
quently, the effect of coordination on the π* level of the
ligand is less significant. Interestingly, the enhancement

Effect of Coordination on Quadratic Hyperpolarizability
Organometallics, Vol. 21, No. 1, 2002 169
Ta ble 5. Absor p tion (λa ) a n d Em ission (λe) Solva toch r om ic Deter m in a tion of âCT a t 1.34 µm a n d
Com p a r ison w ith EF ISH â1.34
d
â
CT at 1.34 µm,d
EFISH â1.34,
molecule
λa (λe),a nm
fb
∆µeg,c D
10-30 cm5 esu-1
10-30 cm5 esu-1
4-trans-NC5H4CHdCHC6H4-4′-NMe2
[Zn(CH3CO2)2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2]
374 (454)
376 (460)
0.39
1.08
9.8
13.4
61
230
(132)f
175
(123)f
3.0
55e
48
cis-[PtCl2(4-trans-NC5H4CHdCHC6H4-4′-NMe2)2]
375 (455)
0.82
13.5
41
5-Me2N-Phen
281
328 (464)
0.31
0.14
1.9
5.0
7.2
6.3
9.3
[Zn(CH3CO2)2(5-Me2N-Phen)]
290
297
344 (524)
0.04
0.04
0.10
15.6
6.9
21.3
3.7
1.8
26.2
31.7
58
71
43
110
33
5-(trans-CHdCHC6H4-4′-NMe2)-Phen
381 (508)
419 (580)
410 (562)
432 (590)
0.23
0.39
0.29
0.37
15.1
7.5
6.5
41
77
46
[Zn(CH3CO2)2(5-(trans-CHdCHC6H4-4′-NMe2)-Phen)]
[Cd(CH3CO2)2(5-(trans-CHdCHC6H4-4′-NMe2)-Phen)]
[Zn(CH3CO2)2(5-(trans-(CHdCH)2C6H4-4′-NMe2)Phen]
10.1
112
a
b
Experimental values in CHCl3. f is the transition oscillator strength, obtained from the experimental integrated absorption coefficient
in CHCl3.8 c ∆µeg is the difference between excited- and ground-state molecular dipole moments, obtained from solvatochromic data.8 d In
CHCl3. e From ref 3. f Calculated using only the projection of µg along the charge-transfer direction.
CF3CO2, the red shifts are 36 and 46 nm, respectively).
As expected from the lack of enhancement of EFISH
(see section 2). The absorptions at higher energy (281
nm in the ligand and 290 and 297 nm in its Zn(II)
complex) show a smaller contribution to âCT, as expected
for πfπ* transitions of the unsaturated aromatic ring.
For nonchelated systems with two ligands, such as
complexes of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 with
“Zn(CH3CO2)2” and “cis-PtCl2”, âCT at 1.34 µm is much
higher than EFISH â1.34. In these latter molecules the
dipole moment is directed along the binary axis, thus
bisecting the angle formed by the directions of the two
equivalent ILCT charge transfers located on the two
stilbazole ligands (see section 1.2.1), while in the free
stilbazole ligand, which shows a far better agreement
between âCT at 1.34 µm and EFISH â1.34, the dipole
moment axis and the ILCT direction are substantially
coincident. However, when only the component of µg
along the charge-transfer direction is considered for both
“Zn(CH3CO2)2” and “cis-PtCl2”, the resulting value of âCT
becomes only slightly higher than twice that of the free
ligand itself. This result would confirm an irrelevant
enhancement of âCT of the free ligand upon coordination
to “Zn(CH3CO2)2” or “cis-PtCl2”, as expected from the
irrelevant red shift of the ILCT transition in both
complexes and the comparable value of EFISH â1.34 of
Zn(II) and Pt(II) complexes and of the free ligand.
â1.34, a similar irrelevant red shift of the ILCT transition
occurs by coordination to “cis-PtCl2” (∆λmax ) 1 nm). The
EFISH â1.34 enhancement factor due to coordination of
two stilbazole ligands to the harder Zn(II) centers is
generally lower than that originated by interaction of
one stilbazole ligand with “M(CO)2Cl” (M ) Rh, Ir)3 or
10
BF3 (Table 3). Interestingly, a positive effect due to
chelation can be pointed out. The µâ1.34 value (1420 ×
22
10-30 D cm5 esu-1
)
of [ZnCl2(4,4′-bis(trans-CHd
CHC6H4-4′-NBu2)-2,2′-bipyridine)] is much higher than
that of the related nonchelated complex [ZnCl2(4-trans-
NC5H4CHdCHC6H4-4′-NMe2)2] (706 × 10-30 D cm5
esu-1). This could be due to the more planar arrange-
ment of the chelated ligand.
