Scholarly article on 2-(3,4,5-trisdodecyloxyphenyl)propan-2-ol 934823-80-0 from Chemistry - A European Journal p. 1674

DOI: 10.1002/chem.200601389

Tuning the Polarization Along Linear Polyaromatic Strands for Rationally

Inducing Mesomorphism in Lanthanide Nitrate Complexes

Emmanuel Terazzi,^[a]Laure GuØnØe,^[a]Pierre-Yves Morgantini,*^[b]

GØrald Bernardinelli,^[c]Bertrand Donnio,^[d]Daniel Guillon,^[d]and Claude Piguet*^[a]

Abstract: The opposite orientation of

tional theory (DFT) and time-depend-

ant DFTcalculations performed in the

gas phase show that the inversion of

thumb, an alternation of electron-poor

and electron-rich aromatic rings favors

intermolecular interactions between

the rigid cores and consequently meso-

morphism, a situation encountered for

the ester spacers in the rodlike ligands

L4^C12

and

(benzimidazole-OOC-phenyl)

L5^C12

(benzimidazole-COO-

the ester spacers has considerable ef-

G

phenyl) drastically changes the elec-

tronic structure of the aromatic sys-

tems, without affecting their meridional

tricoordination to trivalent lanthanides,

Ln^III, and their thermotropic liquid

crystalline (i.e., mesomorphic) behav-

iors. However, the rich mesomorphism

fects on the electronic structure and

polarization of the aromatic groups

along the strands, which control residu-

al intermolecular interactions responsi-

ble for the formation of thermotropic

liquid-crystalline phases. As a rule of

L4^C12, L5^C12, [Ln

not for [Ln (NO₃)₃]. The interca-

(L5^C12

lation of an additional electron-rich di-

phenol ring on going from [Ln

(L5^C12)-

(NO₃)₃] to [Ln (NO₃)₃] restores

(L6^C12

A

)

A

)

ACHTREUNG

A

G

)

ACHTREUNG

mesomorphism despite an unfavorable

orientation of the ester spacers, in

agreement with our simple predictive

model.

exhibited by the complexes [Ln

(NO₃)₃] (Ln=La–Lu) vanishes in [Ln-

(L5^C12

(NO₃)₃], despite superimposable

A

Keywords: electrostatic potentials ·

intermolecular interactions · lantha-

nides · liquid crystals · metallomes-

ogens

ACHTREUNG

A

)

ACHTREUNG

molecular structures and comparable

photophysical properties. Density func-

Introduction

polyaromatic systems), to which long and poorly polarizable

A

flexible alkyl, alkoxy, or siloxy chains (i.e. the second part)

are connected. For these molecules in the crystalline state,

the minimum free energy results from a micro-segregation

process, in which the close packing of the rigid core maxi-

mizes intermolecular (mainly transient) electrostatic interac-

tions, while the peripheral flexible chains optimally fill the

According to a chemical point of view, the usual design of

thermotropic liquid crystals relies on the preparation of mol-

ecules containing two contradictory (i.e., antinomic) parts

separated by a molecular interface.^[1]The first part corre-

sponds to a polarizable rigid core (often made up of flat

[a] Dr. E. Terazzi, Dr. L. GuØnØe, Prof. Dr. C. Piguet

Department of Inorganic, Analytical and Applied Chemistry

University of Geneva, 30 quai E. Ansermet, 1211 Geneva 4 (Switzer-

land)

[d] Dr. B. Donnio, Dr. D. Guillon

Institut de Physique et Chimie des MatØriaux de Strasbourg-IPCMS

Groupe des MatØriaux Organiques

UMR 7504 (CNRS/UniversitØ Louis Pasteur)

23 rue du Loess, B.P. 43, 67034 Strasbourg Cedex 2 (France)

Fax : (+41)22-379-6830

E-mail: Claude.Piguet@chiam.unige.ch

Supporting information for this article is available on the WWW

under http://www.chemeurj.org/ or from the author. It contains addi-

tional tables (Tables S1–S16) and figures (Figure S1–S13) for structur-

al and spectroscopic analyses, and crystallographic data for com-

[b] Dr. P.-Y. Morgantini

Department of Physical Chemistry

University of Geneva, 30 quai E. Ansermet, 1211 Geneva 4 (Switzer-

land)

pounds 2, L5^C1, [Eu(2)

(NO₃)₃](CH₃CN)₂(18) in CIF formats.

A

(H₂O)]

G

N

Fax : (+41)22-379-3216

A

ACHTREUNG

E-mail: Pierre-Yves.Morgantini@chiphy.unige.ch

[c] Dr. G. Bernardinelli

Laboratory of X-ray Crystallography

University of Geneva, 24 quai E. Ansermet, 1211 Geneva 4 (Switzer-

land)

1674

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

FULL PAPER

interstitial voids.^[2]A thermotropic liquid-crystalline phase,

often referred to as a thermotropic mesophase, forms upon

warming up the crystals to the melting temperature T_m, for

which the flexible peripheral chains are molten and adopt a

liquidlike behavior. Since the interactions between poorly

polarizable chains is weak (i.e. DH^c_m^hainsis small), but the re-

lease of configurational entropy upon melting is consider-

dentate Schiff bases^[7]and in planar macrocylic phthalo-

cyanines.^[8]

2) A large number of diverging flexible chains are attached

to the central aromatic binding units in order to give

poly

ACHTREUNG

DS^c_m^hains

,

a

enough to promote liquid-crystalline behavior.^[9]

3) The rigid aromatic cores responsible for the residual in-

termolecular cohesion in the mesophase are spatially de-

coupled from the bulky lanthanide unit, thus preserving

DH^c_m^oreslarge enough to give high-temperature clearing

processes.^[10]

A

temperature T_mꢀDT_m^chains=DH_m^chains/DS^c_m^hainsis low. Further

heating of the mesophase produces a second phase transi-

tion, characterized by the clearing temperature T_cꢀ

DT^c_m^ores=DH_m^cores/DS^c_m^ores, at which the rigid polarizable cores

are completely decorrelated, and a liquid is formed.^[2]Evi-

dently, these successive melting and clearing processes are

not strictly decoupled and a partial dissociation of the inter-

molecular interactions involving the rigid polarizable cores

already occurs during the melting process leading to the

mesophase. However, the simplistic consideration of a ther-

motropic mesophase as containing residual polarizable clus-

ters made up of packed rigid aromatic cores dispersed in a

continuum of the liquidlike molten flexible chains, helps in

correlating microscopic molecular structures and macroscop-

ic thermal behaviors.^[2]With this model in mind, the intro-

duction of bulky molecular objects with considerable three-

dimensional extension close to the rigid core is harmful for

the mesogenic properties, because DH^c_m^oresdecreases to such

an extent that the clearing and melting processes merge, and

the liquid-crystalline phase vanishes.^[3]The latter effect rep-

resents a serious limiting factor for producing metallomeso-

gens, that is, metal-containing liquid crystals, because the

bulky metallic unit is usually introduced within the aromatic

rigid core.^[4]In this context, the trivalent lanthanide cations,

Ln^III, correspond to an extreme case because their extended

coordination spheres are poorly compatible with the specific

molecular anisometries required for an efficient packing of

the cores.^[5]For instance, the mesogenic ligand L1^C12

These considerations suggest that DS^c_m^hainsis currently the

only rationally tunable parameter for programming lantha-

nido mesogens (second strategy), while some concomitant,

but empirical control of DH^c_m^oresresults from the minimiza-

tion of the pertubation brought by the metallic core (first

and third strategies). A fourth unexplored strategy may ben-

efit from recent improvements in the understanding and the

modeling of aromatic p–p stacking interactions,^[11]which

allows some simple electrostatic modeling of the intermolec-

ular packing occurring between rigid aromatic cores in mes-

ophases.^[12]A rational tuning of DH^c_m^oresmay thus result from

a judicious exploitation of the two main characteristics of in-

teracting p systems.

1) The dominant interaction in these systems correspond to

p-electron repulsion. The introduction of electron-at-

tracting atoms or substituents decreases p–p electron re-

pulsion and stabilizes intermolecular packing, thus in-

creasing DH^c_m^ores

.

2) When the p systems are polarized, like polarizations

repel and opposite polarizations attract. An alternation

of electron-poor and electron-rich p system along the

strand is thus recommended for producing intermolecu-

^ꢁC

(Cr ¹³²SmA ¹⁸⁸I) and L2^C12(Cr ¹⁴⁴SmA ¹⁹³I) lose

lar self-cohesion and large DH_m^cores

.

ꢀꢀꢀ!

their mesomorphism upon complexation of the central tri-

dentate 2,6-bis(benzimidazol-2-yl)pyridine to Ln

(NO₃)₃.^[6]

These points are at the origin of the empirical rule of

thumb proposed by Nguyen and co-workers in the context

of banana-shaped mesogens; Nguyen et al. claim that an al-

ternate polarization of successive aromatic rings along bent

strands strongly favors mesomorphism in the final materials,

^ꢁC

as illustrated in L3^C12(Cr ¹⁵⁷SmC ¹⁶²N ¹⁶⁸I).^[12]Inter-

ꢀꢀꢀ!

ꢀꢀꢀ! ꢀꢀꢀ!

Among the rare successes obtained in designing lanthani-

domesogens, that is, lanthanide-containing mesogens,^[2c]

three main strategies can be recognized.

1) Ln^IIIor LnX₃are embedded in a cocoon of wrapped aro-

matic rings, which ensures a significant enthalpic contrib-

ution DH^c_m^oresto the clearing process, as found in mono-

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1675

C. Piguet, P.-Y. Morgantini et al.

estingly, the recent induction of mesomorphism in the bent

complexes [Ln (NO₃)₃] along the lanthanide series

(L4^C12

L5^C12and its complexes [Ln

A

G

U

)

A

ester spacers are reversed (Scheme 1b). A deeper under-

standing of the electronic structures combined with the ex-

plicit modeling of the electrostatic potential at the surface

of the molecule may help in rationalizing intermolecular co-

hesion forces between aromatic cores and, eventually,

DH^c_m^ores. The recent use of predictable aromatic donor–ac-

ceptor p–p stacking interactions between hemidisk mole-

cules for building homo- and

(Ln=La–Lu) might be ascribed to similar electrostatic con-

siderations (Scheme 1a).^[9c,d]To explore the origins and the

consequences of an alternation of electrostatic aromatic po-

larizations along the ligand strands on 1) the intermolecular

interactions and 2) the formation of thermotropic liquid-

crystalline phases, we have synthesized the isomeric ligand

heteroleptic columns in colum-

nar mesophases is related to

the same global approach,^[13]as

are the switches between nem-

atic and smectic organizations

observed upon inversion of the

ester spacers in thermotropic

polyaromatic 1,1’-disubstituted

ferrocene-containing

liquid

crystals and in extended dicate-

nar bipyridines.^[14]

Results

Syntheses and molecular struc-

tures of the ligands L5^Cn(n=0,

1, 12): A mild and selective

KMnO₄oxidation performed

under catalytic phase-transfer

conditions,^[15]followed by alka-

line hydrolysis transform the

starting tridentate unit 2,6-

bis(1-ethyl-5-methoxymethyl-

benzimidazol-2-yl)pyridine (1)^[6]

into 2,6-bis(1-ethyl-5-carboxy-

benzimidazol-2-yl)pyridine (3)

(Scheme 2). The preparation of

the air-sensitive electron-rich

substituted tetraphenol partner

6 is more delicate and requires

a careful Grignard reduction of

the gallic ester 4,^[9d]followed by

a Baeyer–Villiger type oxida-

tive rearrangement, eventually

leading to the target tetraphe-

nol 6.^[16]The coupling of the

central tridentate binding unit 3

with two appended tetraphenols

6 in the presence of 1-[3-(dime-

thylamino)propyl]-3-ethylcarbo-

diimide (EDCI) and 4-(diethyl-

amino)pyridine (DMAP) yields

the extended rodlike ligands

L5^Cn, which possess reversed

ester spacers compared with

the isomeric ligands L4^Cn

(Scheme 1).^[9d]

Scheme 1. Schematic structures, conformations, and polarizations of the tridentate ligands a) L4^Cnand b)

L5^Cn, and of their complexes with Ln(NO₃)₃.

1676

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

Table 1. ¹H NMR chemical shifts [ppm] of the aromatic protons in Li^C12

and [Lu(Li^C12)(NO₃)₃] (i=4, 5) in CD₂Cl₂at 298 K (numbering in

Scheme 2).

H1

H2

H6

H7

H9

H13

L4^C12

8.05

8.19

8.51

8.57

8.32

8.51

8.23

8.31

7.48

7.67

7.65

7.75

7.18

8.25

7.41

8.37

7.64

8.73

8.12

9.16

7.43

6.53

7.52

6.62

L5^C12

[Lu(L4^C12)(NO₃)₃]

[Lu(L5^C12)(NO₃)₃]

both ligands, in agreement with their connection to an elec-

tron-poor pyridine ring, which is not significantly affected

by the orientation of the ester spacers.

In agreement with NOE measurements recorded in solu-

tion, the crystal structures of the tridentate synthon 2 (Fig-

ure 1a) and of the ligand L5^C1(Figure 1b) show the aromat-

ic 2,6-bis(benzimidazol-2-yl)pyridine units to adopt the

usual trans–trans arrangement (i.e. both N2 and N4 atoms

adopt a trans orientation with respect to N1, Scheme 1 and

Figure 1).^[6,9d]This conformation provides an approximate

rodlike shape for the tridentate rigid core with small devia-

tions from planarity between the aromatic rings (interplanar

pyridine benzimidazole–pyridine angles: 20.28 in 2 and 11.1–

13.98 in L5^C1, Tables S1 and S2, Supporting Information).

