ABAB homoleptic bis(phthalocyaninato)lanthanide(III) complexes: Original octupolar design leading to giant quadratic hyperpolarizability

Article abstract of DOI:10.1021/ic4031188

The octupolar cube, a T_d symmetry cube presenting alternating charges at its corners, is the generic point charge template of any octupolar molecule. So far, transposition into real molecular structures has yet to be achieved. We report here a first step toward the elaboration of fully cubic octupolar architectures. A series of octupolar bis(2,3,16,17-tetra(hexylthio) phthalocyaninato)lanthanide double-decker complexes [Pc₂Ln], Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5), are described, whose original three-dimensional structures display the required alternation of ABAB type for one face of the cube and the delocalization between the two rings approximating to the electronic interaction along the edges of the cube. Synthesis, X-ray crystal structure, and study of the optical properties and of the first molecular hyperpolarizability β are reported. The size of the lanthanide (III) central ion modulates the ring-to-ring distance and the degree of coupling between the two phthalocyanine rings. As a consequence, the optical properties of these octupolar chromophores and in particular the strong near-infrared absorption due to the intervalence transition between the two rings also depend on the central lanthanide (III) ion. The first oxidized and reduced states of the complexes, while keeping a similar octupolar structure, display considerably changed optical properties compared to the neutral states. Second-order nonlinear properties were determined by nonpolarized harmonic light scattering in solution at 1907 nm. Exceptionally large dynamic molecular first hyperpolarizabilities √ βHLS21907, among the highest ever reported, were found that showed a strong dependence on the number of 4f electrons.

Full text of DOI:10.1021/ic4031188

Article
pubs.acs.org/IC
ABAB Homoleptic Bis(phthalocyaninato)lanthanide(III) Complexes:
Original Octupolar Design Leading to Giant Quadratic
Hyperpolarizability
Mehmet Menaf Ayhan,†,∥ Anu Singh,⊥ Erwann Jeanneau,§ Vefa Ahsen,† Joseph Zyss,⊥
Isabelle Ledoux-Rak,⊥ Ayse Gul Gurek,† Catherine Hirel,*,† Yann Bretonniere,*,∥ and Chantal Andraud∥
̧
̀
̈ ̈
†Department of Chemistry, Gebze Institute of Technology, P.O. Box 141, 41400 Gebze, Kocaeli, Turkey
§Centre de Diffractomet
rie Henri Longchambon, Universite Lyon I, 43 boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex,
France
́
́
⊥ENS-Cachan, Institut D’Alembert, CNRS UMR 8537, ENS Cachan, 61 avenue du Pres
́
ident Wilson, 94235 Cachan Cedex, France
∥Laboratoire de Chimie de l’ENS de Lyon, CNRS UMR 5182, Universite
́
Lyon I, ENS de Lyon, 46 allee d’Italie, 69364 Lyon cedex
́
07, France
S
* Supporting Information
ABSTRACT: The octupolar cube, a Td symmetry cube presenting
alternating charges at its corners, is the generic point charge template of
any octupolar molecule. So far, transposition into real molecular
structures has yet to be achieved. We report here a first step toward the
elaboration of fully cubic octupolar architectures. A series of octupolar
bis(2,3,16,17-tetra(hexylthio)phthalocyaninato)lanthanide double-deck-
er complexes [Pc2Ln], Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5),
are described, whose original three-dimensional structures display the
required alternation of ABAB type for one face of the cube and the
delocalization between the two rings approximating to the electronic
interaction along the edges of the cube. Synthesis, X-ray crystal
structure, and study of the optical properties and of the first molecular
hyperpolarizability β are reported. The size of the lanthanide (III)
central ion modulates the ring-to-ring distance and the degree of
coupling between the two phthalocyanine rings. As a consequence, the optical properties of these octupolar chromophores and in
particular the strong near-infrared absorption due to the intervalence transition between the two rings also depend on the central
lanthanide (III) ion. The first oxidized and reduced states of the complexes, while keeping a similar octupolar structure, display
considerably changed optical properties compared to the neutral states. Second-order nonlinear properties were determined by
nonpolarized harmonic light scattering in solution at 1907 nm. Exceptionally large dynamic molecular first hyperpolarizabilities
2
⟨βHLS
⟩
1907 , among the highest ever reported, were found that showed a strong dependence on the number of 4f electrons.
became known as octupolar and more generally multipolar
molecules.2 Octupolar molecules in which symmetry con-
straints impose a cancellation of the dipole moment have
considerably extended the possibilities of molecular design by
enlarging the dimensionality of the charge transfer. At a
structural level it was shown that the most generic template for
octupolar molecules was a cube of symmetry Td presenting
alternating charges at the corners (Figure 1).2,3 More specific
arrangements eventually with reduced dimensionality can be
deduced from this basic cubic point charge template by
projecting the charges on the center of the cube along the C3
diagonal axis or by fusion of one type of charge in the
barycenter of the cube. The transposition of these point charge
INTRODUCTION
■
The design of second-order nonlinear optical molecules mainly
focused on one-dimensional (1D) dipolar push−pull molecules
with a charge transfer polarized along the molecular axis z.
Extensive molecular engineering around this [donor]−[elec-
tronic conjugated pathway]−[acceptor] structure combining
the increase of the donor or acceptor strength, the modification
of the electronic structure of the conjugated path, or the
adequate combinations of the end donor and acceptor groups
with the conjugated bridge led to very large molecular factors of
merit μβ.1 This restricted template, however, where only the
on-axis component βzzz is nonzero, failed to take into account
the three-dimensional (3D) nature of the hyperpolarisability
tensor β. Studies based on group theory and quantum
mechanics have helped set conceptual foundations to the
domain of nondipolar molecules for nonlinear optics, which
Received: December 24, 2013
Published: March 5, 2014
© 2014 American Chemical Society
4359
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
compounds has been the subject of long debate. Electron
spin resonance and theoretical calculations on lutetium(III)
bis(phthalocyaninate) correlated to absorption spectroscopy
experiments18 demonstrated the existence of an unpaired
electron completely delocalized over both rings. The now
commonly accepted structure of lanthanide bis-
(phthalocyaninates) is a neutral radical in which one macro-
cycle is twice negatively charged, whereas the other is a doubly
deprotonated π radical [Ln(Pc)(Pc•)] where the term Pc
stands for doubly deprotonated phthalocyanine ring. The
unpaired electron is delocalized over the two cycles on the
semi-occupied SOMO.18b,19
X-ray diffraction studies on [NdIII(Pc)(Pc•)] and [LuIII(Pc)-
(Pc•)]20 showed that the two rings in the sandwich complex
were staggered by around 45°, giving a noncentrosymmetric
structure of D4d symmetry. We put forward that lanthanide
complexes with crosswise ABAB macrocycles featuring
alternating electron donor and electron acceptor groups could
give an almost exact representation of the octupolar cube
(Figure 1), the delocalization between the two rings
approximating to the electronic interaction along the edges of
the cube. This idea was illustrated in a preliminary
communication,21 in which we described the synthesis, the
molecular structure, and the measure of the molecular first-
order hyperpolarizability β of such a complex, with the example
of the lutetium sandwich complex of 2,3,16,17-tetra(hexylthio)-
phthalocyanine, an ABAB crosswise phthalocyanine, bearing
electron donor (thioalkyl) groups on two opposite rings.
Although we were aware that there were no real electron
accepting groups, the required alternation of ABAB type for
one face of the cube was present as thioalkyl, and hydrogen
atoms do not have the same electronic effect. An exceptionally
high value of 5750 × 10−30 esu for the molecular first-order
hyperpolarizability β was measured at 1907 nm, stressing that
this molecular design could lead to very efficient and original
octupolar molecules.
