Polytypism, homochirality, interpenetration, and hydrogen-bonding in transition metal (Mn(II), Ni(II), Cu(II), Zn(II)) 5-hydroxyisophthalate coordination polymers containing 4,4′-bipyridyl

Article abstract of DOI:10.1039/b716454h

We report the synthesis of five coordination polymers of divalent transition metals combined with 5-hydroxyisophthalic acid (HIP) and 4,4′-bipyridyl (bipy). Mn forms two polytypic two-dimensional coordination polymers, Mn(HIP)(bipy)?3H₂O and Mn(HIP)(bipy)?1/ 4bipy?2H₂O (I and II, with ABAB and ABC stacking sequences, respectively); Ni forms a chiral hexagonal three-dimensional coordination polymer with two interpenetrating trigonal sublattices exhibiting the dual quartz topology, Ni(HIP)(bipy)(H₂O) (III); Cu forms a one-dimensional coordination polymer containing arrays of infinite hydrogen-bonded molecular ribbons, Cu(HIP)₂(bipy) (IV); and Zn forms a two-dimensional coordination polymer with a stair-stepped layered structure, Zn₂(HIP)₂(bipy)(H₂O)₂?H ₂O (V). The M-HIP-M connections are perpendicular to the M-bipy-M connections in all structures where they are present. The Royal Society of Chemistry.

Full text of DOI:10.1039/b716454h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Polytypism, homochirality, interpenetration, and hydrogen-bonding in
transition metal (Mn(II), Ni(II), Cu(II), Zn(II)) 5-hydroxyisophthalate
coordination polymers containing 4,4ꢀ-bipyridyl†‡
Russell K. Feller and Anthony K. Cheetham*
Received 31st October 2007, Accepted 8th January 2008
First published as an Advance Article on the web 19th February 2008
DOI: 10.1039/b716454h
We report the synthesis of five coordination polymers of divalent transition metals combined with
5-hydroxyisophthalic acid (HIP) and 4,4ꢀ-bipyridyl (bipy). Mn forms two polytypic two-dimensional
1
4
coordination polymers, Mn(HIP)(bipy)·3H2O and Mn(HIP)(bipy)· bipy·2H2O (I and II, with ABAB
and ABC stacking sequences, respectively); Ni forms a chiral hexagonal three-dimensional
coordination polymer with two interpenetrating trigonal sublattices exhibiting the “dual quartz”
topology, Ni(HIP)(bipy)(H2O) (III); Cu forms a one-dimensional coordination polymer containing
arrays of infinite hydrogen-bonded molecular ribbons, Cu(HIP)2(bipy) (IV); and Zn forms a
two-dimensional coordination polymer with a stair-stepped layered structure,
Zn2(HIP)2(bipy)(H2O)2·H2O (V). The M–HIP–M connections are perpendicular to the M–bipy–M
connections in all structures where they are present.
markedly different structures in the corresponding coordination
polymers.
Introduction
Hybrid inorganic–organic framework structures have attracted
widespread attention in recent years. Since 1990, many com-
pounds of this type have been discovered which also incorporate
Experimental
polyfunctional organic molecules as key structural elements,
bridging the metal centers via covalent bonds.1 The majority
Materials
All reagents were used as purchased; Mn(OAc)2·4H2O (99+%),
of these compounds are prepared through mild hydrothermal
or solvothermal syntheses, at temperatures below 200 ◦C. The
Zn(OAc)2·2H2O (98+%), Ni(OAc)2·4H2O (98%), Cu(OAc)2·H2O
(98+%) and 5-hydroxyisophthalic acid (97%) from Aldrich, and
4,4ꢀ-bipyridyl (98%) from Alfa Aesar. Syntheses were carried out
in 23 mL Parr Teflon-lined stainless-steel autoclaves under auto-
genous pressure. Elemental analysis (C, H, N) was carried out by
the Marine Sciences Institute Analytical Laboratory at UCSB.
flourishing diversity of structures and dimensionalities seen in
these coordination polymers has caused them to grow into a major
field of research.2
5-Hydroxyisophthalic acid has proven to be an effective con-
necting ligand in the construction of coordination polymers. It is
known to form such compounds with alkaline earths,3 transition
metals4 and lanthanides.5 These structures often contain kinked
chain motifs due to the molecule’s bent rod-like geometry. By
incorporating capping ligands such as 2,2ꢀ-bipyridyl, 2-(pyrazol-
3-yl)pyridine or 1,10-phenanthroline to restrict connectivity in
such systems,4d,6 or rigid linkers such as 4,4ꢀ-bipyridyl or 1,2-bis(4-
pyridyl)ethene to increase connectivity,7 compounds with many
different framework topologies have been prepared. In the present
work, coordination polymers containing divalent Mn, Ni, Cu or
Zn combined with 5-hydroxyisophthalate (HIP) and 4,4ꢀ-bipyridyl
(bipy), were synthesized with differing stoichiometries and ioniza-
tion states of 5-hydroxyisophthalic acid. Subtle differences in the
preferred coordination geometries of these four metals have led to
Synthesis of I and II
Yellow blocks of Mn(HIP)(bipy)·3H2O (I) were obtained by
heating a mixture of 1 mmol Mn(OAc)2·4H2O, 1 mmol 5-
hydroxyisophthalic acid, 0.5 mmol 4,4ꢀ-bipyridyl and 5 mL
deionized water for 2 days at 180 ◦C. The product was cooled
to room temperature, vacuum-filtered and rinsed with deionized
1
4
water. Yellow blocks of Mn(HIP)(bipy)· bipy·2H2O (II) were
obtained similarly, but using 1 mmol 4,4ꢀ-bipyridyl in the mixture.
