Dimerization of dendrimeric lanthanide complexes: Thermodynamic, thermal, and liquid-crystalline properties

Inorg. Chem. 2010, 49, 8601–8619 8601

DOI: 10.1021/ic101220v

Dimerization of Dendrimeric Lanthanide Complexes: Thermodynamic, Thermal,

and Liquid-Crystalline Properties

†

ꢀ ꢀ

Thomas Binderup Jensen, Emmanuel Terazzi, Kerry-Lee Buchwalder, Laure Guenee, Homayoun Nozary,

‡

§

,†

ˆ

Kurt Schenk, Benoıt Heinrich, Bertrand Donnio, Daniel Guillon, and Claude Piguet*

^†Department of Inorganic, Analytical and Applied Chemistry, University of Geneva, 30 quai E. Ansermet,

§

ꢀ

CH-1211 Geneva 4, Switzerland, Groupe des Materiaux Organiques, Institut de Physique et Chimie des

ꢀ

Materiaux de Strasbourg (IPCMS), 23 rue du Loess, B.P. 43, F67034 Strasbourg Cedex 2, France, and

‡

ꢀ ꢀ

Laboratory of Crystallography, Ecole Polytechnique Federale de Lausanne, CH-1015 Lausanne, Switzerland

Received June 18, 2010

A series of 10 different mesomorphic semidendrimeric tridentate ligands L5-L14 grafted with terminal cyanobiphenyl

groups have been synthesized. Upon reaction with Ln(NO₃)₃(Ln = trivalent lanthanide), the central 2,6-bis(N-ethyl-

benzimidazol-2-yl)pyridine unit is meridionally tricoordinated to the metal to give rodlike monomeric [Ln(Lk)(NO₃)₃]

and H-shaped dimeric [Ln₂(Lk)₂(NO₃)₆] complexes. For the small Lu^IIIcation, the monomeric complexes are

quantitatively formed in a noncoordinating CD₂Cl₂solution. For larger cations (Ln = Eu, Pr), the thermodynamic

equilibrium 2[Ln(Lk)(NO₃)₃] T [Ln₂(Lk)₂(NO₃)₆] can be evidenced across the complete ligand series. Detailed

thermodynamic studies show that the dimeric complexes result from the formation of primary intermetallic nitrate

bridges whose strength depends on the metallic size. For each complex, secondary nonspecific interstrand van der

Waals interactions produce nonartifactual enthalpy/entropy compensation. In the absence of solvent, only the com-

plexes with the most extended ligands L5 and L6 produce thermotropic mesophases. Layered organizations are

dominant (smectic A) with the induction of nematogenic behavior at high temperature when interstrand interactions are

modulated by methyl substitutions. Correlations between the trend of dimerization and the sequences of thermotropic

mesophases are attempted.

Introduction

as a thermotropic mesophase. When the incriminated amphi-

philic molecule contains a metallic center, i.e., a metallo-

mesogen, the considerable three-dimensional expansion

brought by the coordination sphere of the metal seriously

disturbs intermolecular interactions, and successful micro-

segregation was originally induced for rodlike two-coordinate

or disklike square-planar four-coordinate metallic complexes.²

Only a few metallomesogens have either a tetrahedral³or an

octahedral⁴metal center, because these coordination geo-

metries are difficult to combine with the requirement of

structural anisotropy and efficient intermolecular interactions

Thermotropic liquid crystals are usually produced by the

melting of amphiphilic molecules possessing a rigid polariz-

able aromatic core grafted with long and flexible alkyl chains.

Optimization of theintermolecular interactions in the crystal-

line state results in a microsegregation process, in which the

rigid cores are packed together, while the less polarizable

flexible alkyl chains fill the residual voids of the structure.¹

Upon heating, decorrelation between the flexible alkyl chains

produces a molten continuum, in which the clusters of inter-

acting rigid cores are dispersed, thus forming what is known

(3) (a) Neve, F.; Ghedini, M.; De Munno, G.; Levelut, A. M. Chem. Mater.

1995, 7, 688–693. (b) Douce, L.; El-ghayouri, A.; Skoulios, A.; Ziessel, R. Chem.

*To whom correspondence should be addressed. E-mail: Claude.Piguet@

unige.ch.

ꢀ

(1) (a) Guillon, D.; Skoulios, A. J. Phys. (Paris) 1976, 37, 797–800.

(b) Skoulios, A.; Guillon, D. Mol. Cryst. Liq. Cryst. 1988, 165, 317–332.

(2) (a) Giroud-Godquin, A. M.; Maitlis, P. M. Angew. Chem., Int. Ed.

Engl. 1991, 30, 375–402. (b) Bruce, D. W. J. Chem. Soc., Dalton Trans. 1993,

2983–2989. (c) Hudson, S. A.; Maitlis, P. M. Chem. Rev. 1993, 93, 861–885.

(d) Serrano, J. L. Metallomesogens, Synthesis, Properties and Applications;

VCH: Weinheim, Germany, 1996. (e) Donnio, B.; Bruce, D. W. Struct. Bonding

(Berlin) 1999, 95, 194–247. (f) Donnio, B.; Guillon, D.; Deschenaux, R.; Bruce,

D. W. In Comprehensive Coordination Chemistry; McCleverty, J. A., Meyer,

T. J., Eds.; Elsevier: Oxford, U.K., 2003; Vol. 7, Chapter 79. (g) Date, R. W.;

Fernandez Iglesias, E.; Rowe, K. E.; Elliott, J. M.; Bruce, D. W. Dalton Trans.

2003, 1914–1931.

Commun. 1999, 2033–2034. (c) Gimenez, R.; Manrique, A. B.; Uriel, S.; Barbera, J.;

Serrano, J. L. Chem. Commun. 2004, 2064–2065. (d) Cavero, E.; Uriel, S.; Romero,

P.; Serrano, J. L.; Gimenez, R. J. Am. Chem. Soc. 2007, 129, 11698–11618.

ꢀ

(4) (a) Lattermann, G.; Schmidt, S.; Kleppinger, R.; Wendorf, J. H. Adv.

Mater. 1992, 4, 30–33. (b) Zheng, H.; Swager, T. M. J. Am. Chem. Soc. 1994,

116, 761–762. (c) Swager, T. M.; Zheng, H. Mol. Cryst. Liq. Cryst. 1995, 260,

301–306. (d) Bruce, D. W.; Liu, X. H. Liq. Cryst. 1995, 18, 165–166. (e) Liu,

X.-H.; Abser, M. N.; Bruce, D. W. J. Organomet. Chem. 1998, 551, 271–280.

(f) Walf, G. H.; Benda, R.; Litterst, F. J.; Stebani, U.; Schmidt, S.; Lattermann, G.

Chem. Eur. J. 1998, 4, 93–99. (g) Rowe, K. E.; Bruce, D. W. J. Chem. Soc.,

Dalton Trans. 1996, 3913–3915. (h) Morrone, S.; Guillon, D.; Bruce, D. W.

Inorg. Chem. 1996, 35, 7041–7048.

;

r

2010 American Chemical Society

Published on Web 08/18/2010

pubs.acs.org/IC

8602 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Figure 2. Chemical structures and mesogenic behaviors of the symmetri-

cal dendrimeric complexes [Ln(Lk)(NO₃)₃] (Lk=L5 and L6). The phase

sequences and transition temperatures are given for Ln = Eu as a typical

example (g = glass, SmA = smectic A, N = nematic, and I = isotropic

liquid).^10b

.

for these materials remains to be developed. To the best of

our knowledge and if we exclude the lanthanide doping of

commercially available nematic mesogens,⁷the complexes

[Ln(Lk)(β-diketonate)₃] (Lk = L1-L4) are the only lantha-

nide-containing compounds exhibiting thermotropic nematic

mesophases over a temperature domain larger than 10ꢀ

(Figure 1).⁸As a guideline for inducing nematogenic beha-

vior, the authors introduced some unsymmetrical coordi-

nated ligands in the complexes [Ln(Lk)(β-diketonate)₃] that

are expected to destabilize the residual lateral interactions

responsible for the formation of layers in smectic mesophases

(Figure 1).⁸Parallel contributions by Deschenaux and co-

workers aiming at the encapsulation of bulky isotropic full-

erene into nematic mesophases indeed reach the same con-

clusion withthe introduction of a large carbone allotrope into

unsymmetrical dendrimers.⁹

Inspired by this reasoning, we suspected that nematic lantha-

nidomesogens may also be induced by using symmetrical com-

plexes according to the fact that the periodic organization

found in the smectic phase is disrupted by specific substitutions

that hinder lateral stacking interactions. The dendrimeric

receptor L5 was originally designed for inducing calamitic

mesomorphism in thermotropic lanthanidomesogens thanks

to the connection of four terminal cyanobiphenyl smectogenic

units (Figure 2).¹⁰The nitrato complexes [Ln(L5)(NO₃)₃]

indeed forms smectic A mesophases in the 80-180 ꢀC tem-

perature range with the heavier lanthanides (Ln=Eu-Lu).^10b

Building on these preliminary observations, we showed that the

simple introduction of methyl groups within the cyanobiphenyl

units in L6 seemed to destabilize the smectic A organization,

and a nematic mesophase could be observed over a few degrees

prior to isotropization for the symmetrical complexes

[Ln(L6)(NO₃)₃] (Ln = Eu, Lu; Figure 2).¹¹

Figure 1. Chemical structures of the nematogenic complexes [Ln(Lk)-

(β-diketonate)₃] (Lk=L1-L4). The phase sequences and transition tem-

peratures are given for Ln = Eu as a typical example (g = glass, Cr =

crystal, SmA=smectic A, N=nematic, and I=isotropic liquid).⁸.

compatible with the formation of a mesophase. Because of

stronger geometrical constraints, it is even more difficult to

obtain metallomesogens with coordination numbers larger

than 6, and the number of systems providing thermotropic

lanthanide-containing mesophases (i.e., lanthanidomesogens)

remains limited.^5,6Roughly speaking, rodlike salicylaldimine

Schiff bases (or closely related bidentate ligands) coordinated

to trivalent lanthanides yield lamellar smectic phases, while

disklike polysubstituted phthalocyanine sandwich complexes

or polycatenar complexes give columnar and cubic meso-

phases.^5,6Despite its suitability for practical applications,⁷

the fluid nematic organization has been rarely observed for

thermotropic lanthanidomesogens,⁸and a rational approach

€

(5) For comprehensive reviews, see: (a) Binnemans, K.; Gorller-Walrand,

C. Chem. Rev. 2002, 102, 2303–2346. (c) Piguet, C.; B€unzli, J.-C. G.; Donnio,

B.; Guillon, D. Chem. Commun. 2006, 3755–3768.

(6) For recent reports on lanthanido- and actinidomesogens, see:

(a) Sessler, J. L.; Melfi, P. J.; Tomat, E.; Callaway, W.; Huggins, M. T.;

Gordon, P. L.; Webster, K. D.; Date, R. W.; Bruce, D. W.; Donnio, B.

J. Alloys Compd. 2006, 418, 171–177. (b) Binnemans, K.; Lodewyckx, K.;

Cardinaels, T.; Parac-Vogt, T. N.; Bourgogne, C.; Guillon, D.; Donnio, B. Eur. J.

Inorg. Chem. 2006, 150–157. (c) Cardinaels, T.; Ramaekers, J.; Nockemann, P.;

Driesen, K.; Van Hecke, K.; Van Meervelt, L.; Lei, S.; De Feyter, S.; Guillon, D.;

Donnio, B.; Binnemans, K. Chem. Mater. 2008, 20, 1278–1291.

(9) (a) Deschenaux, R.; Donnio, B.; Guillon, D. New J. Chem. 2007, 31,

1064–1073. (b) Lenoble, J.; Campidelli, S.; Maringa, N.; Donnio, B.; Guillon, D.;

Yevlampieva, N.; Deschenaux, R. J. Am. Chem. Soc. 2007, 129, 9941–9952.

(10) (a) Dardel, B.; Guillon, D.; Heinrich, B.; Deschenaux, R. J. Mater.

Chem. 2001, 11, 2814–2831. (b) Campidelli, S.; Lenoble, J.; Barbera, J.;

Paolucci, F.; Marcaccio, M.; Paolucci, D.; Deschenaux, R. Macromolecules

2005, 38, 7915–7925. (c) Terazzi, E.; Bocquet, B.; Campidelli, S.; Donnio, B.;

Guillon, D.; Deschenaux, R.; Piguet, C. Chem. Commun. 2006, 2922–2924.

(11) Jensen, T. B.; Terazzi, E.; Donnio, B.; Guillon, D.; Piguet, C. Chem.

Commun. 2008, 181–183.

ꢀ

(7) (a) van Deun, R.; De Fre, B.; Moors, D.; Binnemans, K. J. Mater.

Chem. 2003, 13, 1520–1522. (b) Driesen, K.; Moors, D.; Beeckman, J.; Neyts, K.;

€

Gorller-Walrand, C.; Binnemans, K. J. Luminesc. 2007, 127, 611–615 and

references therein.

(8) (a) Cardinaels, T.; Driesen, K.; Parac-Vogt, T. N.; Heinrich, B.;

Bourgogne, C.; Guillon, D.; Donnio, B.; Binnemans, K. Chem. Mater.

2005, 17, 6589–6598. (b) Galyametdinov, Y. G.; Knyazev, A. A.; Dzhabarov,

€

V. I.; Cardinaels, T.; Driesen, K.; Gorller-Walrand, C.; Binnemans, K. Adv.

Mater. 2008, 20, 252–257.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8603

Figure 3. Chemical structures of the symmetrical ligands L5-L14 considered in this contribution.

Interestingly, parallel studies of these molecular complexes

dissolved in noncoordinating dichloromethane showed that

dimerization [equilibrium (1)] is a key characteristic whose

free-energy change depends on both the lanthanide size and

ligand substitution, thus offering some potential modulations

of the global shapes and geometries of the starting complexes

for the design of lanthanidomesogens.

The selective introduction of methyl groups into the cyano-

biphenyl units 1-3 follows a Miyaura-Suzuki strategy,¹²

in which (4-cyanophenyl)boronic acid or (4-cyano-2-methyl-

phenyl)boronic acid¹³is coupled with adequate halomethyl-

phenols (Scheme 2).

All ligands L5-L14 gave satisfying elemental analyses,

displayed molecular peaks by electrospray ionization mass

spectrometry (ESI-MS) corresponding to [Lk þ H]^þ, and

Ln,Lk

dim

1

2½LnðLkÞðNO₃Þ₃ꢀh½Ln₂ðLkÞ₂ðNO₃Þ₆ꢀ

β

ð1Þ

provided H NMR spectra typical for C_2v-symmetrical

organization with a trans-trans conformation of the central

tridentate 2,6-bis(benzimidazole)pyridine unit (Figure 4a).

The reaction of stoichiometric amounts of Lk with Ln-

In this contribution, we report on the systematic thermo-

dynamic investigation of the effect of chain lengths, of chain

branching, of methyl substitutions, and of metallic size on the

formation of the monomeric [Ln(Lk)(NO₃)₃] and dimeric

[Ln₂(Lk)₂(NO₃)₆] complexes in solution with a collection of

10 ligands L5-L14 and two lanthanides Ln = Pr and Eu

(Figure 3). The thermal properties of the mesogenic com-

plexes [Ln(L5)(NO₃)₃] and [Ln(L6)(NO₃)₃] are tentatively

rationalized in light of their affinity for dimerization in a

noncoordinating solvent.

(NO₃)₃xH₂O (Ln=La, Pr, Eu, Gd, Tb, Lu, Y; x=1-3)

3

in acetonitrile/dichloromethane followed by evaporation

yielded insoluble powders whose elemental analyses were

compatible with the formation of complexes [Ln(Lk)-

(NO₃)₃] xH₂O (η=72-89%; Table S1 in the Supporting

Information). Except for the downfield shift of the H

3

1

NMR signal of the proton attached to the para position of

the central pyridine ring (H1), which is diagnostic for the

meridional tricoordination of the bis(benzimidazole)-

Results and Discussion

1

pyridine binding unit to Lu(NO₃)₃,^14,15the H NMR

Syntheses and Characterization of Ligands L5-L14 and

of Their Complexes [Ln(Lk)(NO₃)₃] (Lk = L5-L14 and

Ln = La, Pr, Eu, Tb, Gd, Lu, Y). The syntheses of the

symmetrical ligands L5 and L6 follow a linear strategy, in

which the formations of ester bonds via activation with

carbodiimides (DCC or EDCI) alternate with carboxy-

late and hydroxyl deprotection reactions (Schemes 1 and

S1 and S2 in the Supporting Information).^10,11Because of

the tedious synthetic procedure leading to the final den-

drimeric ligand L5 (16 steps), a single methyl-substituted

analogue L6 (17 steps) has been synthesized (family 1,

Scheme 1), while two groups of smaller ligands with four

different methyl-substituted cyanobiphenyls have been

designed (family 2, short and branched ligands L7-L10,

Scheme S1 in the Supporting Information, 12 steps; family 3,

short and linear ligands L11-L14, Scheme S2 in the Sup-

porting Information, 12 steps).

spectra of the free ligands Lk and of their diamagnetic com-

plexes [Lu(Lk)(NO₃)₃] in CDCl₃are similar (Figure 4a,b).

The detection of a single set of aromatic benzimidazole

(12) (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457–2483.

(b) Stanforth, S. P. Tetrahedron 1998, 54, 263–303. (c) Tobisu, M.; Chatani,

N. Angew. Chem., Int. Ed. 2009, 48, 3565–3568.

(13) Stones, D.; Manky, S.; Lu, X.; Hall, D. G. Chem. Eur. J. 2004, 10,

;

92–100.

€

(14) (a) Nozary, H.; Piguet, C.; Tissot, P.; Bernardinelli, G.; Bunzli, J.-C.

G.; Deschenaux, R.; Guillon, D. J. Am. Chem. Soc. 1998, 120, 12274–12288.

(b) Nozary, H.; Piguet, C.; Rivera, J.-P.; Tissot, P.; Bernardinelli, G.; Vulliermet,

N.; Weber, J.; B€unzli, J.-C. G. Inorg. Chem. 2000, 39, 5286–5298. (c) Nozary, H.;

Piguet, C.; Rivera, J.-P.; Tissot, P.; Morgantini, P.-Y.; Weber, J.; Bernardinelli, G.;

B€unzli, J.-C. G.; Deschenaux, R.; Donnio, B.; Guillon, D. Chem. Mater. 2002,

14, 1075–1090.

