Zinc(ii) coordination polymers, metallohexacycles and metallocapsules - Do we understand self-assembly in metallosupramolecular chemistry: Algorithms or serendipity?

Article abstract of DOI:10.1039/c1ce05884c

Using a strategy of layering solvents and solutions of ligands and metal salts under ambient conditions, we observe the assembly of a discrete molecular metallohexacycle from ZnCl₂ and 4′-(4-ethynylphenyl)-4, 2′:6′,4′′-terpyridine, polycatenated, triply interlocked metallocapsules from ZnI₂ and 4′-(4-pyridyl)-4,2′: 6′,4′′-terpyridine, and 1-dimensional coordination polymers from either ZnCl₂ or ZnI₂ with 4′-{4-(3- chloropyridyl)}-4,2′:6′,4′′-terpyridine. On the basis of these studies and a comparison with related structures in the literature, we urge crystal engineers to be wary of drawing conclusions about self-assembly algorithms in solution using data from single crystal determinations. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c1ce05884c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
CrystEngComm
Cite this: CrystEngComm, 2011, 13, 6864
www.rsc.org/crystengcomm
PAPER
Zinc(II) coordination polymers, metallohexacycles and metallocapsules—do
we understand self-assembly in metallosupramolecular chemistry: algorithms
or serendipity?†
Edwin C. Constable,* Guoqi Zhang, Catherine E. Housecroft* and Jennifer A. Zampese
Received 14th July 2011, Accepted 25th August 2011
DOI: 10.1039/c1ce05884c
Using a strategy of layering solvents and solutions of ligands and metal salts under ambient conditions,
0
we observe the assembly of a discrete molecular metallohexacycle from ZnCl
2
and 4 -(4-ethynylphenyl)-
0
0
0
00
4
4
,2 :6 ,4 -terpyridine, polycatenated, triply interlocked metallocapsules from ZnI and 4 -(4-pyridyl)-
2
0
0
00
0
,2 :6 ,4 -terpyridine, and 1-dimensional coordination polymers from either ZnCl or ZnI with 4 -{4-
2
2
0 0 00
(
3-chloropyridyl)}-4,2 :6 ,4 -terpyridine. On the basis of these studies and a comparison with related
structures in the literature, we urge crystal engineers to be wary of drawing conclusions about self-
assembly algorithms in solution using data from single crystal determinations.
Introduction
solution of ZnCl yields crystals of a highly unusual network
2
consisting of 1-dimensional polycatenanes, but when ZnI2
replaces ZnCl , or when the reaction with ZnCl is carried out
The assembly of coordination polymers crucially depends upon
the compatibility of the coordination preferences of the metal
nodes and the donor abilities of ligand linkers. Together, these
direct the formation of 1-dimensional chains, 2-dimensional
2
2
under solvothermal conditions,
26
a
1-dimensional polymer
2
results. With malate as a co-ligand, the reaction of ZnCl and
0
0 0 00
4
-(4-pyridyl)-4,2 :6 ,4 -terpyridine results in the assembly of
1–5
sheets or 3-dimensional networks. One of the basic tenets of
24
a 3-dimensional network under hydrothermal conditions.
These results prompted us to report our own findings with these
and similar systems: by a process of layering under ambient
conditions, a discrete molecular metallohexacycle assembles
from ZnCl2 and ligand 1 (Scheme 1), polycatenated metal-
2,6–14
metallosupramolecular chemistry
is that an understanding
of the coordination requirements of the metal ion and the donor
properties of the ligands can be used to uniquely define the self-
assembly algorithm. We are increasingly of the opinion that
serendipitous and subtle environmental effects have a profound
influence upon the self-assembly of coordination networks.
The use of 2,4,6-tris(4-pyridyl)-1,3,5-triazine as a building
block has proved extremely rewarding in controlling the
formation of 3-dimensional networks and discrete molecular
locapsules from ZnI and 2, but on going from ligand 2 to 3
2
a 1-dimensional coordination polymer is obtained (Scheme 1).
On the basis of these studies, we urge caution in the interpreta-
tion of data from a single crystal determination to far-reaching
conclusions about self-assembly algorithms in solution.