Higher concentrations of 4,4ꢀ-bipyridyl in the reaction mixture
favor the crystallization of II over I because II can accept guest
4,4ꢀ-bipyridine molecules. Phase-pure samples of I could not be
obtained, so CHN analysis was not performed on this material.
Subsequently, a phase-pure powder of II, as confirmed by powder
X-ray diffraction, was obtained in ∼90% yield by using an excess
(1.5 mmol) of 4,4ꢀ-bipyridyl, and additionally rinsing with acetone.
CHN analysis of this powder suggests that the bipy content of the
bulk powder used for the analysis may be slightly greater than that
in the single crystal, in agreement with the difference in synthesis
conditions. Found (wt%): C, 56.01%; H, 3.50%; N, 8.02%. Calc’d
(wt%): C, 52.80%; H, 3.90%; N, 7.51%.
Materials Research Laboratory, University of California, Santa Barbara,
CA 93106-5121, USA. E-mail: cheetham@mrl.ucsb.edu
† CCDC reference numbers 657326–657330. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b716454h
‡ Electronic supplementary information (ESI) available: Tables of bond
lengths and angles, anisotropic displacement parameters, hydrogen coor-
dinates and hydrogen bond parameters; and powder diffractograms and
thermal analysis for II, IV and V. See DOI: 10.1039/b716454h
2034 | Dalton Trans., 2008, 2034–2042
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Synthesis of III
refined to within three E.S.D.’s of zero, and were thus removed
from the refinements. All non-hydrogen atoms were refined
anisotropically. The Flack parameter for III was −0.05(2).11
CCDC deposition numbers: 657326 for I, 657327 for II, 657328
for III, 657329 for IV, and 657330 for V. See Table 1 for crystal
data and structure refinement parameters.
Bright blue hexagonal pyramids of Ni(HIP)(bipy)(H2O) were
obtained by heating a mixture of 1 mmol Ni(OAc)2·4H2O, 1 mmol
5-hydroxyisophthalic acid, 0.5 mmol 4,4ꢀ-bipyridyl and 5 mL
deionized water for 2 days at 200 ◦C. The product was cooled
to room temperature, vacuum-filtered and rinsed with deionized
water. A yield of several isolated crystals (<5% of the solid
product) was obtained. A slightly better yield (∼10% of the solid)
of polycrystalline clusters was obtained at 220 ◦C using either
0.5 mmol or 1 mmol 4,4ꢀ-bipy, but III still remained a minority
phase. No crystals of the light blue majority phase large enough for
structure determination were obtained, and its powder diffraction
pattern could not be matched. Elemental analysis and a powder
diffractogram of III are not available as a phase pure sample was
not synthesized.
In I and II, riding hydrogens were added to the phenyl carbons,
and water hydrogens were not located due to disorder, as described
below. One of the two rings of the uncoordinated bipy molecule in
II was restrained to planarity, with the other ring being crystallo-
graphically equivalent. An anti-bumping restraint was placed on
its central C–C bond, with other bond distances restrained to be
similar to those in the coordinated bipy molecule. C25, OW2 and
OW4 were lightly restrained to approximate isotropic behavior.
In III, riding hydrogens were added to the phenyl carbons but
the water and phenol hydrogen positions were refined, with all
three O–H distances restrained to be similar and one thermal
parameter used for both water hydrogens. In IV, riding hydrogens
were added to the phenyl carbons, but the phenol and carboxyl
hydrogens were freely refined. In V, all hydrogen atom positions
were refined, with some C–H and O–H distances restrained to
chemically reasonable values. One thermal parameter was refined
for each set of chemically equivalent hydrogens, with those of
the coordinated water molecule differentiated from those of the
uncoordinated water molecule. Atoms of the uncoordinated water
molecule were given occupancy factors of 1/2, as this molecule
appears disordered across an inversion center, as described
below.
The presence of domains of I within II and vice versa due
to stacking faults, along with severe disorder in the guest water
molecules, complicated the structural solutions and refinements
somewhat, with II being more affected. Intensity data collected at
120 K offered no improvement in the final refinement statistics nor
in the location of the disordered water. In addition, because crystal
mosaicity appeared to have increased after cooling, the room
temperature data were used. It was found that refinement statistics
were improved by performing psi-scan absorption corrections
rather than multi-scan corrections.
Synthesis of IV
Deep purple plates of Cu(HIP)2(bipy) were obtained by heating a
mixture of 1 mmol Cu(OAc)2·H2O, 1 mmol 5-hydroxyisophthalic
acid, 1 mmol 4,4ꢀ-bipyridyl and 5 mL deionized water for 2 days
at 150 ◦C. The product was cooled to room temperature, vacuum-
filtered and rinsed with deionized water. Subsequently a phase-
pure powder, as confirmed by powder X-ray diffraction, was
obtained in ∼90% yield by heating 0.75 mmol Cu(OAc)2·H2O,
1.5 mmol 5-hydroxyisophthalic acid, 0.75 mmol 4,4ꢀ-bipyridyl and
5 mL deionized water at 125 ◦C, and additionally rinsing with
acetone. Found (wt%): C, 53.04%; H, 3.05%; N, 4.59%. Calc’d
(wt%): C, 53.65%; H, 3.12%; N, 4.81%.