ꢀ

(15) Terazzi, E.; Torelli, S.; Bernardinelli, G.; Rivera, J.-P.; Benech,

€

J.-M.; Bourgogne, C.; Donnio, B.; Guillon, D.; Imbert, D.; Bunzli, J.-C. G.;

Pinto, A.; Jeannerat, D.; Piguet, C. J. Am. Chem. Soc. 2005, 127, 888–903.

8604 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Scheme 1. Synthesis of Ligands L5 and L6 (DCC = N,N⁰-Dicyclohexylcarbodiimide, DPTS = 4-[(Dimethylamino)pyridinium]-4-toluenesulfonate,

4-Ppy=4-Pyrrolidinopyridine, DMAP=4-(Dimethylamino)pyridine, EDCI=N-[3-(Dimethylamino)propyl]-N⁰-ethylcarbodiimide)

protons points to the existence of a 2-fold axis passing

through the central pyridine ring. The concomitant obser-

vation of enantiotopic signals for the methylene groups

implies additional symmetry planes compatible with a

global C_2vpoint group for both the ligand and its Lu^III

complex on the NMR time scale, a behavior previously

well-documented for closely related complexes with poly-

catenar tridentate ligands L15 and L16 (see Figure 5 for

the structures of these ligands).^14,15

Beyond the expected opposite paramagnetic shifts

=

brought by Ln = Pr (4f², spin expectation factor ÆS_zæ_Pr

-2.97, and Bleaney factor C_Pr= -11.0) or Ln = Eu

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8605

Scheme 2. Synthesis of Compounds 1-3[ⁱProp=2-Propanol and dppf=

1,1⁰-Bis(diphenylphosphino)ferrocene]

shown in equilibrium (1) already in 1992 when analyzing

¹H-¹H COSY NMR spectra of simple nine-coordinate

[(2,6-bis(N-methylbenzimidazol-2-yl)pyridine)Eu(NO₃)₃]

complexes,¹⁷its unambiguous demonstration was de-

layed by more than a decade with the ultimate char-

acterization of the polycatenar complexes [Eu(L15^Cn)-

(NO₃)₃] (bent shape, 5,5⁰-disubstituted) and [Eu(L16^Cn)-

(NO₃)₃] (rodlike shape, 6,6⁰-disubstituted) in solution

(measurements of translational self-diffusion coefficients

by DOSY NMR) and in the solid state (molecular struc-

tures obtained from X-ray diffraction data; Figure 5).¹⁵

van’t Hoff plots derived from analysis of the integra-

tion of variable-temperature ¹H NMR (VT-NMR,

CD₂Cl₂) data yielded ΔHꢀ^E_di^u_m^,L15C12= -54(4) kJ mol^-1

3

and ΔSꢀ_d^E_i^u_m^,L15C12=-162(16) J mol^-1K^-1for the dimeri-

3

zation [equilibrium (1)] operating between two bent poly-

catenar europium complexes, and ΔHꢀ^E_di^u_m^,L16C12=-25(2)

kJ mol^-1and ΔSꢀ^E_di^u_m^,L16C12= -47(5) J mol^-1K^-1for

3

the same process applied to the rodlike analogous com-

plexes (Figure 5).¹⁵Already noticed in 2005,¹⁵the parallel

Eu,Lk

dim

Eu,Lk

dim

increase of ΔH

and ΔS

on going from

[Eu(L15^C12)(NO₃)₃] to [Eu(L16^C12)(NO₃)₃] is reminiscent

of the enthalpy/entropy compensation concept, which

claims that any reduction of the intermolecular interactions

responsible for the formation of the dimer (ΔHꢀ^E_di^u_m^,Lkbecomes

Figure 4. 400 MHz ¹H NMR spectra of (a) L6 and of their complexes

(CDCl₃) (b) [Lu(L6)(NO₃)₃] (CDCl₃), (c) [Eu(L6)(NO₃)₃] (CD₂Cl₂), and

(d) [Pr(L6)(NO₃)₃] (CD₂Cl₂).

(4f⁶, ÆS_zæ_Eu=10.68, and C_Eu=4.0),¹⁶which spread the ¹H

NMR signal over -15 e δ e þ25 ppm (Figure 4c,d), we

observe the coexistence of two complexes in solution,

each possessing average C_2vsymmetry on the NMR time

scale. Because of the location of the H2 protons close to

the paramagnetic center (Figure 4b), their large pseudo-

contact shifts can be exploited for probing speciation in

solution with the detection of two singlets at a low mag-

netic field for Ln=Eu (C_Eu=4.0; Figure 4c) and of two

singlets at a high magnetic field for Ln=Pr (C_Pr=-11.0;

Figure 4d). The existence of only two complexes in solu-

tion is thus confirmed for each complex, and their quan-

tification can be obtained by the simple deconvolution

and integration of the ¹H NMR signals of H2.^10c,11,15

Thermodynamic Parameters for the Dimerization of

[Ln(Lk)(NO₃)₃] in CD₂Cl₂(Lk = L5-14 and Ln = Pr,

Eu). Despite the first mention of the dimerization process

less negative) produces a change in dynamics between the

Eu,Lk 18

dim

two states measured by an increase of ΔS

. Prelimi-

nary results showed that equilibrium (1) also holds for the

dendrimeric complexes [Eu(L5)(NO₃)₃]^10cand [Eu(L6)-

(NO₃)₃],¹¹for which measurement of the diffusion co-

efficients unambiguously demonstrated the coexistence

around room temperature of the monomer and dimer in

CD₂Cl₂(Table 1).^10c,11

The thorough recording of VT-NMR spectra for the 20

complexes [Ln(Lk)(NO₃)₃] (Ln = Pr, Eu; k = 5-14) in

CD₂Cl₂in the 273-318 K range confirms this behavior

and systematically shows the interconversion occurring

between the monomer and dimer according to equilibri-

um (1). Moreover, at the concentration used for NMR

measurements (|Lk|_tot≈ 10 mM), the cumulative stability

Ln,Lk

1,1

Ln,Lk

2,2

constants β

and β

are large enough to prevent

€

(17) Piguet, C.; Williams, A. F.; Bernardinelli, G.; Moret, E.; Bunzli, J.-C.

(16) (a) Golding, R. M.; Halton, M. P. Aust. J. Chem. 1972, 25, 2577–

2581. (b) Pinkerton, A. A.; Rossier, M.; Spiliadis, S. J. Magn. Reson. 1985, 64,

420–425. (c) Bleaney, B. J. Magn. Reson. 1972, 8, 91–100. (d) For a tutorial

review, see: Piguet, C.; Geraldes, C. F. G. C. In Handbook on the Physics and

Chemistry of Rare Earths; Gschneidner, K. A., Jr., B€unzli, J.-C. G., Pecharsky,

V. K., Eds.; Elsevier Science: Amsterdam, The Netherlands, 2003; Vol. 33,

pp 353-463.

G. Helv. Chim. Acta 1992, 75, 1697–1717.

(18) (a) Searle, M. S.; Williams, D. H. J. Am. Chem. Soc. 1992, 114,

10690–10197. (b) Dunitz, J. D. Chem. Biol. 1995, 2, 709–712. (c) Searle, M. S.;

Westwell, M. S.; Williams, D. H. J. Chem. Soc., Perkin Trans. II 1995, 141–151.

(d) Williams, D. H.; O'Brien, D. P.; Bardsley, B. J. Am. Chem. Soc. 2001, 123,

737–738. (e) Levchenko, A. A.; Yee, C. K.; Parikh, A. N.; Navrotsky, A. Chem.

Mater. 2005, 17, 5428–5438.

8606 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Figure 5. (a) Thermodynamic equilibria for the complexation ofL15^Cnor L16^Cnwith Ln(NO₃)₃[R = H (Lk^C0), OCH₃(Lk^C1), and OC₁₂H₂₅(Lk^C12); S =

solvent molecule] and (b) molecular structures of the monomeric [Eu(L15^C0)(NO₃)₃CH₃CN] and dimeric [Eu₂(L16^C0)₂(NO₃)₆] complexes determined by

X-ray diffraction.¹⁵

.

Ln,Lk

dim

detectable decomplexation. The speciation in solution

required for extraction of the thermodynamic data for

each system is thus obtained by the following procedure.

(1) The signals of the paramagnetically shifted protons

H2 in the monomer and dimer are located in the ¹H NMR

spectrum, and their relative intensities upon dilution are

recorded (Figure 6a). Because of the law of mass action,

the quantity of monomer increases upon dilution, which

allows the unambiguous assignment of the signals for the

monomer and dimer. (2) The VT-NMR spectra are recor-

ded in CD₂Cl₂, and the intensity of each signal for H2 is

obtained by deconvoluting the experimental trace with a

single Lorentz function for each peak (Figures 6b and S1

in the Supporting Information). The separated integrated

intensities are then converted into concentrations by using

the mass balance modulated by the number of protons

contributing to each signal. (3) The dimerization constants

are calculated at each temperature β

= (|dimer|/

|monomer|²)c^θby using the standard c^θ=1 M concentra-

tion as the reference state.¹⁹The building of van’t Hoff plots

Ln,Lk

dim

Ln,Lk

dim

(eq 2) eventually gives ΔHꢀ

and ΔSꢀ

as the

slopes and intercepts, respectively, of the linear correla-

tions (Figure 6c).

Ln,Lk

dim

ΔHꢀ

Ln,Lk

dim

Ln,Lk

dim

- R lnðβ

Þ ¼

- ΔSꢀ

ð2Þ

T

For some complexes, severe overlap between the signals

of H2 in the monomer and dimer may limit the precision

of the van’t Hoff analyses (Figure S2a in the Supporting

(19) (a) Martell, A. E. Adv. Chem. Ser. 1967, 62, 272–294. (b) Munro, D.

Chem. Brit. 1977, 13, 100–105. (c) Simmons, E. L. J. Chem. Educ. 1979, 56,

578–579. (d) Chung, C.-S. J. Chem. Educ. 1984, 61, 1062–1064.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8607

Ln,Lk

dim,50%

Lk

crit

Table 1. Thermodynamic Parameters ΔHꢀ^L_diⁿ_m^,Lk, ΔSꢀ^L_diⁿ_m^,Lk, ΔGꢀ^L_diⁿ_m^,Lk, βꢀ^L_diⁿ_m^,Lk, T

, and T for Equilibrium (I) for Ln = Pr and Eu and Ligands L5-L14 (CD₂Cl₂)^a

ΔHꢀ^L_diⁿ_m^,Lk/kJ mol^-1

ΔSꢀ^L_diⁿ_m^,Lk/J mol^-1K^-1

ΔGꢀ^L_diⁿ_m^,Lk/kJ mol^-1

βꢀ

T

^c/ꢀC

T

Ln,Lk b

dim

Ln,Lk

dim,50%

Lk d

/K

crit

3

[Pr(L5)(NO₃)₃]

[Eu(L5)(NO₃)₃]

[Pr(L6)(NO₃)₃]

[Eu(L6)(NO₃)₃]

[Pr(L7)(NO₃)₃]

[Eu(L7)(NO₃)₃]

[Pr(L8)(NO₃)₃]

[Eu(L8)(NO₃)₃]

[Pr(L9)(NO₃)₃]

[Eu(L9)(NO₃)₃]

[Pr(L10)(NO₃)₃]

[Eu(L10)(NO₃)₃]

[Pr(L11)(NO₃)₃]

[Eu(L11)(NO₃)₃]

[Pr(L12)(NO₃)₃]

[Eu(L12)(NO₃)₃]

[Pr(L13)(NO₃)₃]^c

[Eu(L13)(NO₃)₃]

[Pr(L14)(NO₃)₃]

[Eu(L14)(NO₃)₃]

-26(4)

-55(4)

-33(4)

-25(1)

-7.7(6)

-17.9(8)

-2.8(8)

-31.9(5)

f

-43(15)

-166(11)

-72(14)

-58(5)

15(2)

-34(3)

31(3)

-73(2)

f

-12.9(3)

183(21)

9^e

43(7)

-4^e

236

572

208

280

f

-5.5^e

-11.9(2)

-7.93(5)

-12.06(3)

-7.82(3)

-11.92(4)

-10.01(3)

f

-9.93(6)

-12.01(4)

-10.03(3)

-20.3(8)

-8.83(6)

-15.3(2)

-7.3(1)

f

121(11)

24.6(5)

130(2)

23.5(3)

123(2)

56.8(6)

f

29(3)

-11(2)

52(3)

-25(2)

92(26)

12.3(3)

f

770(755)

64(8)

8.2(8)

126(30)

-13(4)

55(3)

2(6)

40(20)

23(3)

55(4)

-5.2(8)

-23.4(8)

-35(8)

-18(2)

-43(4)

-25(3)

f

-25(5)

-62(5)

-17(2)

128(2)

57.2(7)

3525(435)

35.3(9)

487(31)

18.9(8)

f

17(1)

1254(83)

33.3(9)

268

809

545

f

-45(5)

-50(27)

-29(6)

-91(15)

-58(9)

f

-60(17)

-148(15)

-28(6)

-18(5)

f

-19(9)

59(3)

-7.1(2)

-17.7(2)

-8.69(6)

375

-16(4)

^aThe standard deviations for ΔHꢀ^L_diⁿ_m^,Lkand ΔSꢀ^L_diⁿ_m^,Lkwere obtained from the linear least-squares fits, while those for ΔGꢀ_d^L_iⁿ_m^,Lk, βꢀ_d^L_iⁿ_m^,Lk, and T

were

computed by taking into account the interdependence of ΔHꢀ^L_diⁿ_m^,Lkand ΔSꢀ_d^L_iⁿ_m^,Lkgiven by the covariance matrices.^{21 b}The dimerization constant was

Ln,Lk

dim,50%

c

Ln,Lk

dim,50%

Ln,Lk

= ΔHꢀ_d^L_iⁿ_m^,Lk/(ΔSꢀ

dim

Lk

þ R ln C ) is the temperature at

calculated at 298 K by using the standard 1 M concentration as the reference state.

T

tot

which the monomer represents 50% of the ligand distribution in solution (calculated with a total ligand concentration of C^L_to^k_t= 0.01 M used for NMR

measurements; see Appendix 1 in the Supporting Information). ^dCritical temperature T

= (ΔHꢀ

- ΔHꢀ_d^P_i^r_m^,Lk)/ΔSꢀ

- ΔHꢀ

for

Lk

crit

Eu,Lk

dim

Eu,Lk

dim

Pr,Lk

dim

which ΔG^P_di^r_m^,Lk= ΔG

of the VT-NMR spectra.

(estimated for the centroids of enthalpic and entropic contributions). ^eTaken from ref 10b. ^fSpectral overlap prevents analysis

Eu,Lk

dim

Information) or even prevents the estimation of the

thermodynamic parameters (Figure S2b in the Support-

given by the intercepts of the linear correlations

(ΔG (T^Pr= 329 K) = -13.0(1.2) kJ mol^-1and

Pr

dim comp

(ΔG (T^Eu= 278 K) = -9.4(5) kJ mol^-1). Because

3

Ln,Lk

Eu

dim comp

ing Information). The experimental data ΔHꢀ

dim

_E3_u

dim comp

Ln,Lk

dim

Pr

dim comp

Eu

and ΔSꢀ

are collected in Table 1.

obtained for the 18 accessible complexes

(i) (ΔG (T^Pr=329 K)<(ΔG (T

(ii) the computed critical temperature T

=278 K) and

at which

Lk

crit

Ln,Lk

dim

Ln,Lk

Eu,Lk

dim

Plots of ΔHꢀ

versus ΔSꢀ

show two remark-

ΔG^P_di^r_m^,Lk=ΔG

is outside the 280-329 K temperature

dim

Eu,Lk

ably linear relationships given by eqs 3 and 4 including all

ligands interacting with either Ln=Pr (eq 3 and Figure 7a)

or Ln = Eu (eq 4 and Figure 7b).

range, we can deduce that ΔGꢀ^P_di^r_m^,Lk< ΔGꢀ

around

dim

room temperature, so the dimerization process is system-

atically more favorable for the larger praseodymium

cation in CD₂Cl₂(Table 1, column 5). However, prior

to further interpretation of the observed correlations

between the enthalpic and entropic parameters across

this series of ligands, it is worth stressing here that such

apparent compensations seen for several physical or

chemical intermolecular associations may result from a

statistical artifact linked to the van’t Hoff analyses, which

do not independently measure ΔH and ΔS.^23-25More-

over, the narrow experimental temperature range used,

which is required for reasonably considering tempera-

ture-independent values for ΔH and ΔS, also limits the

precision of the separation of enthalpic and entropic

contributions. Beyond the systematic calculations of the

covariance matrix²¹for each van’t Hoff plot, which

indeed display high values, but significantly smaller than

1.0, Krug and co-workers proposed several additional

tests that can be applied to evaluate the possible signifi-

Pr,Lk

dim

ΔHꢀ

¼ 329ð17ÞΔSꢀ^P_di^r_m^,Lk- 13054ð1203Þ J mol^{- 1}

3

ð3Þ

Eu,Lk

dim

ΔHꢀ

¼ 278ð7ÞΔSꢀ^E_di^u_m^,Lk- 9428ð524Þ J mol^{- 1}

3

ð4Þ

Relationships like eqs 3 and 4 are often referred to as

extrathermodynamic because of their lack of basis in the

laws of thermodynamics.²⁰The positive correlation

Ln,Lk

found between ΔHꢀ

and ΔSꢀ

is referred to as

dim

compensation because it tends to minimize changes in

ΔGꢀ_d^L_iⁿ_m^,Lkacross the series.¹⁸The slope of each correlation

has Kelvin units and is called the compensation tempera-

Ln

22

.

ture, T

At this temperature, any variation in the

comp

standard enthalpy of dimerization, ΔHꢀ^L_diⁿ_m^,Lk, across the

series is balanced by an exact compensating variation in

the standard entropy, T

cance of apparent enthalpy/entropy correlations.²⁴First,

Ln

the observed compensation temperature T

comp

must be

ΔSꢀ^L_diⁿ_m^,Lk, thus leading to

processes displaying the same exergonicities, which are

Ln

comp

(23) (a) Liu, L.; Guo, Q. X. Chem. Rev. 2001, 101, 673–695. (b) Sharp, K.