In this article, we wish to extend the chemistry to rare earth
ions of representative ionic size (Nd, Eu, Dy, Y) chosen to
provide a variation in the degree of coupling between the
phthalocyanine rings and as a consequence in the energy of the
intervalence transition, to gain a better understanding of the
second-order nonlinear optical properties observed. Thus, the
synthesis, the X-ray crystal structure, and the study of the
optical properties and of the first molecular hyperpolarizability
β of various (2,3,16,17-tetra(hexylthio)phthalocyaninato)-
lanthanide complexes [LnIII(Pc)(Pc•)], Ln = Nd (1), Eu (2),
Dy (3), Y (4), and Lu (5) are reported. The first oxidized and
reduced states of the complexes presenting very different
optical properties than the neutral states are also investigated.
Figure 1. Bis(phthalocyanines)lanthanide complexes [LnIII(Pc)(Pc•)]
as point-charge template for octupole.
templates into real structures has given rise to a realm of
original octupolar molecules4 that includes purely organic D3h
tri- or hexa-substituted aryl compounds,5 propeller-like D3
molecules built on a functionalized trivalent carbon5c,6 or
nitrogen atom,7 tetrahedral (Td) tetra-substituted carbon,
silicon, and phosphorus derivatives,8 D2d functionalized
biphenyl derivatives,9 or boron subphthalocyanines (C3v),10 to
cite a few. Coordination chemistry further extended the
possibilities of engineering nondipolar architectures. Following
the early work of Zyss on D3 [Ru(bpy)3]2+,11 a large number of
metal complexes and organometallic octupolar structures have
been described. Good examples of these are covalent
tetrahedral (Td) tin derivatives,12 D2d or D3 metal complexes
of bipyridyl ligands13 or lanthanide tris(bipyridine) or
dipicolinate complexes,14 in which not only the template effect
of the metal ion was exploiting to create noncentrosymmetric
structures, but the intrinsic properties of the metal ion and of
the metal-to-ligand specific transition were taken advantage of
to tune the electronic and optical properties of the final object.
Very recently, Ru(II)terpyridyl complexes combined with
zinc(II) porphyrinato units were used to create octupolar
molecules having impressive βHRS (>1000.10−30 esu) values at
telecommunication-relevant wavelength (1300 nm).15 How-
ever, despite the huge number of octupolar structures reported
the translation of the ideal figure into a real “cubic” molecule
(even if deformed) with eight alternated charges at the corner
and delocalization of the charges between the higher and lower
planes has yet to be realized. Taking advantage of the
unconventional through-space intramolecular charge transfer
provided by the paracyclophane framework, Bazan et al.16
described the closest example thus far. However, the cubic
scheme was not complete as only half a cube has been obtained
with substitution consisting of four donor (respectively
acceptor) groups on the paracyclophane moiety only.
EXPERIMENTAL SECTION
■
General. Starting materials were purchased from Aldrich, Fluka,
and Alfa and were used without further purification unless otherwise
stated. Residual water in lanthanide acetate salts Ln(OAc)3 was
removed by heating at 120 °C under vacuum (10−2 mmHg) for one
week. 2,3,16,17-tetra(hexylthio)phthalocyanine was prepared as
previously described.21 1H NMR spectra were recorded on a Bruker
Avance 300 spectrometer. High-resolution mass-spectrometry (HR-
MS) experiments were performed at the center commun de
With various ions and especially with lanthanide(III) ions,17
phthalocyanines form sandwich-type complexes in which the
metal ion holds two macrocycles closer than their van der
Waals distance, resulting in strong π−π interactions and
through-space 3D delocalization. The nature of these
́ ́
spectrometrie de masse from Universite Claude Bernard Lyon I on
a MicroTof-Q Bruker mass spectrometer. Electronic absorption
spectra were recorded on a Jasco V770 spectrophotometer.
(2,3,16,17-Tetra(hexylthio)phthalocyaninato)lanthanide
[LnIII(Pc)(Pc•)], Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5). A
4360
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
Scheme 1. Synthesis of Complexes 1−5: (i) NH3(g), NaOMe, MeOH, reflux, 5 h; (ii) PCl5, o-dichlorobenzene, 110 °C, 1 week;
(iii) (a) Et3N, THF, 0 °C, 12 h; (b) NaOMe, hydroquinone, THF, reflux, 12 h; (iv) Ln(OAc)3·4H2O, DBU, 1-
chloronaphthalene
mixture of dried Ln(OAc)3 (0.08 mmol), 1,8-diazabicyclo[5.4.0]-
undec-7-ene (DBU) (1 mL), and 2,3,16,17-tetra(hexylthio)-
phthalocyanine (H2Pc, 160 mg, 0.16 mmol) in 1-chloronapthalene
(3 mL) was refluxed under nitrogen for 4 h to give a dark-green
solution. After cooling to room temperature, hexane was added
dropwise to precipitate the product. The precipitate was filtered
through a filter paper and dissolved in CH2Cl2. Purifications by
column chromatography on silica gel first and then by preparative thin-
layer chromatography eluting with dichloromethane/hexane 1/1 (v/v)
afforded the pure products as green solid. Yield: 105 mg for 1 (63%),
106 mg for 2 (63%), 91 mg for 3 (54%), 98 mg for 4 (60%), and 102
mg for 5 (60%). HR-MS (ESI-QTOF) calcd for C112H128NdN16S8:
2094.7345; found: 2094.7421. Calcd for C112H128EuN16S8: 2105.7482;
found: 2105.7495. Calcd for C112H128DyN16S8: 2116.7586; found:
2116.7526. Calcd for C112H128YN16S8: 2041.7327; found: 2041.7235.
Calcd for C112H122LuN16S8 2127.7676; found 2127.7665.
Crystallography. Single crystals of 1−5 were obtained by slow
diffusion of methanol into concentrated solution in chloroform.
Suitable crystals were selected and mounted on a Gemini kappa-
geometry diffractometer (Agilent Technologies UK Ltd.) equipped
with an Atlas charge-coupled deviced (CCD) detector and using Mo
radiation (λ = 0.71073 Å). Intensities were collected at 100 K by
means of the CrysalisPro software. Reflection indexing, unit-cell
parameters refinement, Lorentz-polarization correction, peak integra-
tion, and background determination were carried out with the
CrysalisPro software.22 An analytical absorption correction was applied
using the modeled faces of the crystal.23 The structures were solved by
direct methods with SIR97,24 and the least-squares refinement on F2
was achieved with the CRYSTALS software.25 All non-hydrogen atoms
were refined anisotropically. The hydrogen atoms were all located in a
difference map, but were then repositioned geometrically. The H
atoms were initially refined with soft restraints on the bond lengths
and angles to regularize their geometry (C···H in the range of 0.93−
0.98 Å) and Uiso(H) (in the range of 1.2−1.5 times the Ueq value of the
parent atom), after which the positions were refined with riding
constraints. CCDC 913758 (1), 913756 (2), 913755 (3), and 913759
(4) contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif. The structure of complex 5 was already reported in a previous
paper (CCDC 832167).21
Reduction and Oxidation. Reduced species 1b−5b were
obtained in solution by adding NaBH4 (one crystal) to a solution of
complex 1−5 in THF. For the harmonic light scattering (HLS)
experiments, solutions of complexes 1−5 in THF were prepared. Then
one crystal of NaBH4 was added. When the color changed from green
to blue, the solution was diluted with chloroform to the working
concentration and another crystal of NaBH4 was added. The solutions
were filtered through a 0.2 μm membrane filter just before
measurement. Oxidized species 1c−5c were obtained in solution in
chloroform by adding a diluted solution of bromine in chloroform.
Hyper-Rayleigh Scattering. A 10−5 M solution in CHCl3 or 10%
THF in chloroform were prepared as described above and filtered
through a 0.2 μm membrane filter to remove any particles or dust.