Synthesis of V
Pale tan blocks of Zn2(HIP)2(bipy)(H2O)2·H2O were obtained in
∼90% yield by heating a mixture of 1 mmol Zn(OAc)2·2H2O,
1 mmol 5-hydroxyisophthalic acid, 0.5 mmol 4,4ꢀ-bipyridyl and
5 mL deionized water for 2 days at 200 ◦C. The product was cooled
to room temperature, vacuum-filtered and rinsed with deionized
water and then acetone. Powder X-ray diffraction confirmed that
the product is phase-pure under these conditions, and remains so
when treated at temperatures as low as 125 ◦C. Found (wt%): C,
44.30%; H, 2.99%; N, 3.73%. Calc’d (wt%): C, 44.53%; H, 3.17%;
N, 4.00%.
Thermal analysis
TGA/DTA analysis of II, IV and V was performed in air on
a Mettler 851e TG/sDTA up to 600 ◦C. Temperatures were
◦
ramped at 5 C min−1. For II, a gradual mass loss of 3.1%
(7.7% expected for loss of guest water, or 8.4% for loss of guest
bipy), occurred between 100–200 ◦C, accompanied by a very slight
endotherm in the DTA trace. Between 200–400 ◦C, but mostly
after 300 ◦C, a further mass loss of 80.0% occurred, with a large
Structure determinations†
Suitable single crystals were selected under a polarizing micro-
scope and glued to glass fibers. Intensity data were collected at
room temperature on a Siemens SMART CCD diffractometer
◦
corresponding exotherm in the DTA beginning at 350 C. After
˚
(Mo Ka radiation, k = 0.71073 A). The structures were solved
both events, 16.9% of the original mass remained (15.2% expected
for complete transformation to MnO, or 16.4% for Mn3O4).
For IV, a single sharp mass loss of 84.7% occurred between
275–350 ◦C, with a sharp corresponding exotherm in the DTA
beginning at 310 ◦C. After this event, 15.3% of the original mass
remained (13.7% expected for complete transformation to CuO).
It was found that IV expands significantly during decomposition,
which limited the sample size that could be safely used in the
instrument. Accordingly, the apparent mass gain due to sample
buoyancy (∼1.4%) was taken into account in calculating the
using direct methods and difference Fourier synthesis, and refined
against |F|2 using the SHELXTL software package.8 IV was
initially solved in the Cc space group and V was initially
solved in the P1 space group, and both structures were sub-
sequently transformed into the corresponding centrosymmetric
space groups using the PLATON software package.9 Psi-scan
absorption corrections for I and II were made using the XPREP
module of SHELXTL, and multi-scan absorption corrections for
III–V were made using SADABS.10 The extinction coefficients all
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2034–2042 | 2035
©

View Article Online
Table 1 Crystal data and structure refinement parameters
Compound
I
II
III
IV
V
Abbreviated formula
Mn(HIP)(bipy)·3H2O
Mn(HIP)(bipy)·
Ni(HIP)(bipy)(H2O)
Cu(HIP)2(bipy)
Zn2(HIP)2(bipy)-
(H2O)2·H2O
701.20
298(2)
Triclinic
¯
P1
7.4720(14)
9.6175(18)
10.1060(19)
100.587(3)
97.165(3)
112.381(3)
644.8(2)
1
1 bipy·2H2O
4
Formula weight
T/K
445.28
298(2)
Monoclinic
P2/c
10.1726(11)
11.6397(13)
16.2535(18)
90
103.310(3)
90
1872.8(4)
4
1.579
466.31
298(2)
Monoclinic
P2/n
11.6900(8)
10.1772(7)
17.9098(12)
90
102.326(2)
90
2081.6(2)
4
1.488
413.02
298(2)
Hexagonal
581.96
298(2)
Monoclinic
C2/c
10.118(3)
11.051(3)
21.051(5)
90
99.029(5)
90
2324.5(10)
4
1.663
Crystal system
Space group
P61
˚
a/A
11.2217(7)
11.2217(7)
25.0290(16)
90
90
120
2729.6(3)
6
1.508
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
Z
qcalc/g cm−3
l/mm−1
1.806
1.938
0.756
0.681
1.103
1.007
F000
916
958
1272
1188
356
Crystal size/mm
0.20 × 0.18 × 0.15
2.06 to 26.91
15276
3893
0.0971
96.0
Psi-scan
320
0.929
0.0581, 0.1313
0.18 × 0.15 × 0.12
2.00 to 26.73
16826
4239
0.0472
95.7
Psi-scan
327
1.071
0.0728, 0.2284
0.25 × 0.15 × 0.15
2.10 to 27.68
15724
3553
0.0433
94.8
SADABS
257
1.055
0.30 × 0.10 × 0.03
1.96 to 26.73
9477
2346
0.0533
95.7
SADABS
187
1.048
0.0525, 0.1152
0.40 × 0.30 × 0.20
2.37 to 26.02
5163
2485
0.0238
97.5
SADABS
240
1.065
0.0439, 0.1041
h range/◦
Refls collected
Ind. reflections
Rint
Completeness (%)
Abs. correction
# of parameters
G.O.F. on |F|2
R1, wR2 [for all I >
2r(I)]
0.0497, 0.1137
R1, wR2 [for all
reflections]
0.1083, 0.1486
0.1040, 0.2428
0.0608, 0.1193
0.0851, 0.1316
0.0560, 0.1119
Largest diff. peak and
hole/e A
0.698, −0.325
1.234, −0.616
1.375, −0.306
0.877, −0.300
0.703, −0.321
−3
˚
percentages above, but was negligible for II and V. For V, a mass
loss of 7.2% (7.7% expected for loss of all water), occurred between
75–175 ◦C, accompanied by a small endotherm in the DTA trace.