4. Solva toch r om ic In vestiga tion of th e Secon d -
Or d er NLO Resp on ses. Solvatochromic studies are a
useful way to investigate the electronic origin of the
second-order NLO response of molecular species.8 The
quadratic hyperpolarizability âCT, obtained by solvato-
chromic measurements, is usually different from EFISH
âλ. Only when the charge-transfer transition involved
in the NLO response is close in direction to the axis of
the dipole moment may they be compared. For instan-
cee, for ligands such as 5-X-phenanthrolines and trans-
4-(dimethylamino)-4′-stilbazole8 and Ir(I) and Rh(I)
pseudolinear complexes of trans-4-(dimethylamino)-4′-
stilbazole9 solvatochromic âCT values at 1.34 µm are
comparable to, although slightly lower than, EFISH â1.34
values. With these limits in mind, we carried out a
solvatochromic investigation of the second-order NLO
response of some ligands and of their Zn(II), Pt(II), and
Cd(II) complexes, studied in this work (Table 5). First
we noted that for 5-Me2N-1,10-phenanthroline the major
contribution to âCT is given by the absorption band at
lower energy (328 nm in the free ligand and 344 nm in
its Zn(II) complex) characterized by the highest ∆µeg
(difference between excited- and ground-state molecular
dipole moments) value. This result is in agreement with
the assignment of this absorption to a ILCT transition
For free phenanthrolines (in particular 2-4) and their
complexes with “Zn(CH3CO2)2” and for the complex [Cd-
(CH3CO2)2(5-(trans-CHdCHC6H4-4′-NMe2)-1,10-phenan-
throline)] there is a very good correspondence between
â
CT at 1.34 µm and EFISH â1.34 (Table 5), in agreement
with a satisfactory coincidence of the direction of the
ILCT charge transfer and of the dipole moment axis,
already suggested in section 3.1. A detailed analysis
shows that when the donor group NMe2 is not directly
linked to the phenanthroline ring (ligands 3 and 4) the
appreciable increase of âCT upon coordination to “Zn-
(CH3CO2)2” is due to a high oscillator strength and to
the low energy of the ILCT transition rather than to
an increase of ∆µeg, which in contrast decreases from
15.1 to 7.5 D upon coordination (Table 5), while when
the donor group NMe2 is directly linked to the phenan-
throline ring (ligand 2), the relevant enhancement of
âCT is mainly due to a large increase of ∆µeg (from 5.0
(22) Hilton, A.; Renouard, T.; Maury, O.; Le Bozec, H.; Ledoux, I.;
Zyss, J . Chem. Commun. 1999, 2521.

170 Organometallics, Vol. 21, No. 1, 2002
Roberto et al.
to 21.3 D). In the Cd(II) complex the lack of increase of
EFISH â1.34 on coordination, despite the red shift of λmax
of the ILCT transition, is explained by a small decrease
of the energy of this latter transition and by the
relatively small increase of its oscillator strength, which
do not compensate the unexpected high decrease of ∆µeg
(from 21.3 to 6.5 D), therefore producing in total an
irrelevant enhancement of âCT upon coordination (Table
5).
âCT at 1.34 µm and EFISH â1.34 of tetrahedral “Zn-
(CH3CO2)2” and square-planar “cis-PtCl2” complexes
with two stilbazole ligands with a NMe2 group in the
para position are significantly different because the
dipole moment axis is not coincident with the direction
of the ILCT transition of the ligand. It follows that the
two-level model16 cannot be safely applied to EFISH â1.34
values of these complexes in order to evaluate their
static quadratic hyperpolarizability. However, the sol-
vatochromic analysis of the second-order NLO response
along the single ligand direction has confirmed the lack
of a significant increase upon coordination to Zn(II) or
Pt(II) of the quadratic hyperpolarizability of the single
stilbazole ligands, as suggested by EFISH measure-
ments and by the irrelevant red shift of the ILCT
transition. This is an unexpected result, because usually
there is a significant increase of EFISH â1.34 of this kind
of ligand upon coordination to hard or soft metal
centers.3,4,10 In comparison with the high increase of
EFISH â1.34 of related chelated bipyridines in a planar
arrangement, this is probably due to the nonplanar
arrangement of the two stilbazole ligands, as confirmed
by the identification of the best geometry by MNDO
calculations. Therefore, the combined use of EFISH and
solvatochromic measurements is an excellent way to
analyze the electronic and steric origin of the second-
order NLO response of these kinds of complexes.