15
capsules. The metal-binding sites of 2,4,6-tris(4-pyridyl)-
1
,3,5-triazine are exclusively the peripheral pyridine donors.
20–26
16–19
0 0 00
Experimental
Similarly, we
and others
have observed that 4,2 :6 ,4 -
0
terpyridine behaves as a divergent N,N -bidentate ligand with the
1
13
H and C NMR spectra were recorded on a DRX-500 NMR
central pyridine ring remaining uncoordinated and presenting
spectrometer (chemical shifts referenced to residual solvent
peaks, TMS ¼ d 0 ppm) and infrared spectra on a Shimadzu
FTIR-8400S spectrophotometer fitted with a Golden Gate Dia-
mond ATR accessory for solid samples. A Bruker esquire
two monodentate nitrogen donors with an expectation value of
ꢀ
1
20 subtended by the two N–M vectors. To date, the coordi-
0
0
0
00
nation chemistry of 4,2 :6 ,4 -terpyridine and 4 -substituted
derivatives with zinc(II) has been largely restricted to the
3
000 plus mass spectrometer was used to record electrospray
17,18,20–23
assembly of 1-dimensional chains.
A recent publication
(
ESI) mass spectra.
from Dehnen and coworkers reports that layering of a CHCl3
00
19
Ligand 1 was prepared as previously reported.
0
0
0
solution of 4 -(4-pyridyl)-4,2 :6 ,4 -terpyridine with a MeOH
Ligand 2
Department of Chemistry, University of Basel, Spitalstrassse 51, CH-4056
Basel, Switzerland. E-mail: edwin.constable@unibas.ch, catherine.
housecroft@unibas.ch; Fax: +41 61 267 1018; Tel: +41 61 267 1008
27
The synthesis of 2 has been reported. We describe a higher
yielding route; fully assigned NMR spectroscopic data are
reported here for the first time. 4-Acetylpyridine (1.21 g,
10.0 mmol) was added to a solution of 4-pyridinecarbaldehyde
† CCDC reference numbers 832650–832655. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/c1ce05884c
6
864 | CrystEngComm, 2011, 13, 6864–6870
This journal is ª The Royal Society of Chemistry 2011

View Article Online
solid, cm ): 1596s, 1581s, 1558s, 1525m, 1481m, 1398s, 1320w,
ꢁ1
(
1
7
282m, 1175w, 1117w, 1066s, 1029m, 996s, 904w, 852m, 826s,
+
48s, 724m. ESI-MS (CH Cl /MeOH) m/z 345.2 [M + Na] (base
2
2
peak, calc. 345.1). Found C 69.53, H 3.95, N 16.19; C20
requires C 69.67, H 3.80, N 16.25%.
H13ClN
4
2 6
[{ZnCl (1)} ]
A solution of 1 (16.7 mg, 0.050 mmol) in CHCl (6.0 mL) was
3
placed in a long test tube (inner diameter 1.5 cm ꢂ height 15 cm).
MeOH (3.0 mL) was then layered on the top of the solution, and
2
then a solution of ZnCl (6.7 mg, 0.050 mmol) in MeOH (5.0 mL)
was layered carefully on the top of the MeOH layer. The test tube
was sealed with parafilm and allowed to stand for 10 d at room
temperature. Colourless crystals formed on the glass surface and
were isolated by filtration to give solvated [{ZnCl
2 6
(1)} ]
(
15.6 mg) readily lose solvent when removed from the mother
1
liquor. H NMR (500 MHz, DMSO-d6) d 8.77 (d , J ¼ 6.1 Hz,
AB
A2
B3
A3
4
H, H ), 8.52 (s, 2H, H ), 8.37 (d , J ¼ 5.7 Hz, 4H, H ), 8.15
AB
C2/C3 C2/C3
), 7.69 (d, J ¼ 8.2 Hz, 2H, H
(
(
(
(
(
d, J ¼ 8.3 Hz, 2H, H
), 4.39
C^CH
13
A2
s, 1H, H ). C NMR (126 MHz, DMSO-d6) d/ppm 154.4
B2
B4
A4
C1
C ), 150.2 (C ), 149.2 (C ), 145.5 (C ), 131.1 (C ), 132.4
Scheme 1 Structures of ligands 1–3 with numbering for NMR spec-
C2/C3
C2/C3
C4
A3
B3
), 123.1 (C ), 121.4 (C ), 119.1 (C ), 83.1
C
C
), 127.8 (C
troscopic assignments.
