HAL Id: hal-00159946
https://hal.archives-ouvertes.fr/hal-00159946
Submitted on 5 Oct 2017
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Synthesis, magnetic behaviour, and X-ray structures of
dinuclear copper complexes with multiple bridges.
Ecient and selective catalysts for polymerization of
2,6-dimethylphenol
R. Murugavel, R. Pothiraja, N. Gogoi, R. Clérac, L. Lecren, R.J. Butcher, M.
Nethaji
To cite this version:
R. Murugavel, R. Pothiraja, N. Gogoi, R. Clérac, L. Lecren, et al.. Synthesis, magnetic behaviour, and
X-ray structures of dinuclear copper complexes with multiple bridges. Ecient and selective catalysts
for polymerization of 2,6-dimethylphenol. Dalton Transactions, Royal Society of Chemistry, 2007, 6
(6), p. 2405-2410. �10.1039/b618559b�. �hal-00159946�
Synthesis, magnetic behaviour, and X-ray structures of dinuclear copper
complexes with multiple bridges. Efficient and selective catalysts for
polymerization of 2,6-dimethylphenol†‡
Ramaswamy Murugavel,*
a
Ramasamy Pothiraja,
a
Nayanmoni Gogoi,
a
Rodolphe Cl
´
erac,
b
Lollita Lecren,
b
Ray J. Butcher
c
and Munirathinam Nethaji
d
The use of a potentially tridenta
te mono-anionic bridging ligand, 1,3-bis(3,5-dimethylpyrazol-1-yl)-
propan-2-ol (bdmpp-H), in assembling new dimeric copper complexes with interesting magnetic
properties has been investigated. The reaction of copper hydroxide or copper acetate with phenyl
phosphinic acid or diphenyl phosphinic acid in the presence of bdmpp-H produces the dinuclear
complexes [Cu(bdmpp)(ppi)]
2
(1) and [Cu(bdmpp)(dppi-H)]
2
(dppi)
2
(2) (ppi-H = phenylphosphinic
acid; dppi-H = diphenylphosphinic acid), respectively. The products have been characterized with the
help of analytical, thermal, and spectroscopic (IR, UV-vis, and EPR) techniques. Single crystal X-ray
diffraction studies of 1 and 2 reveal that the two bdmpp ligands hold together the dimeric copper unit
in each complex through l-O alkoxide and the pyrazolyl nitrogen ligating centers. Two phenyl
phosphinate ligands additionally bridge the dicopper core in 1 to result in octahedral coordination
geometry around each metal ion. The diphenyl phosphinic acid acts as a terminal ligand in 2, and thus
completes a square pyramidal geometry around each copper ion. Both complexes show a very short
Cu ···Cu separation (3.001 and 3.065 A
˚
for 1 and 2, respectively). The investigation of the magnetic
properties reveals the efficiency of the double alkoxide bridge between the two paramagnetic copper
ions to mediate strong antiferromagnetic interactions [J/k
B
=−620(5) K (−431(4) cm
−1
)and−685(5)
K(−476(4) cm
−1
)for1 and 2, respectively]. Compounds 1 and 2, along with a few other copper
phosphate complexes, were tested as catalysts for the oxidative polymerization of 2,6-dimethylphenol; 1
and 2 were found to be efficient catalysts with an increased selectivity for the formation of the
polyphenylene ether. However a related mononuclear octahedral copper complex [Cu(imz)
4
(dtbp)
2
]
(dtbp-H = di-tert-butylphosphate) was found to be more efficient.
Introduction
Assembling dimeric, tetrameric, and other oligomeric copper
complexes with the aid of suitably designed polydentate ligands
with multiple functional groups has been of great interest in
recent times for a variety of reasons.
1
In particular, there have been
several studies reporting synthesis of dinuclear copper complexes,
with short non-bonded contacts between the metal centers, as
structural models for DNA cleavage reactions through phosphate
ester hydrolysis.
