Hydrothermal syntheses, structures and properties of terephthalate-bridged polymeric complexes with zig-zag chain and channel structures

Article abstract of DOI:10.1039/b102888j

Hydrothermal reactions of 1,4-dicyanobenzene, phen and M(OAc)₂ (M = Co, Zn) or MCl₂ (M = Cu, Mn, Cd) gave rise to the hydrolysis of 1,4-dicyanobenzene into terephthalic acid and the formation of five polymeric complexes [M(ta)(phen)(

Full text of DOI:10.1039/b102888j

View Article Online / Journal Homepage / Table of Contents for this issue
Hydrothermal syntheses, structures and properties of
terephthalate-bridged polymeric complexes with
zig-zag chain and channel structures
Daofeng Sun, Rong Cao,* Yucang Liang, Qian Shi, Weiping Su and Maochun Hong*
State Key Laboratory of Structural Chemistry, Fujian Institute of the Research on the Structure
of Matter, Chinese Academy of Sciences, Fujian, Fuzhou 350002, China
Received 29th March 2001, Accepted 11th June 2001
First published as an Advance Article on the web 31st July 2001
Hydrothermal reactions of 1,4-dicyanobenzene, phen and M(OAc)2 (M = Co, Zn) or MCl2 (M = Cu, Mn, Cd) gave
rise to the hydrolysis of 1,4-dicyanobenzene into terephthalic acid and the formation of five polymeric complexes
[M(ta)(phen)(H2O)]n (M = Co, 1; Zn, 3), [Cu(ta)(phen)]n (2), [Mn(ta)(phen)]n (4), [Cd(ta)1/2(phen)Cl]n (5)
(ta = terephthalate, phen = 1,10-phenanthroline). Crystal structure analyses revealed that complexes 1–3 have 1D zig-
zag chain structures, and 4 and 5 possess 3D network structures with channels. Magnetic susceptibility measurements
for 1, 2 and 4 were consistent with the occurrence of weak antiferromagnetic interaction through the ta bridge.
Introduction
With the aim of understanding the correlation between struc-
ture and magnetism, much attention has been paid to the archi-
tecture of polymeric magnetic systems, especially two- and
three-dimensional systems.1 The key step of the architecture of
polymeric transition metal complexes is to select suitable multi-
dentate bridging ligands.2 Accordingly, multi-carboxylic ligands
with suitable spacers are good choices for such architectures,
especially benzoic acid-based ligands;3–5 for instance, tereph-
thalates (ta) have been used in many synthetic systems because
they can form short bridges via one carboxylato end or long
bridges via the benzene ring, leading a great variety of struc-
tures.6,7 Scheme 1 summarizes some typical coordination modes
Scheme 1 Typical coordination modes for the ta ligand.
for ta.2
Although a series of polymeric complexes containing the
57.02; H, 3.32; N, 6.65%. Found: C, 57.03; H, 3.41; N, 6.76%.
ta ligand has been reported, uncharacterised polymers were
IR (KBr, cmϪ1): 3385(br, vs), 1572(vs), 1369(vs), 1147(m),
839(s), 812(s), 723(s).
often obtained in conventional solution reactions. Considering
the advantage of hydrothermal synthesis,8–10 we recently began
work on the construction of polymeric complexes by hydro-
[Cu(ta)(phen)]n 2. A mixture of CuCl2ؒ2H2O (0.08 g, 0.5
thermal reactions of ta, phen and metal salts; unfortunately,
mmol), 1,4-dicyanobenzene (0.065 g, 0.5 mmol), 1,10-
only uncharacterised precipitates or microcrystals were isol-
phenanthroline (0.1 g, 0.5 mmol) and H2O (16 ml) in the molar
ated. Because the cyano group on the phenyl ring is easily trans-
ratio ca. 1 : 1 : 1 : 1777 was sealed in a 25 ml stainless-steel
formed to a carboxylic group via hydrolysis reaction,6,11 we
reactor with Teflon liner and was heated at 130 ЊC for 72 h.
After cooling, block-shaped, black–blue crystals of 2 were col-
lected by filtration. Yield: 65%. Anal. Calc. for C20H12CuN2O4:
replaced terephthalate by 1,4-dicyanobenzene and isolated a
dinuclear complex [Cu2(phen)4(ta)][ClO4]2 from the reaction
involving Cu(ClO4)2.12 After introducing metal halides and
C, 58.89; H, 2.94; N, 6.87%. Found: C, 58.72; H, 2.94; N, 6.98%.
acetates into the reactions, five polymeric complexes [Co(ta)-
(phen)(H2O)]n 1, [Cu(ta)(phen)]n 2, [Zn(ta)(phen)(H2O)]n 3,
[Mn(ta)(phen)]n 4, and [Cd(ta)1/2(phen)Cl]n 5 were successfully
IR (KBr, cmϪ1): 3385(br, vs), 1614(vs), 1605(vs), 1516(s),
1421(s), 1371(s), 1344(vs), 850(m), 825(s), 742(s), 723(m).
isolated. Reported herein are the details of their syntheses and
characterizations.