Interestingly, the carbonyl group of the spacer is coplanar

with the aromatic ring to which it is attached, which pro-

A

Scheme 2. Reagents: i) KMnO₄, T EBAC, CHCl₂, 71%; (ii) KOH, EtOH,

2

carbonyl group is almost perpendicular to the benzimidazole

ring in L4^C1(interplanar angle=63.9–68.08),^[9d]but coplanar

with this ring in L5^C1(interplanar angle=5.2–8.88, Figure 1,

Table S2 and Figure S1 Supporting Information). However,

the electrostatic repulsion operating between the carbonyl

group and the phenol derivative to which it is connected,

produces similar cross-hatched organization of the appended

100%; iii) CH₃MgI, Et₂O, 90 % (n=12); iv) NaBO₃, BF₃-Et₂O, THF,

81% (n=12); v) EDCI, 4-DMAP, CH₂Cl₂, 71% (n=12).

1

The H NMR spectra of L4^C12and L5^C12display six aro-

matic CH signals (H1, H2, H6, H7, H9, H13, numbering in

Scheme 2) and one enantiotopic A₂X₃spin system for the

ethyl residue connected to the

benzimidazole ring, compatible

with dynamically average C_2v

symmetry for the aromatic rigid

core on the NMR timescale

(Table 1). The lack of NOE ef-

fects between the methylene

group and H2 confirms a trans–

trans arrangement of the 2,6-

bis(benzimidazol-2-yl)pyridine

unit (Scheme 1),^[6,9d]while the

strong diamagnetic shielding of

H13 (Dd=À0.90 ppm) and the

concomitant deshielding of H6

(Dd=+0.19 ppm), H7 (Dd=

1.07 ppm), and H9 (Dd=+

1.09 ppm, Table 1) on going

from L4^C12to L5^C12support the

intuitive polarization of the aro-

matic groups depicted in

Scheme 1. On the other hand,

Figure 1. Views of the tridentate ligands in the crystal structures of a) 2 and b) L5^C1·0.5CH₂Cl₂with numbering

the pyridine protons H1 and

H2 are detected at low field for schemes. Ellipsoids are represented at the 50% probability level.

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1677

C. Piguet, P.-Y. Morgantini et al.

phenyl rings with respect to the central tridentate binding

unit in L4^C1(interplanar benzimidazole–phenyl angles=

68.8–75.58)^[9d]and in L5^C1(interplanar benzimidazole–

phenyl angle=83.1–89.48, Figure 1 and Table S2, Supporting

Information). Consequently, except for the orientation of

the ester spacers, the molecular structures of L4^C1and L5^C1

are almost superimposable (Figure S1, Supporting Informa-

tion), and the bond lengths and bond angles are standard.^[17]

C1

À

A close scrutiny at the aromatic C C bonds in L4 and

L5^C1shows only marginal differences, which cannot be con-

sidered as probes for electrostatic polarization along the

strand (Table S3, Supporting Information).

In the crystal structure of L4^C1, we have previously shown

that each molecule is involved in four intermolecular p–p

stacking interactions: two benzimidazole–phenyl (interpla-

nar angle=18.48, d=3.34 ), one phenyl–phenyl (interpla-

nar angle=08, d=3.67 ), and one benzimidazole–benzim-

idazole (interplanar angle=08, d=3.56 ).^[9d]In the crystal

structure of the isomeric ligand L5^C1, we also observe four

intermolecular p–p interactions responsible for the crystal

packing: two benzimidazole–benzimidazole (interplanar

angle=11.38, d=3.43 , Figure S2, Supporting Information)

and two pyridine–benzimidazole (interplanar angle=13.98,

d=3.68 , Figure S2, Supporting Information). It is worth

noting that the terminal electron-rich substituted tetraphe-

nol rings in L5^C1are not involved in phenyl–phenyl p-stack-

ing interactions, while the related, but electron-poor, gallic

acids in L4^C1participate in intermolecular packing, in agree-

ment with the basic rules established for the stabilization of

interacting p systems.^[11]According to the same rules, the in-

teractions between electron-poor benzimidazole rings occur

at short distances in both ligands L5^C1and L4^C1(Figure S2,

Supporting Information), and in the intermediate 2 (Figure

S3, Supporting Information). Altogether, each isomer L4^C1

and L5^C1takes part in four intermolecular stacking interac-

tions, which ensure comparable cohesion of the rigid cores

in the solid state.

Figure 2. Walsh diagram for selected Kohn–Sham orbitals close to the

HOMO–LUMO gap in ligands L4^C1and L5^C1. Energies are given in eV

(1 eV=8065.5 cm^À1).

have been calculated for sets of 3D points corresponding to

the Connolly surfaces for L4^C1and L5^C1and are shown in

Figure 3. Both potentials (MEP) display an electron-defi-

cient pyridine ring connected to electron-rich benzimidazole

Ground-state electronic structures and thermal properties of

the ligands L5^Cn, (n=0, 1, 12): To explore the origin of the

intermolecular interactions involving the aromatic rings

along the ligand strands and to further rationalize photo-

physical properties, we have performed time-dependent

DFT (TD-DFT) calculations for unraveling the electronic

structure of the ground state, together with the lowest

ligand-centered singlet and triplet excited states. The molec-

ular structure of L4^C1and L5^C1optimized in the gas phase,

are very similar to those observed in the crystal structures,

and Figure 2 shows selected Kohn–Sham orbitals lying close

to the HOMO–LUMO gap. The associated computed ener-

gies of the 120 lowest singlet and triplet electronic states, to-

gether with their energy diagrams are given in Tables S4, S5,

and in Figure S4 (Supporting Information).

The atomic charges have been fitted to reproduce the

electrostatic potentials calculated according to the Merz–

Singh–Kollmann scheme (Tables S6 and S7, Supporting In-

formation).^[18]The molecular electrostatic potentials (MEP)

Figure 3. Color-coded representation of DFTmolecular electrostatic po-

tentials (MEP) computed on the Connolly surfaces around a) L4^C1and

b) L5^C1in their optimized gas phase geometries. MEP values are given in

atomic units.

1678

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

rings. The major difference between the two ligands con-

cerns the terminal phenyl rings, which display a more nega-

tive potential for L5^C1than for L4^C1, in agreement with in-

tuitive polarization derived from standard electronic effects

brought by the substituents connected to the aromatic rings

(Scheme 1). However, the differences in the electrostatic po-

tentials for L4^C1and L5^C1remain faint and marginal

(Figure 3), and we predict only minor variations in the inter-

molecular electrostatic interactions operating in condensed

phases, as exemplified by the packing observed in the crystal

structures of L4^C1and L5^C1. It is thus not surprising that the

hexagonal columnar mesophase Col_hdetected in the temper-

ature range 40–718 for L5^C12(characterized by differential

scanning calorimetry (DSC, Table 2) and small-angle X-ray

scattering (SAXS, Table S8, Supporting Information) resem-

C

N

ACHTREUNG

(NO₃)₃

G

À

A

G

CN¼10

Table 2. Phase-transition temperatures with enthalpy and entropy changes, and geometrical characteristics of

the mesophases for the ligands L5^C12, L6^C12, and their complexes [Lu(L5^C12)(NO₃)₃] and [Lu(L6^C12)(NO₃)₃].

1.187 in 17 and

1.044 in 18, in good agree-

ment with related ionic radii re-

ported for the analogous com-

[a]

Transition

T

[8C]

DH

DS

Geometrical

parameters

Ref.

[9d]

[kJmol^À1

]

[J mol^À1K^À1

]

[b]

L4^C12

Cr!g

0

25

61

40

71

plexes

(CH₃CN)]

and [Yb =

(L4^C0)(NO₃)₃] (R^C_Y^N_b^¼9

[Eu

(L4^C0)

ACHTREUNG

[b]

g!Col_h

Col_h!I

g!Col_h

Col_h!I

Cr!Col_h

Col_h!I

g!Col_h

Col_h!Dec

g!I

a=42.40

2

3.3

10

S=1555

E

(R

Eu

[b]

L5^C12

a=41.69

this work

[9d]

A

N

2

4.0

29.8

12

81

S=1505

1.046 ), in which the lantha-

nides are similarly ten- and

L6^C12

93

a=54.21

2

159

160

223

120–130

230

170

250

3.8

9

S=2545

[b]

[Lu(L4^C12)(NO₃)₃]

[Ln(L5^C12)(NO₃)₃]

[Lu(L6^C12)(NO₃)₃]

a=31.20

nine-coordinated,

respective-

2

ly.^[9d]The detailed arrangement

of the bidentate nitrate groups

around Ln^IIIis slightly different

on going from 18 to 17 in order

to accommodate the extra

water molecule around Eu^III

(Figure 4 and Figure S5a, Sup-

porting Information). However,

each pair of ten-coordinate

S=842

[b]

this work

I!Dec

g!Col_h

Col_h!I/dec

a=31.38

2

S=853

[a] Cr=crystal, g=glass, Col_h=hexagonal columnar phase, I=isotropic fluid; first-order transition tempera-

tures are given as the onset of the peak observed during the second heating processes (Seiko DSC 220C differ-

ential scanning calorimeter, 58Cmin^À1, under N₂); the liquid crystalline phases were identified from their opti-

cal textures and from SAXS studies. [b] Glass transition determined by polarizing light microscopy (PLM).

complexes

(NO₃)₃(H₂O)] (Figure S5b), and of

nine-coordinate complexes [Yb (NO₃)₃] and [Yb-

(L4^C0)

(L5^C0)

(NO₃)₃] (Figure S5c, Supporting Information) are

[Eu

G

bles that previously described for L4^C12(hexagonal colum-

nar mesophase Col_hin the temperature range 25–618,

(CH₃CN)] and [Eu(2)

E

ACHTREUNG

A

ACHTREUNG

[9d]

Table 2).

G

ACHTREUNG

almost superimposable, except for the orientation of the car-

bonyl groups of the spacer, which are systematically copla-

nar with the aromatic ring to which they are connected, as

Syntheses, molecular structures, and thermal properties of

the complexes [Ln

of stoichiometric quantities of L5^C12(1 equiv) in dichloro-

methane with Ln(NO₃)₃·xH₂O (1 equiv, x=2–4, Ln=Sm–

Lu) in acetonitrile gives 80% of microcrystalline powders,

the elemental analyses of which correspond to [Ln

(L5^C12)-

(NO₃)₃]·nH₂O (Ln=Sm, n=3.5, 7; Ln=Eu, n=2.5, 8; Ln=

A

N

ACHTREUNG

Table 3. Selected bond lengths [] and bite angles [8] in

[Eu(2)(NO₃)₃(H₂O)] (17) and [Yb(L5^C0)(NO₃)₃] (18).

AHCTREUNG

ACHTREUNG

17

18

17

18

Gd, n=3, 9; Ln=Tb, n=0, 10; Ln=Dy, n=0, 11; Ln=Ho,

n=1.5, 12; Ln=Er, n=0, 13; Ln=Tm, n=1, 14; Ln=Yb,

n=0.5, 15; Ln=Lu, n=1.5, 16; Table S9, Supporting Infor-

mation). For the larger Ln^III(Ln=La–Nd), mixtures of com-

plexes can be precipitated, in which a significant excess of

ligand is detected (with respect to the expected 1:1 stoichi-

ometry). These last complexes have both been investigated

further in this contribution. Moreover, we were unable to

À

Ln N1

2.657(2)

2.510(2)

2.488(2)

2.526(2)

2.542(2)

2.501(2) Ln O1b

2.402(2) Ln O2b

2.400(2) Ln O1c

2.393(2) Ln O2c

2.521(2)

2.644(2)

2.554(2)

2.470(2)

2.444(2)

2.435(2)

2.348(2)

2.353(2)

2.438(2)

–

À

Ln N2

À

Ln N4

À

Ln O1a

À

Ln O2a

2.370(2) Ln O1w

N1-Ln-N2

N1-Ln-N4

62.97(6)

62.86(7)

N2-Ln-N4 125.78(7)

65.76(7)

65.84(7)

131.37(7)

O1a-Ln-O2a 50.21(6)

O1b-Ln-O2b 49.27(7)

O1c-Ln-O2c 51.08(7)

54.14(7)

53.19(9)

53.01(8)

obtain monocrystals for the lipophilic complexes [Ln

(L5^C12)-

A

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1679

C. Piguet, P.-Y. Morgantini et al.

AHCTREUNG

ACHTREUNG

AHCTREUNG

deduce that the intermolecular

cohesion between the aromatic

cores in the solid state is larger

for the complexes [Ln

A

AHCTREUNG

supported by the melting pro-

cesses occurring around 120–

1308C for [Ln

N

after the removal of the inter-

stitial water molecules (Table 3)

and around 140–1608C for [Ln-

A

)

A

Moreover,

the melting process in the last

complexes leads to organized

cubic or columnar hexagonal

mesophases, in which significant

directional intermolecular inter-

actions between the polarized

cores operate.^[9d]For [Ln-

A

N

the melting and clearing pro-

cesses coincide, thus leading to

the isotropic liquid, in which

any directional intermolecular

interactions are removed. The

behavior of [Ln

N

in isotropic medium has been

studied in CH₂Cl₂:CH₃CN (1:1)

solvent mixtures. Spectrophoto-

metric titrations of L5^C12

Figure 4. Views of a) [Eu(2)(NO₃)₃(H₂O)] in the crystal structure of 17 and b) [Yb(L5^C0)(NO₃)₃] in the crystal

structure of 18 with numbering schemes. Ellipsoids are represented at the 50% probability level.

(10^À4m) with Ln

(NO₃)₃·xH₂O

A

previously described for the free ligands (i.e., phenyl for

(x=2–4, Ln=Eu, Y, Lu) in the range Ln:L5^C12=0.1–2.0

show a single end point for Ln:L5^C12=1.0 (Figure S8, Sup-

porting Information), as previously reported for L4^C12in the

same conditions.^[9d]The experimental data can be satisfying-

ly fitted with non-linear least-squares techniques^[21]to the

equilibrium shown in Equation (1) (Table 4).

L4^C0and benzimidazole for L5^C0).