Between 375–450 ◦C, a further mass loss of 68.4% occurred,
with◦a large corresponding exotherm in the DTA beginning at
410 C. After both events, 24.2% of the original mass remained
(23.2% expected for complete transformation to ZnO). See ESI‡
for full data.
Results and discussion
Structures of I, Mn(HIP)(bipy)·3H2O, and II,
1
4
Mn(HIP)(bipy)· bipy·2H2O
Compound I (Fig. 1) crystallizes in the space group P2/c,
and II (Fig. 2) crystallizes in P2/n. Both are two-dimensional
coordination polymers containing the same layer motif (Fig. 3),
with layers lying in (002) planes, but the layer-inversion patterns
and the directions and sequences of stack-slipping differ. Layers
consist of dimers of distorted trans-MnN2O4 octahedra, which
form eight-membered M–OCO–M–OCO rings. These dimers are
connected in one direction by pairs of HIP ligands, and in
another by bipy ligands. Similar layers are seen in Mn–bipy–
isophthalate and Cd–bipy–isophthalate frameworks which have
also been reported in P2/c with nearly identical cell parameters.12
Each octahedron has one pair of oxygens splayed apart (O–Mn–O
Fig. 1 Thermal ellipsoid plot (50% probability) of I, with the atoms of
one asymmetric unit labeled. Disordered water molecules and minority
(“B”) disorder components are omitted for clarity.
angle 124.7(1)◦ in I), and the opposite pair pinched together (O–
Mn–O angle 56.4(1)◦ in I). The pairs of HIP ligands are oriented in
such a way as to create two symmetry-equivalent C–H · · · O close
◦
˚
contacts (d(C · · · O) = 3.508(4) A in I, C–H · · · O angle = 169.8(2)
in I), possibly indicative of Ar–H · · · O hydrogen bonding. Similar
This journal is The Royal Society of Chemistry 2008
2036 | Dalton Trans., 2008, 2034–2042
©

View Article Online
Fig. 4 ABAB-type stacking of layers in I, viewed from the side, with
layers colored orange or green according to the stacking sequence.
Extra-framework molecules are omitted for clarity.
Fig. 2 Thermal ellipsoid plot (50% probability) of II, with the atoms of
one asymmetric unit labeled. Disordered water molecules are omitted for
clarity.
Fig. 5 ABC-type stacking of layers in II, viewed from the side and
to the same scale as in Fig. 4, with layers colored orange, purple or
green according to the stacking sequence. Extra-framework molecules are
omitted for clarity.
˚
in I, and 8.75 A in II, and the densities differ correspondingly:
1.579 g cm−3 and 1.488 g cm−3, respectively.
Fig. 3 A portion of one Mn(HIP)(bipy) layer as found in I and II. Pairs
of Mn2+ cations are bridged by pairs of HIP ligands in one direction, and
by pairs of bipy ligands in another.
Both I and II contain disordered uncoordinated water
molecules, which could be located at several sites between the
layers, whose occupancies were all fixed at 1/2, and some of
which are in positions that suggest interlayer hydrogen bonding.
˚
pairs of HIP ligands are seen in several of the other structures
previously referred to.4d,6
Water in I occupies 2 × 5 A channels running along c, and water
˚
in II occupies 4 × 3 A channels running along [101]. In both
Compound I exhibits ABAB stacking (Fig. 4), with layers
alternatingly shifted 1/6 along b and with stack-slipping along
a. Compound II exhibits a pseudo-ABC stacking (Fig. 5), with
a shift of nearly 1/3 per layer along a due to stack-slipping, and
180◦ rotation of every other layer. In I, the stacking slips parallel
to the M–HIP–M connections, but in II the stacking slips parallel
compounds, these channels are defined by the rectangular gaps
belonging to consecutive layers; they run perpendicular to the
layers in I, but are obliquely oriented in II due to its different
stacking sequence. The channels in II also host an uncoordinated
bipy molecule whose site occupancy was similarly fixed at 1/2;
it shares the space with the water molecules. The uncoordinated
bipy molecules stack between the coordinated bipy groups as seen
˚
to the M–bipy–M connections. The interlayer distances are 7.91 A
This journal is The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2034–2042 | 2037
©

View Article Online
in He and coworkers’ cobalt–bipy–HIP,7a but with the long axis
rotated approximately 20◦ rather than 80◦.
In I, the HIP ligands are disordered between upward and down-
ward positions, across planes perpendicular to b. The occupancy
factors for the disordered components are approximately 1/3 and
2/3. The amplitude of the disorder appears much smaller in II,
and was not resolvable into two components. This difference may
be because pairs of HIP groups from consecutive layers are much
˚
˚
farther apart in II (3.4 A face-to-face) than in I (2.9 A), where
they experience stronger van der Waals interactions. Additionally,
one ring of every bipy ligand in I is torsionally disordered
between two positions differing by 44◦ in rotation, because there
are no neighboring moieties to confine it. In II, both rings of
the bipy ligands are confined by other groups, so no torsional
disorder is present. Similar reported structures to I and II, such
as the aforementioned bipy–isophthalates, show analogous types
of disorder. Tao and coworkers’ Cd–bipy–isophthalate12b shows
significant up/down disorder in its isophthalate rings, but little
disorder of the bipy ligands, whereas Ma and coworkers’ Mn–
bipy–isophthalate12a is reported with all the isophthalate rings
in one position, but the thermal ellipsoids suggest unresolved
torsional disorder of one ring of the bipy ligands, as in I. Both
of these compounds have stacking patterns analogous to I, and
neither exhibits the 180◦ rotation of alternating layers seen in II.