Fluorescence quartz cells were used for the experiments. Sealable cells
purged with argon were used to measure the reduced and oxidized
species. The details of the method as well as the experimental set up
have been presented previously.13d
RESULTS AND DISCUSSION
■
Synthesis and X-ray Crystal Structure. 2,3,16,17-Tetra-
(hexylthio)phthalocyanine was obtained in 8% yield by the
selective condensation of 5,6-bis(hexylthio)isoindoline-1,3-
diimine with 1,3,3-trichloroisoindolenine, followed by oxidation
with hydroquinone, as reported.21 This method was reported to
be regioselective toward the formation of the crosswise ABAB
phthalocyanine.26 In our hands however significant amounts of
2,3,9,10,16,17-hexa(hexylthio)phthalocyanine (AB3) and traces
of the symmetric derivative 2,3,9,10,16,17,23,24-octa-
(hexylthio)phthalocyanines (B4) were produced that neces-
sitate thorough chromatographic separation from the desired
product. The bis(phthalocyanine)lanthanide complexes 1−5
4361
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
Figure 2. (a) Complex 1 viewed along the b axis (H atoms and disordered C atoms on the hexyl chains are omitted for clarity). (b) Coordination
polyhedron around the Nd3+ ion. The Niso atoms N(2) and N(6) (N atom of the unsubstituted isoindole ring and of the isoindole ring bearing the
electron donating groups, respectively) are pictured in different colors to show the parallel between the coordination polyhedron and the octupolar
cube of Figure 1.
Table 1. Selected Crystallographic and Refinement Data for [LnIII(Pc)(Pc•)]; Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5)
1
2
3
4
5
formula
C112H128NdN16S8·
2(CHCl3)
C112H128EuN16S8·
2(CHCl3)
C112H128DyN16S8·
2(CHCl3)
C112H128YN16S8·
2(CHCl3)
C112H128LuN16S8·
2(CHCl3)
MW (g mol−1
crystal system
space group
a (Å)
)
2337.88
orthorhombic
Fddd
2345.60
orthorhombic
Fddd
2356.14
orthorhombic
Fddd
2282.54
orthorhombic
Fddd
2352.48
orthorhombic
Fddd
23.275(1)
29.404(1)
31.015(2)
90
23.289(1)
29.342(2)
31.977(1)
90
23.1443(2)
29.482(3)
32.219(2)
90
23.1860(10)
29.276(2)
32.081(2)
90
23.206(5)
29.202(2)
31.760(2)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
90
90
90
90
90
90
90
90
90
90
21910(2)
8
21851(2)
8
21984(3)
8
21776(2)
8
21523(5)
8
Z
T (K)
Dx (g cm−3
100
100
100
100
100
)
1.417
1.426
1.424
1.392
1.452
R[F2 > 2σ(F2)]
wR(F2)
0.086
0.067
0.055
0.079
0.059
0.206
0.127
0.127
0.223
0.170
GOF
0.84
1.07
0.94
0.99
0.96
p
min/pmax (e Å−3
)
−4.49/4.50
6931
−4.17/2.59
6887
−2.81/1.73
6733
−1.99/1.90
6973
−1.59/2.10
6787
No. reflections
No. parameters
400
401
400
400
400
were obtained in good yield from the ABAB phthalocyanine
and lanthanide acetate in refluxing 1-chloronaphthalene, using
DBU as a base (see Scheme 1).
Eight double-decker molecules [LnIII(Pc)(Pc•)] as well as 16
disordered solvents molecules (CHCl3) constitute the unit cell.
The double-decker [LnIII(Pc)(Pc•)] entity consists of two
slightly domed phthalocyanine rings flanking a central Ln3+ ion.
The two macrocycles are staggered with respect to each other,
with a staggering angle α of around 45°, which is remarkably
constant from Nd to Lu. The entity [LnIII(Pc)(Pc•)] is of
approximate symmetry D2 and is therefore chiral. The two
enantiomers are present in the crystal structure (Supporting
Information, Figure SI-1). The Ln3+ ion is octacoordinated by
the eight nitrogen atoms of the isoindole Niso (N(2) and
N(6)), the polyhedron of the Ln coordination sphere being
X-ray quality crystals of compounds 1−5 can be isolated by
letting methanol slowly diffuse in a concentrated chloroform
solution, at room temperature for a few days. The X-ray
crystallographic analysis revealed that all the complexes are
isostructural and identical to the lutetium complex previously
described; accordingly, only the crystal structure of 1 (Nd) is
shown in Figure 2. Selected crystallographic and refinement
data are summarized in Table 1, and important structural
parameters are given in Table 2.
4362
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
Table 2. Selected Structural Data for [LnIII(Pc)(Pc•)]; Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5)
1
2
3
4
5
a
r8, pm
110.9
106.6
102.7
101.9
97.7
2.656
b
di, Å
2.956
2.841
2.775
2.747
c
c
Ln−N(2), Å
Ln−N(6), Å
αN(2), deg
2.477(4)
2.466(4)
45.30
2.445(4)
2.428(4)
45.30
2.402(3)
2.411(3)
45.95
2.400(3)
2.399(3)
45.64
2.3666(3)
2.356(3)
45.37
cd
,
ce
,
αN(6), deg
44.14
44.10
43.55
44.05
44.43
a
b
Taken from reference 28. Distance between the upper and lower planes formed by the four isoindole nitrogen atoms Niso (N(2) and N(6)) of
c
each phthalocyanine ring. N(2) and N(6) are the two isoindole nitrogen atoms Niso. N(2) is the Niso of the isoindole moiety bearing the two
thioether groups. N(6) is the Niso of the isoindole moiety bearing the hydrogen atoms. Angle between the two planes connecting the metal ion Ln
and the two N(2) of each phthalocyanine ring. Angle between the two planes connecting the metal ion Ln and the two N(6) of each
d
e
phthalocyanine ring.
Figure 3. (a) Scatter plot of the ring-to-ring distance (di) and of the wavenumber of the intervalence band (IV) against the ionic radius of the eight-
coordinated Ln3+ ion (r8). (b) Scatter plot of the red vibronic (RV) band and Q band (Q) against the ionic radius of the eight-coordinated Ln3+ ion
(r8). (c) Electronic absorption spectra of the neutral [LnIII(Pc)(Pc•)] (Ln = Nd, Eu, Dy, Y, Lu) species 1−5.
best described as a slightly distorted square antiprism. The
increase in the size of the metal ion from Lu3+ to Nd3+ is
reflected in the increase of the Ln−Niso distances (from
2.356(3) Å for the shortest distance Lu−N(6) to 2.477(4) Å
for the longest distance Nd−N(2)), which is very similar to
what was observed for eight-coordinated lanthanide ions in a
4363
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
Table 3. Electronic Absorption Data (in CHCl3) and Dynamic Hyperpolarizabilities
Soret band
BV band
Q band
RV band
IV band
a
bc
,
complex
λmax, nm
λmax, nm
λmax, nm
λmax, nm
λmax, nm
βxyz
βHLS
1
2
3
4
5
328
329
329
327
327
575
563
556
554
545
699
691
687
686
682
922
928
934
934
938
1843
4300
7180
7435
3980
7620
3250
5430
5620
3010
5760
1650
1459/1598
1455/1586
1431
For an effective octupolar D2 molecular symmetry: βHLS = [⟨βZZZ2⟩ + ⟨βYZZ2⟩] = (4/7)β 2. βHLS, (×10−30 esu) determined by hyper-Rayleigh
a
2
b
xyz
experiments at 1907 nm in CHCl3 for the mixed-valent complexes [LnIII(Pc)(Pc•)], where Ln = Nd (1), Eu (2), Dy (3), Y (4), and Lu (5).
c
Reference ethyl violet, β = 170 × 10−30 esu at 1907 nm. Estimated error: 15%.