2
Dinuclear copper compounds have also been
implicated as active catalysts for the polymerization of substituted
phenols to yield poly(phenylene ether) (PPE).
3
While both these
aspects have already been investigated with a variety of ligands
a
Department of Chemistry, Indian Institute of Technology-Bombay, Powai,
Mumbai, 400076, India. E-mail: rmv@iitb.ac.in; Fax: +91-25723480
b
Universite
´
Bor deaux 1; CNRS, Centre de R echer che Paul Pascal-UPR8641,
115 av. du Dr. Albert Schweitzer, 33600, Pessac, France
c
Department of Chemistr y , How ar d University, W ashington DC, 20059, USA
d
Department of Inorganic and Physical Chemistry, Indian Institute of
Science, Bangalore, 560012, India
around a dicopper core, bridged by either l-oxo or l-hydroxo
ligands, the use of multidentate ligands containing alkoxy or
phenoxy groups in multi-metal compounds has continued to
attract considerable attention among synthetic inorganic chemists
in recent years.
We have been interested for a while in the synthesis of a variety
of copper and other transition metal phosphate complexes,
4
containing structurally diverse auxiliary ligands, which are useful
as open framework solids or models to framework solids,
4a
complexes having interesting magnetic properties,
4b
and model
compounds for phosphate diester hydrolysis.
4c
In the present study
we have used a pyrazole based tridentate N
2
O
−
ligand, bdmpp,
and synthesized two new l-alkoxo bridged dicopper complexes
with interesting structural, magnetic, and catalytic properties. The
results of this investigation are reported in this paper.
Results and discussion
Synthesis
Methylene bridged (–(CH
2
)
n
–) bis-pyrazolyl and bis-imidazolyl
compounds have been used as convenient bridging or chelating
ligands in the coordination chemistry of first row transition metal
ions.
5
While a variety of complexes have been synthesized with
1
this type of ligand, it is often difficult to predict the type of
structure (e.g. chelate or bridging mode of coordination) apriori
due to the free rotation of the C–C or C–N bonds in the ligand
backbone. In order to evaluate any such conformational changes in
the presence of additional chelating ligands, we have investigated
the reactions of Cu(
II) salts with bdmpp-H in the presence of
phosphinic acids. Thus, the reaction between equimolar amounts
of copper hydroxide, phenylphosphinic acid, Ph(H)PO
2
Hand
bdmpp-H results in the formation of [Cu(bdmpp)(ppi)]
2
(1)in
good yield (Scheme 1). [Cu(bdmpp)(dppi-H)]
2
(dppi)
2
(2)hasbeen
synthesized by a similar reaction from copper acetate and dppi-H
in the presence of bdmpp-H (Scheme 1).
Scheme 1 Syntheses of 1 and 2.
Characterization
Complexes 1 and 2 have been characterized with the aid of
elemental analysis, IR and UV-visible spectroscopic studies
(Table 1). Both the compounds yielded correct elemental analysis.
Strong IR bands observed at 1157, 1132, and 1018 cm
−1
for 1
and 1197, 1127 and 1044 cm
−1
for 2 indicate the presence of
POO
−
stretching and M–O–P vibrations. The characteristic P–
H stretching vibration of the phenyl phosphinate in 1 appears as
a strong band at 2284 cm
−1
. The UV-visible spectra of both 1 and
2 show a strong absorption in the ultraviolet region (∼345 cm
−1
)
and a weak absorption in the visible region (∼630 cm
−1
). While
the 345 cm
−1
absorption is readily assignable to the p–p*orn–p*
transition arising out of the pyrazole groups in bdmpp, the weak
absorption in the visible region is attributable to the symmetry
forbidden d–d transition of the Cu(
II) ion.