[Zn(ta)(phen)(H2O)]n 3. A mixture of Zn(CH3CO2)2ؒ2H2O
(0.11 g, 0.5 mmol), 1,4-dicyanobenzene (0.065 g, 0.5 mmol),
1,10-phenanthroline (0.1 g, 0.5 mmol) and H2O (16 ml) in the
molar ratio of ca. 1 : 1 : 1 : 1777 was sealed in a 25 ml stainless-
steel reactor with Teflon liner and was heated at 170 ЊC for 72 h.
After cooling, block, colorless crystals of 3 were collected by
filtration. Yield: 59%. Anal. Calc. for C20H14ZnN2O5: C, 56.17;
H, 3.27; N, 6.54%. Found: C, 56.22; H, 3.30; N, 6.63%. IR
(KBr, cmϪ1): 3385(br, vs), 1612(vs), 1391(s), 1358(s), 1284 (m),
1038(m), 768(s), 517(m).
Experimental
Preparation of complexes
[Co(ta)(phen)(H2O)]n 1. A mixture of Co(CH3CO2)2ؒ4H2O
(0.25 g, 1 mmol), 1,4-dicyanobenzene (0.13 g, 1 mmol), 1,10-
phenanthroline (0.4 g, 2 mmol) and H2O (16 ml) in the molar
ratio of ca. 1 : 1 : 2 : 889 was sealed in a 25 ml stainless-steel
reactor with Teflon liner and was heated at 170 ЊC for 72 h.
After cooling, rhombic, violet–red crystals of 1 were collected
by filtration. Yield: 70%. Anal. Calc. for C20H14CoN2O5: C,
[Mn(ta)(phen)]n 4. A mixture of MnCl2ؒ4H2O (0.1 g, 0.5 mmol),
1,4-dicyanobenzene (0.07 g, 0.5 mmol), 1,10-phenanthroline
DOI: 10.1039/b102888j
J. Chem. Soc., Dalton Trans., 2001, 2335–2340
This journal is © The Royal Society of Chemistry 2001
2335

View Article Online
Table 1 Structure determination summary for 1–5
Compound
Empirical formula
Formula weight
Space group
1
2
3
4
5
C20H14CoN2O5
421.27
P1
C20H12CuN2O4
407.86
Cc
C20H14ZnN2O5
427.70
C20H12MnN2O4
399.26
Cc
C16H10CdClN2O2
410.13
C2/c
P1
a/Å
b/Å
c/Å
α/Њ
9.279(1)
10.3777(9)
11.271(3)
112.207(2)
94.849(3)
114.170(4)
879.2(2)
2
15.286(7)
10.143(3)
10.490(4)
90
93.77(3)
90
1622.9(11)
4
1.669
9.1894(4)
10.5765(5)
11.5636(6)
114.6210(10)
91.562(2)
115.1470(10)
895.91(7)
2
17.510(6)
10.394(3)
9.373(4)
90
95.88(2)
90
1696.9(10)
4
1.563
23.099(3)
10.5089(5)
16.804(2)
90
133.783(3)
90
2844.6(4)
8
1.915
β/Њ
γ/Њ
V/Å3
Z
Dc/g cmϪ3
T /K
1.591
193(2)
1.013
1.585
293(2)
1.406
293(2)
1.377
293(2)
0.808
193(2)
1.732
µ/mmϪ1
λ/Å (Mo-Kα)
F(000)
2θ range/Њ
Independent reflections
Observed reflections
Variables
R(Fo)
0.71069
430
4–55
3961
3450
253
0.031
0.041
2.52
0.71073
828
2.41–25.03
1999
1998
244
0.0405
0.0798
1.063
0.71073
436
2.00–25.03
3135
3134
309
0.0444
0.0557
1.053
0.840, Ϫ0.680
0.71073
812
2.28–25.05
1706
1705
244
0.0454
0.1056
1.054
0.71069
1608
4–55
3416
3178
199
0.015
0.025
1.89
Rw(Fo)
S
Largest difference peak and hole/e ÅϪ3 0.39, Ϫ0.21
0.312, Ϫ0.261
0.442, Ϫ0.424
0.33, Ϫ0.29
(0.1 g, 0.5 mmol) and H2O (16 ml) in the molar ratio of ca.