Protein Sci. 2001, 10, 661–667.

(24) (a) Krug, R. R.; Hunter, W. G.; Grieger, R. A. J. Phys. Chem. 1976,

80, 2335–2341. (b) Krug, R. R.; Hunter, W. R.; Grieger, R. A. J. Phys. Chem.

1976, 80, 2341–2351. (c) Krug, R. R.; Hunter, W. G.; Grieger, R. A. Nature 1976,

261, 566–567. (d) Linert, W.; Jameson, R. F. Chem. Soc. Rev. 1989, 18, 477–505.

(25) Leung, D. H.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc.

2008, 130, 2798–2805.

(20) Starikov, E. B.; Norden, B. J. Phys. Chem. B 2007, 111, 14431–14435.

(21) For each least-squares fit analysis, we used the macro LS1 for

computing the covariance matrix and the macro Propagation for estimating

the standard deviations. De Levie, R. Advanced Excel for Scientific Data

Analysis, 2nd ed.; Oxford University Press: Oxford, U.K., 2008.

(22) Leffler, J. E. J. Org. Chem. 1955, 20, 1202–1231.

8608 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Figure 6. Experimental determination of thermodynamic enthalpic

contributions to equilibrium (1) by exploit-

Ln,Lk

dim

ΔH^L_diⁿ_m^,Lkand entropic ΔS

ing the VT-NMR spectra recorded for [Ln(L6)(NO₃)₃] (Ln = Pr, Eu;

CD₂Cl₂): (a) concentration and (b) temperature dependences of the inten-

sity of the signals of proton H2 in the monomer and dimer; (c) associated

van’t Hoff plots (R²= coefficient of determination).

Figure 7. Plots of enthalpy versus entropy values for the dimerization

process shown in equilibrium (1) and occurring in CD₂Cl₂for (a) [Pr(Lk)-

(NO₃)₃] (Lk = L5-L14 except L9 and L13; see Table 1) and (b) [Eu(Lk)-

(NO₃)₃] (Lk = L5-L14; R²= coefficient of determination). The red

points correspond to those previously reported for [Eu(L15^C12)(NO₃)₃]

significantly different from the average experimental

= 329 K can be

_

Pr

comp

temperature T ^L_exⁿ_p. For Ln = Pr, T

and [Eu(L16^C12)(NO₃)₃].¹⁵

.

_

Pr

exp

compared with T

= 285 K, while for Ln = Eu, we

_

Eu

obtain T

comp

Eu

exp

Pr

= 278 K and T

= 303 K (Figure 6). The

whereby this complex displays the larger deviation from

the linear trendline. Because the three above-mentioned

criteria are obeyed, we conclude that the observed corre-

lations between enthalpic and entropic values for the

dimerization of [Ln(Lk)(NO₃)₃] in CD₂Cl₂have true

chemical significance for Ln=Pr and Eu across the com-

plete ligand series. A close scrutiny at the linear enthalpy/

entropy compensations shown in Figure 7 does not reveal

any clear trend associated with either methyl substitution,

the nature of the ligand backbone, or the molecular

weights of the complexes. Interestingly, consideration of

_

two absolute gaps ΔT^Pr= |T

- T | = 44 K and

Pr

exp

comp

_

ΔT^Eu= |T

than the uncertainties affecting T

- T | = 25 K are significantly larger

Eu

exp

comp

Ln

comp

(eqs 3 and 4) and

are of magnitude comparable to those found by Raymond,

Bergman, and co-workers for significant enthalpy/entropy

compensation controlling host-guest encapsulations.²⁵

Surprisingly, these authors suggest a second test, which

claims that, when plotted together, the van’t Hoff cor-

relations -R ln(β

Ln,Lk

dim

) versus T^-1for a set of related reac-

tions should intersect at a common temperature.²⁵How-

ever, this second criterion is a consequence of the linearity

of eqs 3 and 4, because all reactions considered within a

Eu,L15C12

dim

Eu,L16C12

dim

the ÆΔSꢀ^E_di^u_m^,L15C12, ΔHꢀ

æÆΔSꢀ^E_di^u_m^,L16C12, ΔHꢀ

æ

pairs reported for the polycatenar complexes matches the

linear correlation for Ln = Eu (red points in Figure 7b),

which further supports the minor effects of the primary

ligand structure on the compensating trend. However, the

existence of two different series for Ln = Pr and Eu

indicates that the minor change in the ionic radius has

a global impact on dimerization, leading to [Ln₂(Lk)₂-

(NO₃)₆], a behavior that can be tentatively assigned to the

preference of the larger cation for 10-coordination in the

Ln

linear series display the same ergonicity at T

produce identical stability constants β

and thus

at this tempe-

comp

Ln,Lk

dim

rature. Figure S3 (Supporting Information) shows collec-

tions of the van’t Hoff plots for Ln=Pr and Eu across the

ligand series, which obviously confirm this statement. The

larger deviation found for [Pr(L11)(NO₃)₃] (Figure S3a in

the Supporting Information) can be easily predicted when

considering the enthalpy/entropy plots of Figure 7a,

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8609

nitrato-bridged dimer (Figure 5).¹⁵Following the classical

thermodynamic desolvation/association mechanism,²⁶each

complexation process can be then separated into two

successive equilibria (5) and (6) displaying opposite enthalpic

and entropic contributions (S = solvent molecule).²⁶

2½LnðLkÞðNO₃Þ₃ðSÞ_nꢀ h 2½LnðLkÞðNO₃Þ₃ðSÞ_{n - p}ꢀ þ 2pS

Ln,Lk

desolv

ΔG

¼ ΔH_desolv

^Ln,Lk- TΔS^Ln,Lk

ð5Þ

desolv

2½LnðLkÞðNO₃Þ₃ðSÞ_{n - p}ꢀ h ½Ln₂ðLkÞ₂ðNO₃Þ₆ðSÞ_{2ðn - pÞ}ꢀ

Ln,Lk

ΔG

¼ ΔH

- TΔS

ass:

ð6Þ

ass:

In the weakly polar CD₂Cl₂solvent, the desolvation step

(eq 5) can be reasonably assumed to be similar within a

series of complexes of similar molecular weight, differing

only by peripheral methyl substitutions (family 1, L5 and

Figure 8. Representation of a Lennard-Jones (12,6) potential (full trace)

with interpretation of the ε and r₀parameters and its harmonic approx-

imation (eq 9, dashed trace) modeling the intermolecular interactions

responsible for the formation of the dimer [Ln₂(Lk)₂(NO₃)₆].

L6; family 2, L7-L10; family 3, L11-L14). Conse-

Ln,Lk

dim

eventually gives κ = [2/(1 - 2^1/6)²]ε/(r₀)²(κ > 0; see

Appendix 2 in the Supporting Information).

quently, within each family, variation of ΔG

=

Ln,Lk

.

Ln,Lk

desolv

Ln,Lk

ass.

ΔG

þ ΔG

essentially reflects changes in ΔG

ass.

Because lanthanide-ligand bonds are mainly electro-

static, the association process shown in eq 6 can be

approached by a simple (nonquantum) model catching

the physical forces of interest responsible for the forma-

tion of a dimer from two molecules of monomer.²⁷As

demonstrated by Ford,^27bthe enthalpic and entropic

components of such an association process are given by

eqs 7 and 8, respectively (u_minis the minimum of the

binding potential, k_bis Boltzmann’s constant, κ is the

force constant of the interaction, and c^θ= 1/V^θis the

concentration of the reference state).

ε

2

V_harmðrÞ ¼ ½ - 2^{- 7=6}ðr=r₀Þ þ ðr=r₀Þ þ 2^{- 7=6}- 1ꢀ

R

with R ¼ 1 - 2¹⁼⁶þ 2^{- 5=6}- 2^{- 7=6}

ð9Þ

Introducing the expression found for u_minand κ into

eqs 7 and 8, respectively, shows that the observed en-

thalpy/entropy compensation within a series occurs when

ε varies and r₀is fixed. Consequently, within a family of

ligands, the trend in the association process leading to

[Ln₂(Lk)₂(NO₃)₆] can be thought of as an intermolecular

cohesion, for which the magnitude of the interaction

) is modulated by the peripheral methyl substi-

Ln,Lk

ass.

3

(ΔH

ΔH_ass:¼ u_min

þ

k_bT

ð7Þ

ð8Þ

2

tution, but the optimum intermolecular separation is

fixed by the choice of the metal, which consequently

provides entropic compensation due to changes in the

dynamics between the two states (i.e., two monomers

versus one dimer; Figure 8). In simple chemical words, we

can say that the transformation of two monomers

[Ln(Lk)(NO₃)₃] into the dimer [Ln₂(Lk)₂(NO₃)₆] relies

on a primary interaction that depends on the size of the

trivalent cation, which fixes the average intermolecular

separation r₀in our simple harmonic model. For a given

!

ꢀ

ꢁ

3=2

ð2πeÞ

3

K

k_bT

ΔS_ass:¼ k_bln

- k_bln

V^θ

2

According to eqs 7 and 8, the enthalpy/entropy compen-

sation (i.e., a positive correlation) in ΔG^L_asⁿ_s^,_.^Lkimplies that

Ln,Lk

ass.

u_minand κ move in opposite directions, so that ΔH

Ln,Lk

ass.

and ΔS

both decrease and increase, respectively.^27b

metallic center, additional nonspecific interstrand inter-

actions modulate ΔG

Following this simple reasoning, the intermolecular di-

merization shown in equilibrium (6) can be modeled with

Ln,Lk

dim

within the frame of enthalpy/

a simple Lennard-Jones (12,6) potential V=4ε{(r₀/r)¹²

-

entropy compensation. According to the molecular structures

established for [Eu(L15^C0)(NO₃)₃(CH₃CN)], [Lu(L16^C0)-

(NO₃)₃], and [Eu₂(L16^C0)₂(NO₃)₆] (Figure 5),¹⁵we assign

the primary interaction fixing r₀to the formation of

nitrate bridges between the metallic parts of the com-

plexes, a process expected to be favored for the larger

(r₀/r)⁶}, which only resorts to two parameters: (i) the mini-

mum of the attractive well depth ε and (ii) the intermole-

cular separation r₀at which the potential of the inter-

action is zero [V(r₀)=0; Figure 8]. While the correlation

between the well depth and the binding potential is trivial,

u_min= -ε (u_min< 0), an analytical formulation for the

force constant requires application of the harmonic

approximation (eq 9), from which we can write κ(r₀-

cation Pr^III, for which 10-coordination is easily adopted.

Ln,Lk

dim

The additional modulation of ΔH^L_diⁿ_m^,Lkand ΔS

with-

in the frame of strict enthalpy/entropy compensation for

each metal results from nondirectional van der Waals inter-

actions between the polarizable parts of the semidendri-

meric ligand strands. Increasing length (family 3 f family

2 f family 1), varying branching (family 1 vs family 2), or

specific methyl substitution of the terminal cyanobiphe-

nyl groups does influence the organization of the ligand

strands and their interactions in the dimer, but we cannot

2

^1/6r₀)²/2=ε . A straightforward algebric transformation

(26) (a) Choppin, G. R. Lanthanide Probes in Life, Chemical and Earth

Sciences; Elsevier: Amsterdam, The Netherlands, 1989; Chapter 1. (b) Piguet, C.;

B€unzli, J.-C. G. Chem. Soc. Rev. 1999, 28, 347–358.

(27) (a) Luo, H. B.; Sharp, K. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,

10399–10404. (b) Ford, D. M. J. Am. Chem. Soc. 2005, 127, 16167–16170.

8610 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Table 2. Phase-Transition Temperatures and Enthalpy and Entropy Changes for

Ligands L5-L14

Table 3. Phase-Transition Temperatures and Enthalpy and Entropy Changes for

Complexes [Ln(Lk)(NO₃)₃] (Lk = L5-L14)

compound transition^aT/ꢀC ΔH /kJ mol^-1ΔS /J mol^-1K^-1

ΔH /

ΔS /

3

compound

transition^a

T/ꢀC

100^c

kJ mol^-1J mol^-1K^-1

3

L5

g f N

N f I

g f N

N f I

g f I

g f N

N f I

g f N

N f I

g f N

N f I

g f N

N f I

52

238

11.0

9.7

21.5

20.5

[Lu(L5)(NO₃)₃]

g f SmA2

SmA2 f PCr ^b150^c

L6

50-70^b

197

118

PCr f SmA2

SmA2 f I/Dec 200

g f SmA2

90^c

SmA2 f I/Dec 190

g f SmA2

85^c

SmA2 f I/Dec 1880

g f SmA2

80^c

170^c

L7

L8

L9

L10

L11

3.3

6.9

90

85

84

[Tb(L5)(NO₃)₃]

[Gd(L5)(NO₃)₃]

[Eu(L5)(NO₃)₃]

[Pr(L5)(NO₃)₃]

[Lu(L6)(NO₃)₃]

15.0

9.1

32.3

19.8

24.2

-3

70-80^b

324^b

70-80^b

265^b

70-80^b

260^b

70-80^b

205^b

L12

L13

L14

SmA2 f I/Dec 186

11.1

g f SmA2

SmA2 f Dec

g f L

80^c

190^c

90-100^c

130

144

90-110^c

134

144

90-100^c

L f N

0.8

6.9

1.9

16.4

N f I/Dec

g f SmA2

SmA2 f N

N f I/Dec

g f SmA2

^ag = glass, N = nematic phase, and I = isotropic fluid. Tempera-

tures are given as the onset of the peak observed during the second

heating processes. ^bThe liquid-crystalline phases were identified from

their optical textures and from SA-XRD studies.

[Y(L6)(NO₃)₃]

[Tb(L6)(NO₃)₃]

[Gd(L6)(NO₃)₃]

[Eu(L6)(NO₃)₃]

1.5

4.5

3.7

10.7

SmA2 f SmAd 137

SmAd f I/Dec 144

2.8

1.9

6.9

4.5

detect trivial correlations between the chemical structures

and the thermodynamic parameters of the interaction.

Thermal and Liquid-Crystalline Properties of the Ligands

L5-L14 and of Their Complexes [Ln(Lk)(NO₃)₃]. The

thermal behaviors of the ligands L5-L14 (Table 2) and of

their complexes (Table 3) have been investigated by a

combination of thermogravimetric analysis (TGA), dif-

ferential scanning calorimetry (DSC), polarized optical

microscopy (POM), and small-angle X-ray diffraction

(SA-XRD). First, the TGA traces systematically show

the release of interstitial cocrystallized water molecules

during the first heating process in the temperature range

80-140 ꢀC. At higher temperature, decomposition of the

free ligand starts around 350 ꢀC, while the nitrato com-

plexes [Ln(Lk)(NO₃)₃] rapidly decompose upon entering

the isotropic liquid phase around 180-200 ꢀC (Tables 2

and 3 and Figure S4 in the Supporting Information).

Therefore, the thermal behaviors are systematically dis-

cussed for the second heating/cooling cycle following an

initial 25-100-25 ꢀC cycle, which ensures the removal of

all cocrystallized water molecules in the samples.

For ligands L5, L6, and L11-L14, for which the termi-

nal cyanobiphenyl groups are connected to a linear 1-4

phenyl spacer (Figure 3), typical Schlieren textures are

observed by POM in agreement with the formation of

nematic mesophases (Table 2 and Figure S5 in the

Supporting Information). The temperature of the N f I

phase transition (Figure S6 in the Supporting Informa-

tion) systematically decreases with the connections of an

increasing number of methyl groups onto the terminal

cyanobiphenyl units [T_c(L5) > T_c(L6) and T_c(L11) >

T_c(L12) ≈ T_c(L13) > T_c(L14)]. This behavior suggests

some reduction of the residual intermolecular cohesion

in the nematic phases resulting from some limitation of

the π overlaps between methyl-substituted cyanobiphenyl

groups (ΔH_cindeed decreases by 13% on going from L5

to L6). In line with the previous detection of enthalpy/

entropy compensation for the nondirectional association

between the ligand strands of the dimeric complexes in a

noncoordinating solvent, this reduction of ΔH_cis accom-

panied by a concomitant reduction in ΔS_cat the N f I

g f SmA2

SmA2 f SmAd 139

SmAd f I/Dec 144

90-110^c

2.3

3.4

5.5

8.2

g f SmA2

SmA2 f N

N f I

90-110^c

138

2.4

4.9

5.8

11.8

144

90-110^c

147

[Pr(L6)(NO₃)₃]

[La(L6)(NO₃)₃]

g f SmA2

SmA2 f I

g f SmA2

SmA2 f I

17.0

18.5

40.4

42.8

90-110^c

159

[Ln(Lk)(NO₃)₃] ^dg f I

[Ln(Lk)(NO₃)₃] ^eg f I/Dec

110-120^c

180-200^c

^ag = glass and L = lamellar phase shown in Figure 9b, SmA2 =

bilayer smectic A phase, SmAd = partially interdigitated smectic A

phase, N = nematic phase, I = isotropic fluid, and Dec = decomposi-

tion. Temperatures are given as the onset of the peak observed during the

second heating processes. ^bRe-entrant partially crystallized phase (PCr).