C^CH
C^CH
ꢁ1
), 82.6 (C
). IR (solid, cm ): 1616s, 1603s, 1558w,
1
537w, 1506w, 1420w, 1396m, 1219w, 1063m, 1030m, 831s,
(
0.54 g, 5.0 mmol) in EtOH (40 mL). KOH pellets (0.62 g,
1 mmol) were added followed by aqueous NH (32%, 20 mL)
729w, 679s, 650s, 621m. ESI-MS (DMSO/MeOH) m/z 825.5
+
+
1
3
[ZnCl
calc. 334.1). Found
$6MeOH requires C 57.45, H 3.82, N 8.38%.
2
(1) + Na] (calc. 825.1), 334.3 [1 + H] (base peak,
and the resulting red solution was stirred at room temperature
for 24 h. The precipitate was collected by filtration, washed with
C
57.13, 3.30, 8.76;
H
N
C
H90Cl12N18Zn
138 6
H
2
O and EtOH, and dried in vacuo over P
2
O
5
. Compound 2 was
ꢀ
isolated as an off-white solid. Yield: 1.10 g (71%). Mp > 300 C
1
[
Zn12
I
24(2)
8
]
(
decomp.). H NMR (500 MHz, CDCl ) d/ppm 8.64 (d, J ¼ 5.7
3
C2 A2
Hz, 2H, H ), 8.61 (d, J ¼ 5.8 Hz, 4H, H ), 8.03 (d, J ¼ 5.9 Hz,
H, H ), 8.01 (s, 2H, H ), 7.62 (d, J ¼ 5.9 Hz, 2H, H ).
NMR (126 MHz, CDCl ) d/ppm 155.5 (C ), 149.96 (C ), 149.93
6 4 2
A solution of 2 (31.0 mg, 1.00 mmol) in 1,2-C H Cl /MeOH
A3
B3
C3
13
C
4
(10 mL, 4 : 1, v/v) was placed in a long test tube and MeOH
(4.0 mL) was layered on top, followed by a layer of a solution of
B2
C4
C2
C3
3
A2
B4
A4
(
(
C ), 148.4 (C ), 146.2 (C ), 145.9 (C ), 122.0 (C ), 121.6
B3
C ), 119.1 (C ). IR (solid, cm ): 1589s, 1558w, 1529m, 1495w,
ZnI (47.6 mg, 1.50 mmol) in MeOH (8.0 mL). The test tube was
2
A3
ꢁ1
sealed with parafilm and was left to stand for 10 d at room
temperature. Colourless crystals formed on the walls of the tube
and were isolated by decanting the solvent. Crystals of solvated
1
8
430w, 1398s, 1326w, 1266w, 1216w, 1061m, 992m, 892m, 850m,
+
2 2
29m, 814s, 748m. ESI-MS (CH Cl /MeOH) m/z 311.2 [M + H]
+
base peak, calc. 311.1), 333.1 [M + Na] (calc. 333.1). Found C
(
8
[Zn12I24(2) ] (52.1 mg) were washed with MeOH and dried in air.
ꢁ1
7
7.18, H 4.79, N 17.79; C20
8.05%.
H
14
N
4
requires C 77.40, H 4.55, N
IR (solid, cm ): 1613s, 1591s, 1538m, 1437w, 1398s, 1320w,
1219s, 1062s, 1024s, 836s, 817s, 668s, 648s. Satisfactory
elemental analysis was not obtained due to facile loss of solvent.
1
Ligand 3
[
{ZnI (3)} ]
2 n
4
3
-Acetylpyridine (1.21 g, 10.0 mmol) was added to a solution of
-chloroisonicotinaldehyde (0.54 g, 5.0 mmol) in EtOH (40 mL).