1 : 1 : 1 : 1777 was sealed in a 25 ml stainless-steel reactor with
Teflon liner and was heated at 170 ЊC for 72 h. After cooling,
prism-shaped yellow crystals of 4 were collected by filtration.
Yield: 57%. Anal. Calc. for C20H12MnN2O4: C, 60.17; H, 3.03;
N, 7.02%. Found: C, 60.76; H, 3.14; N, 7.28%. IR (KBr, cmϪ1):
3385(br, vs), 1599(vs), 1387(vs), 847(m), 806(m), 750(s), 729(m),
505(m).
coordinates of the non-hydrogen atoms. The positions of H
atoms were generated geometrically, assigned isotropic thermal
parameters, and allowed to ride on their parent carbon atoms
before the final cycle of refinement. The structure was refined
by full-matrix, least squares minimization of Σ(Fo Ϫ Fc)2 with
anisotropic thermal parameters for all atoms except the H
atoms. For 1 and 5, the structures were solved by direct methods
and all calculations were performed on a Silicon Graphics
computer station with the teXsan program package.14b The
positions of H atoms were generated geometrically and allowed
to ride on their parent carbon atoms. The structure was refined
by full-matrix, least squares with anisotropic thermal param-
eters for all atoms except the H atoms. Table 1 summarizes the
important crystal data and Table 2 gives the selected bond
lengths and angles for 1–5.
[Cd(ta)1/2(phen)Cl]n 5. A mixture of CdCl2ؒ5H2O (0.12 g,
0.5 mmol), 1,4-dicyanobenzene (0.07 g, 0.5 mmol), 1,10-
phananthroline (0.1 g, 0.5 mmol) and H2O (16 ml) in the molar
ratio of ca. 1 : 1 : 1 : 1777 was sealed in a 25 ml stainless-steel
reactor with Teflon liner and was heated at 170 ЊC for 72 h.
After cooling, block, colorless crystals of 5 were collected by
filtration. Yield: 60%. Anal. Calc. for C16H10CdN2O2Cl: C,
46.86; H, 2.46; N, 6.83%. Found: C, 46.94; H, 2.46; N, 6.86%.
IR (KBr, cmϪ1): 3385(br, vs), 1570(vs), 1398(vs), 1144(m),
858(s), 744(s), 729(s), 507(m).
CCDC reference numbers 151561–151565.
See http://www.rsc.org/suppdata/dt/b1/b102888j/ for crystal-
lographic data in CIF or other electronic format.
Physical measurements
Crystallography
The infrared spectra were taken on a Magna 750 FTIR spectro-
photometer as KBr pellets in the 4000–400 cmϪ1 region. The
elemental analyses were performed in this institute. Variable-
temperature magnetic susceptibility data for polycrystalline
samples of complexes 1, 2 and 4 were obtained in an external
field of 10.0 kG on a Quantum Design PPMS Model 6000
magnetometer from 275 to 4 K.
Single crystals of complexes 1–5 with approximate dimensions
0.25 × 0.25 × 0.15 (1), 0.50 × 0.20 × 0.16 (2), 0.40 × 0.20 × 0.10
(3), 0.40 × 0.20 × 0.20 (4), 0.20 × 0.20 × 0.15 (5) mm were used
for data collections. For 2–4, data collections were performed
on a Siemens Smart CCD diffractometer with graphite-
monochromated Mo-Kα radiation (λ = 0.71073 Å) at room
temperature. Empirical absorption corrections were applied
using the SADABS program13 for the Siemens area detector.
For 1 and 5, because the thermal parameters of some atoms
are not very well behaved, different data sets were collected on
Siemens Smart CCD diffractometer at room temperature and
Rigaku Rasa-7 CCD diffractometer at Ϫ80 ЊC. The latter data
sets were used in this paper and empirical absorption correction
was made with the 4th3D ϩ 4th2D method for the Rigaku area
detector. The structures of 2–4 were solved by direct methods
and all calculations were performed by using the SHELXL PC
program on a Legend computer.14a For 2 and 3, the coordinates
of the heavy atoms were obtained from an E-map; successive
difference Fourier syntheses gave all the coordinates of other
atoms, including H atoms. The structures were refined by full-
matrix, least squares minimizations of Σ(Fo Ϫ Fc)2 with aniso-
tropic thermal parameters for all atoms except the H atoms.