similar structure.27 The interplanar distance di, the distance
between the two mean planes formed by the four isoindole
nitrogen of each macrocycle, increase monotonically with the
metal size from 2.656 Å for complex 5 (Lu) to 2.956 Å for
complex 1 (Nd). The linear correlation between the di distance
and the ionic radius of eight-coordinated metal ion (r8)
illustrated in Figure 328 was already evidenced in
diphthalocyaninato(2−) metalate complexes of rare earth as
well as bismuth, antimony, and indium K[MIII(Pc)2] (M = La,
Ce, Pr, Sm, Bi, Sb, In).29 The crystal packing of complexes 1−5
consist of alternating layers of molecules [LnIII(Pc)(Pc•)] along
the packing axis b (as shown in Supporting Information, Figure
S.I.-2). As expected the distance between two consecutive
layers decreases from 1 (Nd) to 5 (Lu) according to the
decrease in the ionic radii. On the other hand, the Ln−Ln
distance of two enantiomeric molecules [LnIII(Pc)(Pc•)] from
two consecutive layers, as well as the Ln−Ln distance between
two molecules within the same layer, is not directly related to
the size of the metal ion, suggesting that other parameters than
size govern the crystal packing. The disordered solvent
molecules incorporated into the layer of molecules occupying
the empty spaces between two disordered alkyl chains (C6H13)
of two adjacent molecules.
Spectroscopic Properties. The electronic absorption
spectra of the double-decker complexes 1−5 were recorded
in CHCl3. The data are summarized in Table 3. Figure 3
compares the absorption spectra in the overall 300−2000 nm
range. The electronic absorption of lanthanide bis-
(phthalocyaninates) has been extensively studied and inter-
preted in terms of mixed-valent compounds. The spectra
obtained for compounds 1−5 are typical of such compounds.
The Soret bands appear at 328 nm, whereas the intense Q
bands are observed in the range of 680−690 nm with
vibrational shoulder at 612−618 nm. The Q band is actually
the sum of the overlapping Q band of each Pc2− and (Pc•)−
species.19b A small hypsochromic shift is observed for the Q
band when the ionic radius of the cation decreases from Nd (r8
= 110.9 pm) to Lu (r8 = 97.7 pm). Compared to simpler
unsubstituted [LnIII(Pc)(Pc•)], the presence of eight electron-
donating thioether groups induces a small red shift of all bands.
The main characteristic of the electronic absorption spectra of
lanthanide bis(phthalocyaninates) is a broad absorption band in
the near-infrared region ascribed to an intervalence transition
(IV) between the delocalized b1→a2 HOMO and SOMO.17,19a
For complexes 1−5, this intervalence transition appears as a
more or less structured band with maxima ranging from 1431
nm for Lu to 1843 nm for Nd. As for the Q band, a linear
correlation also exists between the absorption maxima and the
ionic radii r8 (as shown in Figure 3), and therefore the ring-to-
ring separation distance di, confirming what was previously
reported for simple phthalocyaninato,19b porphyrinato,30 or
mixed porphyrinato−naphthalocyaninato31 lanthanide double-
decker complexes. Transition 1eg(π) →a1u(π*) from the
SOMO-to-LUMO (RV) appears around 925−940 nm and
finally transition (BV) from deeper π-bonding 2eg(π) level to
the SOMO, 2eg(π)→a1u(π*), gives a band at around 540−560
nm. For these two bands, a small bathochromic shift and a
small hypsochromic shift are observed, respectively, when the
ring-to-ring distance decreases (Figure 3).
Oxidation and Reduction. The redox properties of
double-decker phthalocyanine lanthanide complexes have
been extensively studied, and seven32 oxidation states
[LnIIIPc2]z, with z ranging from −4 to +2, and six one-electron
redox-reactions, all involving electrons of the macrocyclic rings,
have been characterized in solution.33 Besides the neutral
mixed-valent compound [LnIII(Pc)(Pc•)], at least two redox
states, the one-electron reduced anion [LnIII(Pc)2]− and the
one-electron oxidized cation [LnIII(Pc•)2]+ are easily accessible
chemically. If the structure of the one-electron oxidized cation
[LnIII(Pc•)2]+ has never been reported, quantum chemical
study on [TbIII(Pc•)2]+ based on density functional theory
(DFT) revealed a shortened ring-to-ring distance but a
conserved staggered conformation on the two phthalocyanine
rings.34 Detailed studies were realized on the one-electron
reduced anion structure in solution and in the solid state.29,35
Koike et al. demonstrated from crystallographic studies of
(nBu4N)[MIII(Pc)2] complexes that, with increasing interplanar
distance, the eclipsed conformation of the two cycles would be
favored and that the skew angle can even reach 0° if the
interplanar distance was longer than the van der Waals
distance.35a Analysis of the paramagnetic shifts in proton
NMR, correlated with UV−visible spectroscopy, revealed
highly symmetrical structures in solution for the nonradical
anions of simple phthalocyanine [LnIII(Pc)2]−, probably a
staggered form belonging to the D4d point group.35c Homborg
et al. showed not only that with large ion size such as Nd3+,
large di distances were sufficient to repulse the steric ligand−
ligand interaction but also that the equilibrium between the
staggered and the eclipsed conformation was governed by
subtle equilibrium forces within the crystal packing.29 In
solution then, one can reasonably assume that the octupolar D2
structure of the neutral species would be conserved at least for
the one-electron reduced species. However, in terms of
octupolar character, a possible eclipsed conformation would
be highly favorable if the substituted isoindole rings are
confronted with unsubstituted ones, because this arrangement
would then lead to pure Td symmetry.
4364
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
Figure 4. (a) Scatter plot of the wavenumber of the band Q band (Q1 and Q2) and of the Q band split against the ionic radius of the eight-
coordinated Ln3+ ion (r8). (b) Electronic absorption spectra of the reduced [LnIII(Pc)2]− (Ln = Nd, Eu, Dy, Y, Lu) species 1b−5b with magnified
view of the 850−1300 nm range.
The reduced species [LnIII(Pc)2]− 1b−5b (Ln = Nd, Eu, Dy,
Y, or Lu), in which the two phthalocyanine rings are twice
negatively charged, were obtained in THF or 10% THF−
CHCl3 solution by reduction of the neutral compounds 1−5
with an excess of NaBH4. The solution turned from green to
deep blue. The spectral changes associated with chemical
reduction are illustrated in Figure 4, and the data are
summarized in Table 4. The main spectral feature is the
disappearing of all the electronic transitions associated with the
phthalocyanine π radical. Therefore the spectra only display
three bands: the B band around 335−340 nm, the Q band
around 600−700 nm, and a very weak band appearing in the
foot of the Q band (Figure 4). The Q band, structured and split
into two components, the main band Q1 and the sideband Q2,
as already described for simple [LnIII(Pc)2]− anions,36 also
presents a vibronic shoulder at higher energy. As observed for
the neutral species, there is a hypsochromic shift of the main Q1
band and a small bathochromic shift of the side Q2 band
correlated with the ionic radius of the central cation. As a
consequence, the splitting of the Q band also correlates well
with the interplane distance36 and is much more prominent for
the late lanthanide. It has been attributed to increased exciton
interaction occurring between the two cycles when the ionic
radii decrease. For lighter lanthanide, not only the interaction
between the two cycles decreases, but the conformation
adopted by the two cycles probably changes.
Upon addition of sodium borodeuteride (NaBD4) as a
1
reducing agent, well-resolved H NMR spectra were obtained
for the reduced form of bis(phthalocyaninato)lanthanide(III)
complexes of the diamagnetic cation Y3+ (4b) and Lu3+ (5b) in
which both macrocyclic ligands become diamagnetic dianions.
