A 1,8-naphthalenediol-based unsymmetrical dinucleating ligand

Article abstract of DOI:10.1039/b507579c

A facile synthesis of 2-formyl-1,8-naphthalenediol is reported. Its potential as a general precursor for the preparation of unsymmetrical multidentate chelating ligand systems based on 1,8-naphthalenediol is demonstrated by the synthesis of the dinucleating ligand L^4- (H ₄L = N,N′-bis(2-(1,8-naphthalenediol)methylidene) propylenediamine). Reaction of H₄L with copper acetate results in the formation of the unsymmetrical dinuclear Cu^II complex [LCu ₂] (3), which has been structurally characterized by single-crystal X-ray diffraction. One Cu^II ion is coordinated by a N ₂O₂ compartment of L^4- and the other Cu ^II ion is coordinated by an O₄ compartment of L ^4- while they are bridged by two aryloxide functions of L ^4-. A dimerization of two molecules of 3 to a tetranuclear entity 3₂ occurs through formation of weak apical Cu-O interactions. Analysis of the temperature dependent magnetic susceptibility measurements (2-290 K) established a strong intradimer exchange coupling J₁₂ = -371 cm^-1. This strong superexchange interaction fits nicely in a magneto-structural correlation which has been established for dinuclear bis(phenoxide)-bridged Cu^II complexes demonstrating the electronic equivalence of the aryloxides of a phenol and 1,8-naphthalenediol. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b507579c

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
A 1,8-naphthalenediol-based unsymmetrical dinucleating ligand
a
a
b
Thorsten Glaser,* Ioannis Liratzis and Roland Fr o¨ hlich
a
Institut f u¨ r Anorganische und Analytische Chemie, Westf a¨ lische
Wilhelms-Universit a¨ t M u¨ nster, Wilhelm-Klemm-Straße 8, D-48149, M u¨ nster, Germany.
E-mail: tglaser@uni-muenster.de
Organisch-Chemisches Institut, Westf a¨ lische Wilhelms-Universit a¨ t M u¨ nster, Correnstr. 40,
D-48149, M u¨ nster, Germany
b
Received 31st May 2005, Accepted 5th July 2005
First published as an Advance Article on the web 26th July 2005
A facile synthesis of 2-formyl-1,8-naphthalenediol is reported. Its potential as a general precursor for the preparation
of unsymmetrical multidentate chelating ligand systems based on 1,8-naphthalenediol is demonstrated by the
4
−
ꢀ
synthesis of the dinucleating ligand L (H
Reaction of H
4
L = N,N -bis(2-(1,8-naphthalenediol)methylidene)propylenediamine).
II
4
L with copper acetate results in the formation of the unsymmetrical dinuclear Cu complex [LCu
2
] (3),
2
II
which has been structurally characterized by single-crystal X-ray diffraction. One Cu ion is coordinated by a N
2
O
4
−
II
4−
compartment of L and the other Cu ion is coordinated by an O
4
compartment of L while they are bridged by two
aryloxide functions of L . A dimerization of two molecules of 3 to a tetranuclear entity 3 occurs through formation
of weak apical Cu–O interactions. Analysis of the temperature dependent magnetic susceptibility measurements
4
−
2
−
1
(
2–290 K) established a strong intradimer exchange coupling J12 = −371 cm . This strong superexchange interaction
II
fits nicely in a magneto-structural correlation which has been established for dinuclear bis(phenoxide)-bridged Cu
complexes demonstrating the electronic equivalence of the aryloxides of a phenol and 1,8-naphthalenediol.
Introduction
the most famous ligand systems are mono-nucleating salen-
type ligands, which have a long history in synthetic inorganic
Phenolates are ubiquitous as coordinating groups in ligand sys-
tems. This is in contrast to the ‘one ring- and one donor-increased’
7,8
chemistry and a burgeoning significance in bioinorganic chem-
9
istry and homogenous catalysis. However, numerous metallo
1
,8-naphthalenediol unit, which has only been sparingly used
enzymes make use of the cooperative action of two (or even
more) proximate metal ions within their active site, thus enabling
very efficient catalytic transformations of biologically relevant
for coordination to metal ions. Wuest and coworkers used the
dianion of 1,8-naphthalenediol to coordinate Ti . However,
this is the only structurally characterized complex of the parent
IV
1
10
substrate molecules. While salen-type ligands primarily form
1
,8-naphthalenediol. Phenolates act only in a limited number
mononuclear complexes, we thought it might be interesting to
study the cooperative action in dinuclear complexes of extended
salen-type ligands. Accordingly, we tested the potential of
the new precursor II for the synthesis of an extended salen-
type ligand based on the 1,8-naphthalenediol backbone, which
of complexes as monodentate ligands. Usually, the phenolate
is part of a chelating multidentate ligand system. To synthesize
such multidentate ligand systems, suitable functionalized phenol
precursor compounds must be available. In order to create a
multidentate ligand system incorporating 1,8-naphthalenediol,
provides a salen-like N
2
O
2
ligand compartment and an O ligand
4
2
Vidali et al. used 2-acetyl-3,6-dimethyl-1,8-naphthalenediol
compartment. The bridging aryloxides allow for strong coop-
erative interactions between the two metal ions. Thus, besides
the synthesis of the building block 2-formyl-1,8-naphthalenediol
II we report herein the preparation of the unsymmetrical
and Robson et al. 2,7-diacetyl-3,6-dimethyl-1,8-naphthalenediol
3
as precursor. While no structural characterization using the
former precursor has been provided, only mono-Schiff base
products have been obtained by Robson et al. upon conden-
sation with mono-primary amines. These problems might be
attributed to the general observation that ketones react more
slowly than aldehydes with amines to form imines.
In this respect, we described recently the synthesis of
4−
ꢀ
dinucleating ligand L which is the tetraanion of N,N -bis(2-
1,8-naphthalenediol)methylidene)propylenediamine (H L). Its
capability to form unsymmetrical dinuclear complexes is demon-
3
(
4
4
II
strated by the synthesis of its dinuclear Cu complex [LCu ] (3)
2
whose structural and magnetic properties are described.