6 4 2
A solution of 3 (34.4 mg, 1.00 mmol) in 1,2-C H Cl /MeOH
3
KOH pellets (0.62 g, 11 mmol) were added followed by aqueous
NH (32%, 25 mL) and the resulting orange-red solution was
stirred at room temperature for 18 h. The precipitate that formed
was collected by filtration, washed with H O and EtOH, and
dried in vacuo over P . Compound 3 was isolated as white
solid. Yield: 1.11 g (64.5%). Mp 256–258 C. H NMR
(10 cm , 4 : 1, v/v) was placed in a long test tube. MeOH (4.0 mL)
was layered carefully on the top of the first solution, followed by
3
2
a layer of a solution of ZnI (47.6 mg, 1.50 mmol) in MeOH
2
(8.0 mL). After sealing with parafilm, the test tube was allowed to
stand for 1 month at room temperature. Colourless crystals
formed on the walls of the tube and were isolated by decanting
the solvent. They were washed with MeOH and then dried in the
air. Yield of [{ZnI (3)} ]: 59.2 mg (89.4% based on ligand). IR
2 5
O
ꢀ
1
C2
(
(
(
500 MHz, CDCl ) d/ppm 8.78 (s, overlapping, 1H, H ), 8.77
3
A2
C6
dd, J ¼ 1.6, 4.7 Hz, 4H, H ), 8.64 (d, J ¼ 4.9 Hz, 1H, H ), 8.03
dd, J ¼ 1.6, 4.5 Hz, 4H, H ), 7.91 (s, 2H, H ), 7.37 (d, J ¼ 4.9
) d/ppm 155.5 (C ),
2
n
A3
B3
ꢁ1
(solid, cm ): 1613s, 1439w, 1391m, 1322w, 1286w, 1226w, 1063s,
1024s, 856m, 833s, 731m, 689m, 650s. Found C 36.11, H 1.99, N
C5 13
B2
A4
Hz, 1H, H ). C NMR (126 MHz, CDCl
3
A2
C2
C6
B4
1
1
50.9 (C ), 150.8 (C ), 148.6 (C ), 146.9 (C ), 145.5 (C ),
C4
44.7 (C ), 130.0 (C ), 124.7 (C ), 121.3 (C ), 120.7 (C ). IR
8.35; C20
2 4 2
H13ClI N Zn$0.25H O requires C 35.93, H 2.04, N
C3
C5
A3
B3
8.38%.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 6864–6870 | 6865

View Article Online
[
{ZnCl (3)} ]
2
with I > 2s(I) converged at final R ¼ 0.0341 (R all data ¼
n
1
1
0
.0351), wR ¼ 0.0823 (wR all data ¼ 0.0828), gof ¼ 1.134.
2
2
The method and volumes of solvents were as above, starting with
(34.4 mg, 1.00 mmol) and ZnCl (20.1 mg, 1.50 mmol).
{ZnCl (3)} ] was isolated as colourless crystals after 1 month.
Yield: 41.0 mg (85.8%). IR (solid, cm ): 1614s, 1582m, 1557m,
3
2
[
2
n
Results and discussion
ꢁ
1
0
0
00
The 4,2 :6 ,4 -terpyridine ligand 1 ref. 19) contains a coor-
0
1
7
439w, 1389s, 1285w, 1223w, 1065s, 1026s, 886w, 850s, 841s,
28w, 689m, 651s. Found 49.18, 2.84, 11.07;
Zn requires C 49.93, H 2.72, N 11.65%.
dinatively innocent 4 -alkynyl substituent. Layering of a MeOH
C
H
N
solution of ZnCl
2
over sequential layers of neat MeOH and
20 3 4
C H13Cl N
a CHCl solution of ligand 1 resulted in the growth, after 10 days,
3
of colourless crystals. Structural analysis revealed the product to
Crystallography
Data were collected on a Stoe IPDS diffractometer. The data
be solvated [{ZnCl (1)} ]. The crystals readily lose solvent when
6
2
they are removed from the mother liquor. Crystal growth was
also performed with CH Cl or 1,2-dichloroethane in place of
28
reduction, solution and refinement used Stoe IPDS software
2
2
29
CHCl . Screening of cell dimensions showed that the products
3
and SHELXL97. Structures were analysed using Mercury v.