^cThe liquid-crystalline phases were identified from their optical textures

and/or from SA-XRD studies. ^dLn = La, Pr, Eu, and Lu and Lk =

L7-L10. ^eLn = La, Pr, Eu, and Lu and Lk = L11-L14.

phase transition (Table 2, entries 2 and 3). In order to gain

further insight into the residual interstrand interactions

responsible for the formation of the nematic phase, SA-

XRD data have been collected in the mesophase of L6 at

150 ꢀC upon alignment of the sample in a magnetic field

(1 T; Figure S7 in the Supporting Information). Beyond the

˚

standard diffuse reflection observed at ≈4.5 A and as-

signed to the average separation between the molten alkyl

chains of the rodlike ligands, an additional orthogonal

˚

diffuse band is detected at d ≈ 17 A, which is attributed to

some nonspecific residual intermolecular interactions

between the aromatic segments responsible for the global

alignment along the director n (Figure 9a). Such observa-

B

tions in thermotropic nematic phases obtained by phase

transitions from either lamellar crystalline or liquid-

crystalline phases are very common; they generally cor-

respond to the existence of residual cybotactic groups.²⁸

Because L6 contains four flexible interaromatic segments

(28) Mamlouk, H.; Heinrich, B.; Bourgogne, C.; Donnio, B.; Guillon, D.;

Felder-Flesch, D. J. Mater. Chem. 2007, 17, 2199–2205.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8611

the formation of nematic phases, and the DSC trace

shows a vitrous transition around 90-120 ꢀC (Figure S8,

in the Supporting Information), leading to the disappea-

rance of any specific texture by POM. The associated

˚

SA-XRD profiles do not exhibit any reflection in the 8 A e

˚

d e 140 A angular domain, and we conclude that L7-L10

exist as amorphous solids, which simply melt to give iso-

tropic liquid phases. The associated complexes [Ln(Lk)-

(NO₃)₃] (Lk = L7-L10; Ln = La, Pr, Eu, Lu) display

similar behaviors with transitions around 110-120 ꢀC

(Figure S9 in the Supporting Information) prior to decom-

position occurring at 170-175 ꢀC (Table 3). The complexes

[Ln(Lk)(NO₃)₃] (Lk=L11-L14; Ln=La, Pr, Eu, Lu) are

not mesomorphic and simply decompose when entering the

isotropic liquid phase at 180-200 ꢀC (Table 3). We thus

focus our attention on the mesogenic properties of the

complexes [Ln(Lk)(NO₃)₃] (Ln = La-Lu; Lk = L5 and

L6) with the largest dendrimeric ligands.^10b,11As found for

the free ligands L5 and L6, the associated complexes

[Ln(Lk)(NO₃)₃] (Ln = La-Lu) enter a birefringent liquid-

crystalline phase around 80-100 ꢀC without displaying a

pronounced peak in the DSC traces (Figure 10). SA-XRD

profiles recorded for the complexes [Ln(L5)(NO₃)₃] (Ln =

Eu, Gd, Tb, Lu) in the mesophase show two low-angle

reflections with a 1:2 ratio, which are indexed as the (00l)

˚

reflections of a smectic phase with d₀₀₁= 120 A and d₀₀₂

=

˚

60 A (Figure S10 and Table S2 in the Supporting Infor-

mation). The large interlayer distance given by the first

harmonic of the smectic phase agrees with the approxi-

˚

mate 119 A length estimated by a simple molecular

modeling of the monomeric complex [Ln(L5)(NO₃)₃] in

its extended conformation (Figure S11 in the Supporting

Information).^10b

2

˚

The molecular area (in A ) occupied by the rodlike

cylinder at the interface between the layers in the smectic

phases can be estimated with eq 10 (V is the volume of the

unit cell for Z = 1, MM is the molecular weight of the

complex in g mol^-1, N_avis Avogadro’s number in mol^-1

,

3

F is the density in g cm^-3taken as F=1.0 g cm^-3for the

3

mesophase at 25 ꢀC, and δ = V_CH(T)/V_CH(T⁰) is a

correcting factor for the dependence of F with²tempera-

2

ture, where V_CH(T) = 0.02023T þ 26.5616 (T is the

2

experimental temperature in ꢀC and T ⁰= 25 ꢀC).²⁹

V

MMδ 1

N_AvF d₀₀₁

A ¼

¼

ꢁ 10²⁴

ð10Þ

d₀₀₁

Figure 9. Proposed organizations in (a) the nematic phase of L6, (b) the

layered mesophase of [Lu(L6)(NO₃)₃], and (c) the SmA2 phase of

[Ln(L6)(NO₃)₃] (Ln = La, Pr, Eu) as deduced from POM and SA-

XRD data. Color code: red, aromatic groups; blue, aliphatic chains;

violet, coordinated benzimidazole-pyridine-benzimidazole units.

The experimental values measured for [Ln(L5)(NO₃)₃]

˚

2

(Ln = Eu, Gd, Tb, Lu) at T = 100 ꢀC (50 e A e 53 A ;

Table S2 in the Supporting Information) are only slightly

2

˚

larger than 44 A expected for the cross section of two

closely packed parallel cyanobiphenyl segments.³⁰We

deduce that the rodlike complexes are arranged roughly

perpendicularly to the interlayer interface, thus forming a

bilayer smectic A2 mesophase in the 100-190 ꢀC tempe-

rature range.^8a,31The methylated complexes [Ln(L6)-

(NO₃)₃] (Ln = La, Pr, Eu, Gd, Tb, Y) also form smectic

A2 phases around 90-110 ꢀC (glass transition detected

by POM and SA-XRD; Table 3 and Figure S12, in the

regularly spaced along the strand, we can estimate an

˚

nematic mesophase, which is much shorter than ca. 119 A

average total length of 4 ꢁ 17 ≈ 68 A for the ligand in the

estimated by molecular modeling of L5 or L6 in their

extended form (Figure 9a).^10b,11This discrepancy can be

reasonably assigned to the combination of (i) the shrink-

ing of the flexible alkyl segments for fitting the large

molecular area fixed by the rigid aromatic rods of the

ligand and (ii) the fluctuation of the alignment of the

mesogens along the director in the nematic phase.²⁸

The direct connection of the cyanobiphenyl groups to the

divergent 1,3-phenyl spacer in ligands L7-L10 removes

(29) Gehringer, L.; Bourgogne, C.; Guillon, D.; Donnio, B. J. Am. Chem.

Soc. 2004, 126, 3856–3867.

(30) Guillon, D.; Skoulios, A. Mol. Cryst. Liq. Cryst. 1983, 91, 341–352.

(31) Guillon, D.; Skoulios, A. J. Phys. (Paris) 1984, 45, 607–621.

8612 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

Figure 11. Birefringent texture observed by POM for [Eu(L6)(NO₃)₃] in

the nematic mesophase.

liquid around 147-159 ꢀC (Figure 10b), a clearing tem-

perature approximately 30 ꢀC lower than that found for

the analogous nonmethylated complexes, which suggests

less efficient intermolecular cohesion within the meso-

phase upon methylation of the terminal cyanobiphenyl

groups. For the smaller lanthanides (Ln = Eu, Gd, Tb,

Y), the initial SmA2 phase transforms into a second

mesophase through a first-order phase transition prior

to isotropization (Table 3 and Figures 10c and S13 in the

Supporting Information). The breakdown of the d₀₀₁and

d₀₀₂SA-XRD reflections combined with the appearance

of Schlieren textures by POM (Figure 11) demonstrate

the formation of a nematic phase for Ln=Eu, Y prior to

isotropization.

Please note that the diffuse band attributed to some

nonspecific residual intermolecular interactions between

the aromatic segments responsible for the global align-

ment along the director nB in the nematic phase is shifted

˚

by 1-5 A on going from L6 (d ≈ 17 A) to its complexes

˚

[Lu(L6)(NO₃)₃] (d ≈ 18-22 A; Table S3 in the Supporting

Information), in line with a rigidification of the central

aromatic core by metal complexation. For Ln = Gd and

Tb, the DSC traces (Figure S13 in the Supporting In-

formation) and the SA-XRD profiles (Figure S14 in the

Supporting Information) imply first-order SmA2 f

SmAd transitions,^8,28in which the interlayer separation

abruptly decreases by 10% because of partial interdigita-

tion of the terminal cyanobiphenyl units (Table S3 in the

Supporting Information). Interestingly, the complex with

the smallest lanthanide [Lu(L6)(NO₃)₃] apparently shows

the expected sequence g(90 ꢀC)SmX(130 ꢀC)N(144 ꢀC)I

(Figure 10d) but with an interlayer separation in the lamel-

˚

lar phase of only d₀₀₁= 60 A (Table S3 and Figure S12 in

Figure 10. DSC traces recorded for the complexes (a) [Eu(L5)(NO₃)₃]

(a single heating process), (b) [Eu(L6)(NO₃)₃], (c) [Pr(L6)(NO₃)₃], and

(d) [Lu(L6)(NO₃)₃] (second heating/cooling cycle; SmA2 = bilayer smectic

phase, N = nematic phase, L = lamellar phase, and I = isotropic liquid).

the Supporting Information). This 50% reduction of the

interlayer separation follows the trend initiated with Ln=

Gd and Tb in the SmAd mesophase, and it is accompa-

nied by the appearance of a diffuse reflection around d =

˚

Supporting Information) with similar interlayer separa-

9.0 A (Figure S15 in the Supporting Information). From

2

the molecular area A = 116 A estimated with eq 10 for

˚

tions (100 A e d₀₀₁e 120 A) but slightly larger molecular

˚

2

˚

areas (60 A e A e 66 A ) because of the expansion

brought by the methyl groups in the packing of the

terminal cyanobiphenyl groups (Table S3 in the Support-

ing Information). For the large lanthanides (Ln=La, Pr),

the SmA2 phase is directly transformed into the isotropic

[Lu(L6)(NO₃)₃] in this lamellar phase (L in Table S3 in the

Supporting Information), we deduce that the ionic (i.e.,

metal-containing) and mesogenic (i.e., the cyanobiphenyl

termini) parts of a pair of adjacent complexes separa-

˚

ted by 9.0 A contribute to the interface of interdigitated

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8613

molecules, forming the lamellar phase (Figure 9b). The

˚

total lack of a reflection at 9.0 A for all other lanthanides

between terminal methyl substitution (structural input) and

the phase sequence in these dendrimeric complexes (thermo-

dynamic output) would require the completion of family 1,

initiated with L5 and L6, with the systematic introduction of

all accessible methyl-substituted cyanobiphenyl termini.

(Ln = La-Tb; Figure S15 in the Supporting Informa-

tion), or an extremely weak reflection for [Pr(L6)(NO₃)₃]

(Table S3 in the Supporting Information), strongly sug-

gests a larger separation between the packed rods within

the layers of the SmA2 mesophase, in line with the packing of

H-shaped dimers as schematically shown in Figure 9c.

This observation is further supported by the detailed

thermodynamic studies in a noncoordinating solvent,

which show an increased propensity to form dimers for

the larger lanthanides.

Experimental Section

Chemicals were purchased from Fluka AG and Aldrich

and used without furtherpurification unless otherwise stated.

The starting compounds 2,6-bis(1-ethyl-5-hydroxybenzim-

idazol-2-yl)pyridine,^14a(4-cyano-2-methylphenyl)boronicacid,¹³

4-[(dimethylamino)pyridinium]-4-toluenesulfonate (DPTS),³²

10-bromodecan-1-ol,³³4-(10-hydroxydecyloxy)benzoic acid,^10a

5-(tert-butyldimethylsiloxy)isophthalic acid,³⁴dimethyl-5-

hydroxyisophthalate,³⁵and ligands L11^10band L5^10bwere pre-

Conclusion

Upon complexation of Ln(NO₃)₃to the tridentate 5,5⁰-

disubstituted 2,6[bis(N-ethylbenzimidazol-2-yl)pyridine] bind-

ing unit of the ligands L5-L14 in noncoordinating dichloro-

methane, the monomeric complex [Ln(Lk)(NO₃)₃] is exclusively

formed for the smallest lanthanide (Ln=Lu). For midrange

(Ln = Eu) and for large lanthanides (Ln= Pr), the thermo-

dynamic dimerization equilibria 2[Ln(Lk)(NO₃)₃] T [Ln₂(Lk)₂-

(NO₃)₆] can be investigated across the complete ligand series.

Within the frame of a simple harmonic model, the formation

of the dimer primarily results from the lanthanide-size-

dependent building of intermetallic nitrato bridges, which

are completed by nondirectional van der Waals interactions

betweenthe polarizable groups of the semidendrimeric ligand

strands. Specific branching and methyl substitution of the

ligand strands do influence the intermolecular interactions

between the monomeric subunits, but enthalpy/entropy com-

pensation strongly limits the effects of these structural changes

on the global free energy of cohesion. In the absence of solvent,

the situation completely changes and the ligands L5-L14,

and the complexes [Ln(Lk)(NO₃)₃] with L5-L6 display

variable liquid-crystalline behaviors, which are dominated

by the formation of smectic A and nematic phases. Detailed

studies of the dendrimeric complexes [Ln(L5)(NO₃)₃] and

[Ln(L6)(NO₃)₃] show that the large lanthanide cations (Ln=

La, Pr) only produce bilayer smectic mesophases in the 90-

180 ꢀC range. Interestingly, the connection of methyl groups

onto the terminal cyanobiphenyl groups in [Ln(L6)(NO₃)₃] is

responsible for the appearance of either an additional inter-

digitated lamellar phase (Ln = Gd, Tb, Lu) or a nematic

phase (Ln=Eu, Y, Lu) with midrange and smalllanthanides.

The intermolecular shift of the strands responsible for the

decrease of the interlayer separation in the lamellar phases

seems to be correlated with the propensity of the rodlike

monomeric complexes to form dimer in solution. In other

words, [Lu(L6)(NO₃)₃], which exists as a monomer in di-

chloromethane at millimolar concentration, leads to a lamel-

lar mesophase in the pure compound with large inter-

penetration and short interlayer separations. On the other

hand, [Ln(L6)(NO₃)₃] (Ln = Eu, Pr), which exists as a

mixture of monomers and dimers in solution, leads to bilayer

SmA2, with large interlayer distances matching the global

shape of the extended complexes. We thus suspect that the

disruption of the lateral intermolecular cohesion required for

transforming a lamellar phase into a nematic phase in the

pure complexes drastically depends on peripheral methyl

substitution of the terminal cyanobiphenyl groups, while a

similar effect is masked in solution by enthalpy/entropy com-

pensation. Despite our reticence to develop tedious multistep

synthetic strategies, it is clear that an unambiguous correlation

pared by published procedures. The nitrate salts Ln(NO₃)₃

3

xH₂Owerepreparedfromthecorresponding oxides(Aldrich,

99.99%).³⁶The Ln content ofthesolid salt was determinedby

complexometric titrations with Titriplex III (Merck) in the

presence of urotropine and xylene orange.³⁷Acetonitrile and

dichloromethane were distilled over calcium hydride. Silica-

gel plates Merck 60 F₂₅₄were used for thin-layer chromato-

graphy, and Fluka silica gel 60 (0.04-0.063 mm) was used

for preparative column chromatography. Abbreviations:

DCC=N,N⁰-dicyclohexylcarbodiimide, DPTS=4-[(dimethyl-

amino)pyridinium]-4-toluenesulfonate, 4-Ppy=4-pyrrolidino-

pyridine, DMAP=4-(dimethylamino)pyridine, EDCI= N-[3-

(dimethylamino)propyl]-N⁰-ethylcarbodiimide, and dppf =

1,1⁰-bis(diphenylphosphino)ferrocene.

Preparation of 4⁰-Hydroxy-3⁰-methyl-1,1⁰-biphenyl-4-carbo-

nitrile (1).A solution of 4-iodo-2-methylphenol (580 mg, 2.48 mmol),

(4-cyanophenyl)boronic acid (0.40 g, 2.72 mmol, 1.1 equiv), and

K₃PO₄H₂O (1.64 g, 7.12 mmol, 2.9 equiv) in 2-propanol (16 mL),

3

and H₂O (5 mL) was flushed with N₂for 10 min. Pd(dppf)Cl₂(40mg,

0.049 mmol, 0.02 equiv) was added, and the solution was stirred

under N₂for 3 h at 70 ꢀC. The solution was filtered, 40 mL of CH₂Cl₂

was added, and the phases were separated. The aqueous phase was

extracted with 40 mL of CH₂Cl₂, and the combined organic phases

were dried over Na₂SO₄, filtered, and evaporated to dryness. The

crude product was purified by column chromatography (silica;

CH₂Cl₂/CH₃OH =100/0 f 99.5/0.5) to yield 0.43 g (83%) of 1.

¹H NMR (CDCl₃): δ 7.69 (d, 2H, ³J = 8.5 Hz), 7.63 (d, 2H,

³J=8.6 Hz), 7.38 (d, 1H, ⁴J=2.7 Hz), 7.33 (dd, 1H, ³J=8.3 Hz,

⁴J=2.7 Hz), 6.87 (d, 1H, ³J=8.2 Hz), 4.86 (s, 1H), 2.33 (s, 3H).

ESI-MS: m/z 210.3 ([M þ H]^þ), 227.5 ([M þ H₂O þ H]^þ), 436.5

([2M þ H₂O þ H]^þ). Elem anal. Calcd for C₁₄H₁₁NO 0.03H₂O:

3

C, 80.15; H, 5.31; N, 6.68. Found: C, 80.18; H, 5.33; N, 6.64.

Preparation of 4⁰-Hydroxy-2⁰-methyl-1,1⁰-biphenyl-4-carbo-

nitrile (2). A solution of 4-bromo-3-methylphenol (1.16 g, 6.2

mmol), (4-cyanophenyl)boronic acid (1.00 g, 6.8 mmol, 1.1 equiv),

and K₃PO₄H₂O (4.10 g, 17.8 mmol, 2.9 equiv) in 2-propanol

3

(40 mL) and H₂O (12 mL) was flushed with N₂for 15 min.

Pd(dppf)Cl₂(100 mg, 0.12 mmol, 0.02 equiv) was added, and the

solution was stirred under N₂for 16 h at 70 ꢀC. The solution was

filtered, 2-propanol was evaporated, and 100 mL of half-saturated

aqueous NaCl was added. The solution was extracted with 4 ꢁ

50 mL of CH₂Cl₂, and the combined organic phases were dried

(32) Noore, J. S.; Stupp, S. I. Macromolecules 1990, 23, 65–70.

(33) Chong, J. M.; Heuft, M. A.; Rabbat, P. J. Org. Chem. 2000, 65, 5837–

5838.

(34) Miller, T. M.; Kwock, E. W.; Neenan, T. X. Macromolecules 1992,

25, 3143–3148.

(35) Li, F.-Y.; Zhu, L.-L.; Ma, X.; Wang, Q.-C.; Tian, H. Tetrahedron

Lett. 2009, 50, 597–600.

(36) Desreux, J. F. In Lanthanide Probes in Life, Chemical and Earth

Sciences; B€unzli, J.-C. G., Choppin, G. R., Eds.; Elsevier Publishing Co.:

Amsterdam, The Netherlands, 1989; Chapter 2.

(37) Schwarzenbach, G. Complexometirc Titrations; Chapman and Hall:

London, 1957; p 8.