2
,7-diformyl-1,8-naphthalenediol which is the ‘one ring- and one
5
donor-increased’ derivative of 2,6-diformylphenol. The latter
was introduced in 1970 by Robson as building block for the syn-
thesis of dinucleating ligands (compartmental ligands, Robson
Experimental
6
type ligands). Herein, we report a streamlined synthesis for
All manipulations were performed in an atmosphere of dry
argon by means of standard Schlenk techniques. All reagents
were obtained from commercial suppliers and used without fur-
the previously unknown building block 2-formyl-1,8-naphthal-
enediol II as an extension of salicylaldehyde I (Scheme 1).
ther purification unless otherwise noted. Et
distilled under argon from Na/benzophenone and CaH
tively, prior to use. After addition of benzene and water, DMF
was first fractionally distilled and then distilled from CaH under
2
O and CH
2
Cl
2
were
2
, respec-
2
ꢀ
ꢀ
argon. N,N,N ,N -Tetramethylethylenediamine (TMEDA) was
purified by vacuum distillation from Na/benzophenone. 1,8-
Bis(methoxymethoxy)naphthalene 1 was synthesized as previ-
5
ously described.
Scheme 1
−
1
Infrared spectra (400–4000 cm ) of solid samples were
recorded on a Bruker Vector 22 spectrometer as KBr disks.
Elemental analyses were obtained with a Vario EL III Elemental
Salicylaldehyde I is a precursor for the preparation of a large
number of ligand systems in coordination chemistry. One of
2
8 9 2
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5

View Article Online
Analyzer at the Institut f u¨ r Anorganische und Analytische
Chemie der Westf a¨ lischen Wilhelms-Universit a¨ t M u¨ nster. UV-
Vis-NIR absorption spectra of solutions were measured on a
Varian Cary 50 spectrophotometer in the range 190–1100 nm
at ambient temperatures. EI and MALDI-TOF mass spectra
were recorded on a Varian MAT 212 and a Bruker Reflex IV
mass spectrometer, respectively. H and C NMR spectra were
measured on Bruker ARX300, Bruker AMX400, or Varian
Unity plus 600 spectrometers using the solvent as internal
standard. Temperature-dependent magnetic susceptibilities of
finely grounded crystals were measured by using a SQUID
magnetometer (Quantum Design) at 1.0 T (2.0–290 K). For
261 (21900). Anal. Calc. for C11
C 69.81, H 4.19%.
H
8
O
3
: C 70.21, H 4.28. Found:
ꢀ
N,N -Bis(2-(1,8-naphthalenediol)methylidene)propylenediamine
(H L)
4
1
13
A solution of 1,3-diaminopropane (0.08 M, 15 mL, 1.2 mmol)
in EtOH (15 mL) was added to a solution of 2-formyl-1,8-
naphthalenediol (464 mg, 2.47 mmol) in CHCl (27 mL). The
reaction mixture was allowed to stir overnight, during which a
precipitate formed. This was filtered off. Another crop of pre-
cipitate was obtained from the filtrate after slow evaporation of
the solvent. The combined precipitates were washed three times
3
calculation of the molar magnetic susceptibility, v , the mea-
M
with Et
suitable for a single-crystal X-ray diffraction study were obtained
by slow evaporation of a CHCl –EtOH solution. Yield: 490 mg
2
O resulting in a brownish-yellow solid. Crystals of H
4
L
sured susceptibilities were corrected for the sample holder, the
underlying diamagnetism of the sample by using tabulated
−
6
3
−1
3
Pascal’s constants (3
2
: vdia = 508 × 10 cm mol ), and for
1
3
(
98%). H NMR (599.84 MHz, d
8
-THF): d 2.19 (q, J = 6.9 Hz,
the temperature independent paramagnetism (vTIP) which was
3
2
8
6
H; CH
2
CH
2
CH
2
), 3.71 (m, 4 H; NCH
2
CH
2
), 6.58 (d, J =
obtained by a fitting procedure.
3
4
.9 Hz, 2H; H4), 6.59 (dd, J = 7.9 Hz, J = 1.0 Hz, 2H; H7),
3
3
4
.84 (d, J = 8.9 Hz, 2H; H3), 6.87 (dd, J = 7.9 Hz, J =
3
2
-Formyl-1,8-bis(methoxymethoxy)naphthalene (2)
A solution of n-BuLi (2.5 M in hexane, 8.86 mL, 22.2 mmol)
and TMEDA (2.57 g, 22.2 mmol) in Et O (130 mL) was added
1.0 Hz, 2H; H5), 7.31 (dd, J = 7.9 Hz, 2H; H6), 7.99 (d, J =
1
2.6 Hz, 2H; N=C–H), 12.02 (br s, 2H; OH), 14.05 (s, 2H; OH);
C NMR (150.85 MHz, d -THF): d 32.7 (CH CH CH ), 48.7
NCH CH ), 109.0 (C8a), 112.1 (C7), 115.8 (C4), 117.7 (C5),
118.6 (C2), 129.7 (C3), 134.0 (C6), 141.2 (C4a), 162.9 (C8), 164.5
1
3
8
2
2
2
2
(
2
2
dropwise to a solution of 1,8-bis(methoxymethoxy)naphthalene
◦
(
5.00 g, 20.1 mmol) in Et
2
O (135 mL) at 0 C. After stirring the
+
(
C1), 185.0 (N=C); MS-MALDI-TOF: m/z 415.4 [M + H] ;
mixture for 6 h at this temperature, DMF (3.11 mL, 40.2 mmol)
−
1
FT-IR (KBr): m˜ /cm 1623 (C
(
=
N); UV-Vis (CH CN) k /nm
3
max
◦
was added and the resulting mixture stirred overnight (0 C to
−
1
−1
e/M cm ) 450 (29000), 431 (28000), 274 (78000). Anal. Calc.
: C 69.43, H 5.59, N 6.48. Found: C 69.33, H
r.t.). After addition of water (40 mL), the pH of the reaction
solution was decreased to 5–6 by addition of dilute HCl. The
for C25
H
24
N
2
O
5
5.37, N 6.38%.
aqueous solution was extracted with Et
organic extracts were combined, washed with brine and water,
and dried over anhydrous MgSO . Volatiles were removed under
reduced pressure to obtain 2 as an ochre solid. Yield: 4.89 g
88%). Crystals of 2 suitable for a single-crystal X-ray diffraction
study were obtained by slow evaporation of a n-hexane solution.