Polytypism between coordination polymers has been referred
to previously, with different examples deviating by varying extents
from the strictest definition of the phenomenon. b-CuNCS, which
might be considered an extremely simple coordination polymer,
has been reported by Smith and Saunders to exhibit ABAB
and ABC polytypes reminiscent of wurtzite and sphalerite.13
Polytypism has been claimed between the chain-like compounds
diethylammonium lead(II) iodide and alaninium tin(II) iodide,14
but in addition to their largely differing chemical constituents,
these two compounds differ both in their intra-chain topology and
in their chain packing arrangement. However, multiple packings
of the same chain motif, which might be considered a form
of “rod polytypism”,15 occur both in Groves and coworkers’
lanthanum bis(phosphonomethyl)piperazinium chlorides16 as well
as Kurmoo and coworkers’ and Chen and coworkers’ nickel 1,4-
cyclohexanedicarboxylates.17 While I and II in the present work
differ partially in their guest molecules, they are the truest example
of polytypism among coordination polymers in that their principal
structural difference is in the stacking of the same layer motif.
Fig. 6 Thermal ellipsoid plot (50% probability) of III, with the atoms of
one asymmetric unit labeled.
Fig. 7 One layer of nickel–bipy chains in III, lying in a (006) plane.
from below (−c) by both oxygens of one HIP carboxyl group,
and from above (+c) by one oxygen of the other carboxyl group
of a different HIP ligand, with one aqua ligand occupying the
remaining coordination site above. Alternating layers of the Ni–
bipy chains are covalently connected by HIP ligands, with no
bonding between consecutive layers, hence two sublattices are
formed (Fig. 8). Within one sublattice, consecutive layers are
rotated 120◦, with three layers per cell, and thus each sublattice
possesses 31 symmetry. The collinearity of the two 31 axes and the
equal spacing of contributions from the two sublattices give rise to
the overall 61 symmetry. The aqua ligand donates an intralattice
hydrogen bond to the non-coordinating carboxyl oxygen, as well as
an interlattice hydrogen bond to one of the coordinating carboxyl
oxygens on the nickel cation one layer above (Fig. 9). Additionally,
the phenol group donates an interlattice hydrogen bond to the
non-coordinating carboxyl oxygen belonging to the nickel cation
one layer below. The projection of the Ni–HIP helices alone onto
the ab plane creates a Kagome-type lattice, with 6 trigonal helices
surrounding each hexagonal gap in the lattice (Fig. 10). The bipy
Structure of III, Ni(HIP)(bipy)(H2O)
Compound III (Fig. 6) crystallizes in the space group P61 and is a
three-dimensional chiral coordination polymer consisting of two
interpenetrating sublattices with only weak interactions between
them. Nickel cations sit in trans-NiN2O4 octahedra with one short
˚
˚
(2.000(3) A) and one long (2.181(3) A) Ni–O bond, and the long
bond is bent approximately 25◦ out of perpendicularity to its
neighbors. The trans coordination of the bipy ligands generates
layers of parallel Ni–bipy chains (Fig. 7) which lie in (006)
planes, while the HIP ligands connect these layers perpendicularly,
winding about the c direction. (A copper–bipy–succinate in P61
has been reported by Zheng and Kong to contain similar sheets
connected by succinate anions along the c direction, but all
layers are covalently connected.18) Each nickel is coordinated
2038 | Dalton Trans., 2008, 2034–2042
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 10 A portion of the Kagome-like lattice in III formed by nickel–HIP
helices propagating along c, with atoms color-coded by sublattice and
hydrogen atoms omitted for clarity.
Fig. 8 Packing diagram of III, with atoms color-coded by sublattice
of origin. Alternating layers of nickel–bipy chains belong to alternating
sublattices, with HIP ligands from each sublattice inserting through gaps
in the other.
pyridyl)ethyne, provides all of the connections, and in that the
sublattices differ slightly from each other.
A fivefold-interpenetrating zinc–bis(pyridyl)ethane–HIP in P41
was reported by Li et al.7b In this compound, half of the 41 axes are
encircled by atoms of only one sublattice, with sections of the other
four merely cutting through, while the other half of the 41 axes
are encircled by atoms contributed from four different sublattices.
Thus, there is no superposition of screw axes to generate higher
overall symmetry, as in III. Li et al. have also described interpene-
trating but centrosymmetric coordination polymers of Co, Ni, and
Cd with HIP and 1,3-bis(4-pyridyl)propane;21 the Ni compound
exhibits Ni(II) in a coordination environment nearly identical to
that seen in III. Hu and Tong reported the only other chiral
hexagonal multiply-interpenetrating coordination polymer that
we are aware of: a bis(pyridyl)propane-templated copper iodide
in P6522 consisting of three trigonally-symmetric sublattices.22 In
this compound, the overall 65 symmetry arises from the collinearity
of pairs of 32 axes from different sublattices, themselves related via
perpendicular 2-fold axes. This differs slightly from III in that
the combination of trigonal sublattices into a hexagonal structure
occurs via other symmetry elements, whereas in III it occurs merely
by the relative positioning of the sublattices.