8614 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

over Na₂SO₄, filtered, and evaporated to dryness. The crude pro-

duct was purified by column chromatography (silica; CH₂Cl₂/

CH₃OH =100/0 f 99.5/0.5) to yield 0.87 g (67%) of 2.

heat to room temperature and stirred for 16 h. After filtration

and evaporation, the crude product was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) to

yield 1.06 g (90%) of 19.

¹H NMR (CDCl₃): δ 7.69 (d, 2H, ³J = 8.4 Hz), 7.40 (d, 2H,

³J=8.3 Hz), 7.08 (d, 1H, ³J=8.3 Hz), 6.78 (d, 1H, ⁴J=2.6 Hz),

6.75 (dd, 1H, ³J=8.3 Hz, ⁴J=2.6 Hz), 4.81 (s, 1H), 2.23 (s, 3H).

ESI-MS: m/z 227.4 ([M þ H₂O þ H]^þ), 436.3 ([2M þ H₂O þ

¹H NMR (CDCl₃): δ 8.15 (d, 2H, ³J = 9.0 Hz), 7.58 (d, 1H,

⁴J=1.5 Hz), 7.54 (dd, 1H, ³J=8.0 Hz, ⁴J=1.5 Hz), 7.24 (d, 1H,

³J = 7.9 Hz), 7.15 (m, 1H), 7.10 (m, 2H), 6.98 (d, 2H, ³J = 9.0

Hz), 4.05 (t, 2H, ³J=6.6 Hz), 3.65 (t, 2H, ³J=6.3 Hz), 2.12 (s,

3H), 2.04 (s, 3H), 1.76-1.88 (m, 2H), 1.54-1.62 (m, 2H),

1.44-1.52 (m, 2H), 1.29-1.42 (m, 10 H). ESI-MS: m/z 500.5

([M þ H]^þ), 277.6 ([HO(CH₂)₁₀OC₆H₄CO]^þ). Elem anal. Calcd

H]^þ). Elem anal. Calcd for C₁₄H₁₁NO 0.11H₂O: C, 79.59; H,

3

5.36; N, 6.63. Found: C, 79.61; H, 5.42; N, 6.46.

Preparation of 4⁰-Hydroxy-2,2⁰-dimethyl-1,1⁰-biphenyl-4-carbo-

nitrile (3). A solution of 4-bromo-3-methylphenol (1.06 g, 5.65

mmol), (4-cyano-2-methylphenyl)boronic acid (1.00 g, 6.21 mmol,

for C₃₂H₃₇NO₄0.14H₂O: C, 76.54; H, 7.48; N, 2.79. Found: C,

3

1.1 equiv), and K₃PO₄H₂O (3.91 g, 17 mmol, 3.0 equiv) in

76.54; H, 7.54; N, 2.74.

3

2-propanol (30 mL) and H₂O (10 mL) was flushed with N₂for

30 min. Pd(dppf)Cl₂(90 mg, 0.113 mmol, 0.02 equiv) was added,

and the solution was stirred under N₂for 14 h at 70 ꢀC. The

solution was filtered, 2-propanol was evaporated, the aqueous

phase was extracted with 4 ꢁ 50 mL of CH₂Cl₂, and the combi-

ned organic phases were dried over Na₂SO₄, filtered, and

evaporated to dryness. The crude product was purified by

column chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f

99.5/0.5) to yield 0.75 g (59%) of 3.

Preparation of Compound 21. To a solution of 17 (1.46 g, 3.01

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added

4-carboxybenzaldehyde (0.45 g, 3.01 mmol), DCC (1.30 g, 6.31

mmol), DPTS (0.88 g, 3.01 mmol), and a spatula tip of 4-PPy.

The reaction mixture was allowed to heat slowly to room

temperature and stirred for 16 h. After filtration, the solution

was evaporated to dryness and the residue was purified by

column chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f

99.5/0.5) to yield 1.46 g (78%) of 21.

¹H NMR (CDCl₃): δ 7.56 (d, 1H, ⁴J=1.7 Hz), 7.50 (dd, 1H,

³J=8.2 Hz, ⁴J=1.7 Hz), 7.20 (d, 1H, ³J=7.9 Hz), 6.91 (d, 1H,

³J=8.0 Hz), 6.77 (d, 1H, ⁴J=2.8 Hz), 6.73 (dd, 1H, ³J=8.3 Hz,

⁴J=2.8 Hz), 4.86 (s, 1H), 2.09 (s, 3H), 1.98 (s, 3H). ESI-MS: m/z

¹H NMR (CDCl₃): δ 10.10 (s, 1H), 8.20 (d, 2H, ³J=8.2 Hz),

8.17 (d, 2H, ³J=8.8 Hz), 7.96 (d, 2H, ³J=8.4 Hz), 7.73 (d, 2H,

³J=8.6 Hz), 7.68 (d, 2H, ³J=8.4 Hz), 7.49 (d, 1H, ⁴J=2.2 Hz),

7.46 (dd, 1H, ³J=8.6 Hz, ⁴J=2.2 Hz), 7.25 (d, 1H, ³J=8.6 Hz),

7.00 (d, 2H, ³J=9.0 Hz), 4.36 (t, 2H, ³J=6.7 Hz), 4.06 (t, 2H,

³J=6.7 Hz), 2.30 (s, 3H), 1.75-1.87 (m, 4H), 1.30-1.53 (m, 12H).

ESI-MS: m/z 409.5 (CHOC₆H₄COO(CH₂)₁₀OC₆H₄CO]^þ,

618.3 ([M þ H]^þ), 635.2 ([M þ H₂O]^þ), 1253.3 ([2M þ H₂O þ

H]^þ). Elem anal. Calcd for C₃₉H₃₉NO₆: C, 75.83; H, 6.36; N,

2.27. Found: C, 75.50; H, 6.40; N, 2.27.

222.3 ([M - H]^-). Elem anal. Calcd for C₁₅H₁₃NO 0.05H₂O: C,

3

80.37; H, 5.89; N, 6.25. Found: C, 80.37; H, 5.88; N, 6.22.

Preparation of Compound 17. To a solution of 1 (1.08 g, 5.16

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added 4-(10-

hydroxydecyloxy)benzoic acid (1.52 g, 5.16 mmol), DCC (2.06 g,

10 mmol), DPTS (1.52 g, 5.16 mmol), and a spatula tip of 4-PPy.

The reaction mixture was allowed to heat slowly to room tem-

perature and stirred for 48 h. After filtration, the solution was

evaporated to dryness and the residue was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) to

yield 2.28 g (91%) of 17.

Preparation of Compound 22. To a solution of 18 (1.49 g, 3.07

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added

4-carboxybenzaldehyde (0.46 g, 3.07 mmol), DCC (1.24 g, 6.00

mmol), DPTS (0.90 g, 3.07 mmol), and a spatula tip of 4-PPy.

The reaction mixture was allowed to heat slowly to room tem-

perature and stirred for 48 h. After filtration, the solution was

evaporated to dryness and the residue was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99.5/0.5)

to yield 1.22 g (64%) of 22.

¹H NMR (CDCl₃): δ 8.18 (d, 2H, ³J = 9.0 Hz), 7.73 (d, 2H,

³J=8.6 Hz), 7.68 (d, 2H, ³J=8.7 Hz), 7.49 (d, 1H, ⁴J=1.8 Hz),

7.46 (dd, 1H, ³J=8.3 Hz, ⁴J=2.2 Hz), 7.25 (d, 1H, ³J=8.5 Hz),

6.99 (d, 2H, ³J=9.0 Hz), 4.06 (t, 2H, ³J=6.5 Hz), 3.65 (t, 2H,

³J = 6.5 Hz), 2.31 (s, 3H), 1.79-1.87 (m, 2H), 1.53-1.61 (m,

2H), 1.44-1.52 (m, 2H), 1.30-1.40 (m, 10H). ESI-MS: m/z

277.4 ([HO(CH₂)₁₀OC₆H₄CO]^þ), 486.5 ([M þ H]^þ), 503.5 ([M þ

H₂O]^þ), 508.6 ([M þ Na]^þ), 989.0 ([2M þ H₂O]^þ), 994.0 ([2M þ

¹H NMR (CDCl₃): δ 10.10 (s, 1H), 8.20 (d, 2H, ³J=8.3 Hz),

8.15 (d, 2H, ³J=8.8 Hz), 7.95 (d, 2H, ³J=8.4 Hz), 7.72 (d, 2H,

³J=8.3 Hz), 7.45 (d, 2H, ³J=8.4 Hz), 7.24 (d, 1H, ³J=8.2 Hz),

7.15 (d, 1H, ⁴J=2.3 Hz), 7.11 (dd, 1H, ³J=8.5 Hz, ⁴J=2.4 Hz),

6.98 (d, 2H, ³J=9.0 Hz), 4.36 (t, 2H, ³J=6.7 Hz), 4.05 (t, 2H,

³J=6.6 Hz), 2.27 (s, 3H), 1.75-1.87 (m, 4H), 1.31-1.53 (m, 12H).

ESI-MS: m/z 409.5 (CHOC₆H₄COO(CH₂)₁₀OC₆H₄CO]^þ, 618.5

([M þ H]^þ), 635.7 ([M þ H₂O]^þ), 1253.3 ([2M þ H₂O þ H]^þ).

Na]^þ). Elem anal. Calcd for C₃₁H₃₅NO₄0.18H₂O: C, 76.16; H,

3

7.29; N, 2.87. Found: C, 76.07; H, 7.23; N, 2.83.

Preparation of Compound 18. To a solution of 2 (0.83 g, 3.97

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added 4-(10-

hydroxydecyloxy)benzoic acid (1.17 g, 3.97 mmol), DCC (1.55 g,

7.5 mmol), DPTS (1.17 g, 3.97 mmol), and a spatula tip of 4-PPy.

The reaction mixture was allowed to heat slowly to room tem-

perature and stirred for 3 days. After filtration, the solution was

evaporated to dryness and the residue was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) to

yield 1.59 g (82%) of 18.

Elem anal. Calcd for C₃₉H₃₉NO₆0.19H₂O: C, 75.41; H, 6.39;

3

N, 2.25. Found: C, 75.41; H, 6.41; N, 2.18.

Preparation of Compound 23. A solution of 4-carboxybenzal-

dehyde (0.31 g, 2.06 mmol) and 19 (1.02 g, 2.04 mmol) in 200 mL

of dry CH₂Cl₂was cooled in an ice bath. DCC (0.90 g, 4.37

mmol), DPTS (0.60 g, 2.04 mmol), and a spatula tip of 4-PPy

were added, and the reaction mixture was slowly heated to room

temperature and stirred for 18 h. The solution was filtered, the

solvent was evaporated, and the crude product was purified by

column chromatography (silica; CH₂Cl₂/CH₃OH=100/0 f 99/1)

to yield 1.16 g (90%) of 23.

¹H NMR (CDCl₃): δ 8.15 (d, 2H, ³J = 9.0 Hz), 7.72 (d, 2H,

³J=8.4 Hz), 7.45 (d, 2H, ³J=8.4 Hz), 7.24 (d, 1H, ³J=8.4 Hz),

7.15 (d, 1H, ⁴J=2.1 Hz), 7.11 (dd, 1H, ³J=8.5 Hz, ⁴J=2.4 Hz),

6.98 (d, 2H, ³J=8.8 Hz), 4.05 (t, 2H, ³J=6.5 Hz), 3.65 (q, 2H,

³J = 6.1 Hz), 2.27 (s, 3H), 1.78-1.87 (m, 2H), 1.53-1.62 (m,

2H), 1.43-1.53 (m, 2H), 1.29-1.41 (m, 10H). ESI-MS: m/z

486.6 ([M þ H]^þ). Elem anal. Calcd for C₃₁H₃₅NO₄: C, 76.67; H,

7.26; N, 2.88. Found: C, 76.39; H, 7.29; N, 2.88.

¹H NMR (CDCl₃): δ 10.10 (s, 1H), 8.20 (d, 2H, ³J=8.5 Hz),

8.15 (d, 2H, ³J=8.9 Hz), 7.95 (d, 2H, ³J=8.4 Hz), 7.58 (d, 1H,

⁴J=1.6 Hz), 7.55 (dd, 1H, ³J=8.2 Hz, ⁴J=1.6 Hz), 7.24 (d, 1H,

³J=8.1 Hz), 7.15 (m, 1H), 7.10 (m, 2H), 6.98 (d, 2H, ³J=8.9 Hz),

4.36 (t, 2H, ³J=6.7 Hz), 4.05 (t, 2H, ³J=6.6 Hz), 2.12 (s, 3H), 2.05

(s, 3H),1.74-1.87 (m, 4H), 1.42-1.53(m,2H),1.31-1.42 (m, 10 H).

ESI-MS: m/z 632.5 ([M þ H]^þ), 409.4 ([CHOC₆H₄COO(CH₂)₁₀-

Preparation of Compound 19. To a solution of 4-(10-hydro-

xydecyloxy)benzoic acid (0.70 g, 2.38 mmol) and 3 (0.53 g, 2.37

mmol) in 75 mL of freshly distilled CH₂Cl₂cooled in an ice bath

was added DCC (1.20 g, 5.83 mmol), DPTS (0.70 g, 2.38 mmol),

and a spatula tip of 4-PPy. The reaction mixture was allowed to

OC₆H₄CO]^þ). Elem anal. Calcd for C₄₀H₄₁NO₆0.97H₂O: C,

3

74.01; H, 6.67; N, 2.16. Found: C, 74.00; H, 6.55; N, 2.16.

Article

Preparation of Compound 25. To a solution of 21 (1.38 g, 2.23

Inorganic Chemistry, Vol. 49, No. 18, 2010 8615

and a spatula tip of 4-PPy. The reaction mixture was allowed to

heat slowly to room temperature and stirred for 48 h. After

filtration, the solution was evaporated to dryness and the resi-

due was purified by column chromatography (silica; CH₂Cl₂/

CH₃OH=100/0 f 98.5/1.5) to yield 0.85 g (62%) of L13.

¹H NMR (CDCl₃): δ 8.36 (d, 2H, ³J = 8.3 Hz), 8.34 (d, 4H,

³J=8.8 Hz), 8.20 (d, 4H, ³J=8.4 Hz), 8.15 (d, 4H, ³J=8.9 Hz),

8.08 (t, 1H, ³J=8.1 Hz), 7.72 (d, 4H, ³J=8.4 Hz), 7.71 (d, 2H,

⁴J=2.1 Hz), 7.52 (d, 2H, ³J=9.0 Hz), 7.45 (d, 4H, ³J=8.4 Hz),

7.25 (d, 2H, ³J=8.6 Hz), 7.24 (d, 2H, ³J=8.2 Hz), 7.14 (d, 2H,

⁴J=2.6 Hz), 7.11 (dd, 2H, ³J=8.5 Hz, ⁴J=2.6 Hz), 6.98 (d, 4H,

³J=8.7 Hz), 4.82 (q, 4H, ³J=7.2 Hz), 4.39 (t, 4H, ³J=6.7 Hz),

4.06 (t, 4H, ³J = 6.5 Hz), 2.27 (s, 6H), 1.78-1.88 (m, 8H), 1.41

(t, 6H, ³J = 7.3 Hz), 1.32-1.52 (m, 24H). ESI-MS: m/z 1631.4

([M þ H]^þ). Elem anal. Calcd for C₁₀₁H₉₅N₇O₁₄: C, 74.38; H,

5.87; N, 6.01. Found: C, 74.28; H, 5.93; N, 5.98.

mmol) in 75 mL of THF and 75 mL of H₂O was added NaClO₂

(0.90 g, 10 mmol) and H₂NSO₃H (0.97 g, 10 mmol). After

stirring for 2 h, THF was evaporated and the aqueous phase

was extracted with CH₂Cl₂(3 ꢁ 50 mL). The combined organic

extracts were dried over Na₂SO₄, filtered, and evaporated to

dryness to yield 1.41 g (91%) of 25.

¹H NMR (DMSO-d₆): δ 13.37 (s, 1H), 8.10 (d, 2H, ³J =

8.5 Hz), 8.06 (s, 4H), 7.94 (d, 2H, ³J=9.1 Hz), 7.91 (d, 2H, ³J=

9.0 Hz), 7.76 (d, 1H, ⁴J=2.2 Hz), 7.66 (dd, 1H, ³J=8.3 Hz, ⁴J=

2.2 Hz), 7.33 (d, 1H, ³J=8.7 Hz), 7.12 (d, 2H, ³J=8.6 Hz), 4.30

3

(t, 2H, J = 6.6 Hz), 4.09 (t, 2H, J = 6.5 Hz), 2.23 (s, 3H),

1.68-1.79 (m, 4H), 1.26-1.46 (m, 12H). ESI-MS: m/z 632.7 ([M

- H]^-). Elem anal. Calcd for C₃₉H₃₉NO₇0.46H₂O: C, 72.97;

H, 6.27; N, 2.18. Found: C, 72.97; H, 6.29; N, 2.12.

3

Preparation of Compound 26. To a solution of 22 (1.06 g, 1.72

mmol) in 75 mL of THF and 75 mL of H₂O was added NaClO₂

(0.72 g, 8.0 mmol) and H₂NSO₃H (0.78 g, 8.0 mmol). After

stirring for 2 h, THF was evaporated and the aqueous phase was

extracted with CH₂Cl₂(4 ꢁ 50 mL). The combined organic

extracts were dried over Na₂SO₄, filtered, and evaporated to

dryness to yield 1.09 g (100%) of 26.

Preparation of Ligand L14. To a solution of 27 (1.05 g, 1.62

mmol) and 2,6-bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine

(0.32 g, 0.81 mmol) in 200 mL of dry CH₂Cl₂stirred at 0 ꢀC was

added DCC (0.85 g, 4.13 mmol), DPTS (0.48 g, 1.63 mmol), and

a spatula tip of 4-PPy. The reaction mixture was allowed to heat

to room temperature and stirred for 18 h. After filtration, the

solution was evaporated to dryness and the residue was purified

by column chromatography (silica; CH₂Cl₂/CH₃OH=100/0 f

99/1) to yield 0.42 g (31%) of L14.