The intermolecular interactions for [Eu(2)

N

(17) in the solid state are limited to one pair of benzimida-

zole–benzimidazole p stacking (interplanar angle=4.28, d=

3.69 , Figure S6, Supporting Information). However, the

extended aromatic strands in [Yb

G

L 5^C12þ LnðNO₃Þ₃Ð ½LnðL 5^C12ÞðNO₃Þ₃

Ln-L 5C12

b

11

ð1Þ

ACHTREUNG

actions involving pyridine and phenyl rings on one hand (in-

terplanar angle=14.78, d=3.70 ), and pairs of benzimida-

zole rings (interplanar angle=10.88, d=3.80 ) on the other

hand (Figure S7, Supporting Information). As previously

noted, in the crystal structure of L5^C1, the terminal electron-

rich tetraphenol rings do not show self-complementarity for

intermolecular p stacking, in contrast to the closely related

efficient intermolecular interactions occurring between pairs

Ln

11

Comparisons

between

log(b )

obtained

for

[Ln(Li^C12)(NO₃)₃] (i=4, 5) in the same conditions show

only a marginal stabilization for the complexes with L5^C12

(Table 4), and we conclude that the reverse orientation of

the ester spacers has negligible effect on the affinity of the

ligands for Ln^III.

The ¹H NMR spectrum of [Lu(L5^C12)(NO₃)₃] shows the

expected formation of a monomeric complex displaying dy-

namically average C_2vsymmetry on the NMR time scale

(6CH aromatic signals and one quartet for the enantiotopic

methylene protons connected to the benzimidazole rings,

of electron-poor gallic acid residues in [Yb

(interplanar angle=08, d=3.36 ).^[9d]Altogether, each [Yb-

(L5^C0)

(NO₃)₃] complex in 18 is involved in four intermolec-

ular p-stacking interactions (two benzimidazole–benzimid-

A

N

A

ACHTREUNG

1680

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

rﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

Table 4. Stability constants obtained for the formation of the complexes

~

D_dimer

D_mono

n_monoÁ MM_mono

3

[Ln(Li^C12)(NO₃)₃] (i=4, 5) in CH₂Cl₂:CH₃CN (1:1) at 293 K.

ð3Þ

¼

~

n_dimerÁ MM_dimer

Ln

R

[]^[a]

log(b

)

log(b

)

CN¼9

Ln-L 4C12 [b]

Ln-L 5C12

Ln

11

Variable-temperature NMR data in CD₂Cl₂(273–313 K)

show that the [Eu₂(L5^C12)₂(NO₃)₆]/[Eu(L5^C12)(NO₃)₃] ratio

smoothly decreases with increasing temperature, thus lead-

Eu

Y

Lu

1.12

1.08

1.03

5.3 (2)

5.9 (1)

6.1 (1)

6.3(1)

6.3(2)

6.4(4)

[a] Ionic radius for nine-coordinate Ln^III

ence [9d].

.

[b] Taken from refer-

[19]

2

ing to dimerization constants K^E_d^{u-L 5C12}=jdimerj/jmonoj

[Eq. (2)], which match

a linear vanꢁt Hoff plot with

DH_d^{Eu-L 5C12}=À27(2) kJ·mol^À1

and

DS^E_d^{u-L 5C12}

=

À98(4) Jmol^À1K^À1[Eq. (4)].

Table 1 and Figure S9a, Supporting Information), as previ-

ously established for the related complex

DH^E_d^{u-L iC12}

ÀRlnðK^E_d^{u-L iC12}Þ ¼

ÀDS^E_d^{u-L iC12}

ð4Þ

[Lu(L4^C12)(NO₃)₃].^[22]The protons H13, attached to the ter-

minal phenyl ring, is drastically shielded (Dd=À0.9 ppm) on

going from [Lu(L4^C12)(NO₃)₃] to [Lu(L5^C12)(NO₃)₃], while

H9 is concomitantly deshielded (Dd=+1.04 ppm) as a

result of the connection of the attracting carbonyl group to

the benzimidazole ring in the latter complex (numbering is

given in Scheme 2). It is therefore tempting to consider

these strong variations in the NMR shifts as direct probes

for assigning specific electronic densities in the aromatic

rings of [Lu(L4^C12)(NO₃)₃] and [Lu(L5^C12)(NO₃)₃]. Howev-

T

The thermodynamic data obtained for [Eu(L5^C12)(NO₃)₃]

significantly differ from DH^E_d^{u-L 4C12}=À54(4) kJmol^À1and

DS_d^{Eu-L 4C12}=À162(16) Jmol^À1K^À1

reported

for

[Eu(L4^C12)(NO₃)₃] in the same conditions.^[9d]We ascribe the

dramatic decrease of the enthalpic driving force of the dime-

rization process transforming [Eu(L5^C12)(NO₃)₃] into

[Eu₂(L5^C12)₂(NO₃)₆], to the coplanar orientation of the car-

bonyl groups of the ester spacers with the benzimidazole

rings (Figure 4b and Figure S5c, Supporting Information),

which hinders the approach of the second complex when

forming the double nitrate bridge in the dimer.^[9d]

er,

a

close scrutiny at the crystal structures of

[Yb(L5^C0)(NO₃)₃] (Figure 4b) shows that the coplanar ar-

rangement of the carbonyl group and the benzimidazole

ring puts H9 in the deshielding region of the C=O moiety.^[23]

This effect may easily overcome standard inductive effects.

Moreover, ring current in cyclic conjugated p-system is also

known to balance opposite inductive effects, because an in-

creased electron density in the aromatic ring results in a de-

shielding of the aromatic protons due to local diamagnetic

Ground state electronic structures of the complexes

[Ln(L4^Cn)(NO₃)₃] and [Ln(L5^Cn)(NO₃)₃] (Ln=Eu, Gd, Tb,

Lu): The geometries of the complexes [Lu(L4^C1)(NO₃)₃] and

[Lu(L5^C1)(NO₃)₃] optimized by DFTin the gas phase show

the expected cis–cis orientation of the bound 2,6-bis(benz-

A

anisotropies.^[23]We conclude that the H NMR shifts of the

coordination to Lu^III. The gas-phase and solid-state geome-

tries are similar except for a less pronounced cross-hatched

arrangement of the appended phenyl rings with respect to

the central tridentate unit (Figure S10, Supporting Informa-

tion). Figure 5 shows selected Kohn–Sham orbitals lying

close to the HOMO–LUMO gap, and the associated com-

puted energies of the 120 lowest singlet and triplet electron-

ic states, together with their energy diagrams are given in

Tables S12 and S13 (Supporting Information).

We have also calculated the DFTfitted atomic charges ac-

cording to the Merz–Singh–Kollman scheme (Tables S14,

and S15, Supporting Information)^[18]and the MEP on the

Connolly surface (Figure 6) for [Lu(Li^C1)(NO₃)₃] (i=4, 5).

Compared with the same approach applied to the free li-

gands (Figure 3), the complexation to Lu(NO₃)₃produces a

clearer discrimination of the electrostatic potentials between

a strongly positive region in the back of the 2,6-bis(benzim-

1

aromatic

protons

in

[Lu(L4^C12)(NO₃)₃]

and

in

[Lu(L5^C12)(NO₃)₃] can be hardly used as a safe guide for

probing electron densities in the aromatic rings, and intui-

tive deductions (Scheme 1) must be confirmed by DFTcal-

1

culations. Interestingly, the H NMR spectrum of the analo-

gous complex [Eu(L5^C12)(NO₃)₃], in which Yb^IIIhas been

replaced with the larger Eu^IIIcation, shows the coexistence

of two different complexes in thermodynamic equilibrium

according to Equation (2) (Figure S9, Supporting Informa-

tion), which corresponds to the dimerization process previ-

ously established for [Eu(L4^C12)(NO₃)₃] in the same condi-

tions.^[9d]

2 ½EuðL 5^C12ÞðNO₃Þ₃ Ð ½Eu₂ðL 5^C12Þ₂ðNO₃Þ₆

K^E_d^{u-L 5C12}

ð2Þ

A

around the metallic Lu(NO₃)₃unit (Figure 6). A close scruti-

ny at the computed electrostatic potentials along the strand

shows that the appended phenyl rings are more electron-

rich in [Lu(L5^C1)(NO₃)₃], while the coordinated tridentate

aromatic core is more electron-deficient, as predicted by the

intuitive charge assignment given in Scheme 1. Consequent-

ly, [Lu(L4^C1)(NO₃)₃] possesses a larger number of polariza-

tion alternances (4) along the strand (phenyl(+)-phenyl(À)-

This mechanism is supported by the experimental ratio of

the auto-diffusion coefficients D_mono/D_dimer=1.3(1) deduced

from NMR diffusion measurements in CD₂Cl₂,^[24]which

translates [Eq. (3)] into

a ratio of molecular weights

MM_dimer/MM_mono=2.2(1) close to 2.0, the value predicted by

equilibrium (2) (it is assumed that the specific partial vol-

umes of the two complexes are similar n˜_monoꢀn˜_dimer).^[9d]

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1681

C. Piguet, P.-Y. Morgantini et al.

Figure 5. Walsh diagram for selected Kohn–Sham orbitals close to the HOMO–LUMO gap in the complexes [Lu(L4^C1)(NO₃)₃] and [Lu(L5^C1)(NO₃)₃].

Energies are given in eV (1 eV=8065.5 cm^À1).

imidazole(+)-pyridine(+)-imidazole(+)-phenyl(À)-

phenyl(+), Scheme 1a and Figure 6a) than [Lu(L5^C1)(NO₃)₃]

(2 alternances: phenyl(À)-phenyl(+)-imidazole(+)-pyri-

dine(+)-imidazole(+)-phenyl(+)-phenyl(À), Scheme 1b and

Figure 6b), which agrees with Nguyenꢁs rule of thumb^[12]for

empirically justifying why the lipophilic analogue

[Lu(L4^C12)(NO₃)₃] is mesogenic, while [Lu(L5^C12)(NO₃)₃] is

not (Table 3). In terms of intermolecular packing involving

the aromatic cores, the increased alternance of polarization

along the strand maximizes the possibility of favorable p-

stacking interactions occurring between the aromatic cores

of closely spaced identical molecules. It thus helps in creat-

ing a thermodynamic micro-segregation between the melting

processes associated with the flexible and the rigid parts of

the packed molecules.

Figure 6. Color-coded representation of DFTMEP computed on the

Connolly surfaces around a) [Lu(L4^C1)(NO₃)₃] and b) [Lu(L5^C1)(NO₃)₃]

Consequences of the reversal connection of ester spacers

in their optimized gas-phase geometries. MEP values are given in atomic

units.

on the excited states and photophysical properties of the li-

gands Li^Cnand of their complexes [Ln(Li^Cn)(NO₃)₃] (i=4,

5; Ln=Eu, Gd, Tb, Lu): When the donor oxygen atoms of

the ester spacers are connected to the benzimidazole rings

1682

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

in L4^C1, the HOMO (105a) corresponds to a p-orbital cen-

tered on the tridentate binding unit (Figure 2, left). The

second highest occupied orbital (SHOMO, 102b) still dis-

plays p character, but its location is shifted toward the pe-

ripheral aromatic rings, a trend that continues for the next

highest occupied orbitals. On the other hand, the LUMO

(103b) and SLUMO (106a) orbitals in L4^C1are p orbitals

mainly centered on the pyridine ring. Consequently, the

lowest allowed singlet–singlet electronic transition (105a!

103b, Figure 2, left) mainly involves the 2,6-bis(benzimid-

4, 5) in CH₃CN/CH₂Cl₂(1:1) agree with this prediction since

the maximum of the lowest p!p* envelope is centered at

30580 cm^À1for L4^C12and at 30865 cm^À1for L5^C12(DE=

285 cm^À1, Table 5, Figure 7a). TD-DFT calculations also

show that the related “emissive” triplet excited states locat-

ed on the tridentate binding unit arise at lower energy than

the related singlet excited state (computed singlet–triplet

energy gap between the two lowest excited states:

DE^SÀT(L4^C1)=E(¹pp*, level 5 in Table S4)ÀE(³pp*, level 1

in Table S4, Supporting Information)=1970 cm^À1and

DE^SÀT(L5^C1)=E(¹pp*, level 19 in Table S5)ÀE(³pp*, level

17 in Table S5, Supporting Information)=3285 cm^À1. These

characteristics dominate the experimental electronic emis-

sion spectra of both ligands with E(¹pp*) ꢀ25000 cm^À1(0–0

phonon band) and E(³pp*) ꢀ21000 cm^À1(0–0 phonon

band) for the lowest emitting excited states, which are

mainly centered on the tridentate 2,6-bis(benzimidazol-2-

A

is significant (transition 5 in Table S4, Supporting Informa-

tion). Upon reversal of the ester spacers in L5^C1, the Walsh

diagram clearly shows the expected destabilization of the p

orbitals centered on the phenyl rings and the concomitant

stabilization of those centered on the 2,6-bis(benzimidazol-

2-yl)pyridine unit (Figure 2, right). Consequently, 1) the

HOMO (105a) is centered on the lateral phenyl rings in

L5^C1, 2) the LUMO is still centered on the pyridine ring

and 3) the HOMO–LUMO gap dramatically shrinks (Table

S5, Figure S4b, Supporting Information). The associated ex-

cited states thus possess strong

[6,26]

yl)pyridine unit (Table 5, Figure 7b,c).