Fig. 9 Helical hydrogen bonding in III connecting nickel centers from
alternating sublattices, with atoms color-coded by sublattice and hydrogen
bonds dotted in dark blue.
ligands are stacked directly on the 61 axes, and these stacks occupy
the gaps in the Kagome lattice.
Structure of IV, Cu(HIP)2(bipy)
While the particular crystal used in our structure determination
is chiral, with no indication of racemic twinning, we expect that the
sample overall is a conglomerate, with half of the crystals having
grown in the opposite-handed space group P65. Each crystal’s
chirality is imparted not by any of the individual components,
which are all achiral, but by the helical arrangement of these
components.
The Schla¨fli vertex symbol for each sublattice of III was
determined using the TOPOS software package19 to be (75;9), cor-
responding to the dense “dual quartz” topology, with each nickel
center connecting to four nearest neighbors in a distorted see-saw
geometry. Carlucci et al. have reported a triply-interpenetrating
framework structure with the same sublattice topology,20 but
this structure differs from III in that only one linker, bis(4-
Compound IV (Fig. 11) crystallizes in the space group C2/c and
is a one-dimensional coordination polymer that also contains
infinite hydrogen-bonded molecular ribbons (Fig. 12). It is also
anhydrous, which is unusual for a coordination polymer obtained
at 125 ◦C but common in those synthesized at significantly
higher temperatures.23 The chains consist of highly Jahn–Teller
distorted trans-CuN2O4 octahedra connected linearly along b by
bipy ligands, as in I–III. We suspect that it is primarily due to this
distortion that no Cu(II) isomorphs to any of the other phases were
ever produced, even under otherwise identical conditions. The
four equatorial oxygens belong to two equivalent non-bridging
HIP ligands, with the Cu1–O3 bond normal in length (1.965(2)
˚
˚
A) and the Cu1–O4 bond both elongated (2.685(3) A) and bent
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2034–2042 | 2039
©

View Article Online
bond. This particular pairing of HIP groups, with hydrogen
bonds exchanged between carboxyl and hydroxyl oxygens, is
seen in Burchell and coworkers’ 1,4-diazabicyclo[2.2.2]octane–
5-hydroxyisophthalic acid complex,27 in which the HIP groups
are monoanionic as in IV. At the other end, the protonated
carboxyl group exchanges two equivalent hydrogen bonds with
the equivalent carboxyl group of another neighboring anion, in a
typical carboxylic acid dimer arrangement. Thus, the HIP ligands
are connected into linear ribbons by pairs of hydrogen bonds, and
the ribbons run along two differently inclined directions related by
the two-fold rotation axes in the Cu–bipy chains (Fig. 13). Because
many different Cu–bipy chains contribute their HIP ligands to
occupy each interchain space, the HIP ribbons stack tightly, with
three per cell along the b direction and a ribbon-to-ribbon spacing
˚
of 3.40 A. Thus, along the c axis, sheets of Cu–bipy chains alternate
with sheets of stacked HIP ribbons, which themselves alternate in
their direction of inclination (Fig. 14).
Fig. 11 Thermal ellipsoid plot (50% probability) of IV, with the atoms
of one asymmetric unit labeled.
Fig. 13 A section of the structure of IV, viewed toward a layer of
copper–bipy chains. Only every third HIP ribbon is shown for clarity,
with hydrogen bonds dotted in dark blue.
Fig. 12 (a) One copper–bipy chain with pendant HIP ligands, and (b)
one doubly hydrogen-bonded ribbon of HIP ligands in IV, with hydrogen
bonds dotted in dark blue.
approximately 35◦ out of perpendicularity to its neighbors. The
HIP ligands are related by the two-fold rotation axis through the
Cu–bipy chain and torqued oppositely about their C–C single
bonds, which leaves them inclined approximately 30◦ in opposite
directions out of the ac plane.
Infinite hydrogen-bonded arrays in IV are facilitated by the
protonation of the non-coordinating carboxyl groups of the HIP
ligands. Such HIP monoanions are much less common than HIP
dianions, but are reported in a distorted Zr(II) complex in which the
deprotonated carboxyl group coordinates asymmetrically through
both oxygens,24 and in a square-pyramidal Cu(II) complex in
which only one oxygen coordinates.25 Within the present family,
protonation of carboxyl groups is unique to IV and allows each
anion to participate in pairs of hydrogen bonds with symmetry-
equivalent anions at either end, as seen in certain organic
molecular crystals.26 At one end, the anion donates a hydrogen
bond from its phenol oxygen to a coordinating carboxyl oxygen of
a neighboring anion, and accepts a second equivalent hydrogen
Fig. 14 Packing diagram of IV, viewed along the copper–bipy chains,
approximately perpendicular to the HIP ribbons, with hydrogen bonds
dotted in dark blue.