3
0,31
were the same in each case. As the structural description below
reveals, the infinite lattice contains extremely large voids. Using
2
.4.
TwinRotMat program in Platon. A HKLF 5 was generated
from the original file (118 838 measured reflections, Rint
.1470) using the twin law (ꢁ1,0,0/ꢁ0.5,0,1/ꢁ0.5,1,0) to give
a modified file (16 722 reflections, Rint ¼ 0.0000) that was used in
SHELXL97 with a basf of 0.1649.
2 6
Twinning in [{ZnCl (1)} ] was dealt with using the
32
32
Platon, the occupied space in the solvent-free lattice is calcu-
lated to be 45%. The residual electron density within the voids is
smeared and we were unable to identify individual solvent
molecules which are heavily disordered. The structure was
¼
0
29
32
subsequently refined using the program SQUEEZE. Structural
analysis revealed the formation of a centrosymmetric hexameric
metallocycle (Fig. 1a). The asymmetric unit contains one half of
the macrocycle. Each Zn atom is tetrahedral sited with Zn–N and
[
{ZnCl (1)} ]
2 6
C
138
H90Cl12
N
18Zn
ꢀ
6
, M ¼ 2818.03, colourless block, triclinic,
ꢁ
space group P1, a ¼ 8.793(2), b ¼ 23.662(6), c ¼ 24.672(6) A, a ¼
Zn–Cl bond distances in the ranges 2.024(4) to 2.060(4) and
2.2114(15) to 2.228(17) Aꢁ , respectively. Pairs of symmetry related
Zn atoms are 21.970(6), 24.676(7) and 28.492(6) Aꢁ apart across
ꢀ
ꢁ
3
6
8.897(19), b ¼ 87.31(2), g ¼ 80.27(2) . U ¼ 4719(2) A , Z ¼ 1,
ꢁ3
ꢁ1
D
c
¼ 0.991 Mg m , m (Mo-K
a
) ¼ 0.960 mm , T ¼ 173 K. Total
1
6 722 reflections, 16 722 unique. Refinement of 14 711 reflec-
the macrocyclic cavity. Working around the ring, ligands 1 adopt
a sequential three-up/three-down arrangement producing a chair
conformation (Fig. 1b). Molecules of [{ZnCl (1)} ] pack into
tions (785 parameters) with I > 2s(I) converged at final R
1
¼
0
.0816 (R
1
all data ¼ 0.0885), wR
2
¼ 0.2271 (wR all data ¼
2
2
6
0
.2349), gof ¼ 1.040.
[
Zn I (2) ]
1
2 24
8
C
160
H
112
I
24
N
32Zn12, M ¼ 6313.02, colourless block, rhombo-
ꢀ
ꢁ
hedral, space group R3, a ¼ b ¼ 38.903(5), c ¼ 16.736(3) A. U ¼
ꢁ
3
ꢁ3
2
3
1936(6) A , Z ¼ 3, D
c
¼ 1.434 Mg m , m (Mo-K
a
) ¼
ꢁ
1
.534 mm , T ¼ 173 K. Total 96 624 reflections, 8619 unique,
R
int ¼ 0.0879. Refinement of 6592 reflections (344 parameters)
with I > 2s(I) converged at final R all data ¼
¼ 0.1181 (R
.1442), wR all data ¼ 0. 3115), gof ¼ 1.957.
¼ 0.2993 (wR
1
1
0
2
2
[
{ZnI
2 n 2
(3)} ]$0.25H O
C H ClI N O Zn, M ¼ 668.46, colourless block, mono-
2
0
13.5
2
4
0.25
clinic, space group P2 /n, a ¼ 9.8708(17), b ¼ 10.5239(11), c ¼
1
ꢁ
ꢀ
ꢁ
3
2
2
2
3
0.888(4) A, b ¼ 101.277(13) . U ¼ 2127.9(6) A , Z ¼ 4, D
c
¼
ꢁ3
ꢁ1
.085 Mg m , m (Mo-K
a
) ¼ 4.196 mm , T ¼ 173 K. Total
7 117 reflections, 4166 unique, Rint ¼ 0.0823. Refinement of
840 reflections (262 parameters) with I > 2s(I) converged at
final R
1
¼ 0.0343 (R
1
all data ¼ 0.0382), wR
2
¼ 0.0844 (wR
2
all
data ¼ 0.0868), gof ¼ 1.099.