Close scrutiny of the Kohn–Sham orbitals computed for

the complexes [Lu(L4^C1)(NO₃)₃] and [Lu(L5^C1)(NO₃)₃]

show negligible mixing of ligand- and metal-centered orbi-

phenyl(HOMO)!pyridine(LU-

MO) charge-transfer character,

which strictly limits overlap

density and stabilizing exchange

interactions.^[25]We therefore

predict negligible oscillating

strengths for electronic transi-

tions originating from high-

energy filled orbitals centered

on the appended side arms, and

reaching low-energy empty or-

bitals centered on the pyridine

ring (Table S5, Supporting In-

formation). The low-energy

part of the absorption and

emission electronic spectra of

L5^C1thus mainly reflects the al-

lowed p!p* transitions cen-

tered on the tridentate binding

units (103a!103b, Figure 2

right and transition 19 in Table

S5, Supporting Information). If

we now compare the computed

energy of the latter transition

with that of the related transi-

tion in L4^C1(105a!103b; tran-

sition 5 in Table S4, Supporting

Information, Figure 2 left), we

predict that the first intense

low-energy electronic p!p*

transition occurs only at slightly

higher energy for L5^C1(DE=

335 cm^À1). Experimentally, the

absorption spectra of the analo-

Table 5. Ligand-centered absorption and emission properties for ligands Li^C12and for complexes

[Ln(Li^C12)(NO₃)₃] (i=4, 5; Ln=Eu, Gd, Tb, Lu).^[a]

Compound

Absorption

e

Emission

Lifetime

[cm^À1] p!p*

[m^À1cm^À1

]

[cm^À1] ¹pp*

[cm^À1] ³pp*

[ms] t(³pp*)

L4^C12

34365

30580

47200

46900

26000 sh

24210

20700 sh

19765

431(5)

44(2)

23200 sh

21700 sh

26810 sh

25940

18700

L5^C12

39215

30865

82080

48000

20600 sh

20120

31(1)

5.4(4)

24539

18640

17400 sh

21100 sh

20265

[Lu(L4^C12)(NO₃)₃]

35088

32787

27779

39580

40280

29920

25160

24180

23120 sh

98(2)

19000

18330

21345

[d]

[b]

[c]

[d]

[Lu(L5^C12)(NO₃)₃]

[Gd(L4^C12)(NO₃)₃]

[Gd(L5^C12)(NO₃)₃]

[Eu(L4^C12)(NO₃)₃]

[Eu(L5^C12)(NO₃)₃]

[Tb(L4^C12)(NO₃)₃]

[Tb(L5^C12)(NO₃)₃]

39060

31545

29100

35090

33113

28736

39060

31450

28330

35090

33335

28330

39065

31450

28330

35090

33010

28650

39370

31150

27930

57180

33080

26390

43120

43240

31000

58000

33250

27100

35560

35010

25540

57180

33030

27490

40350

39650

27540

57260

31780

27800

26900 sh

26450

23500

25280 sh

24200

[b]

[c]

25300 sh

24150

22625

21505

23560

22250

20555

22830

23420

22270

19120

[a] Absorption spectra recorded for 10^À4m solution in CH₃CN/CH₂Cl₂(1:1) at 295 K. Luminescence and life-

time data at 77 K. sh=shoulder. [b] Quenched by traces of Eu^IIIand Tb^III, see text.^[30][c] ³pp* luminescence

quenched by transfer to Ln ion. [d] The weak luminescence intensity prevented determination of reliable lumi-

gous lipophilic ligands Li^C12(i= nescence decay times.

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1683

C. Piguet, P.-Y. Morgantini et al.

[Lu(L4^C1)(NO₃)₃] and in [Lu(L5^C1)(NO₃)₃] (Figure 5), a sit-

G

uation previously encountered for L5, but not for L4 in the

free ligands (Figure 2). Consequently, the lowest allowed

singlet–singlet electronic transitions possess strong phe-

A

which strictly limits overlap density and oscillating strengths

(Tables S12 and S13, Supporting Information). The lowest

electronic transitions displaying significant intensities in the

complexes thus involve the HOMO centered on the triden-

tate binding unit (146a in [Lu(L4^C1)(NO₃)₃] and 145a in

[Lu(L5^C1)(NO₃)₃]) and the LUMO centered on the pyridine

ring (142b, Figure 5). The latter transition is predicted to

occur at slightly lower energy for [Lu(L4^C1)(NO₃)₃] than for

[Lu(L5^C1)(NO₃)₃] (DE=1747 cm^À1), because of the inver-

sion of the ester spacers; a trend previously found for the

free ligands, and experimentally supported by the electronic

absorption spectra of the complexes [Lu(Li^C12)(NO₃)₃] (i=

4, 5; DE=1320 cm^À1in CH₂Cl₂:CH₃CN (1:1), Table 5 and

Figure 8a). Moreover, the electron-attracting effect of the

bound Lu(NO₃)₃core in the complexes reduces the energy

1

of the lowest ligand-centered emissive pp* states (level 19

Figure 7. a) Absorption spectra of L4^C12(thin line) and L5^C12(bold line)

in CH₂Cl₂/CH₃CN (1:1) at 298 K. b) Emission spectra of L4^C12at 77 K

(n˜_exc=29240 cm^À1). Thin line=fluorescence, bold line=phosphorescence

(delay time=1 ms, intensity 10). c) Emission spectra of L5^C12at 77 K

(n˜_exc=29410 cm^À1). Thin line=fluorescence, bold line=phosphorescence

(delay time=1 ms, intensity 10).

tals, in agreement with the electrostatic character of metal–

ligand bonds in lanthanide complexes.^[27]Therefore, the

binding of Lu(NO₃)₃to the ligand roughly corresponds to

the connection of a strong electron-attractive substituent to

the three coordinated nitrogen atoms. This produces a

global decrease of about 1 eV in the energy of the orbitals

centered on the tridentate binding unit on going from the

free ligands Li (Figure 2) to the associated complexes

[Lu(Li^C1)(NO₃)₃] (Figure 5). On the other hand, the orbitals

located on the appended phenyl rings are only marginally

affected. Consequently, the HOMO (148a) and the occupied

orbitals close to it (147a, 141b, 140b) are p orbitals centered

on the appended phenyl rings, while the LUMO (142b) and

SLUMO (149a) are centered on the central pyridine ring in

Figure 8. a) Absorption spectra of [Lu(L4^C12)(NO₃)₃] (thin line) and

[Lu(L5^C12)(NO₃)₃] (bold line) in CH₂Cl₂/CH₃CN (1:1) at 298 K. b) Fluo-

rescence spectra of [Lu(L4^C12)(NO₃)₃] (thin line, n˜_exc=29410 cm^À1) and

[Lu(L5^C12)(NO₃)₃] (bold line, n˜_exc=30300 cm^À1intensity 10) in the solid

state at 298 K. The inset shows the phosphorescence spectrum of

[Lu(L5^C12)(NO₃)₃] at 77 K (n˜_exc=29080 cm^À1, delay time=1 ms).

1684

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

!

for [Lu(L4^C1)(NO₃)₃] in Table S12 and level 28 for

[Lu(L5^C1)(NO₃)₃] in Table S13, Supporting Information),

and we calculate red shifts of 4547 and 3125 cm^À1on going

from the free ligands to the complexes. Experimental red

shifts of 2800 and 1770 cm^À1are indeed observed in CH₂Cl₂/

CH₃CN for the low-energy p–p* transitions centered on the

tridentate binding units on going from L4^C12to

[Lu(L4^C12)(NO₃)₃] and from L5^C12to [Lu(L5^C12)(NO₃)₃]

(Figures 7a and 8a; Table 5). Interestingly, the TD-DFT

computed singlet–triplet energy gap between the two lowest

Upon ligand-centered irradiation through the p* p tran-

sitions, the pure Eu^IIIcomplexes [Eu(Li^C12)(NO₃)₃] (i=4, 5)

display standard Eu(⁵D₀! F_J) (J=0–4) emission bands do-

7

5

7

minated by the hypersensitive electric dipolar D₀! F₂tran-

sition (Figure S12, Supporting Information). Both spectra

are almost superimposable and correspond to the low-sym-

metry nine-coordinate EuN₃O₆environment previously es-

tablished for the anhydrous model complex [Eu(2,6-bis(1,1’-

doctyl-benzimidazol-2-yl)pyridine)(NO₃)₃].^[30]The Eu(⁵D₀)

lifetime t^E₇₇^u_K^{-L 5C12}=1.04(2) ms is monoexponential and fairly

matches those found 1) for the model complex [Eu(2,6-

1

3

excited states pp* and pp* is smaller for L5^C1(DE^SÀT

=

E(¹pp*, level

3

in Table S5, Supporting Informatio-

bis(1-octyl-benzimidazol-2-yl)pyridine)(NO₃)₃]

(t=1.11–

n)ÀE(³pp*, level 1 in Table S5)=5 cm^À1, [Lu(L5^C1)(NO₃)₃]

(DE^SÀT=75 cm^À1calculated between the related electronic

levels in Table S13, Supporting Information) and

[Lu(L4^C1)(NO₃)₃], (DE^SÀT=E(¹pp*, level 3 in Table S12,

Supporting Information)ÀE(³pp*, level 1 in Table S12, Sup-

1.15 ms,

4–295 K)^[30]

and

2)

for

the

isomer

[Eu(L4^C12)(NO₃)₃] (t^Eu-L4C12=1.25–1.29 ms, 10–295 K).^[9d]We

thus deduce that no OH oscillator (i.e., water molecule) lies

in the first coordination sphere. Consequently, the sixfold

decrease

in

quantum

yields

on

going

from

porting Information)=220 cm^À1), than for L4^C1(DE^SÀT

=

[Eu(L4^C12)(NO₃)₃] (F_abs=6(2)%) to [Eu(L5^C12)(NO₃)₃]

(F_abs=0.9(1)%) in acetonitrile/dichloromethane (1:1) can

be safely assigned to the reverse connection of the ester

spacers, which appear to slightly improve the sensitization

1970 cm^À1calculated between the related electronic levels in

Table S4). Since the reversal connection of the ester spacers

on going from L4^C1to L5^C1or the complexation of

Ln(NO₃)₃to the ligands significantly decreases the energy of

ligand-centered charge-transfer excited states, we expect a

reduced overlap density of the two singly occupied orbitals

in the triplet states of L5^C1(105a–106a in Figure 2 right), of

[Lu(L4^C1)(NO₃)₃] (148a–142b in Figure 5 left), and of

[Lu(L5^C1)(NO₃)₃] (148a–142b in Figure 5 right). This limits

DE^SÀTfor these systems (DE^SÀT=5–220 cm^À1), while the

contrasting considerable overlap density existing between

the related orbitals 105a and 103b in L4^C1(Figure 2, left) ef-

ficiently stabilizes the triplet state with respect to the singlet

state (DE^SÀT=1970 cm^À1), in complete agreement with a

simple Heitler–London interpretation using the active-elec-

process with L4^C12

.

The analogous Tb^IIIcomplexes

[Tb(Li^C12)(NO₃)₃] (i=4, 5) are poorly luminescent, because

of efficient Tb(⁵D₄)!Li(³pp*) thermally-activated energy

back transfers operating between the almost two resonant

levels, which dramatically quenches Tb^IIIemission at room

temperature.^[30]At 293 K in solution, we measure F_abs

=

0.6(4)% for [Tb(L4^C12)(NO₃)₃] and F_abs

[Tb(L5^C12)(NO₃)₃].

< 0.1% for

Discussion

tron approximation.^[25]For the open-shell 4f⁷

A

From a structural point of view, the opposite orientation of

the ester spacers in L4^Cnand L5^Cnonly affects the spatial ar-

rangement of the carbonyl groups, which are coplanar with

the aromatic rings to which they are connected (i.e., the ter-

minal phenyl ring in L4^Cnand the benzimidazole ring in

L5^Cn). Except for the dimerization process occurring in solu-

tion with large Ln^III[Eq. (2)], which is enthalpically less fa-

vorable for [Ln(L5^C12)(NO₃)₃] because of the presence of

sterically demanding in-plane carbonyl groups, these minor

geometrical changes have negligible structural consequences

and the global cross-hatched organization of the appended

phenyl rings with respect to the central planar 2,6-bis(benz-

complexes [Gd(Li^C12)(NO₃)₃] (i=4, 5), in which no metal-

centered excited levels are accessible for intramolecular

energy transfers from the ligand-centered ¹pp* or ³pp*

levels,^[28]we expect some improvements of 1) the ¹pp*!

³pp* intersystem crossing process (isc) and 2) the oscillator

3

strength of the spin-forbidden transition implying the pp*

level because the Coulomb interactions between the elec-

trons of the ligands and of the metal ions mix the ligand-

centered triplet and singlet wavefunctions.^[29]Surprisingly,

these effects cannot be detected because the phosphores-

cence spectra of [Gd(Li^C12)(NO₃)₃] (i=4, 5) are dominated

by the Eu- and Tb-centered emissions arising from traces

A

imidazol-2-yl)pyridine is maintained in the ligands Li^Cnand

0.01%) of [Eu(Li^C12)(NO₃)₃] (detected at room tempera-

ture, Figure S11a, Supporting Information) and

[Tb(Li^C12)(NO₃)₃] (detected below 77 K, Figure S11b, Sup-

porting Information) in the “pure” (>99.99%) Gd sample.

This phenomenon, originally discovered and investigated in

details for the parent complex [Gd(2,6-bis(1-methyl-benz-

in their Ln^IIIcomplexes [Ln(Li^C12)(NO₃)₃] (i=4, 5). If these

futile structural variations do not justify such detailed inves-

tigations, the changes in the electronic structure and their

expression in the mesogenic and photophysical properties

do bring novel tools for rationalizing and for programming

functional metallomesogens. Firstly, the ester spacers act as

ambivalent electron-attracting and electron-donating sub-

stituents for the aromatic rings, depending on their connec-

tion through their C or O atoms, respectively. Although, the

¹H NMR signals of the aromatic protons are dramatically af-

fected by the orientation of the spacers, they cannot be used

A

of extremely efficient, thermally activated, ligand-centered,

energy-migration processes in the solid state, which are not

perturbated by the appended aromatic side arms and by the

ester spacers.^[30]

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1685

C. Piguet, P.-Y. Morgantini et al.

as simple diagnostic probes for electron densities borne by

the aromatic rings, because shielding or deshielding effects

result from balances between inductive effect and through-

space diamagnetic anisotropies associated with ring-current

mechanisms.^[23]However, sophisticated DFTcalculations

performed on optimized molecular structures in the gas

phase show electrostatic potentials compatible with some

simple predictions considering an O-bound ester as a donor

group and a C-bound ester as an acceptor group (Scheme 1,

Figures 3 and 6). When calculated on the Connolly surface

of a molecule (ligand or complex), the electrostatic potential

can be used as a guide for estimating the efficiency of p-

stacking interactions occurring in condensed phases accord-

ing to the rules of Sanders and Hunter: 1) p-stacking inter-

actions are globally repulsive, 2) like polarizations repel and

3) opposite polarizations attract.^[11]Following Nguyenꢁs em-

pirical suggestion,^[12]we confirm here that the formation of

mesophases is favored by an alternation of electron-poor

and electron-rich p-systems along the rigid core, which helps

in providing self-complementarity and strong cohesion in

condensed phases. Such alternation exists for both L4^C12and

L5^C12, although the electrostatic potentials calculated in the

gas phase indicate some leveling resulting from intramolecu-

lar polarization effects. We thus indeed observe comparable

intermolecular p-stacking interactions in the crystals (each

molecule is involved in four interactions) and the formation

of similar hexagonal columnar mesophases close to room

temperature. For the complexes [Ln(L4^C12)(NO₃)₃] and

[Ln(L5^C12)(NO₃)₃], the strong electro-attractive polarization

brought by the central trivalent lanthanide drastically de-

creases the electron density on the central tridentate binding

unit, and consequently reduces the alternation between elec-

tron-rich and electron-poor aromatic rings along the strand.