2
O (3 × 50 mL). The
II
[LCu
2
] (3)
4
Solid H L (50 mg, 0.12 mmol) and solid [Cu (OAc) (OH ) ]
4
2
4
2 2
(
(72 mg, 0.18 mmol) were each placed in one arm of a H-
shaped tube. The tube was filled carefully with DMF allowing
for diffusion of the two solutions between the two arms. In
the course of 12 days, black crystals deposited. Yield: 45 mg
1
H NMR (598.99 MHz, CDCl
3
): d 3.59 (s, 3 H; CH ), 3.61 (s, 3
3
3
H; CH ), 5.24 (s, 2 H; CH ), 5.36 (s, 2 H; CH
7
7
3
2
2
), 7.20 (dd, J =
4
3
3
−1
.56 Hz, J = 1.50 Hz, 1H; H7), 7.50 (dd, J = 7.56 Hz, J =
(69%). FT–IR (KBr): m˜ /cm 1604 (C=N). Anal. Calc. for
3
4
.90 Hz, 1H; H6), 7.51 (dd, J = 7.90 Hz, J = 1.50 Hz, 1H;
C H N O Cu : C 55.86, H 3.38, N 5.21. Found: C 55.63, H
2
5
18
2
4
2
3
3
H5), 7.61 (dd, J = 8.6 Hz, J = 0.9 Hz, 1H; H4), 7.85 (d, J =
3.33, N 5.42%.
1
3
8
(
1
.6 Hz, 1H; H3), 10.61 (d, J = 0.9 Hz, 1H; CHO); C NMR
150.63 MHz, CDCl ): d 56.6 (CH ), 58.2 (CH ), 95.9 (CH ),
), 111.8 (C7), 119.6 (C8a), 122.7 (C3 or C5), 122.8
3
3
3
2
X-Ray crystallographic data collection and refinement of the
structures
02.1 (CH
2
(
(
2
4
C3 or C5), 125.2 (C4), 127.3 (C2), 129.4 (C6), 140.5 (C4a), 154.3
+
X-Ray diffraction data were collected on a Bruker AXS APEX
diffractometer equipped with a rotating anode using Mo-Ka
radiation (2) or on Nonius Kappa-CCD diffractometers using
C8), 159.2 (C1), 191.1 (CHO); EI-MS: m/z (%) 276 (16) [M] ,
+
+
30 (4) [M − CH
2
OCH
3
] , 200 (100) [M − CHO − CH
2
OCH ] ,
3
+
−1
5 (56) [CH OCH
2
3
] ; FT-IR (KBr): m˜ /cm 1678 (C=O); UV-
−
1
−1
Cu-Ka radiation (H L) or using Mo-Ka radiation (3), in case
4
Vis (CH
3
CN) kmax/nm (e/M cm ) 361 (5400), 298 (5500), 255
of Mo-radiation equipped with a rotating anode generator.
Programs used: data collection SMART (2) and COLLECT
(
38300). Anal. Calc. for C15 : C 65.21, H 5.84. Found: C
H
16
O
5
11
12
64.95, H 6.11%.
11
13
(
H
4
L, 3), data reduction SAINT (2) and Denzo-SMN (H
4
L,
11
14
3
), absorption correction SADABS (2) and Denzo (3), struc-
15
2
-Formyl-1,8-naphthalenediol (II)
ture solution SHELXS-97, structure refinement SHELXL-
16
97. Details of data collection and structure refinements are
An argon purged solution of HCl (5–6 M in isopropanol,
◦
summarized in Table 1.
3
1
8 mL) was added dropwise at 0 C to a solution of 2-formyl-
,8-bis(methoxymethoxy)naphthalene (1.55 g, 5.61 mmol) in
CCDC reference numbers 273592–273594.
See http://dx.doi.org/10.1039/b507579c for crystallographic
data in CIF or other electronic format.
CH
was removed under reduced pressure. The residue was dissolved
in CHCl and dried over anhydrous MgSO . Volatiles were
2
Cl
2
(78 mL). After stirring the mixture for 6 h, the solvent
3
4
removed under reduced pressure to obtain II as an brownish-
yellow solid. Yield: 1.04 g (98%). H NMR (300.14 MHz,
Results and discussion
1
Synthesis and characterization
3
4
CDCl
3
): d 6.95 (dd, J = 8.0 Hz, J = 0.9 Hz, 1H; H7), 7.27 (dd,
J = 8.0 Hz, J = 0.9 Hz, 1H; H5), 7.32 (d, J = 8.7 Hz, 1H; H4),
.39 (d, J = 8.7 Hz, 1H; H3), 7.55 (dd, J = 8.0 Hz, 1H; H6), 9.41
s, 1H; OH), 9.86 (s, 1H; CHO), 14.30 (br s, 1H; OH); C NMR
100.63 MHz, CDCl ): d 111.9 (C7), 113.2 (C3), 113.3 (C8a),
18.9 (C5), 120.5 (C4), 125.8 (C2), 132.7 (C6), 139.3 (C4a), 157.8
3
4
3
Salicylaldehydes may be obtained by cleavage of the correspond-
3
3
7
(
(
1
(
ing salicylaldehyde ethers prepared through directed ortho-
1
3
17
lithiation of the appropriate phenyl ethers and subsequent
18
3
reaction with formaldehyde. This reaction sequence has been
applied broadly for the formation of aromatic compounds
+
5,19
C8), 163.7 (C1), 195.8 (CHO); EI-MS: m/z (%) 188 (100) [M] ;
with an ortho-hydroxyformyl unit.5 In this respect, we have
−
1
FT-IR (KBr): m˜ /cm 1641 (C=O); UV-Vis (CH
CN) kmax/nm
chosen the di-MOM derivative 1 of 1,8-naphthalenediol as
starting material for the preparation of II. An ethereal solution
3
−
1
−1
(
e/M cm ) 423 (9900), 404 (9300), 333 (4000), 323 (4000),
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8
2 8 9 3

View Article Online
Table 1 Crystal data and structure refinement for 2, H
4
L and 3
2
H
4
L
3
Empirical formula
C
276.28
173(2)
15
H
16
O
5
C
414.45
223(2)
1.54178
Monoclinic
P2 /c (no. 14)
14.731(1)
8.650(1)
16.954(1)
109.91(1)
2031.2(3)
4
25
H
22
N
2
O
4
C
537.49
198(2)
0.71073
Monoclinic
P2 /n (no. 14)
25 18 2 4 2
H N O Cu
M
r
T/K
k/A˚
0.71073
Monoclinic
P2 /c (no. 14)
4.5003(7)
13.713(2)
21.761(4)
93.422(3)
1340.6(4)
4
Crystal system
Space group
a/A˚
1
1
1
8.612(1)
b/A˚
c/A˚
18.755(1)
12.243(1)
96.89(1)
1963.2(3)
4
◦
b/
V/A˚
3
Z
−
3
D
c
/g cm
l/mm
1.369
0.103
1.355
0.753
1.819
2.206
−
1
Data/restraints/param.