¹H NMR (DMSO-d₆): δ 13.36 (s, 1H), 8.07 (d, 2H, ³J =

9.0 Hz), 8.06 (s, 4H) 7.93 (d, 2H, ³J=8.4 Hz), 7.61 (d, 2H, ³J=

8.2 Hz), 7.31 (d, 1H, ³J=8.3 Hz), 7.26 (d, 1H, ⁴J=2.2 Hz), 7.19

(dd, 1H, ³J=8.2 Hz, ⁴J=2.5 Hz), 7.11 (d, 2H, ³J=9.0 Hz), 4.30

¹H NMR (CDCl₃): δ 8.37 (d, 2H, ³J = 8.1 Hz), 8.34 (d, 4H,

³J=8.7 Hz), 8.20 (d, 4H, ³J=8.3 Hz), 8.15 (d, 4H, ³J=8.9 Hz),

8.09 (t, 1H, ³J=8.2 Hz), 7.72 (d, 2H, ⁴J=1.7 Hz), 7.58 (m, 2H),

7.53 (d, 2H, ³J=8.1 Hz), 7.53 (d, 2H, ³J=7.9 Hz), 7.26 (d, 2H,

³J = 8.5 Hz), 7.23 (d, 2H, ³J = 7.7 Hz), 7.15 (m, 2H), 7.10 (m,

4H), 6.99 (d, 4H, ³J=8.9 Hz), 4.82 (q, 4H, ³J=7.3 Hz), 4.38 (t,

4H, ³J=6.7 Hz), 4.06 (t, 4H, ³J=6.6 Hz), 2.12 (s, 6H), 2.04 (s,

6H), 1.70-1.87 (m, 8H), 1.44-1.53 (m, 8H), 1.41 (t, 6H, ³J=7.2

Hz), 1.32-1.41 (m, 16 H). ESI-MS: m/z 1659.8 ([M þ H]^þ).

3

(t, 2H, J = 6.8 Hz), 4.08 (t, 2H, J = 6.5 Hz), 2.26 (s, 3H),

1.68-1.78 (m, 4H), 1.27-1.46 (m, 12H). ESI-MS: m/z 632.7

([M - H]^-). Elem anal. Calcd for C₃₉H₃₉NO₇0.45H₂O: C,

3

72.98; H, 6.27; N, 2.18. Found: C, 72.99; H, 6.36; N, 2.02.

Preparation of Compound 27. To a solution of 23 (1.16 g, 1.84

mmol) in 50 mL of THF and 50 mL of H₂O was added NaClO₂

(0.90 g, 10 mmol) and H₂NSO₃H (0.97 g, 10 mmol). After

stirring for 2 h, THF was evaporated and the aqueous phase

was extracted with 3 ꢁ 50 mL of CH₂Cl₂. The combined organic

extracts were dried over Na₂SO₄and evaporated to yield 1.14 g

(96%) of 27.

Elem anal. Calcd for C₁₀₃H₉₉N₇O₁₄0.61H₂O: C, 74.08; H, 6.05;

3

N, 5.87. Found: C, 74.08; H, 6.24; N, 6.06.

Preparation of Compound 28. To a suspension of 60 g of

K₂CO₃in 250 mL of acetone was added 10-bromodecan-1-ol

(13.35 g, 56.3 mmol), dimethyl 5-hydroxyisophthalate (11.83 g,

56.3 mmol), and a spatula tip of KI. The reaction mixture was

refluxed for 68 h, the acetone was evaporated, and the aqueous

phase was extracted with 3 ꢁ 100 mL of ethyl acetate. The

combined organic extracts were dried over Na₂SO₄, filtered, and

evaporated to dryness. The crude ester was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) and

dissolved in 225 mL of EtOH. To this was added 150 mL of 1 M

KOH, and the solution was stirred for 2 h. After evaporation of

ethanol, the aqueous solution was washed with 2 ꢁ 100 mL of

CH₂Cl₂and acidified to pH = 1 with 2 M HCl. The resulting

white precipitate was filtered off, washed with aqueous 0.1 M

HCl, and dried in vacuum to yield 6.15 g (32%) of 28.

¹H NMR (DMSO-d₆): δ 13.31 (s, 1H), 8.07 (d, 2H, ³J =

8.8 Hz), 8.06 (s, 4H), 7.83 (d, 1H, ⁴J=1.5 Hz), 7.73 (dd, 1H, ³J=

8.3 Hz, ⁴J=1.5 Hz), 7.33 (d, 1H, ³J=8.4 Hz), 7.25 (m, 1H), 7.16

(m, 2H), 7.11 (d, 2H, ³J=8.6 Hz), 4.30 (t, 2H, ³J=6.5 Hz), 4.08

(t, 2H, ³J=6.4 Hz), 2.07 (s, 3H), 2.01 (s, 3H), 1.68-1.78 (m, 4H),

1.26-1.45 (m, 12 H). ESI-MS: m/z 646.9 ([M - H]^-). Elem anal.

Calcd for C₄₀H₄₁NO₇0.93H₂O: C, 72.30; H, 6.50; N, 2.11.

3

Found: C, 72.30; H, 6.50; N, 2.05.

Preparation of Ligand L12. To a solution of 25 (0.89 g, 1.40

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added 2,6-

bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine (0.28 g, 0.70

mmol), DCC (0.62 g, 3.0 mmol), DPTS (0.41 g, 1.40 mmol),

and a spatula tip of 4-PPy. The reaction mixture was allowed to

heat slowly to room temperature and stirred for 48 h. After

filtration, the solution was evaporated to dryness and the resi-

due was purified by column chromatography (silica; CH₂Cl₂/

CH₃OH =100/0 f 98.5/1.5) to yield 0.82 g (72%) of L12.

¹H NMR (CDCl₃): δ 8.36 (d, 2H, ³J = 8.2 Hz), 8.34 (d, 4H,

³J=8.5 Hz), 8.20 (d, 4H, ³J=8.7 Hz), 8.17 (d, 4H, ³J=8.7 Hz),

8.08 (t, 1H, ³J=8.2 Hz), 7.72 (d, 4H, ³J=8.7 Hz), 7.71 (d, 2H,

⁴J=2.2 Hz), 7.67 (d, 4H, ³J=8.6 Hz), 7.52 (d, 2H, ³J=8.9 Hz),

7.49 (d, 2H, ⁴J=2.2 Hz), 7.46 (dd, 2H, ³J=8.2 Hz, ⁴J=2.4 Hz),

7.25 (d, 2H, ³J=8.8 Hz), 7.24 (d, 2H, ³J=8.7 Hz), 7.00 (d, 4H,

³J=8.9 Hz), 4.82 (q, 4H, ³J=7.3 Hz), 4.39 (t, 4H, ³J=6.8 Hz),

4.06 (t, 4H, ³J =6.5 Hz), 2.30 (s, 6H), 1.78-1.88 (m, 8H), 1.41

(t, 6H, ³J = 7.3 Hz), 1.32-1.52 (m, 24H). ESI-MS: m/z 1631.4

4

¹H NMR (DMSO-d₆): δ 13.28 (s, 2H), 8.06 (t, 1H, J = 1.4

Hz), 7.62 (d, 2H, ⁴J=1.4 Hz), 4.06 (t, 2H, ³J=6.6 Hz), 3.36 (t,

2H, ³J = 6.7 Hz), 1.67-1.77 (m, 2H), 1.36-1.46 (m, 4H),

1.22-1.36 (m, 10H). ESI- MS: m/z 337.3 ([M - H]^-). Elem

anal. Calcd for C₁₈H₂₆O₆0.58H₂O: C, 61.97; H, 7.85. Found:

3

C, 61.95; H, 7.51.

Preparation of Compound 4. To a mixture of 28 (1.35 g, 3.99

mmol) and 4⁰-hydroxybiphenyl-4-carbonitrile (1.56 g, 7.99 mmol)

in 200 mL of dry CH₂Cl₂cooled in an ice bath was added DCC

(3.80 g, 18.45 mmol), DPTS (2.35 g, 7.99 mmol), and a spatula

tip of 4-PPy. The reaction mixture was allowed to warm to room

temperature and stirred for 18 h. After filtration and evapora-

tion of the solvent, the crude product was purified by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) to

yield 2.24 g (81%) of 4.

([M þ H]^þ). Elem anal. Calcd for C₁₀₁H₉₅N₇O₁₄H₂O: C, 73.57;

3

H, 5.93; N, 5.95. Found: C, 73.57; H, 6.03; N, 6.00.

Preparation of Ligand L13. To a solution of 26 (1.06 g, 1.67

mmol) in 100 mL of dry CH₂Cl₂stirred at 0 ꢀC was added 2,6-

bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine (0.34 g, 0.84

mmol), DCC (0.82 g, 4.0 mmol), DPTS (0.49 g, 1.67 mmol),

¹H NMR (CDCl₃): δ 8.62 (t, 1H, ⁴J = 1.5 Hz), 7.99 (d, 2H,

⁴J=1.4 Hz), 7.75 (d, 4H, ³J=8.3 Hz), 7.70 (d, 4H, ³J=8.4 Hz),

7.67 (d, 4H, ³J=8.6 Hz), 7.38 (d, 4H, ³J=8.6 Hz), 4.13 (t, 2H,

8616 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

3

³J = 6.4 Hz), 3.64 (t, 2H, J = 6.6 Hz), 1.82-1.90 (m, 2H),

Preparation of Compound 9. To a solution of 4-carboxyben-

zaldehyde (0.26 g, 1.73 mmol) and 5 (1.22 g, 1.69 mmol) in 50 mL

of freshly distilled CH₂Cl₂stirred in an ice bath was added DCC

(1.02 g, 4.95 mmol), DPTS (0.50 g, 1.70 mmol), and a spatula tip

of 4-PPy. The reaction mixture was allowed to warm slowly to

room temperature, stirred for 16 h, and filtered. Purification by

column chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f

99.5/0.5) gave 1.01 g (70%) of 9.

1.46-1.61 (m, 4H), 1.30-1.43 (m, 10H). ESI-MS: m/z 693.5

([M þ H]^þ). Elem anal. Calcd for C₄₄H₄₀N₂O₆1.09H₂O: C,

3

74.18; H, 5.97; N, 3.93. Found: C, 74.18; H, 6.02; N, 3.91.

Preparation of Compound 5. To a solution of 28 (1.26 g, 3.72

mmol) and 1 (1.56 g, 7.46 mmol) in 50 mL of dry CH₂Cl₂cooled

in an ice bath was added DCC (4.12 g, 20 mmol), DPTS (2.19 g,

7.45 mmol), and a spatula tip of 4-PPy. After 2 h, the ice bath

was removed and the reaction mixture was stirred for 2 days,

filtered, and evaporated to dryness. The crude product was

purified by column chromatography (silica; CH₂Cl₂/CH₃OH =

100/0 f 99.5/0.5) to yield 1.22 g (46%) of 5.

¹H NMR (CDCl₃): δ 10.09 (s, 1H), 8.66 (t, 1H, ⁴J=1.5 Hz),

8.19 (d, 2H, ³J=8.4 Hz), 8.01 (d, 2H, ⁴J=1.5 Hz), 7.95 (d, 2H,

³J=8.5 Hz), 7.74 (d, 4H, ³J=8.6 Hz), 7.69 (d, 4H, ³J=8.6 Hz),

7.52 (d, 2H, ⁴J=2.2 Hz), 7.49 (dd, 2H, ³J=8.3 Hz, ⁴J=2.0 Hz),

7.28 (d, 2H, ³J=8.3 Hz), 4.35 (t, 2H, ³J=6.7 Hz), 4.13 (t, 2H,

³J = 6.4 Hz), 2.34 (s, 6H), 1.74-1.92 (m, 4H), 1.28-1.55 (m,

12H). ESI-MS: m/z 853.5 ([M þ H]^þ). Elem anal. Calcd for

¹H NMR (CDCl₃): δ 8.66 (t, 1H, ⁴J = 1.5 Hz), 8.01 (d, 2H,

⁴J=1.5 Hz), 7.74 (d, 4H, ³J=8.3 Hz), 7.69 (d, 4H, ³J=8.4 Hz),

7.52 (d, 2H, ⁴J=1.9 Hz), 7.49 (dd, 2H, ³J=8.3 Hz, ⁴J=2.0 Hz),

7.28 (d, 2H, ³J=8.3 Hz), 4.14 (t, 2H, ³J=6.4 Hz), 3.64 (t, 2H,

³J = 6.6 Hz), 2.34 (s, 6H), 1.80-1.92 (m, 4H), 1.30-1.60 (m,

12H). ESI-MS: m/z 721.8 ([M þ H]^þ). Elem anal. Calcd for

C₅₄H₄₈N₂O₈1.30H₂O: C, 74.01; H, 5.82; N, 3.20. Found: C,

3

74.01; H, 5.80; N, 3.05.

Preparation of Compound 10. A solution of 4-carboxybenzal-

dehyde (0.11 g, 0.73 mmol) and 6 (0.52 g, 0.72 mmol) in 50 mL of

freshly distilled CH₂Cl₂was stirred in an ice bath. DCC (0.40 g,

1.94 mmol), DPTS (0.21 g, 0.71 mmol), and a spatula tip of

4-PPy were added. The ice bath was allowed to melt, and the

reaction mixture was stirred for 16 h at room temperature before

being filtered. Purification by column chromatography (silica;

CH₂Cl₂/CH₃OH =100/0 f 99.5/0.5) gave 0.32 g (52%) of 10.

¹H NMR (CDCl₃): δ 10.09 (s, 1H), 8.61 (t, 1H, ⁴J=1.5 Hz),

8.19 (d, 2H, ³J=8.2 Hz), 7.98 (d, 2H, ⁴J=1.5 Hz), 7.95 (d, 2H,

³J=8.3 Hz), 7.73 (d, 4H, ³J=8.1 Hz), 7.46 (d, 4H, ³J=8.2 Hz),

7.27 (d, 2H, ³J=8.3 Hz), 7.19 (d, 2H, ⁴J=2.1 Hz), 7.15 (dd, 2H,

³J=8.2 Hz, ⁴J=2.1 Hz), 4.36 (t, 2H, ³J=6.9 Hz), 4.12 (t, 2H,

³J = 6.5 Hz), 2.29 (s, 6H), 1.74-1.90 (m, 4H), 1.42-1.55 (m,

4H), 1.28-1.42 (m, 8H). ESI-MS: m/z 853.4 ([M þ H]^þ). Elem

C₄₆H₄₄N₂O₆0.40H₂O: C, 75.89; H, 6.20; N, 3.85. Found: C,

3

75.91; H, 6.28; N, 3.60.

Preparation of Compound 6. A solution of 28 (0.29 g, 0.86

mmol) and 2 (0.36 g, 1.72 mmol) in 50 mL of freshly distilled

CH₂Cl₂was cooled in an ice bath. DCC (0.82 g, 3.98 mmol),

DPTS (0.51 g, 1.73 mmol), and a spatula tip of 4-PPy were

added, and the ice bath was removed. After stirring for 24 h at

room temperature, the solution was filtered and evaporated to

dryness. Purification by column chromatography (silica; CH₂Cl₂/

CH₃OH =100/0 f 99.4/0.6) gave 0.52 g (84%) of 6.

¹H NMR (CDCl₃): δ 8.61 (t, 1H, ⁴J = 1.5 Hz), 7.98 (d, 2H,

⁴J=1.5 Hz), 7.73 (d, 4H, ³J=8.1 Hz), 7.46 (d, 4H, ³J=8.2 Hz),

7.27 (d, 2H, ³J=8.5 Hz), 7.19 (d, 2H, ⁴J=2.3 Hz), 7.16 (dd, 2H,

³J=8.3 Hz, ⁴J=2.4 Hz), 4.13 (t, 2H, ³J=6.5 Hz), 3.64 (t, 2H,

³J = 6.7 Hz), 2.29 (s, 6H), 1.75-1.90 (m, 4H), 1.44-1.61 (m,

4H), 1.27-1.43 (m, 8H). ESI-MS: m/z 721.7 ([M þ H]^þ). Elem

anal. Calcd for C₅₄H₄₈N₂O₈1.58H₂O: C, 73.59; H, 5.85; N,

3

3.18. Found: C, 73.61; H, 6.06; N, 2.94.

anal. Calcd for C₄₆H₄₄N₂O₆1.05H₂O: C, 74.68; H, 6.28; N,

Preparation of Compound 11. A solution of 4-carboxybenzal-

dehyde (0.17 g, 1.13 mmol) and 7 (0.84 g, 1.12 mmol) in 75 mL of

dry CH₂Cl₂was cooled in an ice bath. To this solution were

added DCC (0.62 g, 3.01 mmol), DPTS (0.33 g, 1.12 mmol), and

a spatula tip of 4-PPy. After being allowed to slowly heat to

room temperature, the reaction mixture was stirred for 19 h and

filtered. Purification by column chromatography (silica; CH₂Cl₂/

CH₃OH=100/0 f 99.5/0.5) gave 0.83 g (84%) of 11.

3

3.79. Found: C, 74.70; H, 6.28; N, 3.39.

Preparation of Compound 7. A suspension of 3 (0.66 g, 2.96

mmol) and 28 (0.50 g, 1.48 mmol) in 50 mL of freshly distilled

CH₂Cl₂was cooled in an ice bath. After the addition of DCC

(1.50 g, 7.28 mmol), DPTS (0.87 g, 2.96 mmol), and a spatula tip

of 4-PPy, the reaction mixture was allowed to warm slowly and

was stirred for 16 h at room temperature. The filtered solution

was evaporated to dryness and purified by column chromatog-

raphy (silica; CH₂Cl₂/CH₃OH = 100/0 f 99/1) to yield 1.00 g

(90%) of 7.

¹H NMR (CDCl₃): δ 10.10 (s, 1H), 8.62 (t, 1H, ⁴J=1.5 Hz),

8.20 (d, 2H, ³J=8.3 Hz), 7.98 (d, 2H, ⁴J=1.5 Hz), 7.95 (d, 2H,

³J=8.5 Hz), 7.59 (d, 2H, ⁴J=1.7 Hz), 7.55 (dd, 2H, ³J=7.9 Hz,

⁴J=1.7 Hz), 7.25 (d, 2H, ³J=7.9 Hz), 7.19 (m, 2H), 7.11-7.17

(m, 4H), 4.36 (t, 2H, ³J=6.7 Hz), 4.12 (t, 2H, ³J=6.7 Hz), 2.13

(s, 6H), 2.07 (s, 6H), 1.75-1.92 (m, 4H), 1.32-1.55 (m, 12H).