For [Ln(L4^C12)(NO₃)₃], the connection of the ester spacers

through their O-donor atoms to the benzimidazole rings par-

tially overcomes the electron depletion in the six-membered

ring of the coordinated benzimidazole, and mesomorphism

is maintained (each rigid core of the complex is involved in

six intermolecular p-stacking interactions in the solid

state).^[9d]For [Ln(L5^C12)(NO₃)₃], the connection of the ester

spacer through their attracting carbon atom further contrib-

utes to the depletion of the electronic density in the coordi-

nated benzimidazole ring, and the alternation of polariza-

tion is reduced (each rigid core of the complex is involved

in only four intermolecular p-stacking interactions in the

However, the existence of a large amount of low-energy

states in L5^C12provides efficient electronic relays favoring

de-excitation processes, which significantly quenches the

emission intensities in L5^Cnand in its complexes

[Ln(L5^C12)(NO₃)₃] (Ln=Eu, Gd, Tb, Lu).

Conclusion

The reversal connection of the ester spacers on going from

L4^Cnto L5^Cnis harmful for both mesomorphism and lumi-

nescence. However, the investigation into the origin of this

behavior allows us to propose some predictive rules for “ra-

tionally” designing luminescent thermotropic lanthanido-

mesogens with this class of organic receptors.

1) The intermolecular cohesion forces between the polariz-

able aromatic cores, which are responsible for the micro-

segregation processes leading to the formation of the

thermotropic mesophases, can be roughly estimated by

considering the efficiency and the amount of packing in-

teractions occurring in the solid state.

2) An alternance of a sufficient number of electron-rich

and electron-poor aromatic rings along the rigid core

favors efficient intermolecular interactions in condensed

phases, which are partially maintained in the mesophas-

es. This electrostatic effect can be, a priori, estimated by

the electrostatic potential calculated on the Connolly sur-

face of the molecule by DFTin the gas phase.

3) The ester spacers should be oriented in order to maxi-

mize the alternance between electron-rich and electron-

poor aromatic rings along the strand.

4) The overall quantum yields of the Eu^IIIand Tb^IIIcom-

plexes strongly depend on the ligand-centered sensitiza-

tion process, and it thus benefits from a large HOMO–

LUMO gap, which is produced when low-energy charge-

transfer states are avoided.

With the three first rules in mind, we can predict that the

extended ligand L6^C12is a good candidate for restoring

mesomorphism in the complexes [Ln(L6^C12)(NO₃)₃], despite

the “wrong” orientation of the ester groups, which previous-

ly appeared to be harmful for the formation of mesophases

with [Ln(L5^C12)(NO₃)₃]. The major change on going from

L5^C12to L6^C12concerns the introduction of an interstitial

electron-rich p-diphenol ring in between two electron-poor

aromatic moieties provided by the carbon connection of the

ester spacers. This provides an alternance of nine aromatic

portions with opposite polarizations along the strand in

L6^C12, and an alternance of five aromatic portions with op-

posite polarizations in [Lu(L6^C12)(NO₃)₃] (Scheme 3),

whereas only three related aromatic portions exist in

[Lu(L5^C12)(NO₃)₃] (Scheme 1). Preliminary DSC (Table 2)

and SAXS studies (Table S8 and Figure S13, Supporting In-

formation) indeed show that both the free ligand L6^C12and

its complex [Lu(L6^C12)(NO₃)₃] display hexagonal columnar

mesophases (Col_h), while [Lu(L5^C12)(NO₃)₃] simply melts to

solid

state).

No

mesophase

is

detected,

and

[Ln(L5^C12)(NO₃)₃] simply melt to give liquids. If we now

consider potential exploitations of the photophysical proper-

ties brought into the mesophases by the Ln-centered lumi-

nescence,^[31]the TD-DFT calculations show that the concen-

tration of the donor substituents within the terminal phenyl

rings in L5^Cndrastically reduces the HOMO–LUMO gap

and puts the HOMO onto the appended aromatic rings. The

experimental electronic absorption spectra of the systems

with L4^C12and L5^C12are not very sensitive to this change,

because only p!p* transitions centered on the tridentate

units possess enough oscillator strengths to be detected.

1686

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

T_c([Lu(L4^C12)(NO₃)₃])@

T_m([Lu(L5^C12)(NO₃)₃]) suggests

that, among these two enthalpic

contributions, DH^c_m^oresis more

influenced by minor changes in

intermolecular aromatic stack-

ing interactions occurring in

condensed phases. After having

shown that DS^c_m^hainsis the first

thermodynamic parameter that

can be rationally tuned for in-

ducing mesomorphism in lan-

thanidomesogens with these

systems,^[9d]DH_m^coresrepresents a

second opportunity for chemists

to induce and to control the for-

mation of mesophases in lan-

thanidomesogens, and probably

more generally in metallomeso-

gens.

Experimental Section

Solvents, starting materials and syn-

theses: Chemicals were purchased

from Fluka AG and Aldrich, and used

without further purification unless oth-

erwise stated. The synthon 2,6-bis-(1-

ethyl-5-methoxymethyl-benzimidazol-

2-yl)pyridine (1) was prepared accord-

ing to a literature procedure.^[6]The ni-

trate salts Ln(NO₃)₃·xH₂O (Ln=La–

Lu, Y, x=1–4) were prepared from

the corresponding oxides (Rhodia,

99.99%).^[32]The Ln content of solid

salts was determined by complexomet-

ric titrations with Titriplex III (Merck)

in the presence of urotropine and

xylene orange.^[33]Acetonitrile and di-

chloromethane were distilled over cal-

Scheme 3. Synthesis and mesomorphism of the extended ligand L6^C12and of its complex [Lu(L6^C12)(NO₃)₃].

cium hydride, diethyl ether and tetrahydrofuran were distilled from

sodium.

give a liquid. The transition temperatures observed for

L6^C12and its complex [Lu(L6^C12)(NO₃)₃] are slightly higher

than those found for the related systems with L5^C12and

Preparation of 2,6-bis(1-ethyl-5-carboxybenzimidazol-2-yl)pyridinedi-

methyl ester (2): Compound

46.54 mmol), and triethylbenzylammonium

1

(1.06 g, 2.33 mmol), KMnO₄(7.35 g,

chloride (10.60 g,

L4^{C12 [9d]}

in agreement with the expected larger enthalpic

,

46.54 mmol) were refluxed in dichloromethane (100 mL) for 3 d. Excess

KMnO₄was removed by addition of a concentrated aqueous solution of

Na₂S₂O₅until the complete disappearance of the violet color. The organic

phase was separated and the aqueous phase was extracted with dichloro-

methane (3100 mL). The combined organic phases were dried

(Na₂SO₄), filtered, and evaporated to dryness. The crude yellow residue

was purified by column chromatography (silica gel; CH₂Cl₂/MeOH

99.5:0.5!99:1) to give 0.8 g (1.63 mmol, yield 71%) of 2 as a microcrys-

contributions to the melting and clearing processes resulting

from the improved sizes of the rigid cores. Finally, it is

worth noting the considerable contraction of the lattice pa-

rameter a in the Col_hmesophase on going from L6^C12to its

complex [Lu(L6^C12)(NO₃)₃] (Table 2, and Table S8, Support-

ing Information), which reflects the change from rodlike!

bent shapes occurring upon complexation of the ligand

strand (Scheme 3).

1

talline white solid. H NMR (CDCl₃): d=1.35 (t, J³=7.5 Hz, 6H), 3.94 (s,

6H), 4.78 (q, J³=6.9 Hz, 4H), 7.47 (d, J³=8.7 Hz, 2H), 8.07 (dd, J³=

8.7 Hz, J⁴=1.5 Hz, 2H), 8.07 (t, J³=7.8 Hz, 1H), 8.36 (d, J³=7.8 Hz,

2H), 8.56 ppm (d, J⁴=1.2 Hz, 2H); ¹³C NMR ([D₆]DMSO): d=15.66,

52.40 (C_prim); 40.33 (C_sec); 110.17, 123.00, 125.26, 126.46, 138.63 (C_tert);

139.31, 142.62, 149.83, 151.61, 167.70 ppm (C_quat); ESI-MS (CH₂Cl₂/

MeOH 9:1): m/z: 506.2 [M+Na]⁺, 989.4 [2M+Na]⁺; recrystallization

from acetonitrile gave X-ray quality prisms of 2.

We eventually conclude that the judicious design of aro-

matic rings with opposite polarization along the strand rep-

resents a novel and useful tool for chemically tuning the en-

thalpic contributions to the melting and clearing processes

(DH^c_m^hainsand DH_m^cores). However, the observation that

T_m([Lu(L4^C12)(NO₃)₃])ꢀT_m([Lu(L5^C12)(NO₃)₃]),

but

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1687

C. Piguet, P.-Y. Morgantini et al.

Preparation of 2,6-bis-(1-ethyl-5-carboxybenzimidazol-2-yl)pyridine (3):

Compound 2 (0.40 g, 0.83 mmol) and KOH (1.00 g, 17.82 mmol) were re-

fluxed in ethanol/water (50 mL:20 mL) for 45 min. The hot solution was

neutralized with conc. hydrochloric acid (35%) until pH 1.0, and a white

precipitate formed upon cooling. The precipitate was filtered and dried

under vacuum (908C) to give 0.38 g (0.83 mmol, yield 100%) of 3 as a

white solid. ¹H NMR ([D₆]DMSO): d=1.29 (t, J³=7 Hz, 6H), 4.82 (q,

J³=7 Hz, 4H), 7.84 (d, J³=7 Hz, 2H), 7.96 (dd, J³=8 Hz, J⁴=1 Hz, 2H),

8.22–8.35 (m, 3H), 8.40 ppm (d, J³=8 Hz, 2H); ¹³C NMR ([D₆]DMSO):

d=15.64 (C_prim); 40.47 (C_sec); 111.76, 122.15, 125.12, 126.78, 139.59 (C_tert);

125.94, 139.53, 142.53, 149.84, 151.72, 168.35 ppm (C_quat); EI-MS: m/z:

455 [M]⁺.

3.91 (m, 6H), 5.17 (m, 1H), 6.05 ppm (s, 2H); ¹³C NMR (CDCl₃): d=

14.14 (C_prim); 22.73, 26.12, 26.18, 29.36, 29.41, 29.43, 29.46, 29.67, 29.69,

29.74, 29.77, 29.79, 30.27, 31.75, 31.96, 68.98, 73.69 (C_sec); 94.25 (C_tert);

131.89, 151.89, 153.54 ppm (C_quat); EI-MS: m/z: 647 [M]⁺. The same pro-

cedure was followed for the derivatives 6^C1and 6^C0

.

1

Data for 6^Cl: H NMR (CDCl₃): d=3.70 (s, 9H), 5.25 (brs, 1H), 6.05 ppm

(s, 2H); ¹³C NMR (CDCl₃): d=56.30 (C_prim); 94.10 (C_tert); 132.10, 151.65,

154.20 (C_quat); EI-MS: m/z: 185 [M₊].

1

Data for 6^C0: H NMR (CDCl₃): d=5.40 (brs, 4H), 6.12 ppm (s, 2H); EI-

MS: m/z: 143 [M⁺].