Molecular structure of 1·CH
2
Cl
2
·2H
2
O. Compound 1 was
crystallized from a CH
2
Cl
2
/toluene mixture (1 : 1) at 5
◦
C. Blue
single crystals of 1·CH
2
Cl
2
·2H
2
O were obtained after one week by
slow evaporation of the solvent in air. The copper phosphinate
1 crystallizes in the monoclinic C2/m space group with one
molecule of CH
2
Cl
2
and two molecules of water. The molecule
has a crystallographically imposed 2/m symmetry. In the dimeric
structure of 1, each copper ion is firmly coordinated by two
nitrogen atoms and two oxygen atoms of two bdmpp in a distorted
square-planar environment (Fig. 1). The two phenylphosphinate
anions symmetrically bridge the two copper ions on either side of
the square plane formed by the donor atoms of the bdmpp ligands
to result in a tetragonally elongated octahedral geometry for the
copper ions.
Fig. 1 ORTEP of [Cu(ppi)(bdmpp)]
2
(1) (hydrogen atoms removed for
clarity).
The central Cu
2
O
2
ring is planar and there are two types of Cu–
O distances in the molecule. The Cu(1)–O(1) distance (1.931(2) A
˚
)
is considerably shorter than the Cu(1)–O(2) bond (2.767(4) A
˚
) due
to the Jahn–Teller distortion. The Cu(1)–O(1) distance within the
four-membered ring is slightly longer compared to similar Cu–O
distances found in [Cu
2
(bdpo)
2
](ClO
4
)
2
·2MeOH (bdpo = 1,3-
bis(2-benzimidazolyl)-propan-2-oxide) (1.930(2) and 1.923(2) A
˚
).
6
Likewise the Cu(1)–O(2) distance (2.767(4) A
˚
) is one of the longest
Cu–O distances known in the literature,
4d
which can be ascribed to
the combined effect of steric crowding around the copper ions and
the Jahn–Teller elongation along the z-axis of the d
9
-Cu(II) ion.
Table 1 Comparison of selected structural parameters of dicopper complexes with a Cu
2
O
2
core
Compounds Cu ···Cu/A
˚
Cu–O–Cu/
◦
gJ/K J/cm
−1
Reference
[Cu(bdmpp)(ppi)]
2
(1) 3.001 101.95(1) 2.08 −620(5) −431(4) This work
[Cu(bdmpp)(dppi-H)]
2
(dppi)
2
(2) 3.065 103.50(1) 2.08 −685(5) −476(4) This work
[Cu
2
(N-(2-hydroxymethylphenyl)salicylamide)
2
](NBu
4
)
2
3.018(2) 104.3(4) 2.25 −539 −375 7
[Cu
2
(bdmpp)
2
][Cu(MeOH)Cl
3
]
2
2.990(2) 103.9(2), 98.8(2) 2.10 −556 −386 6
[Cu(nhep)Cl]
2
(Hnhep is 1-(2-hydroxyethyl)-pyrazole) 3.055 105.4, 104.2 — <−800 K <−500 8
[PBu
4
]
2
[Cu
2
(N-(2-hydroxyphenyl)salicylamide)
2
] 3.035(2) 102.6(5) 2.15 −322 −224 9
[NPr
4
]
2
[Cu
2
(N-(2-hydroxyphenyl)-3-methoxycarbonyl-
4,6-dimethylsalicylamide)
2
]
3.033(4) 103.7(3) 2.24 −272 −189 9
2
The Cu ···Cu distance in 1 (3.001 A
˚
) is comparable to those found
in other hydroxo or alkoxo bridged dinuclear complexes,
5,7–10
but
is considerably shorter compared to the mono-hydroxo or mono-
alkoxo bridged dinuclear complexes [Cu
2
(L1)(O
2
P(OAr))](PF
6
)
2
(3.773(4) A
˚
),
11
[(L2)Cu
2
(O
2
P(OCH
2
Ph)
2
](ClO
4
)
2
(3.67 A
˚
)
12
and
[Cu
2
(L3)((O
2
P(OAr))](ClO
4
)
2
(3.779(1) A
˚
)
13
(HL1 = 2,6-bis[bis(2-
pyridylethyl)aminomethyl]phenol; Ar = 4-nitrophenyl; HL2 = 2,
6-bis[bis(2-benzimidazolylmethyl)aminomethyl]-4-methylphenol),
HL3 = N,N,N
,N
-tetrakis{(6-methyl-2-pyridyl)methyl}-1,3-di-
aminopropan-2-ol).