For 4, the coordinates of the heavy atoms were obtained from
an E-map; successive difference Fourier syntheses gave all the
Results and discussion
Preparation of the complexes
The reactions of 1,4-dicyanobenzene, 1,10-phananthroline and
Co(), Cu(), Zn(), Mn(), Cd() salts by a hydrothermal
method resulted in complexes 1–5 in yields from 57 to 70%.
According to their crystal structures, they are ta-bridged
polymeric complexes. It is clear that 1,4-dicyanobenzene is
hydrolyzed to terephthalic acid during the reactions, similar to
the work reported by Fun et al.6
Cano et al. have reported reactions using ta as an initial
regent in methanol: they isolated dimmeric Co(), Cu()
2336
J. Chem. Soc., Dalton Trans., 2001, 2335–2340

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for 1–5
1a
4c
Co–O(1)
Co–O(3)
Co–O(5)
2.005(1)
2.040(1)
2.061(1)
Co–N(2)
Co–N(1)
2.094(2)
2.143(2)
Mn–O(4c)
Mn–O(1)
Mn–O(3b)
2.105(10)
2.145(9)
2.144(7)
Mn–O(2a)
Mn–N(1)
Mn–N(2)
2.150(8)
2.336(11)
2.395(9)
O(3)–Co–O(1)
O(3)–Co–O(5)
O(1)–Co–O(5)
O(3)–Co–N(2)
O(1)–Co–N(2)
99.42(6)
89.05(6)
139.32(6)
92.95(6)
94.03(6)
O(5)–Co–N(2)
O(3)–Co–N(1)
O(1)–Co–N(1)
O(5)–Co–N(1)
N(2)–Co–N(1)
125.42(6)
168.01(6)
89.64(6)
89.30(6)
78.15(6)
O(4c)–Mn–O(1)
O(4c)–Mn–O(3b)
O(1)–Mn–O(3b)
O(4c)–Mn–O(2a)
O(1)–Mn–O(2a)
O(3b)–Mn–O(2a)
O(4c)–Mn–N(1)
O(1)–Mn–N(1)
97.8(3)
85.8(4)
174.69(10)
99.55(13)
84.6(3)
98.6(3)
95.1(4)
O(3b)–Mn–N(1)
O(2a)–Mn–N(1)
O(4c)–Mn–N(2)
O(1)–Mn–N(2)
O(3b)–Mn–N(2)
O(2a)–Mn–N(2)
N(1)–Mn–N(2)
86.0(4)
164.9(4)
165.6(4)
85.9(3)
89.7(4)
94.6(3)
2b
70.95(12)
Cu–O(4a)
Cu–O(2)
1.912(9)
1.948(9)
Cu–N(1)
Cu–N(2)
2.003(11)
2.032(10)
89.8(4)
5d
O(4a)–Cu–O(2)
O(4a)–Cu–N(1)
O(2)–Cu–N(1)
95.23(11)
92.2(4)
172.5(4)
O(4a)–Cu–N(2)
O(2)–Cu–N(2)
N(1)–Cu–N(2)
173.2(4)
91.4(4)
81.10(13)
Cd–O(1b)
Cd–O(2)
Cd–N(2)
2.207(1)
2.309(1)
2.345(1)
Cd–N(1)
Cd–Cl(a)
Cd–Cl
2.409(1)
2.6132(4)
2.6671(5)
3a
Zn–O(1)
Zn–O(3)
Zn–O(5)
O(1b)–Cd–O(2)
O(1b)–Cd–N(2)
O(5)–Cd–N(2)
O(1b)–Cd–Cl(a)
O(2)–Cd–Cl(a)
N(2)–Cd–Cl(a)
N(1)–Cd–Cl(a)
O(1b)–Cd–Cl
94.81(5)
171.84(5)
81.97(5)
95.83(4)
88.58(3)
91.60(4)
154.49(4)
89.68(4)
O(1b)–Cd–N(1)
O(2)–Cd–N(1)
N(2)–Cd–N(1)
O(2)–Cd–Cl
N(2)–Cd–Cl
N(1)–Cd–Cl
Cl(a)–Cd–Cl
103.89(5)
105.47(5)
70.01(5)
171.14(3)
94.54(4)
80.75(3)
83.37(1)
2.003(2)
2.023(2)
2.053(3)
Zn–N(1)
Zn–N(2)
2.109(3)
2.178(3)
O(1)–Zn–O(3)
O(1)–Zn–O(5)
O(3)–Zn–O(5)
O(1)–Zn–N(1)
O(3)–Zn–N(1)
97.44(11)
130.95(11)
87.42(12)
101.47(11)
91.60(11)
O(5)–Zn–N(1)
O(1)–Zn–N(2)
O(3)–Zn–N(2)
O(5)–Zn–N(2)
N(1)–Zn–N(2)
127.27(12)
94.70(11)
165.14(11)
91.09(13)
77.57(11)
a Symmetry transformations used to generate equivalent atoms: #1 Ϫx ϩ 1, Ϫy ϩ 3, Ϫz ϩ 1; #2 Ϫx Ϫ 1, Ϫy ϩ 2, Ϫz. b #1 x Ϫ 1/2, Ϫy Ϫ 1/2,
z Ϫ 1/2; #2 x ϩ 1/2, Ϫy Ϫ 1/2, z ϩ 1/2. c #1 x Ϫ 1/2, Ϫy Ϫ 1/2, z Ϫ 1/2; #2 x Ϫ1/2, y ϩ 1/2, z; #3 x, Ϫy, z Ϫ 1/2; #4 x, Ϫy, z ϩ 1/2; #5 x ϩ 1/2,
Ϫy Ϫ 1/2, z ϩ 1/2; #6 x ϩ 1/2, y Ϫ 1/2, z. d #1 Ϫx, y, Ϫz ϩ 1/2; #2 Ϫx, Ϫy ϩ 2, Ϫz ϩ 1; #3 Ϫx Ϫ 1/2, Ϫy ϩ 5/2, Ϫz.
and Mn() complexes accompanied by insoluble polymers.7
Recently, Chen and coworkers isolated three polymeric com-
plexes from the hydrothermal reactions of 4,4Ј-bipy, Na2ta and