Preparation of 2,6-bis-(1-ethyl-5-carboxybenzimidazol-2-yl)pyridine-

di[3,4,5-tris(alkyloxy)benzyl] ester (L5^Cn

; n=0, 1, 12): A catalytic

Preparation of 3,4,5-trialkoxybenzyl methyl ester (4^Cn, n=1, 12): Com-

pound 4^C0(8.00 g, 43.4 mmol), anhydrous potassium carbonate (54.17 g,

391.9 mmol), 1-bromododecane (38.37 g, 130.3 mmol) and potassium

iodide (200 mg, 1.20 mmol) were refluxed in dry acetone (300 mL) for

2 d. The solution was evaporated to dryness, partitioned between di-

chloromethane (200 mL) and water (200 mL). The organic layer was sep-

arated, and the aqueous phase extracted with dichloromethane (3

200 mL). The combined organic phases were dried (Na₂SO₄), filtered,

and evaporated to give a colorless oil, which was triturated with ethanol

(50 mL). The resulting voluminous white solid residue was filtered and

dried under vacuum to give 15.27 g (22.15 mmol, yield 51%) of 4^C12as a

white solid. M.p.=458C; ¹H NMR (CDCl₃): d=0.91 (t, J³=7 Hz, 18H),

1.24–1.42 (m, 48H), 1.46–1.53 (m, 6H), 1.75–1.87 (m, 6H), 3.91 (s, 3H),

4.02–4.05 (m, 6H), 7.27 ppm (s, 2H); ¹³C NMR (CDCl₃): d=14.11 , 50.09

(C_prim); 22.70 , 26.08, 26.10, 29.34, 29.37, 29.41, 29.58, 29.65, 29.67, 29.71,

29.74, 29.76, 30.35, 31.94, 31.95, 69.22, 73.51 (C_sec); 108.08 (C_tert); 124.67,

142.46, 152.84, 166.95 ppm (C_quat); EI-MS: m/z: 688 [M]⁺. The same pro-

cedure was followed for the derivative 4^C1by using methyliodide as the

alkylating agent.

amount of 4-dimethylaminopyridine, compound 3 (0.21 g, 0.46 mmol),

compound 6^C12(0.60 g, 0.93 mmol), and EDCI (0.36 g, 1.85 mmol) were

refluxed for 6 h in CH₂Cl₂(50 mL) under an inert atmosphere. The re-

sulting mixture was washed with water (4100 mL). The organic layer

was separated, dried (Na₂SO₄), and evaporated to dryness. The crude

yellow oil was purified by column chromatography (silica gel; CH₂Cl₂/

MeOH 100:0!99:1) to give 0.56 g (0.33 mmol, yield 71%) of L5^C12as a

pale yellow solid. ¹H NMR (CDCl₃): d=0.82–0.86 (m, 18H), 1.23–1.40

(m, 114H), 1.69–1.86 (m, 12H), 3.91–3.96 (m, 12H), 4.83 (q, J³=7.2 Hz,

4H), 6.45 (s, 4H), 7.54 (d, J³=8.7 Hz, 2H), 8.10 (t, J³=7.8 Hz, 1H), 8.20

(dd, J³=8.7 Hz, J⁴=1.5 Hz, 2H), 8.40 (d, J³=7.8 Hz, 2H), 8.72 ppm (d,

J⁴=1.5 Hz, 2H); ¹³C NMR (CDCl₃): d=14.32, 15.69 (C_prim); 14.32, 15.69,

22.89, 26.29, 26.36, 29.50, 29.57, 29.61, 29.85, 29.90, 29.98, 30.55, 32.13,

32.15, 40.43, 69.31, 73.73 (C_sec); 100.54, 110.37, 123.73, 125.69, 126.59,

138.69 (C_tert); 124.60, 136.03, 139.76, 142.75, 146.95, 149.82, 151.85,

153.57, 165.88 ppm (C_quat); ESI-MS (CH₂Cl₂/MeOH 9:1 + HCl 0.1%):

m/z: 1714.4

[M+H]⁺; elemental analysis calcd (%)

for

C₁₀₉H₁₇₃N₅O₁₀·0.6H₂O (1724.40): C 75.90, H 10.10, N 4.06; found: C

75.61, H 10.26, N 3.94; the same procedure was followed for the ligands

Data for 4^Cl¹H NMR (CDCl₃): d=3.88 (s, 9H), 3.91 (s, 3H), 7.33 ppm

:

L5^C1and L5^C0

Data for L5^C1

.

(s, 2H); ¹³C NMR (CDCl₃): d=56.42, 61.14 (C_prim); 107.56 (C_tert); 124.32,

143.13, 153.14, 166.95 ppm (C_quat); EI-MS: m/z: 226 [M⁺].

:

¹H NMR (CDCl₃): d=1.39 (t, J³=7.2 Hz, 6H), 3.85 (s,

18H), 4.84 (q, J³=6.7 Hz, 4H), 6.50 (s, 4H), 7.55 (d, J³=8.7 Hz, 2H),

8.10 (t, J³=7.8 Hz, 1H), 8.20 (dd, J³=8.7 Hz, J⁴=1.5 Hz, 2H), 8.40 (d,

J³=7.8 Hz, 2H), 8.73 ppm (d, J⁴=1.5 Hz, 2H); ¹³C NMR (CDCl₃): d=

15.68, 56.39, 61.15 (C_prim); 40.44 (C_sec); 99.58, 110.42, 123.76, 125.70,

126.61, 138.68 (C_tert); 124.41, 136.05, 139.84, 142.78, 147.43, 149.82,

151.89, 153.77, 165.87 ppm (C_quat); ESI-MS (CH₂Cl₂/MeOH 9:1): m/z:

810.3 [M+Na]⁺, 1598.2 [2M+Na]⁺; elemental analysis calcd (%) for

C₄₃H₄₁N₅O₁₀·0.6H₂O (798.62): C 64.67, H 5.33, N 8.77; found: C 64.70, H

5.41, N 8.77; recrystallization from CH₂Cl₂/hexane gave X-ray quality

prisms of L5^C12·0.5CH₂Cl₂.

Preparation of 2-(3,4,5-trisdodecyloxyphenyl)propan-2-ol (5^Cn, n=0, 1,

12): Methyl iodide (2,27 g, 16.00 mmol) was added to a suspension of

magnesium (0.39 g, 16.00 mmol) in dry diethyl ether (50 mL). The reac-

tion was initiated by heating and then the temperature was controlled

with a water/ice bath. When solid magnesium had disappeared, a solution

of compound 4^C12(5.00 g, 7.26 mmol) in dry diethyl ether (5 mL) was

slowly added. The resulting mixture was refluxed for 12 h. Addition of a

saturated aqueous ammonium chloride solution (10 mL) produced a vigo-

rous reaction, then water (150 mL) and diethyl ether (150 mL) were

added. The organic layer was separated, and the aqueous phase extracted

with diethyl ether (350 mL). The combined organic phases were dried

(Na₂SO₄), filtered, and evaporated to give a pale yellow oil. Column

chromatography (silica gel; CH₂Cl₂) gave 4.50 g (6.53 mmol, yield 90%)

of 5^C12as a colorless oil, which was poorly stable at RT. ¹H NMR

(CDCl₃): d=0.91 (t, J³=6.9 Hz, 18H), 1.23–1.39 (m, 48H), 1.45–1.53 (m,

6H), 1.58 (s, 6H), 1.73–1.85 (m, 6H), 3.96 (t, J³=6.6 Hz, 2H), 4.01 (t,

J³=6.5 Hz, 4H), 6.70 ppm (s, 2H); EI-MS: m/z: 670 [M⁺]. The same pro-

Data for L5^C0¹H NMR (CDCl₃): d=1.45 (t, J³=7.2 Hz, 6H), 4.90 (q,

:

J³=7.2 Hz, 4H), 7.28–7.40 (m, 6H), 7.47–7.51 (m, 4H), 7.62 (d, J³=

8.6 Hz, 2H), 8.17 (t, J³=7.9 Hz, 1H), 8.29 (dd, J³=8.6 Hz, J⁴=1.6 Hz,

2H), 8.48 (d, J³=7.9 Hz, 2H), 8.82 ppm (d, J⁴=1.3 Hz, 2H); ¹³C NMR

(CDCl₃): d=15.51 (C_prim); 40.32 (C_sec); 110.30, 121.83, 123.48, 125.68,

125.85, 126.57, 138.59 (C_tert); 124.56, 139.48, 151.19, 151.53, 165.53 ppm

(C_quat); ESI-MS (CH₂Cl₂/MeOH 9:1): m/z: 608.2 [M]⁺, 630.2 [M+Na]⁺,

1214.2 [2M]⁺, 1237.2 [2M+Na]⁺; elemental analysis calcd (%) for

C₃₇H₂₉N₅O₄(607.67): C 73.13, H 4.81, N 11.52; found: C 73.23, H 4.71, N

11.56.

cedure was followed for the derivatives 5^C1and 5^C0

Data for 5^Cl¹H NMR (CDCl₃): d=1.59 (s, 6H), 3.78 (s, 9H), 6.70 ppm

(s, 2H); EI-MS: m/z: 208 [M⁺].

.

:

Preparation of (4-hydroxyphenyl)-1-(3,4,5-trimethoxy)benzoic acid ester

(19): A catalytic amount of 4-dimethylaminopyridine, 3,4,5-tri(dodecylox-

y)benzoic acid (3.00 g, 4.44 mmol),^[9d]4-benzyloxyphenol (0.82 g,

4.10 mmol), and dicyclohexylcarbodiimide (DCC, 1.83 g, 8.87 mmol)

were refluxed for 16 h in CH₂Cl₂(100 mL). The resulting insoluble dicy-

clohexylurea was separated by filtration and the resulting mixture evapo-

rated to dryness. The colorless oil was purified by column chromatogra-

phy (silica gel; CH₂Cl₂), then recrystallized from ethanol to give 3.29 g

(3.84 mmol, yield 84%) of (p-benzyloxyphenyl)-1-(3,4,5-trimethoxy)ben-

zoic acid ester as a white solid. M.p.=438C; ¹H NMR (CDCl₃): d=0.85

(t, J³=6.9 Hz, 9H), 1.15–1.60 (m, 54H), 1.68–1.84 (m, 6H), 3.98–4.04 (m,

6H), 5.05 (s, 2H), 6.98 (d, J³=9.3 Hz, 2H), 7.08 (t, J³=9.3 Hz, 2H), 7.28–

7.44 ppm (m, 7H); ¹³C NMR (CDCl₃): d=14.43 (C_prim); 22.91, 26.29,

29.49, 29.58, 29.61, 29.79, 29.85, 29.87, 29.91, 29.94, 30.55, 32.14, 69.41,

70.61, 73.77 (C_sec); 108.65, 115.70, 122.75, 127.70, 128.24, 128.82 (C_tert);

Data for 5^C0¹H NMR (CDCl₃): d=1.60 (s, 6H), 6.72 ppm (s, 2H); EI-

:

MS: m/z: 166 [M⁺].

Preparation of 3,4,5-trisdodecyloxyphenol (6^Cn, n=0, 1, 12): BF₃·Et₂O

(2.06 g, 14.50 mmol) and NaBO₃·4H₂O (0.45 g, 2.90 mmol) were dissolved

in dry THF (20 mL) at 08C. After 30 min stirring at 08C, a solution of

compound 5^C12(1.00 g, 1.45 mmol) in THF (5 mL) was added and the re-

action was left to reach RT, and stirred for a further 12 h. A saturated

aqueous sodium sulfite solution (10 mL) was added together with water

(100 mL). The resulting mixture was extracted with diethyl ether (3

100 mL). The combined organic phases were dried (Na₂SO₄), filtered,

and evaporated to give a brownish solid. Column chromatography (silica

gel; CH₂Cl₂) gave 0.76 g (1.17 mmol, yield 81%) of 6^C12as a air-sensitive

pale yellow solid. M.p.=488C; ¹H NMR (CDCl₃): d_H0.90 (t, J³=6.9 Hz,

18H), 1.28–1.34 (m, 48H), 1.42–1.46 (m ,6H), 1.70–1.82 (m, 6H), 3.86–

1688

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

124.18, 137.02, 143.05, 144.87, 153.13, 156.68, 165.58 ppm (C_quat); EI-MS :

m/z: 857 [M]⁺. (p-Benzyloxyphenyl)-1-(3,4,5-trimethoxy)benzoic acid

ester (2.08 g, 2.43 mmol) was then dissolved in a solution containing a

suspension of Pd/C (0.10 g, 10%) in CH₂Cl₂:CH₃OH (100 mL:10 mL).

The mixture was refluxed at RT under dihydrogen (1 bar) for one hour.

Filtration over Celite, followed by evaporation, and drying under vacuum

gave 1.85 g (2.41 mmol, yield 99%) of 19 as a white solid. M.p.=598C;

¹H NMR (CDCl₃): d=0.84 (t, J³=6.9 Hz, 9H), 1.15–1.60 (m, 54H), 1.70–

1.82 (m, 6H), 3.97–4.04 (m, 6H), 4.77 (s, 1H), 6.82 (d, J³=8.7 Hz, 2H),

7.01 (d, J³=8.7 Hz, 2H), 7.35 ppm (s, 2H); ¹³C NMR (CDCl₃): d=14.34

(C_prim); 22.91, 26.29, 29.48, 29.58, 29.61, 29.78, 29.85, 29.87, 29.91, 29.95,

30.52, 32.14, 69.42, 73.83 (C_sec); 108.64, 116.38, 122.78 (C_tert); 124.06,

143.05, 144.49, 153.13, 153.74, 166.09 ppm (C_quat); EI-MS : m/z: 767 [M]⁺.

cial position (4e). The CH₂Cl₂molecule in L5^C1·0.5CH₂Cl₂was disor-

dered about an inversion center and was refined with a population pa-

rameter of 0.5. In 17, the hydrogen atoms of the methyl group of the ni-

tromethane molecule were refined with restraints on bond lengths and

bond angles.

Spectroscopic and analytical measurements: Electronic spectra in the

UV/Vis region were recorded at 208C from solutions in CH₃CN/CH₂Cl₂

(1:1) with a Perkin–Elmer Lambda 900 spectrometer using quartz cells of

0.1 and 1 mm path length. Spectrophotometric titrations were performed

with a J&M diode array spectrometer (Tidas series) connected to an ex-

ternal computer. In a typical experiment, L5^C12in CH₃CN/CH₂Cl₂(1:1;

50 mL, 2·10^À4moldm^À3

) were titrated at 208C with a solution of

Ln(NO₃)₃·xH₂O (10^À3moldm^À3) in the same solvent under an inert at-

mosphere. After each addition of 0.10 mL, the absorbance was recorded

using Hellma optrodes (optical path length 0.1 cm) immersed in the ther-

mostated titration vessel and connected to the spectrometer. Mathemati-

Preparation of 2,6-bis-(1-ethyl-5-carboxybenzimidazol-2-yl)pyridine-

di(3,4,5-trimethoxybenzoic acid 4-hydroxyphenyl ester) ester (L6^C12): A

catalytic amount of 4-dimethylaminopyridine, compound

3 (0.30 g,

cal treatment of the spectrophotometric data was performed with factor

0.66 mmol), compound 19 (1.00 g, 1.30 mmol), and EDCI (1.01 g,

5.21 mmol) were refluxed for 12 h in CH₂Cl₂(100 mL) under an inert at-

mosphere. The resulting mixture was washed with water (3100 mL).