The two spectra are identical. For these homoleptic
compounds, the nonperipheral and peripheral protons of the
unsubstituted isoindole ring resonate as four sets of signals as
4365
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
multiplets, at δ = 8.03 and 8.08 ppm for the peripheral and 8.88
and 8.95 ppm for the nonperipheral protons, respectively,
whereas two well-resolved singlets at δ = 8.83 and 8.93 ppm
were observed for the nonperipheral protons of the substituted
ring. This indicates that a 2-fold symmetry is present in
solution, confirmed by the observation of two sets of signals for
each CH2 (and the CH3) of the thiohexyl chains. For the
paramagnetic Eu3+ complex (2b) the aromatic signals become
broader and are shifted to lower field at 9.07 pm for the
peripheral and in the range from 11.09 to 11.27 ppm for all the
nonperipheral protons. The SCH2 protons resonate as two sets
of signals at 4.46 and 4.65 ppm. Unfortunately, for the
complexes of the paramagnetic cations Nd3+ 1b and Dy3+ 3b,
no proper spectra could be obtained.
Table 4. Electronic Absorption Data (in THF) and Dynamic
Hyperpolarizabilities
Q band
Soret
band
Q1
Q2
c
ab
,
complex λmax, nm λmax, nm λmax, nm splitting, cm−1 βxyz
βHLS
1b
2b
3b
4b
5b
341
340
338
338
337
647
643
639
638
632
689
697
705
707
718
942
1193
1463
1530
1892
3220
4880
5570
3250
6070
2430
3690
4210
2460
4590
a
βHLS, (×10−30 esu) determined by hyper-Rayleigh experiments at
1907 nm in a mixed solvent system of 10% THF in CHCl3 for the
reduced complexes [LnIII(Pc)2]−, where Ln = Nd (1b), Eu (2b), Dy
(3b), Y (4b), and Lu (5b). Reference ethyl violet, β = 170 × 10−30 esu
The oxidized species 1c−5c (Ln = Nd, Eu, Dy, Y, or Lu)
were obtained in CHCl3 solution by oxidation of the neutral
compounds 1−5 with a slight excess of bromine (Br2), and the
solution turned from green to brown. The most probable
b
c
at 1907 nm. Estimated error: 15%. For an effective octupolar D2
molecular symmetry: βHLS = [⟨βZZZ2⟩ + ⟨βYZZ2⟩] = (4/7)β
.
2
2
xyz
Figure 5. (a) Scatter plot of wavenumber of the red vibronic (RV) band, blue vibronic (BV) band, and Q (Q) band against the ionic radius of the
eight-coordinated Ln3+ ion (r8). (b) Electronic absorption spectra of the oxidized [LnIII(Pc•)2]+ (Ln = Nd, Eu, Dy, Y, Lu) species 1c−5c.
4366
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
structure of the one-electron oxidized species is a complex
formed by two π-radical rings (Pc•)−.27a The absorption spectra
of complexes 1c−5c are shown in Figure 5 and are summarized
in Table 5. Besides the B band at 350−400 nm, the strongest Q
First of all, note that the values, ranging from 3010 × 10−30 esu
for 4 (Y) to 5760 × 10−30 esu for 5 (Lu), are exceptionally large
for hyperpolarizabilities at 1907 nm. The neodymium complex
1 also presents the strongest absorption at 1907 nm but the
weakest hyperpolarizabilty measured for the series at that
wavelength. This suggests that the intervalence transition has
little influence on the quadratic response as it was already
postulated by Shirk et al. for the third-order nonlinear
properties of simple [LnIII(Pc)(Pc•)].39
Table 5. Electronic Absorption Data (in CHCl3) and
Dynamic Hyperpolarizabilities
Soret band BV band
Q band
RV band
c
bb
,
λmax, nm λmax, nm λmax, nm
βxyz
βHLS
The structures of complexes 1−5 in solution were not
a
a
a
a
a
1
1c
2c
3c
4c
5c
n.d.
n.d.
n.d.
n.d.
n.d.
625
614
607
603
592
751
740
733
732
723
1140
1069
1023
1010
967
3690
6020
6200
3700
6250
2790
4550
4690
2800
4720
determined; however, H NMR studies on homoleptic and
heteroleptic diphthalocyaninato(2−) metalate complexes
[LnIII(Pc)2]− of rare earth have shown that the solid-state
structure was preserved in solution at least for the smaller
ions35c even if the two rings are free to rotate with respect to
each other. Thus, we can reasonably assume an effective
octupolar D2 molecular symmetry in solution for complexes 1−
5. In that symmetry group, there are six nonzero tensor
components βxyz contributing to the total HLS signal related to
the observable by the following relationships:
a
b
Under the residual absorption of Br2. βHLS, (×10−30 esu) determined
by hyper-Rayleigh experiments at 1907 nm in CHCl3 for the oxidized
complexes [LnIII(Pc•)2]+, where Ln = Nd (1c), Eu (2c), Dy (3c), Y
(4c), and Lu (5c). Reference ethyl violet, β = 170 × 10−30 esu at 1907
b
c
nm. Estimated error 15%. For an effective octupolar D2 molecular
symmetry: βHLS = [⟨βZZZ2⟩ + ⟨βYZZ2⟩] = (4/7)β
2
2
xyz
4
7
2
2
βHLS = [⟨βZZZ2⟩ + ⟨βYZZ2⟩] =
β
xyz
(1)
band loses intensity and exhibits a small red shift compared to
the neutral form. The transitions characteristic of the π radical
at 590−610 nm (BV) and around 1000 nm (RV) increase
significantly. Accordingly, the intervalence transition prominent
in the neutral form disappeared. Again, for the Q band and the
two transitions associated with the π radical, a good linear
correlation with the ionic radius was found with hypsochromic
shifts observed when the ring-to-ring distance decreases (Figure
5).
Calculation gave values ranging from 3980 × 10−30 esu for 4
(Y) to 5760 × 10−30 esu for 5 (Lu), much higher than those
obtained for any octupolar chromophores. Actually, to the best
of our knowledge, these are the largest values ever reported for
nondipolar compounds, higher than the best octupolar
molecules described to date.15
The second important observation is the increase of the
hyperpolarizability values with the number of f electron. The
enhancement of the second-order nonlinear response as a
function of the number of f electrons was initially reported on
dipolar complexes of dibutylaminophenyl-functionalized anne-
lated terpyridine ligand14a,c and on D3d octupolar complexes of
dipicolinic acid.14b Theoretical calculation involved the polar-
ization of the 4f electrons in these Ln complexes of push−pull
ligand of a rather polarizable ligand to explain the enhancement
observed.14c,40 More recently, study on lanthanide complexes of
hexafluoroacetylacetonate and diglyme, [Ln(hfac)3(diglyme)],
two ligands lacking a charge-transfer transition, confirmed the
tuning of the second-order nonlinear properties by the f
electrons and emphasized the unexpected polarizable character
of f electrons.41 It also showed that the influence of the f
electrons on the molecular quadratic hyperpolarizability was
much higher for the dipolar contribution than for the octupolar
contribution, although the major contribution to the total
quadratic hyperpolarizability was the octupolar one. In our case,
the influence of the number of f electrons on the hyper-
polarizability is perfectly illustrated by the difference observed
between the dysprosium and the yttrium complexes 3 and 4.
The Y3+ and Dy3+ ions have similar ionic radii (r8 = 1.019 and
1.027 Å, respectively),28 and the two complexes have similar
optical properties. However, Yttrium is not a 4f-block element
and therefore Y3+ has a 4f0 electronic configuration compared
to the 4f9 electronic configuration of Dy3+. As a consequence,
the quadratic hyperpolarizability of 4 is 1.9 times lower than
that of 3 in relation to the f-electron polarizable character.