ESI-MS: m/z 898.3 ([M þ NH₄]^þ). Elem anal. Calcd for

¹H NMR (CDCl₃): δ 8.62 (t, 1H, ⁴J = 1.5 Hz), 7.98 (d, 2H,

⁴J=1.5 Hz), 7.59 (s, 2H), 7.55 (d, 2H, ³J=7.9 Hz), 7.25 (d, 2H,

³J=8.1 Hz), 7.19 (m, 2H), 7.11-7.17 (m, 4H), 4.12 (t, 2H, ³J=

3

6.5 Hz), 3.64 (t, 2H, J = 6.7 Hz), 2.13 (s, 6H), 2.07 (s, 6H),

C₅₆H₅₂N₂O₈1.10H₂O: C, 74.67; H, 6.06; N, 3.11. Found: C,

1.81-1.92 (m, 2H), 1.46-1.55 (m, 4H), 1.30-1.42 (m, 10H).

3

ESI-MS: m/z 749.5 ([M þ H]^þ).Elem anal. Calcd for C₄₈H₄₈-

74.66; H, 5.91; N, 3.05.

N₂O₆2.07H₂O: C, 73.32; H, 6.69; N, 3.56. Found: C, 73.31; H,

3

Preparation of Compound 12. To a solution of 8 (2.17 g, 2.63

mmol) in 50 mL of THF and 50 mL of H₂O was added NaClO₂

(1.36 g, 15.0 mmol) and H₂NSO₃H (1.46 g, 15.0 mmol). After

stirring for 3 h, THF was evaporated and the aqueous phase was

extracted with 4 ꢁ 50 mL of CH₂Cl₂. The combined extracts

were dried over Na₂SO₄, filtered, and evaporated to dryness to

yield 2.16 g (98%) of 12.

6.39; N, 3.43.

Preparation of Compound 8. 4-Carboxybenzaldehyde (0.49 g,

3.26 mmol) and 4 (2.28 g, 3.29 mmol) in 200 mL of dry CH₂Cl₂

were stirred in an ice bath. DCC (1.65 g, 8.01 mmol), DPTS (0.96

g, 3.27 mmol), and a spatula tip of 4-PPy were added. The

reaction mixture was allowed to warm slowly to room tempera-

ture, stirred for 19 h, and filtered. Purification by column chro-

matography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99.5/0.5) gave

2.17 g (81%) of 8.

¹H NMR (CDCl₃): δ 8.62 (t, 1H, ⁴J = 1.5 Hz), 8.16 (d, 2H,

³J=8.2 Hz), 8.12 (d, 2H, ³J=8.3 Hz), 7.99 (d, 2H, ⁴J=1.5 Hz),

7.75 (d, 4H, ³J=8.4 Hz), 7.70 (d, 4H, ³J=8.4 Hz), 7.67 (d, 4H,

³J=8.7 Hz), 7.37 (d, 4H, ³J=8.8 Hz), 4.35 (t, 2H, ³J=6.6 Hz),

¹H NMR (CDCl₃): δ 10.10 (s, 1H), 8.63 (t, 1H, ⁴J=1.5 Hz),

8.20 (d, 2H, ³J=8.5 Hz), 7.99 (d, 2H, ⁴J=1.5 Hz), 7.95 (d, 2H,

³J=8.4 Hz), 7.75 (d, 4H, ³J=8.4 Hz), 7.70 (d, 4H, ³J=8.5 Hz),

7.67 (d, 4H, ³J=8.8 Hz), 7.37 (d, 4H, ³J=8.6 Hz), 4.35 (t, 2H,

3

4.13 (t, 2H, J = 6.5 Hz), 1.75-1.90 (m, 4H), 1.30-1.55 (m,

12H). ESI-MS: m/z 839.7 ([M - H]^-). Elem anal. Calcd for

C₅₂H₄₄N₂O₉1.27H₂O: C, 72.31; H, 5.43; N, 3.24. Found: C,

72.32; H, 5.46; N, 3.09.

3

³J = 6.8 Hz), 4.13 (t, 2H, J = 6.6 Hz), 1.75-1.90 (m, 4H),

1.32-1.90 (m, 12H). ESI-MS: m/z 825.7 ([M þ H]^þ). Elem anal.

Calcd for C₅₂H₄₄N₂O₈: C, 75.71; H, 5.38; N, 3.40. Found: C,

75.53; H, 5.42; N, 3.34.

Preparation of Compound 13. A solution of 9 (1.01 g, 1.18

mmol), NaClO₂(0.90 g, 10.0 mmol), and H₂NSO₃H (0.97 g, 10.0

mmol) in 50 mL of THF and 50 mL of H₂O was stirred for 3 h.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8617

THF was evaporated, and the aqueous phase was extracted

with 3 ꢁ 50 mL of CH₂Cl₂. The combined extracts were dried

over Na₂SO₄, filtered, and evaporated to dryness to yield 1.00 g

(98%) of 13.

¹H NMR (CDCl₃): δ 8.66 (t, 2H, ⁴J = 1.4 Hz), 8.61 (d, 2H,

³J=8.4 Hz), 8.32 (d, 4H, ³J=8.5 Hz), 8.26 (t, 1H, ³J=8.4 Hz),

8.20 (d, 4H, ³J=8.4 Hz), 8.02 (d, 4H, ⁴J=1.5 Hz), 7.89 (d, 2H,

⁴J=1.8 Hz), 7.74 (d, 8H, ³J=8.5 Hz), 7.69 (d, 8H, ³J=8.5 Hz),

7.62 (d, 2H, ³J=9.1 Hz), 7.52 (d, 4H, ⁴J=2.0 Hz), 7.49 (dd, 4H,

³J=8.3 Hz, ⁴J=2.0 Hz), 7.38 (dd, 2H, ³J=9.0 Hz, ⁴J=1.9 Hz),

7.29 (d, 4H, ³J=8.4 Hz), 4.85 (q, 4H, ³J=7.0 Hz), 4.38 (t, 4H,

³J=6.7 Hz), 4.15 (t, 4H, ³J=6.4 Hz), 2.34 (s, 12H), 1.77-1.92

(m, 8H), 1.48(t, 6H, ³J=7.2 Hz), 1.30-1.58(m, 24H). ESI-MS:m/z

¹H NMR (CDCl₃): δ 8.65 (t, 1H, ⁴J = 1.4 Hz), 8.15 (d, 2H,

³J=8.4 Hz), 8.11 (d, 2H, ³J=8.4 Hz), 8.01 (d, 2H, ⁴J=1.4 Hz),

7.73 (d, 4H, ³J=8.3 Hz), 7.68 (d, 4H, ³J=8.4 Hz), 7.51 (d, 2H,

⁴J=2.2 Hz), 7.49 (dd, 2H, ³J=8.3 Hz, ⁴J=2.2 Hz), 7.27 (d, 2H,

³J=8.2 Hz), 4.35 (t, 2H, ³J=6.7 Hz), 4.13 (t, 2H, ³J=6.3 Hz),

2.33 (s, 6H), 1.74-1.90 (m, 4H), 1.30-1.56 (m, 12H). ESI-MS: m/z

2101.2 ([M þ H]^þ). Elem anal. Calcd for C₁₃₁H₁₁₃N₉O₁₈1.07H₂O:

3

867.7 ([M - H]^-). Elem anal. Calcd for C₅₄H₄₈N₂O₉1.97H₂O: C,

C, 74.19; H, 5.47; N, 5.94. Found: C, 74.21; H, 5.49; N, 5.75.

Preparation of Ligand L9. A solution of 14 (0.30 g, 0.34 mmol),

2,6-bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine (0.069 g, 0.17

3

71.72; H, 5.79; N, 3.10. Found: C, 71.73; H, 5.91; N, 2.97.

Preparation of Compound 14. To a solution of 10 (0.32 g, 0.37

mmol) in 25 mL of THF and 25 mL of H₂O was added NaClO₂

(0.18 g, 2.0 mmol) and H₂NSO₃H (0.19 g, 2.0 mmol). After

stirring for 2 h at room temperature, THF was evaporated and

the aqueous solution extracted with CH₂Cl₂(3 ꢁ 25 mL). Eva-

poration yielded 0.30 g (92%) of 14.

mmol), EDCI HCl (0.19 g, 1.0 mmol), and DMAP (0.12 g, 1.0

3

mmol) in 50 mL of dry CH₂Cl₂was refluxed under an atmosphere

of N₂for 14 h. After washing with a saturated aqueous NaCl

solution (3 ꢁ 50 mL), the organic phase was dried over Na₂SO₄,

filtered, and evaporated to dryness. Purification by column chro-

matography (silica; CH₂Cl₂/CH₃OH=100/0 f 98.5/1.5) gave 0.07

g (20%) of L9.

¹H NMR (CDCl₃): δ 8.61 (t, 1H, ⁴J=1.5 Hz), 8.11-8.17 (m,

4H), 7.98 (d, 2H, ⁴J = 1.5 Hz), 7.73 (d, 4H, ³J = 8.2 Hz), 7.46

(d, 4H, ³J=8.3 Hz), 7.27 (d, 2H, ³J=8.4 Hz), 7.19 (d, 2H, ⁴J=

2.2 Hz), 7.15 (dd, 2H, ³J=8.4 Hz, ⁴J=2.3 Hz), 4.35 (t, 2H, ³J=

¹H NMR (CDCl₃): δ 8.61 (t, 2H, ⁴J = 1.6 Hz), 8.45 (d, 2H,

³J=8.3 Hz), 8.32 (d, 4H, ³J=8.6 Hz), 8.20 (d, 4H, ³J=8.5 Hz),

8.14 (t, 1H, ³J=8.1 Hz), 7.98 (d, 4H, ⁴J=1.5 Hz), 7.76 (d, 2H,

⁴J=1.6 Hz), 7.72 (d, 8H, ³J=8.4 Hz), 7.55 (d, 2H, ³J=8.8 Hz),

7.45 (d, 8H, ³J=8.5 Hz), 7.29 (dd, 2H, ³J=8.9 Hz, ⁴J=1.9 Hz),

7.27 (d, 4H, ³J=8.5 Hz), 7.19 (d, 4H, ⁴J=2.4 Hz), 7.15 (dd, 4H,

³J=8.4 Hz, ⁴J=2.5 Hz), 4.83 (q, 4H, ³J=6.9 Hz), 4.38 (t, 4H,

³J=6.9 Hz), 4.13 (t, 4H, ³J=6.6 Hz), 2.29 (s, 12H), 1.75-1.92

(m, 8H), 1.42-1.57 (m, 8H), 1.43 (t, 6H, ³J=7.1 Hz), 1.31-1.42

(m, 8H). ESI-MS: m/z 2101.1 ([M þ H]^þ). Elem anal. Calcd for

3

6.7 Hz), 4.12 (t, 2H, J = 6.7 Hz), 2.29 (s, 6H), 1.75-1.89 (m,

4H), 1.42-1.54 (m, 4H), 1.30-1.42 (m, 8H). ESI-MS: m/z 867.7

([M - H]^-). Elem anal. Calcd for C₅₄H₄₈N₂O₉2.58H₂O: C,

3

70.84; H, 5.85; N, 3.06. Found: C, 70.86; H, 6.02; N, 2.95.

Preparation of Compound 15. A solution of 11 (0.83 g, 0.94

mmol), NaClO₂(0.45 g, 5.0 mmol), and H₂NSO₃H (0.49 g, 5.0

mmol) in 50 mL of THF and 50 mL of H₂O was stirred for 4 h.

THF was evaporated, and the aqueous phase was extracted with

3 ꢁ 150 mL of CH₂Cl₂. The combined organic extracts were

dried over Na₂SO₄, filtered, and evaporated to dryness. Yield:

0.79 g (93%) of 15.

C

₁₃₁H₁₁₃N₉O₁₈3.73H₂O: C, 72.56; H, 5.60; N, 5.81. Found: C,

3

72.57; H, 5.58; N, 5.62.

Preparation of Ligand L10. A solution of 2,6-bis(1-ethyl-5-

hydroxybenzimidazol-2-yl)pyridine (0.176 g, 0.44 mmol) and 15

(0.79 g, 0.88 mmol) in 50 mL of freshly distilled CH₂Cl₂was

cooled in an ice bath. After the addition of DCC (0.41 g, 1.99

mmol), DPTS (0.26 g, 0.88 mmol), and a spatula tip of 4-PPy,

the ice bath was allowed to melt and the reaction mixture was

stirred at room temperature for 16 h. The solution was filtered

and evaporated to dryness. The crude product was cleaned by

column chromatography (silica; CH₂Cl₂/CH₃OH=100/0 f 99/1)

to yield 0.18 g (19%) of L10.

¹H NMR (CDCl₃): δ 8.61 (t, 1H, ⁴J = 1.5 Hz), 8.16 (d, 2H,

³J=8.3 Hz), 8.13 (d, 2H, ³J=8.4 Hz), 7.98 (d, 2H, ⁴J=1.4 Hz),

7.59 (d, 2H, ⁴J=1.5 Hz), 7.55 (dd, 2H, ³J=7.7 Hz, ⁴J=1.4 Hz),

7.25 (d, 2H, ³J=7.8 Hz), 7.19 (m, 2H), 7.11-7.17 (m, 4H), 4.35

(t, 2H, ³J=6.9 Hz), 4.12 (t, 2H, ³J=6.5 Hz), 2.13 (s, 6H), 2.07 (s,

6H), 1.77-1.84 (m, 4H), 1.29-1.53 (m, 12H). ESI-MS: m/z

895.1 ([M - H]^-). Elem anal. Calcd for C₅₆H₅₂N₂O₉1.07H₂O:

3

C, 73.40; H, 5.96; N, 3.06. Found: C, 73.43; H, 6.27; N, 2.84.

Preparation of Ligand L7. A solution of 2,6-bis(1-ethyl-5-

hydroxybenzimidazol-2-yl)pyridine (0.20 g, 0.50 mmol), 12

¹H NMR (CDCl₃): δ 8.62 (t, 2H, ⁴J = 1.5 Hz), 8.36 (d, 2H,

³J=7.8 Hz), 8.33 (d, 4H, ³J=8.6 Hz), 8.20 (d, 4H, ³J=8.5 Hz),

8.08 (t, 1H, ³J=7.8 Hz), 7.98 (d, 4H, ⁴J=1.4 Hz), 7.70 (d, 2H,

⁴J=2.1 Hz), 7.59 (d, 4H, ⁴J=1.4 Hz), 7.54 (dd, 4H, ³J=7.6 Hz,

⁴J=1.4 Hz), 7.27 (dd, 2H, ³J=8.9 Hz, ⁴J=1.9 Hz), 7.25 (d, 4H,

³J=7.8 Hz), 7.19 (m, 4H), 7.11-7.17 (m, 8H), 4.82 (q, 4H, ³J=

7.3 Hz), 4.38 (t, 4H, ³J=6.7 Hz), 4.13 (t, 4H, ³J=6.5 Hz), 2.13

(s, 12H), 2.07 (s, 12H), 1.77-1.92 (m, 8H), 1.41 (t, 6H, ³J=7.2

Hz), 1.31-1.57 (m, 24H). ESI-MS: m/z 2157.0 ([M þ H]^þ). Elem

(0.84 g, 1.0 mmol), EDCI HCl (0.60 g, 3.13 mmol), and DMAP

3

(0.38 g, 3.11 mmol) in 50 mL of freshly distilled CH₂Cl₂was

refluxed under N₂for 16 h. After washing with 3 ꢁ 50 mL of a

saturated aqueous NaCl solution, the organic phase was dried

over Na₂SO₄, filtered, and evaporated. Purification by column

chromatography (silica; CH₂Cl₂/CH₃OH = 100/0 f 98.5/1.5)

gave 0.45 g (44%) of L7.

¹H NMR (CDCl₃): δ 8.62 (t, 2H, ⁴J = 1.4 Hz), 8.42 (d, 2H,

³J=8.5 Hz), 8.32 (d, 4H, ³J=8.5 Hz), 8.19 (d, 4H, ³J=8.5 Hz),

8.12 (d, 2H, ³J=8.5 Hz), 7.99 (d, 4H, ⁴J=1.5 Hz), 7.75 (d, 8H,

³J=8.3 Hz), 7.75 (d, 2H, ³J=8.3 Hz), 7.69 (d, 8H, ³J=8.5 Hz),

7.67 (d, 8H, ³J=8.8 Hz), 7.54 (d, 2H, ³J=8.8 Hz), 7.37 (d, 8H,

³J=8.7 Hz), 7.27 (d, 2H, ³J=8.8 Hz), 4.38 (t, 4H, ³J=6.6 Hz),

anal. Calcd for C₁₃₅H₁₂₁N₉O₁₈2.05H₂O: C, 73.89; H, 5.75; N,

3

5.74. Found: C, 73.92; H, 6.01; N, 5.57.

Preparation of Compound 29. To a suspension of 5-(tert-

butyldimethylsiloxy)isophthalic acid (0.70 g, 2.35 mmol) in 50 mL

of freshly distilled CH₂Cl₂stirred in an ice bath was added first a

solution of DCC (3.00 g, 14.5 mmol), DPTS (0.30 g, 1.02 mmol),

and a spatula tip of 4-PPy in 25 mL of CH₂Cl₂and then a

solution of 17 (2.28 g, 4.70 mmol) in 25 mL of CH₂Cl₂. After

stirring for 30 min, the ice bath was removed and the solution

was stirred for 17 h. After filtration, the solution was evaporated

to dryness. The crude product was purified by column chroma-

tography (silica; CH₂Cl₂/CH₃OH = 100/0 f 99.5/0.5) to yield

2.21 g (79%) of 29.

3

4.14 (t, 4H, J = 6.5 Hz), 1.77-1.92 (m, 8H), 1.32-1.55 (m,

24H). ESI-MS: m/z 2044.9 ([M þ H]^þ). Elem anal. Calcd for

C₁₂₇H₁₀₅N₉O₁₈2.72H₂O: C, 72.84; H, 5.32; N, 6.02. Found: C,

72.84; H, 5.18; N, 5.83.