Goodness of fit on F
Final R indices [I > 2r(I)]
3059/0/245
1.024
R1 = 0.0447
wR2 = 0.0947
R1 = 0.0772
wR2 = 0.1045
3209/0/288
1.008
R1 = 0.0484
wR2 = 0.1274
R1 = 0.0849
wR2 = 0.1449
4773/0/326
1.025
R1 = 0.0361
wR2 = 0.0813
R1 = 0.0548
wR2 = 0.0894
2
R indices (all data)
of 1 was treated with 1.1 equivalents of a 1 : 1 mixture
of n-BuLi/TMEDA and subsequently with 2 equivalents of
DMF. Acidic work-up resulted in the formation of 2 as an ochre
solid in 88% yield (Scheme 2).
(Scheme 2). The IR spectrum of II exhibits a C=O vibration at
−1 −1
1641 cm in comparison to 1678 cm in 2. This low energy shift
◦
is indicative of a rotation of the formyl substituent by ∼180 and
the formation of a hydrogen bond between the hydroxyl proton
and the oxygen atom of the formyl group in the unprotected
monoaldehyde II.
Aldehyde 2 exhibits a higher-order NMR spectrum due to
the nearly identical chemical shifts of the aromatic hydrogen
atoms H5 and H6. Therefore, the molecular structure of 2
was established by single-crystal X-ray diffraction (Fig. 1(a)).
The preparation of salen-type ligands is routinely achieved by
the Schiff-base condensation of ethylenediamine or a related
1
8
However, simulation of the 600 MHz H NMR spectrum in
diamine with two equivalents of an aldehyde or ketone. In
conjunction with 2D NMR spectra resulted in a conclusive
evaluation of the coupling constants, an rigorous assignment of
all resonances, and thus an NMR confirmation of the structure
of 2. The molecular structure in crystals of 2 reveals the formyl
group almost coplanar to the plane of the naphthalene ring with
the C=O function tilted away from the OMOM group (Fig. 1a).
this respect, we reacted the new monoaldehyde II with 1,3-
diaminopropane and obtained the unsymmetrical dinucleating
ligand H
based on the 1,8-naphthalenediol backbone (Scheme 2). The
molecular structure of H L was confirmed by single-crystal
4
L as the first member of a new family of ligands
4
X-ray diffraction (Fig. 1(b)). The two naphthalenediol units
are located at opposite ends of the molecule. The two planes
Deprotection of the hydroxy groups was performed by
◦
dissolving 2 in CH
2
Cl
2
and adding a solution of HCl in i-
of the naphthalene rings form an angle of 107.6 . These
PrOH. This procedure resulted in pure II with 98% yield
orientation seems to originate from intermolecular p–p stacking
Scheme 2
2
8 9 4
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8

View Article Online
aprotic polar solvents (acetonitrile, acetone, dichloromethane,
chloroform), as well as in polar protic solvents (methanol,
ethanol, DMF, water). Thus, single-crystals of this complex have
been obtained by slow diffusion techniques of DMF solutions of
H
4
L and copper acetate. Structural characterization by single-
crystal X-ray diffraction enabled the formulation of 3 as [LCu ].
2
Structural characterization of [LCu
2
] (3)
Fig. 2 shows the molecular structure of the dinuclear complex
[
L
LCu
2
] (3) and the labeling scheme used. The dinucleating ligand
provides two tetradentate coordination compartments, one
donor set occupied by Cu1 and one with an O
4
−
with an N
2
O
2
4
donor set occupied by Cu2. Selected interatomic distances and
angles are listed in Table 2.
Fig. 1 Molecular structures of (a) 2 and (b) H L with the atoms
4
represented by thermal ellipsoids at the 50% probability level. Selected
interatomic distances ( A˚ ): 2 O1–C1 1.376(2), O2–C7 1.369(2), O5–C15
1
4
.207(2); H L O1–C5 1.278(2), O2–C7 1.351(3), O21–C25 1.277(2),
O22–C27 1.333(3), N1–C3 1.302(3), N21–C23 1.296(3).
Fig. 2 Molecular structure of 3 with the atoms represented by thermal
ellipsoids at the 50% probability level. Hydrogen atoms have been
omitted for clarity.