metal nitrates.15 In our system, if Na2ta was used as a reactant,
the reactions resulted in insoluble precipitates or microcrystals.
It is reasonable that the unsatisfactory results were induced by
the high coordination ability of the carboxylate in ta2Ϫ, leading
the difficulty of crystal growth. When 1,4-dicyanobenzene
was used, terephthalic acid and the complexes might be
formed simultaneously, and the polymeric complexes in crys-
talline form were successfully isolated in high yield under
hydrothermal conditions.
In an attempt on the preparation of polymers with different
structures, the reactions of MCl2 [or M(OAc)2], 1,4-dicyano-
benzene and phen with different molar ratios were carried out;
however, the same results were obtained although the yields
were different, illustrating that the compounds isolated in this
reaction system are the most stable form under hydrothermal
conditions. It is interesting, when Cu(ClO4)2 was used as a
reactant, that a dinuclear copper compound [Cu2(phen)4(ta)]-
[ClO4]2, which has a similar structure to [Cu2(bipy)4(ta)]-
[ClO4]2,7 was obtained in high yield.12 Although detailed study
is still needed, the difference in the structures may be caused by
different anions in the reactions. When Cu(ClO4)2 is used, Cu2ϩ
is easily coordinated by two phen ligands and then bridged by
ta, leading the formation of the dinuclear complex owing to the
presence of non-coordinating anion [ClO4]Ϫ. When metal halide
or acetate is used as reactant, the presence of coordinating
anions ClϪ or OAcϪ may render the easy coordination of two
phen ligands to a same metal atom, and the coordinated anions
may be replaced by ta to yield the formation of polymeric struc-
tures; the presence of coordinated Cl in 5 may support this
speculation.
Fig. 1 Local coordination environment around metal atoms (a) for 1
and 3, (b) for 2.
Crystal structures
The structures of complexes 1–3 are very similar, being made
up of one-dimensional zig-zag chain structures. Complexes 1
and 3 are isostructural, one coordinated water molecule per
metal ion is present in the structure. The metal environment
[Fig. 1(a)] can be best described as highly distorted trigonal
bipyramidal with two nitrogen atoms from phen, two
J. Chem. Soc., Dalton Trans., 2001, 2335–2340
2337

View Article Online
Fig. 3 Local coordination environment around the Mn atom in 4.
Fig. 2 The 1-D chain structures along the b axis (a) for 1 and 3, (b) for 2.
carboxylato-oxygen atoms from different ta ligands and one
oxygen atom from the coordinated water molecule. One nitro-
gen [N(1)] and one carboxylato-oxygen [O(3)] occupy the axial
positions, whereas another nitrogen [N(2)] and two oxygens
[O(5), O(1)] comprise the equatorial plane; the metal
atom shows slight deviation from the plane (0.1246 Å for 1
and 0.0651 Å for 3). The axial M–N and M–Ota distances
are slightly longer than the equatorial distances, as shown by
Co–N(1) [2.143(2) Å] and Co–O(3) [2.040(1) Å] with respect to
Co–N(2) [2.094(2) Å] and Co–O(1) [2.005(1) Å] in 1, Zn–N(1)
[2.178(3) Å] and Zn–O(3) [2.023(3) Å] relative to Zn–N(2)
[2.109(3) Å] and Zn–O(1) [2.003(2) Å] in 3.