The bdmpp ligands chelate the copper atoms through their
alkoxide oxygen and the pyrazolyl nitrogen to form two six-
membered rings. Thus the core structure is stabilized by the forma-
tion of one four-membered Cu
2
O
2
and four six-membered Cu–O–
C–C–N–N metallocycles (Fig. 1). The intermolecular hydrogen
bonding between the lattice water molecules and phosphinate
oxygen atoms (O–H ···O 168
◦
) leads to the formation of a chain-
like structure as shown in Fig. 2.
Fig. 2 Hydrogen bonding aided polymeric chain formation in
[Cu(ppi)(bdmpp)]
2
CH
2
Cl
2
·2H
2
O(1).
Molecular structure 2. Compound 2 was crystallized from
methanol at room temperature. Blue single crystals were obtained
after one week by slow evaporation of the solvent. The compound
crystallizes in the triclinic P
¯
1 space group with a crystallographi-
cally imposed inversion centre. Each copper ion in 2 is coordinated
by two nitrogen atoms and two oxygen atoms of two bdmpp
ligands in a distorted square-planar environment. A phosphoryl
oxygen atom at the axial position completes the square pyramidal
geometry around each of the copper ions in the molecule. As in
the case of 1, the two copper atoms and two alkoxy oxygen atoms,
which form the central Cu
2
O
2
unit, are lying in the same plane.
There are two types of Cu(1)–O(1) distances in the structure of
2 (Cu–O(1)#1 1.938(2) A
˚
and Cu–O(1) 1.966(2) A
˚
), which are
longer than those observed for 1. The larger difference in the
internal angles (Cu–O–Cu 103.50(1) and O–Cu–O 76.49(7)
◦
)and
the distances account for the deviation of the central ring from a
regular square shape. The Cu–O–Cu angle in 2 (103.50(1)
◦
) is one
of the largest observed for this type of compound.
6–10
Although the
Cu ···Cu distance in 2 (3.065 A
˚
) is comparable to other dinuclear
complexes with a Cu
2
O
2
core, (including 1), it definitely lies on
the upper side of the range observed for compounds with a Cu
2
O
2
core.
Each bdmpp ligand in 2 chelates the two copper atoms through
its alkoxide oxygen and the pyrazolyl nitrogen to form two six-
membered rings as in 1. Thus the core structure is again stabilized
by the formation of one four-membered Cu
2
O
2
and four six-
membered Cu–O–C–C–N–N metallocycles (Fig. 3). The major
difference between the structures of 1 and 2 is the mode in which
Fig. 3 ORTEP of the cationic part of [Cu(dppi-H)(bdmpp)]
2
(dppi)
2
(2)
(C–H hydrogen atoms removed for clarity).
the phosphinic acid ligand is coordinated to the metal. While the
phenyl phosphinate in 1 bridges the two copper ions in a symmetric
fashion, the diphenyl phosphinic acid in 2 prefers to coordinate to
the metal in unidentate fashion, that too through its phosphoryl
oxygen (P
=
O) (Cu(1)–O(2) 2.292(2) A
˚
) on only one side of the
square plane formed by the alkoxide oxygen and pyrazolyl nitrogen
atoms. This leads to a square pyramidal geometry around the
metal, and leaves the complex with a di-positive (+2) charge. For
the charge neutrality, one uncoordinated dppi anion is additionally
present for each copper atom as shown in Fig. 3. It is also of interest
to note that the elongation of the axial site in 2 is considerably
small, compared to 1.