2040 | Dalton Trans., 2008, 2034–2042
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Structure of V, Zn2(HIP)2(bipy)(H2O)2·H2O
Compound V (Fig. 15) crystallizes in the space group P–1, and
is a two-dimensional coordination polymer consisting of stair-
stepped layers in (−111) planes. The zinc exhibits approximately
trigonal bipyramidal coordination, with pairs of coordination
polyhedra sharing edges around inversion centers to form dimers
(Fig. 16). The distortion of the polyhedron is most suggestive of
a capped tetrahedron; the Addison s parameter28 was calculated
to be 0.703 indicating 29.7% square pyramidal character, when
defining angle a as O5–Zn1–N1 and angle b as O6–Zn1–O3. The
˚
three unshared ligand atoms are 1.920(2)–2.073(4) A from Zn1,
˚
and O3 is shared, with a typical bond (1.962(3) A) to one zinc atom
˚
and a long bond (2.741(3) A) to its symmetry equivalent. Similar
double bridging of [4 + 1]-coordinated Zn(II) centers about an
inversion center has been seen both in inorganic solids29 as well
as in molecular complexes,30 however previous examples have not
shown the degree of asymmetry seen in V. The Zn2O6N2 dimers are
connected along [101] by single bipy ligands, and along [110] by
pairs of non-coplanar but parallel HIP ligands (Fig. 17). The HIP
ligands lie parallel to the overall (−111) layers, and the bipy ligands
nearly so. Stair-stepping in the layers arises because the bipy
ligands connect the bottom members of a chain of Zn2O6N2 dimers
to the top members of the next chain. Between the bipy ligands lie
water molecules which appear disordered across inversion centers,
Fig. 17 Plan view of one layer in V, with HIP ligands in green, bipy ligands
in dark blue, and zinc cations and water molecules colored as in Fig. 15
and 16. The HIP and bipy ligands connect zinc dimers along two nearly
perpendicular directions. Ligand hydrogen atoms and uncoordinated water
molecules are omitted for clarity.
˚
with two half-occupied components lying 2.16 A away from each
other (Fig. 18).
Fig. 18 Packing diagram of V, viewed from the side of the stair-stepped
layers, with one layer highlighted in pink and other atoms colored as in
Fig. 15 and 16.
Both carboxyl groups of each HIP ligand coordinate through
only one oxygen, leaving the non-coordinating oxygen available to
accept a hydrogen bond. This, in addition to a higher degree of
hydration than III or IV, allows for the formation of rather exten-
sive interlayer hydrogen bonding in V. Pairs of hydrogen bonds
occur about inversion centers between approximately coplanar
pairs of HIP ligands from consecutive layers, from the phenol
hydrogens to one non-coordinating carboxyl oxygen, forming the
same carboxyl/hydroxyl pairing arrangement as seen in IV. The
aqua ligands supply additional interlayer hydrogen bonds to the
other non-coordinating HIP carboxyl oxygen. Both equivalent
disorder components of the uncoordinated water molecule accept
hydrogen bonds from aqua ligands in one layer and donate
hydrogen bonds to HIP phenol oxygens in the next layer.
Fig. 15 Thermal ellipsoid plot (50% probability) of V, with the atoms of
one asymmetric unit labeled. The occupancy of the uncoordinated water
molecule is 1/2.
Conclusion
By utilizing 5-hydroxyisophthalic acid together with 4,4ꢀ-bipyridyl
to construct transition metal-based coordination polymers, we
Fig. 16 One centrosymmetric Zn2N2O6 dimer in V, showing the signifi-
cant asymmetry of the bond lengths to the bridging oxygen O3.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2034–2042 | 2041
©

View Article Online
have produced a wide and unexpected variety of new framework
structures, including examples of one-, two- and three-dimensional
covalent connectivity. The two Mn phases together provide the
best example to date of polytypism in the field of coordination
polymers. The Ni phase is an intricate interpenetrating 3D chiral
network with unusual topology which, in projection, displays a
classic Kagome lattice motif. The Cu phase is a rare example
of a hybrid framework containing hydrogen-bonded molecular
ribbons within its structure. Finally, the Zn phase is constructed
from a unique infinite stair-stepped building unit. In all cases, the
HIP ligand’s phenolic oxygen remains protonated, but the ligand
can be either singly or doubly deprotonated at its two carboxyl
groups. In addition, this family of compounds has illustrated the
sensitivity of the overall framework structures to fine nuances in
the coordination geometry preferences of the metal cations that
they contain.
7 (a) H.-Y. He, Y.-L. Zhou, Y. Hong and L.-G. Zhu, J. Mol. Struct., 2005,
737, 97; (b) X. Li, R. Cao, D. Sun, W. Bi, Y. Wang, X. Li and M. Hong,
Cryst. Growth Des., 2004, 4((4)), 775.
8 G. M. Sheldrick, SHELXTL-97, A Program for Crystal Structure So-
lution and Refinement, University of Go¨ttingen, Go¨ttingen, Germany,
1997 (Ver. 5.1).
9 A. L. Spek, PLATON—A Multipurpose Crystallographic Tool, Utrecht
University, Utrecht, The Netherlands, 2005.
10 G. M. Sheldrick, SADABS User Guide, University of Go¨ttingen,
Go¨ttingen, Germany, 1995.
11 H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876–881.
12 (a) C. Ma, C. Chen, Q. Liu, D. Liao, L. Li and L. Sun, New J. Chem.,
2003, 27, 890; (b) J. Tao, X.-M. Chen, R.-B. Huang and L.-S. Zheng,
J. Solid State Chem., 2003, 170, 130.
13 (a) D. L. Smith and V. I. Saunders, Acta Crystallogr., Sect. B, 1981, 37,
1807; (b) D. L. Smith and V. I. Saunders, Acta Crystallogr., Sect. B,
1982, 38, 907.