interactions with distances between neighboring naphthalene
˚
planes of ∼3.4 A. Interestingly, difference Fourier synthesis
establishes a double zwitterionic form for ligand H L. The
4
electron density establishes two hydrogen atoms close to the
imine nitrogen atoms N1 and N21 and two hydrogen atoms
close to the phenolic oxygen atoms O2 and O22. That leads to
a formulation with phenolate oxygen atoms O1 and O21 being
deprotonated hydroxy groups. This assignment is corroborated
by the shorter mean O–C bond distances for O1 and O21
The distances of Cu1 can be compared to the analogue
distances in the mononuclear complexes [(salen)Cu] (H salen =
2
ꢀ
pr
N,N -bis(salicylidene)ethylenediamine) and [(salen )Cu]
pr
ꢀ
(H salen
2
=
N,N -bis(salicylidene)propylenediamine). The
21
range of Cu–N distances found in [(salen)Cu] is 1.92–1.96 A˚
pr
22
and found in [(salen )Cu] is 1.94–1.99 A˚ . The Cu1–N distances
˚
˚
of 1.278(2) and 1.277(2) A, respectively, as compared to the
of 1.93 and 1.95 A in 3 fit well into these ranges. On the other
hand, the Cu1–O bond distances of 1.97 and 1.98 A˚ found
analogous mean bond distances for O2 and O22 of 1.351(3)
˚
and 1.333(3) A, respectively. Strong N–H · · · O hydrogen bonds
in 3 exceed the ranges of Cu–O distances found in [(salen)Cu]
˚
˚
21
pr
˚
22
exists between N1 and O1 (distance 2.614 A) and N21 and
(1.89–1.91 A) and found in [(salen )Cu] (1.84–1.90 A). The
˚
O21 (distance 2.620 A). Beside these there are further strong
aryloxides in 3 are bridging ligands (O1 and O3) while they
˚
pr
hydrogen bonds between O1 and O2 (distance 2.530 A) and
are terminal ligands in [(salen)Cu] and [(salen )Cu]. A bridging
˚
O21 and O22 (distance 2.525 A). O1 and N1 form a nearly
ligand donates charge to two metal centers leading to a higher
overall charge donation but a diminished charge donation per
planar five-membered ring with the three interstices carbon
˚
atoms (maximum deviation from best plane 0.004 A) as do O21
˚
◦
and N21 (0.006 A). In this respect it is interesting to note that the
˚
Table 2 Selected interatomic distances (A) and angles ( ) for 3
1
6
00 MHz H NMR spectrum of ligand H
4
L measured in d -THF
8
exhibits a doublet for the imine hydrogen atom with a coupling
constant of 12.6 Hz. 2D NMR spectra exhibit this coupling to
the proton with d 12.02 ppm assigned to the hydrogen atom
of the protonated imine. The correlation of the structural and
NMR spectroscopic data indicate that in the case of such a
Cu1–O1
Cu1–O3
Cu1–N1
Cu1–N2
1.978(2)
1.967(2)
1.950(3)
1.930(3)
2.415(2)
1.944(2)
1.913(2)
1.878(2)
1.853(2)
2.712(2)
O1–C2
O2–C4
O3–C17
O4–C19
N1–C11
N2–C15
Cu1 · · · Cu2
Cu1 · · · Cu2
Cu1 · · · Cu1
Cu2 · · · Cu2
1.337(3)
1.323(3)
1.331(3)
1.308(3)
1.277(5)
1.280(5)
3.012(1)
3.221(1)
5.139(1)
3.534(1)
a
Cu1–O2
Cu2–O1
Cu2–O3
Cu2–O2
Cu2–O4
Cu2–O1
strong coupling a zwitterionic form might be present in solution.
1
a
a
a
The 600 MHz H NMR spectrum of ligand H
4
L measured
in d -DMSO does not show a coupling of the imine proton
6
a
indicating that in DMSO solution not the zwitterionic forms
but the convenient ortho-hydroxyimine tautomer dominates.
O1–Cu1–O3
O3–Cu1–N2
N2–Cu1–N1
N1–Cu1–O1
O1–Cu2–O2
77.77(7)
92.26(11)
99.04(13)
90.35(10)
91.46(7)
O2–Cu2–O4
O4–Cu2–O3
O3–Cu2–O1
Cu2–O1–Cu1
Cu2–O3–Cu1
93.41(7)
95.11(8)
79.89(8)
100.38(8)
101.87(8)
The reaction of H
4
L with 1 or more equivalents of
[
Cu (OAc) (OH ] results in the formation of an olive green
2
4
2 2
)
4
−
II
solid (3) which analyses as one L and two Cu . Coordination
of the imine nitrogen atoms to copper in 3 is readily indicated
−
1
by the red shift of the C=N stretch to 1604 from 1623 cm
a
Symmetry transformations used to generate equivalent atoms: −x, −y,
20
in H
4
L. However, this solid proofed to be insoluble in aprotic
−
z.
nonpolar solvents (hexane, heptane, toluene, ethyl acetate), in
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8
2 8 9 5

View Article Online
Fig. 4 Temperature dependence of the effective magnetic moment, leff
calculated for the tetranuclear assembly 3 at 1 T. The solid line is a fit
to the experimental data using the appropriate spin-Hamiltonian with
the values given in the text. Inset: Spin system of 3 imposed on its
,
2
2
II
molecular structure. Note that the magnetic orbitals (dx2 −y2 ) of each Cu
are oriented in the plane of its dimeric unit 3 perpendicular to the apical
oxygen donor of the other dimeric unit.
2
Fig. 3 Two views on the molecular structure of the dimer of dimer 3 .
Dashed lines indicate the weak apical Cu–O interactions leading to the
dimerization.
Several studies have been dealing with the necessity to incor-
porate non-zero values for the interdimer J terms in dimer of
II
dimer Cu complexes in which dimerization is achieved by weak
apical Cu–ligand interactions. Differing magnitudes and signs
have been reported for those interdimer coupling constants.
Fallon et al. already pointed out that the interdimer interaction
should be very small due to the large interdimer Cu–O distances
2
3
metal–ligand bond. The longer Cu1–O bond distances found
in 3 provide clear evidence for the reduced Cu1–O covalency as
compared to mononuclear salen copper complexes.
24–26
Two dinuclear complexes 3 dimerize by weak apical copper–
aryloxide interactions to a tetranuclear entity 3 . This results in
an overall square pyramidal coordination environment for each
copper center (see Fig. 3). For Cu1 the basal plane is build by
N1, N2, O3 and O1. Cu1 is positioned 0.10 A above the best
plane of these four basal ligand atoms toward the apical ligand
which is formed by a coordinated aryloxide donor (O2#1) of the
other dinuclear complex. The Cu1–O2#1 distance of 2.415(2) A
indicates only a weak apical interaction. On the other hand, Cu2
forms an apical bond with O1#1 of the other dinuclear complex
which is even longer at 2.712(2) A indicating an even weaker
(
apical Cu–O bond of a square-pyramidal coordination geom-
etry) as compared to the shorter intradimer Cu–O distances
equatorial Cu–O bonds of a square-pyramidal coordination
2
(
26
geometry). They were able to fit their magnetic data without
the use of an interdimer coupling constant. Moreover, by
explicitly using the appropriate tetramer model, they obtained
˚
26
an interdimer coupling constant of zero. This argument also
˚
holds for 3 because the interdimer Cu–O distances are larger
2
˚
(
(
2.42 and 2.71 A) as compared to the intradimer Cu–O distances
1.91–1.98 A).