Apart from the fact that the hydrogen atom on O(3) was located
from the difference map and refined, the observed difference in
the P–O distances involving O(2) and O(3)
14
(P(1)–O(2) 1.498(2)
and P(1)–O(3) 1.562(2) A
˚
), supports the inference that the metal
is bound to the P
=
O group rather than the P–O
−
terminal in
2 (Fig. 3). Additional support for this result comes from the
observation that the two P–O distances in the uncoordinated
phosphinate are indeed similar (P(2)–O(4) 1.510(2) A
˚
and P(2)–
O(5) 1.527(2) A
˚
). The phosphinate anions, apart from providing
the charge stabilization, also bridge the dimeric cations through
O–H ···O hydrogen bonding on one side (O–H ···O 173.7
◦
)and
C–H ···O hydrogen bonding on the other side (C–H ···O 165.9
◦
)
as shown in Fig. 4, to form an extended structure. The implications
of the observed structural similarities and differences between the
structure of 1 and 2 on the observed magnetic behaviour of these
complexes are described below.
Magnetic properties of 1 and 2. The temperature dependence
of the susceptibility v (where v is M/H)for1 and 2 are shown
in Fig. 5. The magnetic behaviors of both compounds are very
similar in spite of the observed structural differences. At room
temperature, these compounds possess a very weak paramagnetic
susceptibility that highlights the presence of strong antiferro-
magnetic interaction between copper centers. On lowering the
temperature, the susceptibilities slowly decrease until 150 K where
they reach a minimum. Below 150 K, the susceptibilities increase
3
Fig. 4 Hydrogen bonding pattern in [Cu(dppi-H)(bdmpp)]
2
(dppi)
2
(2).
Fig. 5 Temperature dependence of the magnetic susceptibility for 1 ()
and 2 (䊊) at 1000 G. The solid lines show the best fits obtained. Inset:
Zoom v vs T data in the 100–300 K range for 1 ()and2 (䊊) at 1000 G.
obeying the Curie law. On the basis of the structures described
above, the magnetic properties of these complexes should be
described as a dimer of S = 1/2. The theoretical expression for
the magnetic susceptibility of an antiferromagnetic coupled S =
1/2 dimer is given by the Bleaney–Bowers equation (eqn (1)).
15
v
dimer
=
2Ng
2
l
2
B
k
B
T(3 + exp(−2J/ k
B
T))
(1)
where N is the Avogadro number, l
B
is the magnetic moment and
k
B
is the Boltzmann constant, g is the Land
´
e factor of the given
Cu
2
complex, and J is the magnetic exchange constant between
paramagnetic ions (S = 1/2) in the dimer (H =−2JS
1
·S
2
). A
second superposed magnetic contribution is observed (Fig. 5);
the increase in the susceptibility value in the lowest temperature
region is often seen in these dimer systems (that present a singlet
ground state). This Curie type behavior (second part of eqn (2)) is
associated with the presence of small amounts of paramagnetic
impurities. Therefore the experimental susceptibility has been
fitted to the following expression:
v = (1 − q)v
dimer
+ q
g
imp
Nl
2
B
S
imp
(S
imp
+ 1)
3k
B
T
+ v
dia
(2)
where v
dia
is a fixed diamagnetic contribution calculated from the
Pascal’s constant,
16
q is the fraction of paramagnetic impurity
(with g
imp
= 2andS
imp
= 1/2). The experimental data fit very
well to eqn (2) (Fig. 5, solid line) and the best set of parameters
obtained is J/k
B
=−620(5) K (431(4) cm
−1
)and−685(5) K
(476(4) cm
−1
), g = 2.08(1) and 2.08(1), q = 0.010 and 0.015, for 1
and 2, respectively. These results show that these complexes possess
a diamagnetic ground state, i.e. S = 0. Hence the enhancement of
the susceptibility above 150 K is due to the thermal population
of the triplet excited spin state (S = 1). Within this type of Cu(
II)
dimer unit, the magnetic interaction is governed by the overlap
between the d
x
2
−y
2
orbital of the copper ions and the p orbitals
of the oxygen linkers. Therefore the magnitude of the Cu ···Cu
interaction depends strongly on the Cu–O–Cu angles and the co-
planarity of Cu–O
2
–Cu bridge.