The copper atom in 2 exhibits a distorted square planar
geometry [Fig. 1(b)] with two nitrogen atoms from phen and
two carboxylato-oxygen atoms from different tas. The deviation
of the metal atom from the mean N(1)N(2)O(2)O(4a) plane is
0.0162 Å. However, the Cu–O(1) [2.734(2) Å] and Cu–O(3a)
[2.655(2) Å] distances show some weak interaction between the
copper and uncoordinated oxygen of ta ligands, which may be
viewed as a semi-chelating coordination mode.16 Thus, the
environment of the copper atom in 2 can also be described as a
somewhat distorted octahedron: the two phen nitrogens [N(1)
and N(2)] and two carboxylato-oxygens from two tas [O(2) and
O(4a)] comprise the equatorial plane, while the axial positions
are filled by the other two carboxylato-oxygens [O(1) and
O(3a)]; the axial Cu–O distances [2.734(2) and 2.655(2) Å]
are extremely elongated compared with the equatorial Cu–O
distances [1.912(9) and 1.949(9) Å].
Each ta ligand in 1–3 adopts a µ2-bridging mode (Scheme 1c)
with a typically average M–O distance, and each phen ligand
acting as a terminal ligand with a typical M–N distance and
N–M–N chelating angles. Thus, each ta ligand links two metal
centers and each metal center connects two ta ligands to form
the zig-zag chain structure (Fig. 2); the remaining coordination
positions of the metal atom are saturated by the terminal phen
ligand and the water molecule. Metal–metal distances are
11.040, 10.853 and 11.094 Å for 1–3, respectively.
Fig. 4 Packing structure along the c axis for 4. Carbon atoms of phen
are omitted for clarity.
distorted octahedron geometry (Fig. 3). The phen ligand acts as
a typical chelating ligand terminally coordinating to the Mn()
atom with the Mn–N distances ranging from 2.336(11) to
2.395(9) Å and the N–Mn–N angle being 71.95(12)Њ; the Mn–N
distances are longer than those found in [Mn2(phen)4(ta)-
(H2O)]2ϩ.7 Each ta ligand adopts a µ4-bridging mode (Scheme
1f ), linking four Mn() atoms, and each Mn() attaches to four
ta ligands. The other carboxylato end of the bridging ta ligand
in one Mn2 subunit connects to another subunit, i.e. every two
Mn2 subunits are linked together by sharing their bridging ta
ligands to form a linear chain. The ta ligands outside the chains
further inter-connect the chains both by long and short bridges:
long bridges (through the phenyl ring) connect the chains to
give rise to a two-dimensional layer containing a regular rhom-
bus, in which the apexes are occupied by Mn() atoms and the
edges are formed by ta ligands; the short bridges (carboxylato
end) connect the layers to generate the final three-dimensional
structure with rhombus channels (Fig. 4). The channel has a
Complex 4 possesses a three-dimensional network structure:
the dinuclear [Mn2(phen)2(ta)4] species with interatomic dis-
tance of 4.810(2) Å may be viewed as the basic subunit for the
structure; the Mn() atom is coordinated by two nitrogen atoms
from phen and four oxygen atoms from different ta ligands in a
2338
J. Chem. Soc., Dalton Trans., 2001, 2335–2340

View Article Online
diameter of 10.181 × 11.135 Å based on the Mn–Mn distance
and is filled by phen ligands.