[
2 n
{ZnCl (3)} ]
C H Cl N Zn, M ¼ 481.08, colourless block, orthorhombic,
2
0
13
3
4
space group Pbca, a ¼ 17.353(3), b ¼ 10.3148(16), c ¼ 20.974(4)
ꢁ
ꢁ
3
ꢁ3
A. U ¼ 3754.1(11) A , Z ¼ 8, D
c
¼ 1.702 Mg m , m (Mo-K
a
) ¼
ꢁ1
1
.702 mm , T ¼ 173 K. Total 77 342 reflections, 5068 unique,
2 6
Fig. 1 The molecular structure of [{ZnCl (1)} ] showing (a) the hex-
R
int ¼ 0.0601. Refinement of 4931 reflections (253 parameters)
ametallic macrocycle, and (b) the chair conformation of the metallocycle.
6
866 | CrystEngComm, 2011, 13, 6864–6870
This journal is ª The Royal Society of Chemistry 2011

View Article Online
infinite columns that follow the crystallographic a axis (Fig. 2).
The crystal packing is predominantly: (i) face-to-face p-stacking
of pairs of the central and one outer pyridine rings (separation of
ꢁ
the planes is 3.48 A) between molecules coloured red and blue in
Fig. 2, (ii) face-to-face p-interactions between interdigited
ꢁ
pyC
6
H
4
C^CH units (interplane distance ¼ 3.45 A), and (iii)
extensive CH/Cl non-classical hydrogen bonds. The effective-
ness of the latter is enhanced by the polarity of the Zn–Cl bond.
The voids in the columnar assembly are very large.
33
The unexpected assembly of discrete metallohexacyclic
[
{ZnCl (1)} ] encouraged us to explore the effects of introducing
6
2
0
the coordinatively non-innocent 4 -pyridyl substituent in
0 0 00
4
,2 :6 ,4 -terpyridine. Given that the central N-donor in 2 typi-
19
cally does not bind a metal ion, the ligand can be considered an
5
analogue of 2,4,6-tris(4-pyridyl)-1,3,5-triazine. Coordination
8
Fig. 3 The structure of the molecular capsule [Zn12I24(2) ] looking
networks incorporating 2 have been prepared under sol-
approximately down the 3-fold axis of the cage. The central ring of each
ligand 2 is disordered over 3 positions and only one position for each N is
shown.
24,34
vothermal conditions.
room temperature and pressure conditions, ZnCl
assemble into a network comprising 1-dimensional poly-
However, the observation that, under
2
and 2 can
26
catenanes is more relevant to the crystallization procedures
ꢁ
2.522(3)–2.589(3) A, respectively. There are only two indepen-
that we have previously utilized for the formation of a range
0
0
00
ꢀ
of coordination polymers containing 4,2 :6 ,4 -terpyridine
dent N–Zn–N (102.6(5) and 103.5(4) ) and I–Zn–I (118.21(8)
ꢀ
1
6–19
ligands.
,2-C H Cl /MeOH, neat MeOH and finally a solution of ZnI
2
Sequential layering of
a
solution of
2
in
and 123.69(9) ) angles, the rest being generated by the symmetry.
1
The 12 Zn atoms define an icosahedral cage although the
6
4
2
in MeOH led, after 10 days, to colourless crystals. After isola-
tion, solvent loss from the crystals was problematical and they
were weakly diffracting. Of the ten reflection datasets collected,
the best gave only a poor quality structure due to highly disor-
dered solvent molecules in large voids. Consequently, the crystal
symmetry is reduced from I such that there are three pairs of
h
ꢁ
cross cage Zn/Zn distances of 25.015(5) and 26.184(5) A rather
than the six equal distances inherent in the I symmetry.
h
Fig. 4a depicts the packing of [Zn I (2) ] cages with a view
1
2 24
8
down the c-axis coinciding with a 3-fold axis. The most fasci-
35
nating feature of the lattice is the polycatenation of cages.