The organic layer was separated, dried (Na₂SO₄), and evaporated to dry-

ness. The crude yellow oil was purified by column chromatography (silica

gel; CH₂Cl₂/MeOH 99.8:0.2!99:1) to give 0.75 g (0.38 mmol, yield 58%)

[33]

analysis^[37]and with the SPECFITprogram.

IR spectra were obtained

from KBr pellets with a FT-IR Perkin–Elmer Spectrum One. ¹H and

¹³C NMR spectra were recorded at 258C on a Bruker Avance 400 MHz

and Bruker DRX-500 MHz spectrometers. Chemical shifts are given in

ppm with respect to TMS. Diffusion experiments were recorded at

400 MHz-proton-Larmor frequency at RT. The sequence corresponds to

Bruker pulse program ledbpgp2s^[38]by using stimulated echo, bipolar gra-

dients and longitudinal eddy current delay as z filter. The four 2 ms gradi-

ents pulses have sine-bell shapes and amplitudes ranging linearly from

2.5 to 50 Gcm^À1in 16 steps. The diffusion delay was 100 ms and the

number of scans was 16. The processing was done by using a line broad-

ening of 5 Hz and the diffusion rates were calculated by using the Bruker

processing package. Pneumatically assisted electrospray (ESI-MS) mass

1

of L6^C12as a white solid. H NMR (CDCl₃): d=0.85 (t, J³=6.6 Hz, 18H),

1.15–1.55 (m, 108H), 1.40 (t, J³=7.2 Hz, 6H), 1.69–1.90 (m, 12H), 4.01–

4.05 (m, 12H), 4.84 (q, J³=6.9 Hz, 4H), 7.25 (d, J³=9.0 Hz, 4H), 7.32 (d,

J³=9.0 Hz, 4H), 7.39 (s, 4H), 7.56 (d, J³=8.7 Hz, 2H), 8.11 (t, J³=

8.1 Hz, 1H), 8.23 (dd, J³=8.7 Hz, J⁴=1.2 Hz, 2H), 8.41 (d, J³=7.8 Hz,

2H), 8.76 ppm (m, 2H); ¹³C NMR (CDCl₃): d_C14.33, 15.71 (C_prim); 22.90,

26.28, 26.30, 29.51, 29.58, 29.61, 29.79, 29.85, 29.87, 29.92, 29.95, 30.56,

32.14, 40.46, 69.46, 73.79 (C_sec); 108.74, 110.47, 122.91, 123.78, 125.77,

126.64, 138.72 (C_tert); 123.95, 124.36, 139.84, 142.72, 143.23, 148.65,

148.78, 149.79, 151.88, 153.18, 165.17, 165.63 ppm (C_quat.); ESI-MS

(CH₂Cl₂/MeOH 8:2 + HCl 0.1%): m/z: 1953.9 [M+H]⁺; elemental anal-

ysis calcd (%) for C₁₂₃H₁₈₁N₅O₁₄·1.57H₂O (1982.06): C 74.54, H 9.36, N

3.53; found: C 74.54, H 9.41, N 3.45.

spectra were recorded from 10^À4moldm^À3solutions on

a Finnigan

SSQ7000 instrument. Quantum yields were determined using a Perkin–

Elmer LS50B fluorimeter. The quantum yields were calculated using the

F_x

F_r

A_rð~nÞI_rð~nÞn²_xD_x

equation

=

_{ðnÞn D}, in which x refers to the sample and r to the ref-

2

r

A_xðnÞI_x

~

r

erence; A is the absorbance, n˜ the excitation wavenumber used, I the in-

tensity of the excitation light at this energy, n the refractive index (n=

1.341 for acetonitrile solution and n=1.330 for 0.1 moldm^À3aqueous

tris-buffer solution) and D the integrated emitted intensity. Cs₃[Eu(2,6-

pyridinedicarboxylic acid)₃] (f=9.5% in 0.1 moldm^À3aqueous tris-buffer

solution) was used as reference.^[39]TG were performed with a thermogra-

vimetric balance Seiko TG/DTA 320 (under N₂). DSC traces were ob-

tained with a Seiko DSC 220C differential scanning calorimeter from 3–

5 mg samples (58Cmin^À1, under N₂). The characterization of the meso-

phases were performed with a polarizing microscope Leitz Orthoplan-Pol

with a Leitz LL 20/0.40 polarizing objective, and equipped with a

Linkam THMS 600 variable-temperature stage. The SAXS patterns were

obtained with two different experimental setups, and in all cases, the

crude powder was filled in Lindemann capillaries of 1 mm diameter. The

characterization of the wide-angle region and the measurements of the

periodicities were achieved using a linear monochromatic Cu_Kabeam ob-

tained with a sealed-tube generator (900 W) and a bent quartz mono-

chromator. One set of diffraction patterns was registered with a curved

counter Inel CPS 120, for which the sample temperature was controlled

within ꢂ0.058C; periodicities up to 60 could be measured. The other

set of diffraction patterns was registered on an image plate; the cell pa-

rameters were calculated from the position of the reflection at the small-

est Bragg angle, which was in all cases the most intense. Periodicities up

to 90 could be measured, and the sample temperature was controlled

within ꢂ0.38C. The exposure times were varied from 1 to 24 h depending

on the specific reflections being sought (weaker reflections obviously

taking longer exposure times). Elemental analyses were performed by

Dr. H. Eder from the microchemical Laboratory of the University of

Geneva.

Preparation of the complexes [Ln(L5^C12)(NO₃)₃] (Ln=Sm-Lu): A solu-

tion of compound L5^C12(0.1 g, 5.8410^À5mol) in dichloromethane

(5 mL) was added to Ln(NO₃)₃·xH₂O (Ln=Sm–Lu, x=2–4) in acetoni-

trile (5 mL). After 1 h stirring at RT, the dichloromethane was evaporat-

ed, the solution was cooled to À208C, and the white precipitate was fil-

tered, washed with CH₃CN, and dried to give 80% of

[Ln(L5^C12)(NO₃)₃]·nH₂O (Ln=Sm, n=3.5, 7; Ln=Eu, n=2.5, 8; Ln=

Gd, n=3, 9; Ln=Tb, n=0, 10; Ln=Dy, n=0, 11; Ln=Ho, n=1.5, 12;

Ln=Er, n=0, 13; Ln=Tm, n=1, 14; Ln=Yb, n=0.5, 15; Ln=Lu, n=

1.5, 16). All the complexes were characterized by satisfying elemental

analyses (Table S9, Supporting Information) and IR spectra. X-ray quali-

ty

prisms

of

[Eu(2)(NO₃)₃(H₂O)](CH₃NO₂)

(17)

and

[Yb(L5^C0)(NO₃)₃](CH₃CN)₂(18) were obtained by slow diffusion of di-

ethyl ether into concentrated solutions of the complexes in nitromethane

and acetonitrile, respectively.

Single crystal structure determinations: Summary of crystal data, intensi-

ty measurements and structure refinements for 2, L5^C1·0.5CH₂Cl₂,

[Eu(2)(NO₃)₃(H₂O)](CH₃NO₂) (17) and [Yb(L5^C0)(NO₃)₃](CH₃CN)₂(18)

are collected in Table S16 (Supporting Information). All crystals were

mounted on quartz fibers with protection oil. Cell dimensions and inten-

sities were measured at 200 K on a Stoe IPDS diffractometer with graph-

ite-monochromated Mo_Karadiation (l=0.71073 ). Data were corrected

for Lorentz and polarization effects and for absorption. The structures

were solved by direct methods (SIR97),^[34]all other calculations were per-

[35]

formed with the XTAL

system and ORTEP^[36]programs. CCDC-

611681–611684 (2, L5^C1·0.5CH₂Cl₂, [Eu(2)(NO₃)₃(H₂O)](CH₃NO₂) (17)

and [Yb(L5^C0)(NO₃)₃](CH₃CN)₂(18) respectively) contain the supple-

mentary 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.

Computational details: The geometries of ligands L4^C1, L5^C1and com-

plexes [Lu(L4^C1)(NO₃)₃] and [Lu(L5^C1)(NO₃)₃] were fully optimized at

[40]

the DFTlevel using the B3LYP

hybrid functional as implemented in

the Gaussian 03 package.^[41]The crystal structures of L4^C1

,

L5^C1

,

[9d]

In 2 and 17, the hydrogen atoms were observed and refined with U_iso

=

0.04 ². The atomic positions of the other hydrogen atoms were calculat-

[Yb(L4^C0)(NO₃)₃]^[9d]and [Yb(L5^C0)(NO₃)₃] were used as starting struc-

tural models and geometry optimizations were restricted to the C₂sym-

ed. For 2, the ligand was located on a twofold axis with N1 and C3 in spe-

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1689

C. Piguet, P.-Y. Morgantini et al.

metry point group. For all Gaussian 03 calculations, DZVP double-z

basis set developed by Godbout and co-workers^[42]was used for the H, C,

and O atoms and the lutetium cation was described by the quasi-relativis-

tic pseudopotential of Dolg and co-workers^[43]for the 46 + 4fⁿcore elec-

trons and by a (7 s, 6p, 5d)/(5 s, 4p, 3d) Gaussian basis set for the valence

electrons. Fitted atomic charges were calculated according to the Merz–

Singh–Kollmann scheme^[18]using 25 points per layer. Connolly surfaces

were generated using Molekel 4.3^[44]and were then used to calculate mo-

lecular electrostatic potential (MEP). TD-DFT calculations of singlet–

singlet and singlet–triplet excitation energies^[45]were performed by using

the Amsterdam density functional program (ADF 2005.01).^[46]A Slater-

type orbitals TZ2P all-electron basis set (double-z in core, triple-z in the

valence region + 2 polarization functions) was used and relativistic ef-

fects were introduced using the Zero Order Regular Approximation

(ZORA) formalism.^[47]To ensure a correct asymptotic behavior to the ex-

change-correlation potential, the Statistical Average of Orbital Potential

(SAOP)^[48]was chosen for the TD-DFT calculations.

[8] a) C. Piechocki, J. Simon, J. J. AndrØ, D. Guillon, P. Petit, A. Skou-

lios, P. Weber, Chem. Phys. Lett. 1985, 122, 124–128; b) H. Miwa, N.

Kobayashi, K. Ban, K. Ohta, Bull. Chem. Soc. Jpn. 1999, 72, 2719–

2728; c) K. Binnemans, J. Sleven, S. De Feyter, F. C. De Schryver, B.

Donnio, D. Guillon, Chem. Mater. 2003, 15, 3930–3938.

[9] a) K. Binnemans, K. Lodewyckx, B. Donnio, D. Guillon, Chem. Eur.

J. 2002, 8, 1101–1105; b) K. Binnemans, K. Lodewyckx, Supramol.

Chem. 2003, 15, 485–494; c) E. Terazzi, J.-M. BØnech, J.-P. Rivera,

G. Bernardinelli, B. Donnio, D. Guillon, C. Piguet, Dalton Trans.

2003, 769–772; d) E. Terazzi, S. Torelli, G. Bernardinelli, J.-P.

Rivera, J.-M. BØnech, C. Bourgogne, B. Donnio, D. Guillon, D.

Imbert, J.-C. G. Bünzli, A. Pinto, D. Jeannerat, C. Piguet, J. Am.

Chem. Soc. 2005, 127, 888–903; e) K. Binnemans, K. Lodewyckx, B.

Donnio, D. Guillon, Eur. J. Inorg. Chem. 2005, 1506–1513; f) K.

Binnemans, K. Lodewyckx, T. Cardinaels, T. N. Parac-Vogt, C. Bour-

gogne, D. Guillon, B. Donnio, Eur. J. Inorg. Chem. 2006, 150–157.

[10] a) T. Cardinaels, K. Driesen, T. N. Parac-Vogt, B. Heinrich, C. Bour-

gogne, D. Guillon, B. Donnio, K. Binnemans, Chem. Mater. 2005, 17,

6589–6598; b) E. Terazzi, B. Bocquet, S. Campidelli, B. Donnio, D.

Guillon, R. Deschenaux, C. Piguet, Chem. Commun. 2006 2922–

2924.

Acknowledgements

[11] a) C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112,

5525–5534; b) C. A. Hunter, Chem. Soc. Rev. 1994, 23, 101–109;

c) C. A. Hunter, K. R. Lawson, J. Perkins, C. J. Urch, J. Chem. Soc.

Perkin Trans. 2 2001, 651–669; d) S. L. Cockroft, C. A. Hunter,

K. R. Lawson, J. Perkins, C. J. Urch, J. Am. Chem. Soc. 2005, 127,

8594–8595.

We thank H. Lartigue, K.-L. Buchwalder and B. Bocquet for their techni-

cal assistance and Dr T. Wesolowski for fruitful discussions. Financial

support from the Swiss National Science Foundation is gratefully ac-

knowledged.

[12] a) J. P. Bedel, H. T. Nguyen, J. C. Rouillon, J. P. Marcerou, G.

Sigaud, P. Barois, P. Mol. Cryst. Liq. Cryst. 1999, 332, 163–171;

b) H. T. Nguyen, J. C. Rouillon, J. P. Marcerou, J. P. Bedel, P. Barois,

S. Sarmento, Mol. Cryst. Liq. Cryst. 1999, 328, 177–184; c) J. P.

Bedel, J. C. Rouillon, J. P. Marcerou, M. Laguerre, H. T. Nguyen,

M. F. Achard, Liq. Cryst. 2000, 27, 1411–1421; d) J. P. Bedel, J. C.

Rouillon, J. P. Marcerou, M. Laguerre, M. F. Achard, H. T. Nguyen,

Liq. Cryst. 2000, 27, 103–113; e) J. P. Bedel, J. C. Rouillon, J. P. Mar-

cerou, M. Laguerre, H. T. Nguyen, M. F. Achard, Liq. Cryst. 2001,

28, 1285–1292.

[13] J. J. Reczek, K. R. Villazor, T. M. Swager, B. L. Iverson, J. Am.

Chem. Soc. 2006, 128, 7995–8002.

[14] a) R. Deschenaux, J.-L. Marendaz, J. Santiago, Helv. Chim. Acta

1993, 76, 865–876; b) K. E. Rowe, D. W. Bruce, J. Mater. Chem.