for 1, 2 and 4 were performed on polycrystalline samples. For
complex 4, the χM value increases as the temperature decreases,
reaching a maximum of 0.1140 cm3 molϪ1 at about 11 K, and
then decreases on further cooling; the value of χMT is 4.2159
cm3 K molϪ1 at room temperature and decreases continuously
upon cooling; indicating the occurrence of antiferromagnetic
interaction between the metal ions (Fig. 7). According to the
structural data, the complex may exhibit two types of magnetic
exchange interactions of Mn()–Mn(): type one is through a
short bridge via the carboxolato end of ta accompanying π–π
interactions of phen along the c axis; and type two is through a
long bridge via the phenyl ring of ta. Actually, the magnetic
behavior is well interpreted based on the modified analytical
expression [eqn. (1)],17 which is derived from the Fisher model
Complex 5 also has a three-dimensional structure: the
coordination geometry of Cd() is a distorted octahedron
completed by two nitrogen atoms from phen, two chloride
atoms and two carboxylato-oxygen atoms (Fig. 5). Each phen
acts as a typical chelating ligand terminally coordinating to the
Cd() atom with the average Cd–N distance and N–Cd–N
angle being 2.377(1) Å and 70.01(5)Њ, respectively. Similar to 4,
each ta ligands adopts a µ4-bridging mode (Scheme 1f ), linking
four Cd() centers. However, it is interesting that the Cd()
atom only connects to two ta ligands, and the remaining
coordination position is attached by a Cl atom that acts as a µ2-
bridge linking two Cd atoms. Hence, the dinuclear Cd2(phen)-
Cl2 spices with an interatomic distance of 3.940(1) Å may be
viewed as the subunit of the structure: two Cd2 units are linked
head-to-tail by a short bridge (carboxylato end) of one ta
ligand, forming another subunit [(Cd2(phen)Cl2)2(ta)]; two of
the new subunits are connected face-to-face by two long-
bridged ta ligands to form rectangular pores with a cavity of
10.51 × 11.85 Å. Each ta ligand is shared by four pores, gener-
ating a two-dimensional layer (Fig. 6a), different from the
rectangular sheet in the Cd–ta–4,4Ј-bipy complex.15a The layers
are further interconnected by long-bridged ta ligands, pro-
ducing the three-dimensional structure. If viewed along the c
axis, the structure can also be regarded as rectangular channels
filled by phen ligands (Fig. 6b).
(1)
where µ1 = coth(J1/kT) Ϫ kT/J1, µ2 = coth(J2/kT) Ϫ kT/J2.
By using a least squares method, a very satisfactory fit of the
data at 5–300 K was obtained using the set of parameters of
J1 = Ϫ2.02 K, J2 = Ϫ1.69 K and g = 2.01. The goodness of fit R,
defined as
(2)
is 1.0 × 10Ϫ5. J1 may be assigned to the antiferromagnetic
exchange interaction of type one and J2 to type two, the shorter
Mn–Mn distance along with the π–π interactions of phen
generate a stronger interaction J1.
Magnetic properties
Temperature-dependent magnetic susceptibility measurements
For complexes 1 and 2, Fig. 8 shows the susceptibility curve
of χM in the full temperature range, with the 1/χ plot as an inset.
The χM value decreases as temperature decreases; the values of
χM are 0.0092 and 0.0013 cm3 molϪ1 at room temperature for 1
and 2, respectively. The χM data were fit to a modified Curie–
Weiss law χ = χ0 ϩ C/(T ϩ θ).18 The calculated values of Cm are
2.781, 0.337 and the Weiss temperatures, θ are Ϫ1.361 and
Ϫ7.190 K for 1 and 2, respectively.
Conclusion
We have systematically investigated the hydrothermal reac-
tions of MCl2 (M = Cu, Mn, Cd) or M(OAc)2 (M = Co, Zn),
1,10-phenanthroline and 1,4-dicyanobenzene, and isolated five
ta-bridged polymeric complexes in which the coordination
geometries of the metal atoms are different except for the
Co and Zn complexes. 1,4-Dicyanobenzene was hydrolyzed to
terephthalic acid, resulting in good quality single crystals
obtained in high yields for the five complexes. Also, the metal
halides and acetates are helpful for the formation of polymeric
Fig. 5 Local coordination environment around the Cd atom in 5.
Fig. 6 (a) The layer structure in 5, (b) packing structure along the c axis for 5. Carbon atoms of phen are omitted for clarity.
J. Chem. Soc., Dalton Trans., 2001, 2335–2340
2339

View Article Online
Acknowledgements
The authors thank the financial support from NNSF of China
and the Key Project from CAS.
References
1 (a) H. O. Stumf, I. Oualmb, Y. Pei, D. Grandjean and O. Kahn,
Science, 1993, 261, 447; (b) H. Miyasaka, N. Matsumoto, H. Okawa,
N. Re and E. Floriani, Angew. Chem., Int. Ed. Engl., 1995, 34, 1446;
(c) J. Kim, M. Lim, Y. K. Choi and Y. Do, Angew. Chem., Int. Ed.
Engl., 1996, 35, 993.
2 C. S. Hong, S. K. Son, Y. S. Lee, M. J. Jun and Y. Do, Inorg. Chem.,
1999, 38, 5602.