˚
˚
Besides this distance point of view, the orientation of the
apical interaction. This is corroborated by the fact that in the
case of the coordination environment of Cu2 not the metal ion
is positioned above a best plane of its basal ligands (O1, O2,
O4, O3) but one of the basal ligands (O2) is positioned above
a best plan described by O1, O3, O4, and Cu2. This is due to
the function of O2 as an apical donor for Cu1#1 of the other
dinuclear complex.
magnetic orbitals should also contribute to the fact that the
interdimer couplings are weak. The magnetic orbital of each Cu
II
ion (dx2 −y2 ) is oriented towards the four equatorial ligand atoms
of the dinuclear unit 3. This leads to an effective superexchange
pathway across the bridging oxygen atoms of the dinuclear
complex. On the other hand, the magnetic orbital of each
II
Cu (dx2 −y2 ) is of d-type symmetry with regard to the Cu–O
interaction with its apical ligand. As these oxygen atoms don’t
have d-type symmetry orbitals, there is no interaction with the
Magnetic properties
II
magnetic orbitals of the Cu ions. Hence, the superexchange
27
The magnetic susceptibility was measured in the temperature
pathway through the apical ligand has to be rather weak.
range 2–290 K and analyzed for the tetranuclear entity 3
order to take into account possible interdimer interactions. The
effective magnetic moment, leff, of the tetranuclear assembly
2
in
In order to reduce the parameter space we neglect the effect
of J22ꢀ because this interaction involves two-times the longest
apical Cu–O separation of 2.72 A, while J (and J21ꢀ ) involves
˚
12ꢀ
˚
˚
3
2
has a value of 1.14 l
than the theoretical value of 3.65 l
ions (S = 2.11). Decreasing the temperature leads
= 1/2, g
to a steadily decrease of leff until it reaches at ∼120 K a
plateau of 0.27 l (Fig. 4). This behavior is characteristic for
strong antiferromagnetic exchange interactions between the Cu
centers resulting in a S
B
at 290 K which is significantly lower
one apical Cu–O separations of 2.42 A and only one of 2.72 A.
The magnetic properties were analyzed by using the spin
Hamiltonian in eqn (1) including the isotropic Heisenberg–
Dirac–van Vleck (HDvV) exchange Hamiltonian and the single-
ion Zeeman interaction by means of a full-matrix diagonaliza-
tion approach. The macroscopic magnetization is calculated by
summing the microscopic magnetizations weighted according to
II
B
for four uncoupled Cu
i
i
B
II
t
= 0 spin ground state. The plateau at low
temperature originates from a trace amount of a paramagnetic
impurity (probably a mononuclear Cu species).
the Boltzmann distribution.
II
ꢀ
H = HHDvV
+
[l
B
g
i
S B]
i
The spin topology of 3
relations J12 = J1ꢀ2ꢀ and J12ꢀ = J1ꢀ2 arise by consideration of the
. The dominant intradimer coupling
2
is indicated in the inset of Fig. 4. The
(
1)
i
H
HDvV = −2J12(S
1
S
2
+ S1ꢀ
S
2ꢀ ) − 2J12ꢀ (S
1
S
2ꢀ + S1ꢀ
2
S )
molecular symmetry of 3
2
is presented by J12, whereas J12ꢀ and J22ꢀ represent interdimer
Fitting the data resulted in an intradimer coupling constant
of J12 = −371 cm− with g = 2.11, a paramagnetic impurity with
1
couplings.
2
8 9 6
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8

View Article Online
S = 1/2 (g = 2.00) of 2.4%, and vTIP = 82 × 10− cm mol
6
3
−1
and their derivatives. The straightforward synthesis of the
34
28
II
(
Fig. 4). The value of J12ꢀ showed no effect on the fit.
Simulations using the values given above in combination with
dinuclear complex [LCu
ligand compartment and one Cu in the O ligand compartment
demonstrate the capability of ligand L to form unsymmet-
rical dinuclear complexes. The structural analysis evidenced
the formation of weak axial Cu–O interactions to result in
3 . The ligand L allows for a strong exchange coupling
between the two Cu ions of the dinuclear complex 3 with a
2
] (3) with one Cu ion in the N
2
O
2
II
4
−
1
−1
4−
interdimer couplings J12ꢀ = −10 cm or J12ꢀ = +10 cm lead
to leff vs. T curves which are indistinguishable to that with
J
12ꢀ = 0. The reason for this insensitivity to the interdimer
exchange is easy to derive by analyzing the spin ladder of the
tetranuclear unit 3 . Considering the intradimer exchange J12
371 cm only, results in a S
= 0 spin ground state, two
= 1 spin states at 742 cm , and three degenerate
= 0, S = 1, S = 2) at 1484 cm . This spin
ladder leads to thermal populations of the two first excited S
spin states of 6.55% each at the highest temperature used for
4
−
2
II
2
=
−
1
−1
−
t
coupling constants of J12 = −371 cm . The comparison of
−
1
degenerate S
t
this value to established magneto-structural correlations for
−
1
II
II
spin states (S
t
t
t
Cu (OR) Cu demonstrates the electronic equivalence of the
aryloxide functions of 1,8-naphthalenediol to that of phenols.
The synthesis of other homo- as well as heterodinuclear
complexes with ligand L and related ligands is currently
performed in our laboratory. One interest in such dinuclear
complexes lies in the area of magnetochemistry. Secondly, these
complexes will be evaluated for their catalytic performance in
transformations which are known to be catalyzed by mononu-
clear salen complexes. Especially the possible pre-binding of the
2
t
=
1
4
−
the measurement of the susceptibility data (290 K). The three
highest excited states exhibit only minor populations of 0.06%
(
S
t
= 0), 0.17% (S
Introducing an interdimer coupling of J12ꢀ = −10 cm results
into a splitting of the first excited S
t
= 1) and 0.28% (S
t
= 2).