17
Therefore the contribution of
the O–P–O bridges to J in 1 is difficult to evaluate due to the
difference on the Cu–O–Cu angles between 1 and 2. Nevertheless,
it is important to note that the obtained values of the magnetic
interactions are close to those reported for related compounds
(Table 1).
Polymerization of 2,6-dimethylphenol using 1 and 2 as cata-
lysts. The Cu(
II) catalyzed polymerization of 2,6-dimethylphenol
(DMP) leading to the formation of poly(phenylene ether) (PPE),
an industrial high performance thermoplastic, and small amount
of a dimeric product, 4-(3,5-dimethyl-4-oxo-2,5-cyclohexadienyl-
idene)-2,6-dimethyl-2,5-cyclohexadienone (DPQ), has been
known for over four decades.
18
Historically, a varying combination
of copper salts and amines viz. tertiary alkylamines, pyridine,
or imidazole derivatives have been employed for the industrial
scale production of PPE.
19
Recent mechanistic proposals suggest
a dinuclear copper-phenoxo species as the key intermediate in
the polymerization reaction
20
and therefore it is expected that
dinuclear copper complexes like 1 and 2 can be better catalysts
and give higher selectivity in PPE over the dimer DPQ.
The catalytic activity of both the new complexes
[Cu(bdmpp)(ppi)]
2
(1) and [Cu(bdmpp)(dppi-H)]
2
(dppi)
2
(2)
in the oxidative polymerization reaction of 2,6-dimethylphenol,
has been investigated (Scheme 2). For the sake of comparison,
we also investigated the catalytic activity of two other related
mononuclear copper complexes previously prepared in our
laboratory.
4d
The results of polymerization studies carried out are
summarized in Table 2. Both the complexes 1 and 2 are found to
be excellent catalysts yielding high molecular weight PPE along
with a very small amount of undesired DPQ (1% for complex 1
Table 2 Polymerization of 2,6-dimethylphenol catalyzed by different copper complexes
Catalyst Molar ratio Substrate : Cat. Time/h %PPE %DPQ M
w
M
n
PDI
[Cu
2
(bdmpp)
2
(ppi)]
2
·CH
2
Cl
2
.2H
2
O 1 : 0.05 5 99.0 1.0 14 700 7100 2.0
[Cu(bdmpp)(dppi-H)]
2
(dppi)
2
1 : 0.05 5 97.5 2.5 19 100 10 600 1.8
[Cu(imz)
4
(dtbp)
2
] 1 : 0.05 5 94.6 5.4 271 000 56 000 4.8
[Cu(dpp)
2
(bpy)(H
2
O)]
a
1 : 0.05 5 95.8 4.2 11 900 2320 5.1
a
dpp = diphenylphosphate
4
![Fig. 1 ORTEP of [Cu(ppi)(bdmpp)]2 (1) (hydrogen atoms removed for clarity).](/figures/fig-1-ortep-of-cu-ppi-bdmpp-2-1-hydrogen-atoms-removed-for-235papgi.webp)
![Fig. 4 Hydrogen bonding pattern in [Cu(dppi-H)(bdmpp)]2(dppi)2 (2).](/figures/fig-4-hydrogen-bonding-pattern-in-cu-dppi-h-bdmpp-2-dppi-2-2-35jwg0fl.webp)
![Fig. 3 ORTEP of the cationic part of [Cu(dppi-H)(bdmpp)]2(dppi)2 (2) (C–H hydrogen atoms removed for clarity).](/figures/fig-3-ortep-of-the-cationic-part-of-cu-dppi-h-bdmpp-2-dppi-2-27d4muq6.webp)