3 (a) J. Cano, G. Demunno, J. L. Sanz, R. Ruiz, J. Faus, F. Lloret,
M. Julve and A. Caneshi, J. Chem. Soc., Dalton Trans., 1997, 1915;
(b) E. G. Bakalbassis, J. Mrozaski and C. A. Tsipis, Inorg. Chem.,
1986, 25, 3684; (c) M. Verdaguer, J. Gouteron, S. Jeannin, Y. Jeannin
and O. Kahn, Inorg. Chem., 1984, 23, 4291.
Fig. 7 Experimental χM versus T and χMT versus T curves for complex 4.
4 (a) K. S. Bürger, P. Chaudhuri, K. Weighardt and B. Nuber, Chem.
Eur. J., 1995, 1, 583; (b) P. Chaudhuri, K. Oder, K. Wieghardt,
S. Gehring, W. Haase, B. Nuber and J. Weiss, J. Am. Chem. Soc.,
1988, 110, 3657.
5 (a) C. E. Xanthopoulos, M. P. Sigala, G. A. Katsoulos, C. A. Tsipis,
A. Terzis and M. Mentzafos, Inorg. Chem., 1993, 32, 5433; (b) E. G.
Bakalbassis, A. P. Bozopoulos, J. Mrozinski, P. J. Rentzeperis and
C. A. Tsipis, Inorg. Chem., 1988, 27, 529; (c) E. G. Bakalbassis,
P. Bergerat, O. Kahn, S. Jeannin, Y. Jeannin, Y. Dromzee and
M. Guillot, Inorg. Chem., 1992, 31, 625.
6 H. K. Fun, S. S. S. Raj, R. G. Xiong, J. L. Zuo, Z. Yu and X. Z. You,
J. Chem. Soc., Dalton Trans., 1999, 1915.
7 J. Cano, G. D. Munno, J. L. Sanz, R. Ruiz, J. Faus, F. Lloret,
M. Julve and A. Caneschi, J. Chem. Soc., Dalton Trans., 1997, 1915
and references therein.
8 R. M. Barrer, The Hydrothermal Chemistry of Zeolites, Academic
Press, London, 1982.
9 P. J. Hagrman, D. Hagrman and J. Zubieta, Angew. Chem., Int. Ed.,
1999, 38, 2638.
10 (a) A. Rabenau, Angew. Chem., Int. Ed. Engl., 1985, 24, 1026;
R. A. Ludise, Chem. Eng. News, 1987, (Sept 28), 30; (b) J. Y. Lu,
M. L. Lawandy, J. Li, T. Yuen and C. L. Lin, Inorg. Chem., 1999,
38, 2695 and references therein.
11 O. R. Evans, R. G. Xiong, Z. Wang, G. K. Wong and W. Lin, Angew.
Chem., Int. Ed. Engl., 1999, 38, 536.
12 D. Sun, R. Cao, Y. Liang, M. Hong, W. Su and J. B. Weng, Acta
Crystallogr., Sect. C, 2000, 56, e240.
13 G. M. Sheldrick, SADABS, Program for Empirical Absorption
Correction of Area Detector Data, University of Gottingen,
Germany, 1996.
14 (a) G. M. Sheldrick SHELXL 93, Program for Crystal Structure
Refinement, University of Gottingen, Germany, 1993; (b) teXsan:
Crystal Structure Analysis Package, Molecular Structure
Corporation, The Woodlands, TX, 1985 and 1992.
15 (a) J. Tao, M. L. Tong and X. M. Chen, J. Chem. Soc., Dalton Trans.,
2000, 3669; (b) S. M. F. Lo, S. S. Y. Chui, L. Y. Shek, Z. Lin, X. X.
Zhang, G. Wen and I. D. Williams, J. Am. Chem. Soc., 2000, 122,
6293.
16 G. Guilera and J. W. Steed, Chem. Commun., 1999, 1563.
17 R. Cortes, M. Drillon and S. Solans, Inorg. Chem., 1997, 36, 677.
18 X. X. Zhang, S. S.-Y. Chui and L. D. Williams, J. Appl. Phys., 2000,
87, 6007.
Fig. 8 Experimental χ versus T curves; insert is a plot of 1/χ versus T
data (a) for 1, (b) for 2.
structures. The magnetic property studies suggest that metal–
metal exchange interactions in Co, Cu and Mn complexes are
antiferromagnetic, although in the cases of Co and Cu com-
plexes this may be more complicated.
2340
J. Chem. Soc., Dalton Trans., 2001, 2335–2340