1998, 8, 331–341.

[15] J. H. Markgraf, B. Y. Choi, Synth. Commun. 1999, 29, 2405–2407.

[16] G. W. Kabalka, N. K. Reddy, C. Narayana, Tetrahedron Lett. 1993,

34, 7667–7668.

[17] F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen,

R. Taylor, J. Chem. Soc. Perkin Trans. II 1987, S1-S19.

[18] U. C. Singh, P. A. Kollman, J. Comput. Chem. 1984, 5, 129–145.

[19] R. D. Shannon, Acta Crystallogr. Sect. A 1976, 32, 751–767.

[20] J.-C. G. Bünzli, B. Klein, G. Chapuis, K. J. Schenk, Inorg. Chem.

1982, 21, 808–812.

[21] a) H. Gampp, M. Maeder, C. J. Meyer, A. Zuberbühler, Talanta

1985, 32, 1133–1139; b) H. Gampp, M. Maeder, C. J. Meyer, A. Zu-

berbühler, Talanta 1986, 33, 943–951.

[1] “Thermotropic Liquid Crystals”: K. J. Toyne, Crit. Rep. Appl. Chem.

1987, 22, 28–63.

[2] a) D. Guillon, A. Skoulios, J. Phys. 1976, 37, 797–800; b) E. Terazzi,

S. Suarez, S. Torelli, H. Nozary, D. Imbert, O. Mamula, J.-P. Rivera,

E. Guillet, J.-M. BØnech, G. Bernardinelli, R. Scopelliti, B. Donnio,

D. Guillon, J.-C. G. Bünzli, C. Piguet, Adv. Funct. Mater. 2006, 16,

157–168; c) C. Piguet, J.-C. G. Bünzli, B. Donnio, D. Guillon, Chem.

Commun. 2006, 3755–3768.

[3] a) J. L. Serrano, Metallomesogens, Synthesis, Properties and Applica-

tions, VCH, Weinheim, 1996; b) B. Donnio, D. Guillon, R. Desche-

naux, D. W. Bruce, in Comprehensive Coordination Chemistry, Vol. 7

(Eds.: J. A. McCleverty, T. J. Meyer), Elsevier, Oxford (UK), 2003,

Chapter 7.9, pp. 357–627; c) R. Deschenaux, J. W. Goodby, in Ferro-

cene: Homogeneous Catalysis, Organic Synthesis, Material Sciences

(Eds.: A. Togni, T. Hayashi), VCH, Weinheim (Germany), 1995,

Chapter 9, pp. 471–495.

[4] a) A.-M. Giroud-Godquin, P. M. Maitlis, Angew. Chem. 1991, 103,

370–398; Angew. Chem. Int. Ed. Engl. 1991, 30, 375–402; b) S. A.

Hudson, P. M. Maitlis, Chem. Rev. 1993, 93, 861–885; c) D. W.

Bruce, J. Chem. Soc. Dalton Trans. 1993, 2983–2989; d) A.-M.

Giroud-Godquin, Coord. Chem. Rev. 1998, 178–180, 1485; e) P. Espi-

net, Gold Bull. 1999, 32, 127–225.

[5] For a comprehensive review, see K. Binnemans, C. Gçrller-Walrand,

Chem. Rev. 2002, 102, 2303–2346.

[6] H. Nozary, C. Piguet, P. Tissot, G. Bernardinelli, J.-C. G. Bünzli, R.

Deschenaux, D. Guillon, J. Am. Chem. Soc. 1998, 120, 12274–

12288.

[22] E. Terazzi, J.-P. Rivera, N. Ouali, C. Piguet, Magn. Reson. Chem.

2006, 44, 539–552.

[23] H. Günther, NMR Spectroscopy, Wiley, Chichester, New York, 1980,

pp. 74–86.

[7] a) Y. Galyametdinov, M. A. Athanassopoulou, K. Griesar, O. Khari-

tonova, E. A. Soto Bustamante, L. Tinchurina, I. Ovchinnikov, W.

Haase, Chem. Mater. 1996, 8, 922–926; b) K. Binnemans, Y. G. Ga-

lymetdinov, S. R. Collinson, D. W. Bruce, J. Mater. Chem. 1998, 8,

1551–1553; c) K. Binnemans, Y. G. Galyametdinov, R. Van Deun,

D. W. Bruce, S. R. Collinson, A. P. Polishchuk, I. Bikchantaev, W.

Haase, A. V. Prosvirin, L. Tinchurina, U. Litvinov, A. Gubajdullin,

A. Rakhmatullin, K. Uytterhoeven, L. Van Meervelt, J. Am. Chem.

Soc. 2000, 122, 4335–4344; d) R. Van Deun, K. Binnemans, Liq.

Cryst. 2001, 28, 621–627; e) Y. Galyametdinov, W. Haase, L. Maly-

khina, A. Prosvirin, I. Bikchantaev, A. Rakhmatullin, K. Binnemans,

Chem. Eur. J. 2001, 7, 99–105; f) K. Binnemans, D. Moors, T. N.

Parac-Vogt, R. Van Deun, D. Hinz-Hübner, G. Meyer, Liq. Cryst.

2002, 29, 1209–1216.

[24] a) P. Stilbs, Prog. Nucl. Magn. Reson. Spectrosc. 1986, 19, 1–45;

b) A. R. Waldeck, P. W. Kuchel, A. J. Lennon, B. E. Chapman, Prog.

Nucl. Magn. Reson. Spectrosc. 1997, 30, 39–68; c) P. S. Pregosin, E.

Martinez-Viviente, P. G. A. Kumar, Dalton Trans. 2003, 4007–4014;

d) P. S. Pregosin, P. G. P. G. A. Kumar, I. Fernandez, Chem. Rev.

2005, 105, 2977–2998.

[25] a) O. Kahn, Adv. Inorg. Chem. 1995, 43, 179–259; b) O. Kahn, Mo-

lecular Magnetism, VCH, Weinheim (Germany), 1993, pp. 148–184;

c) J.-M. Senegas, G. Bernardinelli, D. Imbert, J.-C. G. Bünzli, P.-Y.

Morgantini, J. Weber, C. Piguet, Inorg. Chem. 2003, 42, 4680–4695;

1690

www.chemeurj.org

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2007, 13, 1674 – 1691

Liquid Crystals

FULL PAPER

d) J. CurØly, B. Barbara, Struct. Bonding (Berlin) 2006, 122, 207–

[40] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.

250.

[41] Gaussian 03, Revision B.03, M. J. Frisch, G. W. Trucks, H. B. Schle-

gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-

y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-

gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.

Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.

Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,

O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,

J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.

Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.

Ochterski, P.-Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.

Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.

Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,

J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-

slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-

mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,

A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.

Chen, M. W. Wong, C. Gonzalez, J. Pople, J.A.; Gaussian, Inc.,

Wallingford CT, 2004.

[26] H. Nozary, C. Piguet, J.-P. Rivera, P. Tissot, P.-Y. Morgantini, J.

Weber, G. Bernardinelli, J.-C. G. Bünzli, R. Deschenaux, B. Donnio,

D. Guillon, Chem. Mater. 2002, 14, 1075–1090.

[27] J.-C. G. Bünzli, C. Piguet, Chem. Rev. 2002, 102, 1897–1928.

[28] W. T. Carnall, P. R. Fields, K. Rajnak, J. Chem. Phys. 1968, 49,

4443–4446.

[29] a) S. Tobita, M. Arakawa, I. Tanaka, J. Phys. Chem. 1984, 88, 2697–

2702; b) S. Tobita, M. Arakawa, I. Tanaka, J. Phys. Chem. 1985, 89,

5649–5654.

[30] C. Piguet, A. F. Williams, G. Bernardinelli, E. Moret, J.-C. G. Bünzli,

Helv. Chim. Acta 1992, 75, 1697–1717.

[31] a) K. Binnemans, L. Malykhina, V. S. Mironov, W. Haase, K. Drie-

sen, R. Van Deun, L. Fluyt, C. Gçrller-Walrand, Y. G. Galyametdi-

nov, ChemPhysChem 2001, 2, 680–683; b) K. Binnemans, D. Moors,

J. Mater. Chem. 2002, 12, 3374–3376; c) S. Suarez, O. Mamula, D.

Imbert, C. Piguet, J.-C. G. Bünzli, Chem. Commun. 2003, 1226–

1227; d) R. Van Deun, B. De FrØ, D. Moors, K. Binnemans, J. Mater.

Chem. 2003, 13, 1520–1522; e) E. Guillet, D. Imbert, R. Scopelliti,

J.-C. G. Bünzli, Chem. Mater. 2004, 16, 4063–4070; f) S. Suarez, D.

Imbert, F. Gumy, C. Piguet, J.-C. G. Bünzli, Chem. Mater. 2004, 16,

3257–3266; g) L. N. Puntus, K. J. Schenk, J.-C. G. Bünzli, Eur. J.

Inorg. Chem. 2005, 4739–4744.

[42] N. Godbout, D. R. Salahub, J. Andzelm, E. Wimmer, Can. J. Chem.

1992, 70, 560.

[43] a) M. Dolg, H. Stoll, A. Savin, H. Preuss, Theor. Chim. Acta 1989,

75, 173; b) M. Dolg, H. Stoll, A. Savin, H. Preuss, Theor. Chim.

Acta 1993, 85, 441.

[32] J.-C. G. Bünzli, E. Moret, J.-R. Yersin, Helv. Chim. Acta 1978, 61,

762.

[33] G. Schwarzenbach, Complexometric Titrations; Chapman & Hall,

London, 1957, p. 8.

[34] A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Giacovaz-

zo, A. Guagliardi, G. Moliterni, G. Polidori, R. Spagna, J. Appl.

Crystallogr. 1999, 32, 115.

[44] MOLEKEL 4.3, P. Flükiger, H. P. Lüthi, S. Portmann, J. Weber,

Swiss Center for Scientific Computing, Manno (Switzerland), 2000–

2002 (http://www.cscs.ch/molekel/).

[45] a) S. J. A. van Gisbergen, J. G. Snijders, E. J. Baerends, Comput.

Phys. Commun. 1999 118, 119; b) A. Rosa, E. J. Baerends, S. J. A.

van Gisbergen, E. van Lenthe, J. A. Groeneveld J. G. Snijders, J.

Am. Chem. Soc. 1999, 121, 10356.

[35] XTAL 3.2, Userꢁs Manual (Eds.: S. R. Hall, H. D. Flack, J. M. Stew-

art), Universities of Western Australia and Maryland, 1989.

[36] C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge National

Laboratory: Oak Ridge, Tennessee, 1976.

[46] ADF2005.01, SCM, Theoretical Chemistry, Vrije Universiteit, Am-

sterdam (The Netherlands), http://www.scm.com.

[47] E. van Lenthe, A. E. Ehlers, E. J. Baerends, J. Chem. Phys. 1999,

110, 8943.

[37] E. R. Malinowski, D. G. Howery, Factor Analysis in Chemistry,

Wiley, New York, Chichester, 1980.

[48] P. R. T. Schipper, O. V. Gritsenko, S. J. A. van Gisbergen, E. J. Baer-

ends, J. Chem. Phys. 2000, 112, 1344.

[38] D. Wu, A. Chen, C. S. Johnson, Jr., J. Magn. Reson. A 1995, 155,

260.

[39] A.-S. Chauvin, F. Gumy, D. Imbert, J.-C. G. Bünzli, Spectrosc. Lett.

2004, 37, 517.

Received: September 28, 2006

Published online: January 16, 2007

Chem. Eur. J. 2007, 13, 1674 – 1691

ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

1691

Article Doi

Tuning the polarization along linear polyaromatic strands for rationally inducing mesomorphism in lanthanide nitrate complexes

DOI: 10.1002/chem.200601389

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.200601389

Full text of DOI:10.1002/chem.200601389

Products guided by the article

R&D Labs maybe for 934823-80-0

Relevant to this article

Practical syntheses of both enantiomers of cyclopropylglycine and of methyl 2-cyclopropyl-2-N-Boc-iminoacetate

Exceptional stereoselectivity in the synthesis of 1,3,4-trisubstituted 4-carboxy β-lactam derivatives from amino acids

Substituent Effect Studies of Aryl-Assisted Solvolyses. II. The Acetolysis of 2-Phenyl-2-(substituted phenyl)ethyl p-Toluenesulfonates

Silaindene - eine einfache Synthese

Selective NaNO₂ recognition by a simple heteroditopic salt receptor based on l-ornithine molecular scaffold

2,2'-Bipyridine lariat calixcrowns: A new class of encapsulating ligands forming highly luminescent Eu³⁺ and Tb ³⁺ complexes

Article Doi

Tuning the polarization along linear polyaromatic strands for rationally inducing mesomorphism in lanthanide nitrate complexes

DOI: 10.1002/chem.200601389

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.200601389

Full text of DOI:10.1002/chem.200601389

Products guided by the article

R&D Labs maybe for 934823-80-0

Relevant to this article

Practical syntheses of both enantiomers of cyclopropylglycine and of methyl 2-cyclopropyl-2-N-Boc-iminoacetate

Exceptional stereoselectivity in the synthesis of 1,3,4-trisubstituted 4-carboxy β-lactam derivatives from amino acids

Substituent Effect Studies of Aryl-Assisted Solvolyses. II. The Acetolysis of 2-Phenyl-2-(substituted phenyl)ethyl p-Toluenesulfonates

Silaindene - eine einfache Synthese

Selective NaNO2 recognition by a simple heteroditopic salt receptor based on l-ornithine molecular scaffold

2,2'-Bipyridine lariat calixcrowns: A new class of encapsulating ligands forming highly luminescent Eu3+ and Tb 3+ complexes

Selective NaNO₂ recognition by a simple heteroditopic salt receptor based on l-ornithine molecular scaffold

2,2'-Bipyridine lariat calixcrowns: A new class of encapsulating ligands forming highly luminescent Eu³⁺ and Tb ³⁺ complexes