However, it is worth noting that the neodymium complex 1 has
a similar value to 4 when it should be higher (Nd3+ has a 4f4
configuration). This seems to indicate that the second-order
nonlinear responses of complexes 1−5 are not only driven by
SECOND-ORDER NONLINEAR OPTICAL
PROPERTIES
As previously reported for the lutetium complex 5 the dynamic
■
molecular first hyperpolarizabilties ⟨βH2LS
⟩
of compounds
1907
1−4 were determined by nonpolarized HLS (also named
hyper-Rayleigh scattering)37 in chloroform solution, using an
incident wavelength at 1907 nm. The reference used for the
HLS measurements was ethyl violet for which a ⟨βH2LS
⟩
1907
value of 170 × 10−30 esu at 1907 nm was taken.38 If for
complexes 3−5 (Dy, Y, and Lu) the incident wavelength of
1907 nm was higher than or close to the absorption cutoff of
the intervalence band and therefore off-resonance, this is not
the case for the other compounds. The europium complex 2
and the neodymium complex 1 especially display quite a strong
linear absorption at 1907 nm. Therefore for all compounds, the
HLS values were corrected for absorption at 1907 nm by the
factor of 10A , A1907 being the absorption of the compound at
1907
the laser incident wavelength λinc of 1907 nm. Moreover, note
that the scattered light at the second harmonic (953 nm) would
be partially absorbed because of the small sharp peak of the RV
transition displayed by all compounds around 933 nm. The
HLS data were further corrected for by a factor of 10A /2, A953
953
being the absorption of the compound solution at the second
harmonic wavelength λSHG of 953 nm. Nevertheless, to
minimize the total correction factor applied before determining
953/2
β1907 (10A × 10A
< 3), very low concentrations (no
1907
more than 10−5 M) were used. Table 3 lists the dynamic
hyperpolarizability ⟨βH2LS
by HLS obtained at 1907 nm.
⟩
1907
4367
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
the f-electron polarization but also indirectly by geometrical
parameters such as the metal ionic radius.
intervalence transition between the two rings. These potent
nonlinear chromophores exhibit exceptionally large dynamic
The dynamic molecular first hyperpolarizabilties ⟨βH2LS
⟩
molecular first hyperpolarizabilities ⟨βH2LS
⟩
1907 as determined
1907 of all the reduced [LnIII(Pc)2]− and oxidized [LnIII(Pc•)2]+
species 1b−5b and 1c−5c have also been measured by
nonpolarized HLS and are given in Tables 4 and 5. The
absorptions of the compounds at the laser incident wavelength
(λ = 1907 nm) and at the second-harmonic signal (λ = 953
nm) were less problematic than they were for the neutral
compounds. Nevertheless, the measurements were done in
dilute solution (10% THF in CHCl3 or CHCl3), at the same
concentration (10−5 M), to obtain values that can be directly
compared with the previous ones regardless of concentration
effects. For a given metal ion, very similar values were obtained
for the reduced and oxidized form (except for Eu). For all metal
ions, however, the values obtained are somewhat lower than
those of the neutral form. Nevertheless, they remain excep-
tionally large for octupolar molecules. Analyzing the total HLS
by nonpolarized HLS in solution at 1907 nm, among the
highest ever measured. That laser wavelength comes in
resonance with the intervalence band in the case of the larger
ion complexes (Nd3+ and Eu3+). Despite that, the lutetium
complex presents the highest hyperpolarizability over the
lanthanide series, which seems to indicate that the intervalence
transition has little influence on the quadratic nonlinear
response. On the other hand, the dynamic molecular first
hyperpolarizabilities values show the dependence with the
number of the 4f electrons already observed on lanthanide (III)
octupolar complexes of other symmetries attributed to the
polarizable character of f electrons.40 Finally, the one-electron
oxidized and the one-electron reduced complexes were studied.
Owing to the different oxidation states on the macrocycles, the
oxidized and reduced species display considerably changed
optical properties compared to the neutral states while keeping
a similar octupolar structure. The dynamic molecular first
hyperpolarizabilities measured at 1907 nm remain exceptionally
large for such simple structures, emphasizing the interest of this
original octupolar design. This work represents a first step
toward the elaboration of fully cubic octupolar architectures. In
that prospect, work on crosswise phthalocyanine bearing true
electron withdrawing groups has started. The one-electron
reduced species are particularly appealing because, as suggested
by the works by Koike35a and Homborg,29 diphthalocyaninato-
(2−)metallate(III) complexes display conformational hetero-
geneity depending on the size of the central metal ion and of
the countercation. In particular, in the case of ions larger in size
than Nd3+, such as La3+ or Bi3+, eclipsed conformation of the
two cycles can be obtained on the occasion of slightly different
interactions of the anions and the cations. Work in that
direction has also started.
2
value βHLS in an effective D2 molecular symmetry using eq 1
gave values of βxyz in the range from 3220 to 6070 × 10−30 esu
for 1b−5b and 3690 to 6250 × 10−30 esu for 1c−5c of the same
order of magnitude as for the supermolecular chromophores
described by Therien and Clays.15 The hyperpolarizabilities of
the reduced species 1b−5b regularly increased from Nd to Lu
following the 4f orbital filling. The enhancement factor
calculated between 3b (Dy, 4f9) and 4b (Y, 4f0) was somewhat
lower than the 1.9 obtained for the neutral form, but
comparable with the 1.5 reported in the literature.14a,b
However, the value for 4b is the same as the one for 1b
(Nd, 4f3), suggesting that, as already noticed for the neutral
form, the interaction between the two macrocycles (related to
the interplane distance hence the metal size) is another
important parameter controlling the hyperpolarizabilty. The
same observations can be made for the hyperpolarizabilty
values measured for the oxidized species 1c−5c. The
enhancement factor between 3c (Dy, 4f9) and 4c (Y, 4f0) is
1.6. The increase of the hyperpolarizabilty values resembles the
one observed for the neutral form, with a big difference
between neodymium and europium and then a regular increase
to lutetium. In that case, note that the incident wavelength at
1907 nm comes into resonance with one of the π-radical
transitions centered at around 1000 nm. This band is not an
intervalence transition, and hence two-photon enhancement in
the hyperpolarizability cannot be ruled out. Given the blue shift
observed for this transition when the ionic radius decreases, this
effect would be more pronounced for the lighter lanthanides.
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional figures, HR-MS data and spectra, NMR spectra, and
X-ray crystallographic data. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: chirel@gyte.edu.tr (C.H.).
*E-mail: yann.bretonniere@ens-lyon.fr (Y.B.).
■
Funding
CONCLUSION
■
̈
̇
This work was supported by TUBITAK 1001 Project 108M034
and the French Embassy in Turkey (Ph.D. fellowship to
M.M.A.).
A series of octupolar nonlinear optical chromophores have
been synthesized based on double-decker lanthanide (III)
complexes of crosswise ABAB phthalocyanine ligands bearing
thioether electron donor group and hydrogen atoms alter-
natively on the macrocycle. This 3D structure was designed to
closely match the structure of the octupolar cube, the
delocalization between the two rings approximating to the
electronic interaction along the edges of the cube. The D2
noncentrosymmetric structure was confirmed by X-ray
diffraction studies. The size of the lanthanide (III) central ion
modulates the ring-to-ring distance and the degree of coupling
between the two phthalocyanine rings and consequently
modulates the optical properties of the chromophore, in
particular the strong near-infrared absorption due to the
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) (a) Dalton, L. R. J. Phys.: Condens. Matter 2003, 15, R897−R934.
(b) Bentoumi, W.; Mulatier, J.-C.; Bouit, P.-A.; Maury, O.; Barsella, A.;
Vola, J.-P.; Chastaing, E.; Divay, L.; Soyer, F.; Le Barny, P.;
̀
Bretonniere, Y.; Andraud, C. Submitted for publication.