Con clu sion
In this work we added further evidence that coordina-
tion to relatively hard tetrahedral Zn(II) centers pro-
duces, as already reported by Le Bozec et al. for
chelating bipyridine ligands,2,6 a significant enhance-
ment of âCT or EFISH â1.34 of chelating π-delocalized
nitrogen donor ligands such as 5-X-1,10-phenanthro-
lines bearing a NMe2 group linked to position 5 via a
π-delocalized network (ligands 2-4). As for η1 donor
pyridines3 this effect is more significant when the NMe2
group is directly linked to the ring bound to the acceptor
center. However, this is not a general trend: for
instance, such an enhancement does not occur when the
phenanthroline ligand 3 is coordinated to a softer center
such as Cd(II), although a red shift of the ILCT
transition of the free ligand, which is the origin of the
second-order NLO response, occurs. The enhancement
factor upon coordination of the quadratic hyperpolar-
izability of chelating ligands is lower for planar phenan-
throlines than for nonplanar and flexible bipyridines.6
We have attributed this significant difference to the
planarity and rigidity induced by coordination of bipy-
ridines, thus producing a better π* acceptor system for
the ILCT transition and therefore its significant shift
to lower energies.
Finally, the very stable Zn(II) complex with phenan-
throline 4 has a µâ0 value of 450 × 10-30 D cm5 esu-1
,
comparable to that of some organic materials such as
Disperse Red One (µâ0 ) 500 × 10-30 D cm5 esu-1
)
currently used in electrooptic polymers.23 This µâ0 value
is of the same order of magnitude found in the complex
of 4-trans-NC5H4CHdCHC6H4-4′-NMe2 with BF3,10
which, however, is rather unstable, or with some
expensive Rh(I), Ir(I), and Os(II) metal carbonyl cen-
ters.3
The behavior is different when two stilbazole ligands
with a NMe2 group in a para position are coordinated
to tetrahedral “Zn(CH3CO2)2” or square-planar “cis-
PtCl2” acceptor centers. In this case we could not observe
the expected significant increase of EFISH â1.34; even
Ack n ow led gm en t. This work was supported by the
Ministero dell’Universita` e della Ricerca Scientifica e
Tecnologica (Programma di ricerca MURST di tipo
interuniversitario, nell’area delle scienze chimiche (ex
40%) year 1997; title Molecole per Materiali Funzionali
Nanostrutturati) and by the Consiglio Nazionale delle
Ricerche (Progetto Finalizzato Materiali Speciali per
Tecnologie Avanzate II, year 1998; title Sintesi e
sviluppo di composti molecolari organometallici e di
coordinazione con proprieta` di ottica non lineare (NLO)
e con proprieta` elettriche anisotropiche e isotropiche). We
thank Dr. Andrea Busia, Dr. Maurizio E. Gallina, and
Dr. Francesca A. Porta for experimental help and Mr.
Pasquale Illiano and Mr. Americo Costantino for NMR
measurements.
-
when the ancillary CH3CO2 ligand is substituted by
more electronegative ancillary ligands such as Cl- or
CF3CO2-. With these latter ligands the increase of
EFISH â1.34 is much less significant than that occurring
in structurally related Zn(II) complexes with chelated
bipyridines.6 Obviously chelation seems to be a struc-
tural requirement to increase the second-order NLO
response. These unexpected results were interpreted by
a solvatochromic investigation. The lack of increase of
EFISH â1.34 or âCT of phenanthroline 3 upon coordina-
tion to “Cd(CH3CO2)2” is due to a very large decrease of
∆µeg, which is not compensated by a relatively small red
shift, and the increase of the oscillator strength of the
ILCT transition, while in the related Zn(II) complex a
less relevant decrease of ∆µeg is compensated by a more
significant red shift and by a high oscillator strength of
the ILCT transition.
OM010470H
(23) Singer, K. D.; Sohn, E.; King, L. A.; Gordon, H. M.; Katz, H.
E.; Dirk, P. W. J . Opt. Soc. Am. B 1989, 6, 1339.