32
structure was refined using SQUEEZE. Despite these prob-
lems, details of the molecular capsule shown in Fig. 3 are
unambiguous and confirm that a change from Cl to I does not
perturb the assembly of the {Zn X (2) } cage reported by
Adjacent capsules are triply interlocked (i.e. three bonds must be
broken to separate the cages), and catenation occurs exclusively
along the c-axis. The structure mimics that observed for the
1
2
12
8
26
26
Dehnen. This is in contrast to Dehnen’s own results where
chlorido analogue. As seen in Fig. 3, the I atoms are directed
a change from Cl to I is assumed to be the reason for the
away from the surface of the capsule, and weak C–Haromatic/
I
26
assembly of a 1-dimensional polymer [{ZnI (2)} ]. Compound
2
intermolecular interactions are the dominant packing forces
ꢁ
n
[
Zn12
I
24(2)
8
] crystallizes as a solvate in the rhombohedral space
(range CH/I contacts ¼ 3.08–3.32 A compared to the sum of
ꢀ
group R3, and the asymmetric unit contains two independent
ꢁ
the van der Waals radii of 3.35 A). Such interactions have had
ZnI
2
units and 1.33 independent ligands. The central pyridine
less attention in the literature than CH/F, CH/Cl or CH/Br
33
rings of the coordinated ligands are disordered; this arises from
the high symmetry of the molecule, the triangle defined by
contacts. In 1999, Desiraju and Steiner stated: ‘‘available
structural data on covalently bonded I-acceptors hardly allows
36
0
00
the peripheral N,N ,N -donor set having no preferred
for any kind of statistical analysis’’. A search of the Cambridge
2+
30,37
orientation. Each Zn ion is in a tetrahedral environment; Zn–N
and Zn–I bond lengths are in the ranges 2.024(10)–2.106(13) and
Structural Database
for C–H/I–M contacts (M ¼ any
metal) reveals 3778 hits, indicating that such packing interactions
are no longer a curiosity. Although the p-systems of ligands 2 on
adjacent molecules in the direction shown in Fig. 5 are separated
ꢁ
by ꢃ3.5 A, the twisted conformations of the coordinated ligands
lead to inefficient face-to-face p-interactions. Solvent loss from
the large voids in the lattice is extremely facile and the calcu-
32
lated occupied space of the solvent-free lattice is only 43%.
We next looked at how the introduction of a substituent into
0
the 3-position of the 4 -pyridyl ring in ligand 2 would affect the
architectural assembly. Substitution at this site necessarily causes
0
twisting of the 4 -pyridyl ring (Scheme 2) and was expected to
interfere with the assembly of a capsule analogous to
Zn I (2) ]. Ligand 3 (Scheme 1) was prepared by Kr o€ hnke
[
1
2 24
8
38
methodology starting from 4-acetylpyridine and 3-chlor-
2 6
Fig. 2 Unit cell viewed down the a axis: packing of [{ZnCl (1)} ]
oisonicotinaldehyde and was fully characterized by mass spec-
1 13
trometry, H and C NMR and IR spectroscopies, and elemental
molecules produces a porous columnar structure.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 6864–6870 | 6867

View Article Online
0
Scheme 2 Steric interactions cause twisting of the 4 -(3-XC
0
6
H
3
N)
0
00
substituent with respect to the 4,2 :6 ,4 -terpyridine framework.
techniques) were consistent with the structure drawn in Scheme 1
for 3. Reaction of 3 with ZnI was carried out under identical
2
conditions as that of 2 and ZnI described above. Crystal growth
2
was slower but after a month yielded colourless blocks.
Elemental analysis was consistent with a stoichiometry {ZnI (3)}
2
and this was confirmed by the structural determination of
[{ZnI
2 n 2
(3)} ]$0.25H O.