(2) Zyss, J. J. Chem. Phys. 1992, 98, 6583−6599.
(3) (a) Zyss, J. Nonlinear Opt. 1991, 1, 3−18. (b) Zyss, J.; Ledoux, I.
Chem. Rev. 1994, 94, 77−105.
(4) Kim, H. M.; Cho, B. R. J. Mater. Chem. 2009, 19, 7402−7409.
4368
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
(5) (a) Chandra Ray, P.; Kumar Das, P. Chem. Phys. Lett. 1995, 244,
(15) Ishizuka, T.; Sinks, L. E.; Song, K.; Hung, S.-T.; Nayak, A.;
Clays, K.; Therien, M. J. J. Am. Chem. Soc. 2011, 133, 2884−2896.
(16) Bartholomew, G. P.; Ledoux, I.; Mukamel, S.; Bazan, G. C.;
Zyss, J. J. Am. Chem. Soc. 2002, 124, 13480−13485.
153−156. (b) Stadler, S.; Brauchle, C.; Brandl, S.; Gompper, R. Chem.
̈
Mater. 1996, 8, 676−678. (c) Brasselet, S.; Cherioux, F.; Audebert, P.;
Zyss, J. Chem. Mater. 1999, 11, 1915−1920. (d) Cho, B. R.; Park, S. B.;
Lee, S. J.; Son, K. H.; Lee, S. H.; Lee, M.-J.; Yoo, J.; Lee, Y. K.; Lee, G.
J.; Kang, T. I.; Cho, M.; Jeon, S.-J. J. Am. Chem. Soc. 2001, 123, 6421−
6422. (e) Alcaraz, G.; Euzenat, L.; Mongin, O.; Katan, C.; Ledoux, I.;
Zyss, J.; Blanchard-Desce, M.; Vaultier, M. Chem. Commun. 2003,
2766−2767. (f) Brunel, J.; Mongin, O.; Jutand, A.; Ledoux, I.; Zyss, J.;
Blanchard-Desce, M. Chem. Mater. 2003, 15, 4139−4148. (g) Cui, Y.
Z.; Fang, Q.; Lei; Xue, G.; Yu, W. T. Chem. Phys. Lett. 2003, 377, 507−
511. (h) Traber, B.; Wolff, J. J.; Rominger, F.; Oeser, T.; Gleiter, R.;
Goebel, M.; Wortmann, R. Chem.Eur. J. 2004, 10, 1227−1238.
(i) Hennrich, G.; Asselberghs, I.; Clays, K.; Persoons, A. J. Org. Chem.
2004, 69, 5077−5081. (j) Quintiliani, M.; García-Frutos, E. M.;
(17) Weiss, R.; Fischer, J. Lanthanide Phthalocyanine Complexes. In
The Porphyrin Handbook; Kadish, K. M.; Smith, K. M.; Guillard, R.,
Eds.; Academic Press, Elsevier: Amsterdam, Boston, London, New
York, Oxford, Paris, San Diego, San Francisco, Singapore, Sydney,
Tokyo, 2003; Vol. 16, pp 171−246.
(18) (a) Corker, G. A.; Grant, B.; Clecak, N. J. J. Electrochem. Soc.
1979, 126, 1339−1343. (b) Orti, E.; Bred
́
as, J. L.; Clarisse, C. J. Phys.
Chem. 1990, 92, 1228−1235.
(19) (a) VanCott, T. C.; Gasyna, Z.; Schatz, P. N.; Boyle, M. E. J.
Phys. Chem. 1995, 99, 4820−4830. (b) Ostendorp, G.; Homborg, H.
Z. Anorg. Allg. Chem. 1996, 622, 1222−1230.
(20) (a) La Mar, G. N.; de Ropp, J. S.; Smith, K. M.; Langry, K. C. J.
Am. Chem. Soc. 1980, 102, 4835−4836. (b) De Cian, A.; Moussavi, M.;
Fisher, J.; Weiss, R. Inorg. Chem. 1985, 24, 3162−3167.
́
Gouloumis, A.; Vazquez, P.; Ledoux-Rak, I.; Zyss, J.; Claessens, C. G.;
Torres, T. Eur. J. Org. Chem. 2005, 2005, 3911−3915. (k) Piao, M. J.;
Chajara, K.; Yoon, S. J.; Kim, H. M.; Jeon, S.-J.; Kim, T.-H.; Song, K.;
Asselberghs, I.; Persoons, A.; Clays, K.; Cho, B. R. J. Mater. Chem.
2006, 23, 2273−2281. (l) Argouarch, G.; Veillard, R.; Roisnel, T.;
Amar, A.; Boucekkine, A.; Singh, A.; Ledoux, I.; Paul, F. New J. Chem.
(21) Ayhan, M. M.; Singh, A.; Hirel, C.; Gurek, A. G.; Ahsen, V.;
̈
Jeanneau, E.; Ledoux-Rak, I.; Zyss, J.; Andraud, C.; Bretonnier
Am. Chem. Soc. 2012, 134, 3655−3658.
̀
e, Y. J.
(22) CrysAlisPro, Version 1.171.35.21 (release 20−01−2012
CrysAlis171 .NET); Agilent Technologies: Santa Clara, CA, compiled
Jan 23, 2012, 18:06:46.
2011, 35, 2409−2411. (m) Moreno Oliva, M.; Juar
Segura, J. L.; van Cleuvenbergen, S.; Clays, K.; Goodson, I.; Theodore;
Lopez Navarrete, J. T.; Casado, J. J. Phys. Chem. C 2013, 117, 626−
632.
(6) Ratera, I.; Ruiz-Molina, D.; Sporer, C.; Marcen, S.; Montant, S.;
Letard, J.-F.; Freysz, E.; Rovira, C.; Veciana, J. Polyhedron 2012, 22,
1851−1856.
(7) (a) Stadler, S.; Feiner, F.; Brauchle, C.; Brandl, S.; Gompper, R.
́
ez, R.; Ramos, M.;
́
(23) Clark, R. C.; Reid, J. S. Acta Crystallogr. 1995, A51, 887−897.
(24) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.
(25) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487.
́
̈
Chem. Phys. Lett. 1995, 245, 292−296. (b) Stadler, S.; Brauchle, C.;
̈
(26) (a) Idelson, E. M. Novel tetrazaporphins. U.S. Patent
US004061654, 1977.12.6, 1977. (b) Young, J. G.; Onyebuagu, W. J.
Org. Chem. 1990, 55, 2155−2159.
Brandl, S.; Gompper, R. Chem. Mater. 1996, 8, 414−417.
(8) (a) Lambert, C.; Schmalzlin, E.; Meerholz, K.; Brauchle, C.
̈
̈
Chem.Eur. J. 1998, 4, 512−521. (b) Bourgogne, C.; Le Fur, Y.; Juen,
́
(27) (a) Gurek, A. G.; Ahsen, V.; Luneau, D.; Pecaut, J. Inorg. Chem.
P.; Masson, P.; Nicoud, J.-F.; Masse, R. Chem. Mater. 2000, 12, 1025−
̈
2001, 40, 4793−4797. (b) Jiang, J.; Bian, Y.; Furuya, F.; Liu, W.; Choi,
M. T. M.; Kobayashi, N.; Li, H.-W.; Yang, Q.; Mak, T. C. W.; Ng, D.
K. P. Chem.Eur. J. 2001, 7, 5059−5069. (c) Lu, G.; Bai, M.; Li, R.;
Zhang, X.; Ma, C.; Lo, P.-C.; Ng, D. K. P.; Jiang, J. Eur. J. Inorg. Chem.
́ ́
1033. (c) Quintiliani, M.; Perez-Moreno, J.; Asselberghs, I.; Vazquez,
P.; Clays, K.; Torres, T. J. Phys. Chem. C 2010, 114, 6309−6315.