−
1
−
1
t
= 1 states of 20 cm and an
−1
overall splitting of the three highest states of 29.8 cm . The spin
states exhibit the following energies and thermal populations at
substrate by the metal ion in the O compartment seems to be a
good candidate for catalyst improvements.
4
−
1
−1
2
S
2
90 K: S
t
t
= 1 at 733 cm (6.87%), S
= 0 at 1466 cm (0.06%), S
at 1495.8 cm (0.26%).
The splitting of the first two excited S
t
= 1 at 753 cm (6.23%),
−
1
−1
t
= 1 at 1475.8 cm (0.17%), S
t
=
−
1
Acknowledgements
t
= 1 states has no
net effect on the leff vs. T curve because both spin states have
T. G. gratefully acknowledges Professor F. E. Hahn for his
generous support. This work was supported by the Fonds
der Chemischen Industrie, the BMBF, the Dr. Otto R o¨ hm
Ged a¨ chtnisstiftung, and the DFG (Priority Program 1137
Molecular Magnetism’). Dr E. Bill (Max-Planck-Institute for
Bioinorganic Chemistry, M u¨ lheim, Germany) is thanked for the
magnetic measurements and for valuable discussions.
the same magnetic moment. The splitting of the spin states
−
1
around 1500 cm is also negligible because they are thermally
not accessible to a significant level due to their high energy.
An analogous treatment by taking into account a ferromagnetic
interdimer exchange coupling J12ꢀ = +10 cm leads to similar
effects and argumentation. In order to observe an effect on
the leff vs T curve, absolute values of the interdimer coupling
constant in the order of 50 cm are necessary which are too
large considering the associated exchange pathway.
It is interesting to compare the coupling constant (J12
‘
−
1
−
1
References and notes
=
1
M. Poirier, M. Simard and J. D. Wuest, Organometallics, 1996, 15,
296.
P. A. Vigato, U. Casellato, M. Vidali, R. Graziani, D. E. Fenton and
C. M. Regan, Inorg. Chim. Acta, 1979, 32, L27; R. Graziani, M.
Vidali, P. A. Vigato, U. Casellato, D. E. Fenton and C. M. Regan,
Inorg. Chim. Acta, 1980, 38, 85.
3 B. F. Hoskins, C. J. McKenzie and R. Robson, J. Chem. Soc., Dalton
Trans., 1992, 3083.
S. Dayagi and Y. Degani, in The Chemistry of the Carbon–Nitrogen
Double Bond, ed. S. Patai, Interscience Publishers, London, 1970,
p. 64.
−
1
−
371 cm ) obtained for the bis(l-naphthoxide)-bridged dicop-
1
per(II) unit in 3 to analogous bis(l-hydroxy)-, bis(l-alkoxy)-
and bis(l-phenoxide)-bridged dicopper(II) complexes. Using
the well-established magneto-structural relationship for bis(l-
hydroxy)-bridged dicopper(II) complexes (J = 1/2 (−74.53a +
2
−
1
29
7
270) cm with a being the Cu–O–Cu angle results in a range
−
1
for J12 in 3 of −105 to −160 cm . This predicted range implies
4
a weaker exchange coupling as compared the observed value
−
1
of J12 = −371 cm . Inherently stronger exchange interactions
have been established for bis(l-alkoxide)-bridged dicopper(II)
5 T. Glaser and I. Liratzis, Synlett, 2004, 735.
25,30
6
R. Robson, Inorg. Nucl. Chem. Lett., 1970, 6, 125; R. Robson,
Aust. J. Chem., 1970, 23, 2217; N. H. Pilkington and R. Robson,
Aust. J. Chem., 1970, 23, 2225.
complexes.
Recently, a magneto-structural relation for bis(l-
phenoxide)-bridged dicopper(II) complexes has been reported
which exhibits no zero-crossing due to ‘accidental orthogonality’
7
P. Pfeiffer, E. Breith, E. L u¨ bbe and T. Tsumaki, Liebigs Ann., 1933,
03, 84; P. Pfeiffer, T. Hesse, H. Pfitzner, W. Scholl and H. Thielert,
J. Prakt. Chem., 1937, 149, 217; P. Pfeiffer, W. Christeleit, T. Hesse,
H. Pfitzner and H. Thielert, J. Prakt. Chem., 1938, 150, 261.
8 R. H. Holm, G. W. Everett and A. Chakravorty, Prog. Inorg. Chem.,
966, 7, 83; M. Calligaris and L. Randaccio, in Comprehensive
Coordination Chemistry, ed. G. Wilkinson, R. D. Gillard and
J. A. McCleverty, Pergamon, Oxford, 1987, vol. 2, p. 715.
W. Zhang, J. L. Loebach, S. R. Wilson and E. N. Jacobsen, J. Am.
Chem. Soc., 1990, 112, 2801; R. Irie, K. Noda, Y. Ito, N. Matsumoto
and T. Katsuki, Tetrahedron Lett., 1990, 31, 7345; E. N. Jacobsen,
Acc. Chem. Res., 2000, 33, 421.
0 N. Str a¨ ter, W. N. Lipscomb, T. Klabunde and B. Krebs, Angew.
Chem., Int. Ed. Engl., 1996, 35, 2024; B. A. Averill, in Comprehensive
Coordination Chemistry II, ed. J. A. McCleverty and T. J. Meyer,
Elsevier Ltd., Oxford, 2004, vol. 8, pp. 641; R. H. Holm, P. Kennepohl
and E. I. Solomon, Chem. Rev., 1996, 96, 2239.
−
1
31
for reasonable angles (J = 1/2 (−31.95a + 2462) cm ).
5
This relation predicts the strongest inherent antiferromagnetic
II
II
interactions in the series Cu (OR)
2
Cu (R = H < Alk < Ph)
−
1
and results in values for 3 of J12 = −373 to −396 cm . This
1
range compares nicely to the experimental value for 3 of J12
=
−
1
−
371 cm . Hence, the electronic properties of the aryloxide
functions of 1,8-naphthalenediol closely resembles those of
phenols.