14 R. D. Willett, K. R. Maxcy and B. Twamley, Inorg. Chem., 2002, 41,
7024.
ˇ
15 A. Guinier, G. B. Bokij, K. Boll-Dornberger, J. M. Cowley, S. Durovicˇ,
H. Jagodzinski, P. Krishna, P. M. De Wolff, B. B. Zvyagin, D. E.
Cox, P. Goodman, T. Hahn, K. Kuchitsu and S. C. Abrahams, Acta
Crystallogr., Sect. A, 1984, 40, 399.
16 J. A. Groves, P. A. Wright and P. Lightfoot, Inorg. Chem., 2005, 44,
Acknowledgements
1736.
17 (a) M. Kurmoo, H. Kumagai, M. Akita-Tanaka, K. Inoue and S.
Takagi, Inorg. Chem., 2006, 45, 1627; (b) J. Chen, M. Ohba, D. Zhao,
W. Kaneko and S. Kitagawa, Cryst. Growth Des., 2006, 6(3), 664.
18 Y.-Q. Zheng and Z.-P. Kong, Z. Anorg. Allg. Chem., 2003, 629, 1469.
19 (a) V. A. Blatov, IUCr Commission on Crystallographic Computing
Newsletter, (no. 7), Nov. 2006, p. 4 (URL: http://www.iucr.org/iucr-
top/comm/ccom/newsletters/); (b) http://topos.ssu.samara.ru/.
20 L. Carlucci, G. Ciani, P. Macchi and D. M. Proserpio, Chem. Commun.,
1998, 1837.
21 X.-J. Li, R. Cao, W.-H. Bi, Y.-Q. Wang, Y.-L. Wang and X. Li,
Polyhedron, 2005, 24, 2955.
22 S. Hu and M.-L. Tong, Dalton Trans., 2005, 1165.
23 P. M. Forster, A. R. Burbank, C. Livage, G. Fe´rey and A. K. Cheetham,
Chem. Commun., 2004, 368.
24 Y.-J. Zhao, R.-W. Liu and X.-H. Li, Acta Crystallogr., Sect. E, 2005,
61, m535.
25 H.-P. Xiao, W.-D. Wang, W.-B. Zhang and J.-G. Wang, Acta Crystal-
logr., Sect. E, 2005, 61, m841.
26 M. C. Grossel, C. A. Golden, J. R. Gomm, P. N. Horton, D. A. S.
Merckel, M. E. Oszer and R. A. Parker, CrystEngComm, 2001, 3, 170.
27 C. J. Burchell, G. Ferguson, A. J. Lough and C. Glidewell, Acta
Crystallogr., Sect. C, 2000, 56, 1126.
The authors thank Guang Wu, Kinson Kam, James Orton and
Crystal Merrill for valuable discussion. This work was supported
by the U.S. Department of Energy, Office of Basic Energy Sciences,
under Award Number DE-FG02-04ER46176.
References
1 B. F. Hoskins and R. Robson, J. Am. Chem. Soc., 1990, 112, 1546.
2 (a) A. K. Cheetham, C. N. R. Rao and R. K. Feller, Chem. Commun.,
2006, 4780; (b) C. N. R. Rao, S. Natarajan and R. Vaidhyanathan,
Angew. Chem., Int. Ed., 2004, 43, 1466; (c) M. OꢀKeeffe, M. Eddaoudi,
H. Li, T. Reineke and O. M. Yaghi, J. Solid State Chem., 2000, 152, 3.
3 Q. Shuai, S. Chen and S. Gao, Inorg. Chim. Acta, 2007, 360, 1381.
4 (a) X.-H. Sun, Acta Crystallogr., Sect. E, 2005, 61, m2256; (b) H.-P.
Xiao, Acta Crystallogr., Sect. E, 2006, 62, m731; (c) Z.-W. Wang, Y.-Z.
Li, Y. Cai and H.-G. Zheng, Acta Crystallogr., Sect. E, 2006, 62, m734;
(d) M. J. Plater, M. R. St. J. Foreman, R. A. Howie, J. M. S. Skakle,
S. A. McWilliam, E. Coronado and C. J. Go´mez-Garc´ıa, Polyhedron,
2001, 20, 2293.
5 H. Xu and Y. Li, J. Mol. Struct., 2004, 690, 137.
6 (a) J. M. S. Skakle, M. R. St. J. Foreman and M. J. Plater, Acta
Crystallogr., Sect. E, 2001, 57, m373; (b) X.-Y. Cao, J. Zhang, Y. Kang,
J.-K. Cheng, Z.-J. Li, X.-Q. Wang and Y.-G. Yao, Acta Crystallogr.,
Sect. E, 2004, 60, m460; (c) X.-Y. Cao, J. Zhang, Y. Kang, J.-K. Cheng,
Z.-J. Li, X.-Q. Wang, Y.-H. Wen and Y.-G. Yao, Acta Crystallogr., Sect.
C, 2004, 60, m350.
28 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor,
J. Chem. Soc., Dalton Trans., 1984, 1349.
29 R. J. B. Jakeman and A. K. Cheetham, J. Am. Chem. Soc., 1988, 110(4),
1142.
30 M. Barcelo´-Oliver, A. Terro´n, A. Garc´ıa-Raso, J. J. Fiol, E. Molins and
C. Miravitlles, J. Inorg. Biochem., 2004, 98, 1703.
2042 | Dalton Trans., 2008, 2034–2042
This journal is
The Royal Society of Chemistry 2008
©