(9) Blanchard-Desce, M.; Baudin, J.-B.; Ruel, O.; Jullien, L.; Brasselet,
S.; Zyss, J. Opt. Mater. 1998, 9, 276−279.
(10) (a) Sastre, A.; Torres, T.; Díaz-García, M. A.; Agullo-
́
Lop
́
ez, F.;
2001, 3703−3709. (d) Gurek, A. G.; Basova, T.; Luneau, D.; Lebrun,
̈
Dhenaut, C.; Brasselet, S.; Ledoux, I.; Zyss, J. J. Am. Chem. Soc. 1996,
118, 2746−2747. (b) Kang, S. H.; Kang, Y.-S.; Zin, W.-C.; Olbrechts,
G.; Wostyn, K.; Clays, K.; Persoons, A.; Kim, K. Chem. Commun. 1999,
1661−1662. (c) Olbrechts, G.; Wostyn, K.; Clays, K.; Persoons, A.;
Kang, S. H.; Kim, K. Chem. Phys. Lett. 1999, 308, 173−175.
C.; Kol’tsov, E.; Hassan, A. K.; Ahsen, V. Inorg. Chem. 2006, 45, 1667−
1676. (e) Zhou, Y.; Zhang, Y.; Wang, H.; Jiang, J.; Bian, Y.; Muranaka,
A.; Kobayashi, N. Inorg. Chem. 2009, 48, 8925−8933. (f) Katoh, K.;
Komeda, T.; Yamashita, M. Dalton Trans. 2010, 39, 4708−4723.
(28) Shannon, R. D. Acta Crystallogr. 1976, 32, 751−767.
(d) Martín, G.; Rojo, G.; Agullo-
́
Lop
́
ez, F.; Ferro, V. R.; García de
(29) Huckstadt, H.; Tutaß, A.; Goldmer, M.; Cornelissen, U.;
̈
̈ ̈
la Vega, J. M.; Martínez-Díaz, M. V.; Torres, T.; Ledoux, I.; Zyss, J. J.
Phys. Chem. B 2002, 106, 13139−13145. (e) Claessens, C. G.;
Homborg, H. Z. Anorg. Allg. Chem. 2001, 627, 485−497.
(30) Buchler, J. W.; Scharbert, B. J. Am. Chem. Soc. 1988, 110, 4272−
4276.
́ ́ ́
Gonzalez-Rodríguez, D.; Torres, T.; Martín, G.; Agullo-Lopez, F.;
Ledoux, I.; Zyss, J.; Ferro, V. R.; García de la Vega, J. M. J. Phys. Chem.
B 2005, 109, 3800−3806.
(31) Bian, Y.; Wang, D.; Wang, R.; Weng, L.; Dou, J.; Zhao, D.; Ng,
D. K. P.; Jiang, J. New J. Chem. 2003, 27, 844−847.
(11) Zyss, J.; Dhenaut, C.; Chauvan, T.; Ledoux, I. Chem. Phys. Lett.
1993, 206, 409−414.
(32) (a) Yilmaz, I.; Nakanishi, T.; Gurek, A. G.; Kadish, K. M. J.
̈
Porphyrins Phthalocyanines 2003, 7, 227−238. (b) Zhu, P.; Lu, F.; Pan,
N.; Arnold, D. P.; Zhang, S.; Jiang, J. Eur. J. Inorg. Chem. 2004, 2004,
510−517.
(12) Lequan, M.; Branger, C.; Simon, J.; Thami, T.; Chauchard, E.;
Persoons, A. Adv. Mater. 1994, 6, 851−853.
(13) (a) Dhenaut, C.; Ledoux, I.; Samuel, I. D. W.; Zyss, J.;
Bourgault, M.; Le Bozec, H. Nature 1995, 374, 339−342. (b) Evans, C.
C.; Masse, R.; Nicoud, J.-F.; Bagieu-Beucher, M. J. Mater. Chem. 2000,
10, 1419−1423. (c) Feuvrie, C.; Maury, O.; Le Bozec, H.; Ledoux, I.;
Morrall, J. P.; Dalton, G. T.; Samoc, M.; Humphrey, M. G. J. Phys.
(33) L’Her, M.; Pondaven, A. Electrochemistry of Phthalocyanines.
In The Porphyrin Handbook; Kadish, K. M.; Smith, K. M.; Guillard, R.,
Eds. Academic Press: New York, 2003; Vol. 16, pp 117−170.
(34) Takamatsu, S.; Ishikawa, N. Polyhedron 2007, 26, 1859−1862.
(35) (a) Koike, N.; Uekusa, H.; Ohashi, Y.; Harnoode, C.; Kitamura,
F.; Ohsaka, T.; Tokuda, K. Inorg. Chem. 1996, 35, 5798−5804.
(b) Gonidec, M.; Amabilino, D. B.; Veciana, J. Dalton Trans. 2012, 41,
13632−13639. (c) Konami, H.; Hatano, M.; Tajiri, A. Chem. Phys. Lett.
1989, 160, 163−167.
Chem. A 2001, 111, 8980−8985. (d) Maury, O.; Viau, L.; Sen
Corre, B.; Guegan, J.-P.; Renouard, T.; Ledoux, I.; Zyss, J.; Le Bozec,
H. Chem.Eur. J. 2004, 4454−4466.
(14) (a) Senechal, K.; Toupet, L.; Ledoux, I.; Zyss, J.; Le Bozec, H.;
́ ́
echal, K.;
́
́
́
Maury, O. Chem. Commun. 2005, 2180−2181. (b) Tancrez, N.;
(36) Liu, W.; Jiang, J.; Du, D.; Arnold, D. P. Aust. J. Chem. 2000, 53,
131−135.
Feuvrie, C.; Ledoux, I.; Zyss, J.; Toupet, L.; Le Bozec, H.; Maury, O. J.
Am. Chem. Soc. 2005, 127, 13474−13475. (c) Sen
Hemeryck, A.; Tancrez, N.; Toupet, L.; Williams, J. A. G.; Ledoux, I.;
Zyss, J.; Boucekkine, A.; Guegan, J.-P.; Le Bozec, H.; Maury, O. J. Am.
Chem. Soc. 2006, 128, 12243−12255.
́
ec
́
hal-David, K.;
́
(37) (a) Clays, K.; Andre, P. Phys. Rev. Lett. 1991, 66, 2980−2983.
(b) Clays, K.; Persoons, A. Rev. Sci. Instrum. 1992, 83, 3285−3289.
(38) Le Bozec, H.; Le Bouder, T.; Maury, O.; Bondon, A.; Ledoux, I.;
Deveau, S.; Zyss, J. Adv. Mater. 2001, 13, 1677−1681.
́
4369
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370

Inorganic Chemistry
Article
(39) (a) Shirk, J. S.; Lindle, J. R.; Bartoli, F. J.; Boyle, M. E. J. Phys.
Chem. 1992, 96, 5847−5852. (b) Shirk, J. S.; Lindle, J. R.; Bartoli, F. J.;
Hoffman, C. A.; Kafafi, Z. H.; Snow, A. W. Appl. Phys. Lett. 1989, 55,
1287−1288.
(40) Furet, E.; Costuas, K.; Rabiller, P.; Maury, O. J. Am. Chem. Soc.
2008, 130, 2180−2183.
(41) Valore, A.; Cariati, E.; Righetto, S.; Roberto, D.; Tessore, F.;
Ugo, R.; Fragala, I. L.; Fragala, M. E.; Malandrino, G.; De Angelis, F.;
̀ ̀
Belpassi, L.; Ledoux-Rak, I.; Hoang-Thi, K.; Zyss, J. J. Am. Chem. Soc.
2010, 132, 4966−4970.
4370
dx.doi.org/10.1021/ic4031188 | Inorg. Chem. 2014, 53, 4359−4370