3

Preparation of Ligand L8. A solution of 13 (0.52 g, 0.60

mmol), 2,6-bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine

(0.12 g, 0.30 mmol), EDCI HCl (0.38 g, 2.0 mmol), and DMAP

3

(0.24 g, 2 mmol) in 50 mL of dry CH₂Cl₂was refluxed under N₂

for 18 h. The solution was washed with 3 ꢁ 50 mL of a saturated

aqueous NaCl solution, and the organic phase was dried over

Na₂SO₄, filtered, and evaporated. The crude product was purified

by column chromatography (silica; CH₂Cl₂/CH₃OH=100/0 f

98.5/1.5) to yield 0.19 g (30%) of L8.

¹H NMR (CDCl₃): δ 8.28 (t, 1H, ⁴J = 1.5 Hz), 8.17 (d, 4H,

³J=9.0 Hz), 7.72 (d, 4H, ³J=8.7 Hz), 7.67 (d, 4H, ³J=8.9 Hz),

7.67 (d, 2H, ⁴J=1.5 Hz), 7.48 (d, 2H, ⁴J=1.8 Hz), 7.46 (dd, 2H,

³J=8.2 Hz, ⁴J=2.3 Hz), 7.24 (d, 2H, ³J=8.2 Hz), 6.99 (d, 4H,

³J=8.9 Hz), 4.33 (t, 4H, ³J=6.8 Hz), 4.05 (t, 4H, ³J=6.6 Hz),

8618 Inorganic Chemistry, Vol. 49, No. 18, 2010

Jensen et al.

2.30 (s, 6H), 1.74-1.87 (m, 8 H), 1.30-1.53 (m, 24 H), 1.00 (s,

9H), 0.24 (s, 6H). ESI-MS: m/z 1253.4 ([M þ Na]^þ). Elem anal.

4H, ³J=8.6 Hz), 7.48 (d, 2H, ⁴J=2.2 Hz), 7.46 (dd, 2H, ³J=8.3

Hz, ⁴J=2.2 Hz), 7.24 (d, 2H, ³J=8.3 Hz), 6.98 (d, 4H, ³J=9.0

Hz), 6.98 (d, 2H, ³J=9.0 Hz), 4.35 (t, 6H, ³J=6.7 Hz), 4.04 (t,

6H, ³J=6.6 Hz), 2.30 (s, 6H), 1.74-1.89 (m, 12 H), 1.29-1.53

(m, 36 H). ESI-MS: m/z 1540.3 ([M - H]^-). Elem anal. Calcd for

Calcd for C₇₆H₈₆N₂O₁₁Si 0.33H₂O: C, 73.75; H, 7.06; N, 2.27.

3

Found: C, 73.75; H, 7.08; N, 2.33.

Preparation of Compound 31. A solution of 29 (2.24 g, 1.82

mmol) and Zn(BF₄)₂6H₂O (8.00 g, 23.0 mmol) in 160 mL of

C₉₅H₁₀₀N₂O₁₇1.92H₂O: C, 72.38; H, 6.64; N, 1.78. Found: C,

3

72.38; H, 6.75; N, 1.70.

THF and 40 mL of H₂O was stirred at 60 ꢀC for 15 h. THF was

evaporated, and the aqueous phase was extracted with 4 ꢁ 40

mL of CH₂Cl₂. The combined organic phases were dried over

Na₂SO₄, filtered, and evaporated. The solid residue was purified

by column chromatography (silica; CH₂Cl₂/CH₃OH=100/0 f

99/1) to yield 2.01 g (99%) of 31.

Preparation of Ligand L6. A solution of 37 (1.46 g, 0.95

mmol), 2,6-bis(1-ethyl-5-hydroxybenzimidazol-2-yl)pyridine

(189 mg, 0.47 mmol), EDCI HCl (400 mg, 2.1 mmol), and

3

DMAP (250 mg, 2.05 mmol) in 100 mL of freshly distilled

CH₂Cl₂was refluxed under N₂for 17 h. The solution was

extracted three times with 80 mL of H₂O þ 10 mL of saturated

NH₄Cl. The combined aqueous phases were extracted with 50 mL

of CH₂Cl₂, and the organic extracts were dried over Na₂SO₄,

filtered, and evaporated to dryness. The crude product was

purified by column chromatography (silica; CH₂Cl₂/CH₃OH =

100/0 f 99/1). Yield: 830 mg (51%).

¹H NMR (CDCl₃): δ 8.26 (t, 1H, ⁴J = 1.4 Hz), 8.17 (d, 4H,

³J=9.0 Hz), 7.72 (d, 4H, ³J=8.6 Hz), 7.70 (d, 2H, ⁴J=1.4 Hz),

7.67 (d, 4H, ³J=8.6 Hz), 7.48 (d, 2H, ⁴J=2.2 Hz), 7.45 (dd, 2H,

³J=8.2 Hz, ⁴J=2.3 Hz), 7.24 (d, 2H, ³J=8.3 Hz), 6.99 (d, 4H,

³J=9.0 Hz), 5.61 (s, 1H), 4.33 (t, 4H, ³J=6.7 Hz), 4.05 (t, 4H,

³J=6.6Hz), 2.30 (s, 6H), 1.73-1.87 (m, 8 H), 1.30-1.53 (m, 24 H).

ESI-MS: m/z 1116.1 ([M - H]^-). Elem anal. Calcd for C₇₀H₇₂-

¹H NMR (CDCl₃): δ 8.59 (t, 2H, ⁴J = 1.5 Hz), 8.35 (d, 2H,

³J=8.1 Hz), 8.33 (d, 4H, ³J=8.6 Hz), 8.20 (d, 4H, ³J=8.5 Hz),

8.16 (d, 8H, ³J=8.9 Hz), 8.14 (d, 4H, ³J=8.8 Hz), 8.08 (t, 1H,

³J=7.9 Hz), 8.05 (d, 4H, ⁴J=1.5 Hz), 7.72 (d, 8H, ³J=8.4 Hz),

7.71 (d, 2H, ⁴J=2.3 Hz), 7.67 (d, 8H, ³J=8.4 Hz), 7.52 (d, 2H,

³J=9.0 Hz), 7.48 (d, 4H, ⁴J=2.1 Hz), 7.46 (dd, 4H, ³J=8.3 Hz,

⁴J=2.2 Hz), 7.25 (dd, 2H, ³J=9.0 Hz, ⁴J=2.3 Hz), 7.24 (d, 4H,

³J=8.3 Hz), 6.98 (d, 12H, ³J=9.0 Hz), 4.82 (q, 4H, ³J=7.2 Hz),

4.38 (t, 4H, ³J =6.7 Hz), 4.35 (t, 8H, ³J=6.8 Hz), 4.05 (t, 4H,

³J=6.6 Hz), 4.04 (t, 8H, ³J=6.6 Hz), 2.30 (s, 12H), 1.74-1.87

(m, 24 H), 1.30-1.52 (m, 78 H). ESI-MS: m/z 1724.3 ([M þ

N₂O₁₁0.18H₂O: C, 75.02; H, 6.51; N, 2.51. Found: C, 75.02; H,

3

6.58; N, 2.43.

Preparation of Compound 33. To a solution of 4-(10-hydroxy-

decyloxy)benzoic acid (0.53 g, 1.79 mmol) in 100 mL of freshly

distilled CH₂Cl₂stirred in an ice bath was added 31 (2.00 g, 1.79

mmol), DPTS (0.53 g, 1.79 mmol), DCC (0.83 g, 4.00 mmol),

and a spatula tip of 4-PPy. The ice bath was removed, and the

solution was stirred at room temperature for 72 h. After filtra-

tion, the solution was evaporated to dryness and the crude pro-

duct was purified by column chromatography (silica; CH₂Cl₂/

CH₃OH =100/0 f 99/1) to yield 2.44 g (98%) of 33.

2H]^2þ). Elem anal. Calcd for C₂₁₃H₂₁₇N₉O₃₄0.22H₂O: C,

3

74.13; H, 6.35; N, 3.65. Found: C, 74.14; H, 6.41; N, 3.53.

Preparation of [Ln(Lk)(NO₃)₃] nH₂O Complexes. In a typical

¹H NMR (CDCl₃): δ 8.60 (t, 1H, ⁴J = 1.4 Hz), 8.17 (d, 4H,

³J=9.0 Hz), 8.14 (d, 2H, ³J=9.0 Hz), 8.06 (d, 2H, ⁴J=1.4 Hz),

7.72 (d, 4H, ³J=9.0 Hz), 7.67 (d, 4H, ³J=9.0 Hz), 7.48 (d, 2H,

⁴J=2.3 Hz), 7.45 (dd, 2H, ³J=8.7 Hz, ⁴J=2.3 Hz), 7.24 (d, 2H,

³J=8.4 Hz), 6.98 (d, 6H, ³J=8.7 Hz), 4.35 (t, 4H, ³J=6.6 Hz),

4.04 (t, 6H, ³J=6.6 Hz), 3.64 (t, 2H, ³J=6.5 Hz), 2.30 (s, 6H),

1.74-1.86 (m, 10 H), 1.54-1.61 (m, 2 H), 1.29-1.52 (m, 36 H).

ESI-MS: m/z 1411.1 ([M þ H₂O]^þ). Elem anal. Calcd for

3

preparation, 100-150 mg of ligand (Lk) was dissolved in 5 mL

of CH₂Cl₂. A total of 1 equiv of Ln(NO₃)₃nH₂O (Ln=Y, La,

3

Pr, Eu, Gd, Tb, Lu) in 5 mL of CH₃CN was added, and the

solution was stirred for 1 h. After evaporation of the solvent, the

complex was washed with CH₃CN and dried in a vacuum

overnight at 60 ꢀC. Yields ranged from 72 to 89%. Elemental

analyses were collected in Table S1 (Supporting Information)

C₈₇H₉₆N₂O₁₄0.64H₂O: C, 74.36; H, 6.98; N, 1.99. Found: C,

3

1

Spectroscopic and Analytical Measurements. H NMR spec-

74.36; H, 7.14; N, 2.20.

tra were recorded at 25 ꢀC on a Bruker Avance 400 MHz spec-

trometer. Chemical shifts are given in ppm with respect to tetra-

methylsilane. Diffusion experiments were carried out at 400 MHz

Preparation of Compound 35. To a solution of 4-carboxyben-

zaldehyde (74 mg, 0.49 mmol) and 33 (670 mg, 0.49 mmol) in 100

mL of freshly distilled CH₂Cl₂stirred in an ice bath was added

DPTS (30 mg, 0.10 mmol), DCC (300 mg, 1.45 mmol), and a

spatula tip of 4-PPy. After 30 min, the ice bath was removed and

the solution was stirred for 24 h at room temperature. The

solution was filtered and evaporated, and the crude product was

purified by column chromatography (silica; CH₂Cl₂/CH₃OH=

100/0 f 99.5/0.5) to yield 700 mg (96%) of 35.

Larmor frequency. Solutions (CD₂Cl₂, 293 K, |complex|_tot

=

10^-2M) of the complex were prepared in situ and left to equili-

brate for 48 h. The pulse sequence used was the Bruker pulse

program ledbpgp2s,³⁸which employs stimulated echo, bipolar

gradients, and longitudinal eddy current delay as the z filter. The

four 2 ms gradient pulses have sine-bell shapes and amplitudes

ranging linearly from 2.5 to 50 G cm^-1in 32 steps. The diffusion

¹H NMR (CDCl₃): δ 10.10 (s, 1H) 8.59 (t, 1H, ⁴J =1.5 Hz),

8.20 (d, 2H, ³J=8.4 Hz), 8.17 (d, 4H, ³J=8.9 Hz), 8.14 (d, 2H,

³J=9.0 Hz), 8.06 (d, 2H, ⁴J=1.6 Hz), 7.95 (d, 2H, ³J=8.3 Hz),

7.72 (d, 4H, ³J=8.6 Hz), 7.67 (d, 4H, ³J=8.5 Hz), 7.48 (d, 2H,

⁴J=2.3 Hz), 7.46 (dd, 2H, ³J=8.4 Hz, ⁴J=2.3 Hz), 7.24 (d, 2H,

³J=8.4 Hz), 6.99 (d, 4H, ³J=8.9 Hz), 6.98 (d, 2H, ³J=9.0 Hz),

4.36 (t, 2H, ³J =6.7 Hz), 4.35 (t, 4H, ³J = 6.7 Hz), 4.05 (t, 6H,

³J = 6.5 Hz), 2.30 (s, 6H), 1.74-1.87 (m, 12 H), 1.30-1.52 (m,

36 H). ESI-MS: m/z 1543.1 ([M þ H₂O]^þ). Elem anal. Calcd for

3

delay was in the range 60-140 ms depending on the analyte

diffusion coefficient, and the number of scans was 32. Processing

was done using a line broadening of 5 Hz, and the diffusion

coefficients were calculated with the Bruker processing package.

VT-NMR measurements of samples (5-20 mg; c ≈ 10^-2M) in a

CD₂Cl₂solution were measured on a Bruker Avance 400 spec-

trometer equipped with a variable-temperature unit. The integ-

rated intensities of the relevant peaks were obtained by decon-

volution using Matlab or Excel (one Lorentz function per peak)

after Fourier transformation and phasing of the spectrum using

SpinWorks. Fitting of van’t Hoff plots was done using Excel.

The standard deviations of ΔG, K, and T_50%were calculated by

taking into account the strong interdependence of the values of

ΔH and ΔS, derived from the slope and intercept, respectively,

of the van’t Hoff plots. The covariance matrix of ΔH and ΔS

was calculated with the macro LS1, and the standard deviations

C₉₅H₁₀₀N₂O₁₆0.91H₂O: C, 73.98; H, 6.65; N, 1.82. Found: C,

3

73.98; H, 6.60; N, 1.86.

Preparation of Compound 37. A solution of 35 (0.64 g, 0.42

mmol), NaClO₂(0.20 g, 2.2 mmol), and H₂NSO₃H (0.21 g, 2.3

mmol) in 25 mL of THF and 25 mL of H₂O was stirred at room

temperature for 2 h. After evaporation of THF, the aqueous

phase was extracted with 2 ꢁ 50 mL of CH₂Cl₂. The combined

organic phases were dried over Na₂SO₄, filtered, and evaporated

to yield 0.64 g (99%) of 37.

¹H NMR (CDCl₃): δ 8.59 (t, 1H, ⁴J=1.5 Hz), 8.11-8.19 (m,

10H), 8.06 (d, 2H, ⁴J=1.5 Hz), 7.72 (d, 4H, ³J=8.6 Hz), 7.67 (d,

(38) Wu, D.; Chen, A.; Johnson, C. S., Jr. J. Magn. Reson. A 1995, 115,

260–264.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8619

of the other three parameters were calculated using the macro

Propagation.²¹Pneumatically assisted ESI-MS spectra were

recorded from 10^-4M solutions on an Applied Biosystems

API 150EX LC/MS system equipped with a Turbo Ionspray

source. Elemental analyses were performed by Dr. H. Eder from

the Microchemical Laboratory of the University of Geneva.

TGA was performed with a Seiko TG/DTA 320 thermogravi-

metric balance(under N₂). DSC traces were obtained with a

Mettler Toledo DSC1 Star Systems differential scanning calori-

The cell parameters were calculated from the position of the

reflection at the smallest Bragg angle, which was in all cases the

˚

most intense. Periodicities up to 90 A could be measured, and

the sample temperature was controlled within (0.3 ꢀC. The

exposure times were varied from 1 to 24 h, depending on the

specific reflections being sought (weaker reflections obviously

taking longer exposure times). (4) A SAXS system from Mole-

cular Metrology equipped with a Cu KR Bede microsource

conditioned with confocal Max-FluxTM optics and a two-

dimensional multiwire gas detector. A modified temperature

stage from Linkham was used to control the temperatures. The

analysis was performed with MatLab-based open-source soft-

ware from Molecular Metrology.

meter from 3 to 5 mg samples (5 ꢀC min^-1, under N₂). Charac-

3

terization of the mesophases was performed with a Leitz Orthoplan-

Pol polarizing microscope with a Leitz LL 20ꢁ/0.40 polarizing

objective and equipped with a Linkam THMS 600 variable-

temperature stage. The SA-XRD patterns were obtained with

four different experimental setups, and in all cases, the crude

powder was filled in Lindemann capillaries of 1 mm diameter.

(1) A STOE STADI P transmission powder diffractometer

system using a focused monochromatic Cu KR₁beam obtained

from a curved germanium monochromator (Johann-type) and

collected on a curved image-plate position sensitive detector.

A calibration with silicon and copper laurate standards, for

high- and low-angle domains, respectively, was preliminarily

performed. Sample capillaries were placed in the high-tempera-

ture attachment for measurements in the range of desired

temperatures (from -40 to þ170 ꢀC) within (0.05 ꢀC. Periodi-

Acknowledgment. The authors acknowledge contributions

of the Sciences Mass Spectrometry platform at the Faculty of

Sciences, University of Geneva, for mass spectrometry services.

Financial support from the Swiss National Science Foundation

is gratefully acknowledged.

Supporting Information Available: Tables of elemental ana-

lyses (Table S1) and SA-XRD data (Tables S2 and S3) for

complexes [Ln(Lk)(NO₃)₃], schemes for the syntheses of ligands

L7-L14 (Schemes S1 and S2), and figures showing van’t Hoff

analyses (Figures S1-S3), TGA and DSC traces (Figures S4, S8,

S9, and S13), birefringent textures (Figures S5 and S10), SA-

XRD profiles (Figures S7, S10, S12, S14, and S15), and molec-

ular modeling of the complexes [Ln(Lk)(NO₃)₃] (Figure S11).

This material is available free of charge via the Internet at http://

pubs.acs.org.

˚

cities up to 50 A could be measured. The exposure times were

varied from 1 to 4 h. (2) An Inel CPS 120 curved counter using a

linear monochromatic Cu KR₁beam obtained with a sealed-

tube generator (900 W) and a bent quartz monochromator, for

which the sample temperature was controlled within (0.05 ꢀC.

˚

Periodicities up to 60 A could be measured. (3) An image plate.

Article Doi

Dimerization of dendrimeric lanthanide complexes: Thermodynamic, thermal, and liquid-crystalline properties

DOI: 10.1021/ic101220v

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic101220v

Full text of DOI:10.1021/ic101220v

Products guided by the article

R&D Labs maybe for 1245791-66-5

Relevant to this article

Hot Product