9
Conclusions and outlook
1
We have for the first time synthesized 2-formyl-1,8-
naphthalenediol II as the ‘one ring- and one donor-increased’
derivative of salicylaldehyde I where numerous applications as
precursor for the preparation of various ligand systems have
L proves the ability of
aldehyde II as a versatile precursor for multidentate chelating
ligand systems. The tetra-anion L provides a N
compartment closely related to that of salen-type ligands
and an additional O compartment. Other hetero-dinucleating
ligands with a N and an O compartment are based on 3-
formylsalicylic acid or 1-(o-hydroxyphenyl)-1,3-butanedione
11 Bruker AXS, 2000.
been reported. The preparation of H
4
12 COLLECT, Nonius B. V., 1998.
3 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276, 307.
14 Z. Otwinowski, D. Borek, W. Majewski and W. Minor, Acta
Crystallogr., Sect. A, 2003, 59, 228.
5 G. M. Sheldrick, SHELXS-97, Program for solution of crystal
structures, University of G o¨ ttingen, Germany, 1997.
1
4
−
2
O
2
ligand
1
4
2
O
2
4
16 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of G o¨ ttingen, Germany, 1997.
32
33
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8
2 8 9 7

View Article Online
1
1
1
7 V. Snieckus, Chem. Rev., 1990, 90, 879.
24 L. Merz and W. Haase, J. Chem. Soc., Dalton Trans., 1978, 1594; L.
Schwabe and W. Haase, J. Chem. Soc., Dalton Trans., 1985, 1909;
R. Wegner, M. Gottschaldt, H. G o¨ rls, E.-G. J a¨ ger and D. Klemm,
Chem. Eur. J., 2001, 7, 2143; J. Sletten, A. Sorensen, M. Julve and Y.
Journaux, Inorg. Chem., 1990, 29, 5054.
25 L. Merz and W. Haase, J. Chem. Soc., Dalton Trans., 1980, 875.
26 G. D. Fallon, B. Moubaraki, K. S. Murray, A. M. Vandenbergen and
B. O. West, Polyhedron, 1993, 12, 1989.
8 H. Christensen, Synth. Commun., 1975, 5, 65.
9 N. S. Narasimhan, R. S. Mali and M. V. Barve, Synthesis, 1979,
9
06; R. G. Harvey, C. Cortez, T. P. Ananthanarayan and S.
Schmolka, J. Org. Chem., 1988, 53, 3936; A. S. Parsons, J. M.
Garcia and V. A. Snieckus, Tetrahedron Lett., 1994, 35, 7537; H.
Brunner and H. Schießling, Bull. Soc. Chim. Belg., 1994, 103,
1
19; M. M. G. Antonisse, B. H. M. Snellink-Ruel, A. C. Ion,
J. F. J. Engbersen and D. N. Reinhoudt, J. Chem. Soc., Perkin
Trans. 1, 1999, 1211; H.-C. Zhang, W.-S. Huang and L. Pu,
J. Org. Chem., 2001, 66, 481; A. Arroult, J.-Y. M e´ rour, J.-M.
L e´ ger, C. Jarry and G. Guillaumet, Helv. Chim. Acta, 2001, 84,
27 T. Glaser, T. L u¨ gger and R. Fr o¨ hlich, Eur. J. Inorg. Chem., 2004, 394.
28 The routine JULIUS was used for spin Hamiltonian simulations of
the data (C. Krebs, E. Bill, F. Birkelbach and V. Staemmler, unpub-
lished results). For the calculation of the magnetic susceptibility of
the paramagnetic impurity JULIUS uses the molar mass of the main
2
198.
2
2
2
0 R. Hern a´ ndez-Molina and A. Mederos, in Comprehensive Coordina-
tion Chemistry II, ed. J. A. McCleverty and T. J. Meyer, Elsevier Ltd.,
Oxford, 2004, vol. 1, p. 411.
component. Thus the obtained value for the paramagnetic impurity
II
of 3
2
(2.4%) is far overestimated if a mononuclear Cu species is
anticipated.
1 E. N. Baker, D. Hall and T. N. Waters, J. Am. Chem. Soc., 1970,
29 V. H. Crawford, H. W. Richardson, J. R. Wasson, D. J. Hodgson and
W. E. Hatfield, Inorg. Chem., 1976, 15, 2107; D. J. Hodgson, Inorg.
Chem., 1976, 15, 3174.
4
06, 406; H. Milburn, M. R. Truter and B. L. Vickery, J. Chem. Soc.,
Dalton Trans., 1974, 841.
2 G. B. Drew Michael, R. N. Prasad and R. P. Shama, Acta Crystallogr.,
Sect. C., 1985, 41, 1755; R.-G. Xiong, B.-L. Song, J.-L. Zuo and X.-
Z. You, Polyhedron, 1996, 15, 903; B. H. Chen, H. H. Yao, W. T.
Huang, P. Chattopadhyay, J. M. Lo and T. H. Lu, Solid State Sci.,
30 M. Handa, N. Koga and S. Kida, Bull. Chem. Soc. Jpn., 1988, 61,
3853.
31 L. K. Thompson, S. K. Mandal, S. S. Tandon, J. N. Bridson and
M. K. Park, Inorg. Chem., 1996, 35, 3117.
32 H. Okawa, M. Tanaka and S. Kida, Chem. Lett., 1974, 987.
33 D. E. Fenton, S. E. Gayda, U. Casellato, P. A. Vigato and M. Vidali,
Inorg. Chim. Acta, 1978, 27, 9.
34 U. Casellato, P. A. Vigato, D. E. Fenton and M. Vidali, Chem. Soc.
Rev., 1979, 8, 199; D. E. Fenton and H. Okawa, Chem. Ber., 1997,
130, 433.
1
999, 1, 119; M. Hasegawa, K. Kumagai, M. Terauchi, A. Nakao, J.
Okubo and T. Hoshi, Monatsh. Chem., 2002, 133, 285; L. C. Nathan,
J. E. Koehne, J. M. Gilmore, K. A. Hannibal, W. E. Dewhirst and
T. D. Mai, Polyhedron, 2003, 22, 887.
2
3 T. Glaser, B. Hedman, K. O. Hodgson and E. I. Solomon, Acc. Chem.
Res., 2000, 33, 859.
2
8 9 8
D a l t o n T r a n s . , 2 0 0 5 , 2 8 9 2 – 2 8 9 8