Introduction of the chloro-substituent into the ligand switches
the assembly preference from the discrete cage (Fig. 3) to
0
a 1-dimensional polymer in which the 4 -(3-chloropyridyl) group
is non-coordinated (Fig. 6a). The complex crystallizes in the
1 2
monoclinic P2 /n space group with one independent {ZnI (3)} in
2+
the asymmetric unit. The Zn ion is tetrahedrally sited with
ꢁ
typical Zn–N (2.059(3), 2.071(3) A) and Zn–I (2.5253(7),
ꢁ
.5449(6) A) bond lengths. The range of bond angles is as
2
expected on steric grounds (N–Zn–N ¼ 99.83(13), I–Zn–I ¼
ꢀ
1
17.25(2), I–Zn–N range 107.04(10)–111.86(10) ). The Zn/Zn
ꢁ
separations along each chain are 13.190(2) A which is slightly
Fig. 4 (a) Packing of [Zn12
down the c-axis; the colour codes highlight the 3-fold symmetry. (b)
Polycatenation of [Zn12 24(2) ] cages along the c-axis.
8
I24(2) ] molecules with the unit cell viewed
I
8
Fig. 5 Inefficient face-to-face p-interactions between ligands 2 in the
solid state structure of solvated [Zn12 24(2) ]. Angles between the least
I
8
squares planes of the rings in ligands 2 shown in space-filling represen-
ꢀ
tations are 27.9, 13.1 and 31.2 .
analysis. The base peak in the electrospray mass spectrum was
+
observed at m/z 345.2 and was assigned to [M + Na] . The H and
1
Fig. 6 (a) Part of one chain of [{ZnI (3)} ] in [{ZnI (3)} ]$0.25H O, and
2
2
n
2
n
13
C NMR spectra (assigned using COSY, HMQC and HMBC
(b) packing of chains in the unit cell (see text).
6
868 | CrystEngComm, 2011, 13, 6864–6870
This journal is ª The Royal Society of Chemistry 2011

View Article Online
longer than corresponding distances in structurally related
0
Conclusions
0
00
[
{ZnCl L} ] 1-dimensional polymers (L ¼ 4,2 :6 ,4 -terpyridine
2
n
0
0
00
0
30,39
4,2 :6 ,4 -Terpyridines typically function as bidentate bridging
ligands utilizing only the 4-pyridyl nitrogen donors. Rigid self-
or 4 -substituted derivatives thereof).
ꢀ
The 3-chloropyridyl
substituent is twisted 42.3 out of the plane of the pyridine ring to
which it is attached, and adjacent pairs of pyridine rings in the
ꢀ
assembly algorithms predict a 120 angle between the N–M
ꢀ
vectors in coordination compounds. Crystallization experiments
leading to a few single crystals or bulk material cannot be used in
the absence of a profound understanding of the stoichiometric
and structural space and solution speciation to make far-reach-
ing conclusions about subtle influences on solid state structures.
Different crystallization conditions can lead to profoundly
different structures. In the case of ligand 2, it is not possible to
rationalize differing structural features in terms of, for example,
halogen steric requirements.
tpy domain are mutually twisted by 14.6 and 30.8 . The
1
-dimensional polymer chains pass obliquely through the unit
cell. Since the chain is generated by a glide plane, we refrain from
using the descriptor ‘helical’ which implies that the chain is built
up by a screw axis. In Fig. 6b, pairs of chains coloured red and
green are related by 2-fold screw axes, and pairs of chains of the
same colour are related by inversion. Face-to-face p-stacking of
centrosymmetric pairs of 3-chloropyridyl substituents on adja-
cent chains is efficient (highlighted in space-filling representation
ꢁ
in Fig. 6b), the separation of the planes of the rings being 3.65 A.
Additional packing interactions involve CH/Nchloropy hydrogen
bonds and CH/I contacts. The latter lie in the range 3.11–
3
Acknowledgements
ꢁ ꢁ
.14 A, significantly less than 3.35 A which represents the sum of
We thank the Swiss National Science Foundation, the European
Research Council (Advanced Grant 267816 LiLo), and the
University of Basel for financial support. GZ thanks the Novartis
Foundation, formerly Ciba-Geigy Jubilee Foundation for
support. Discussions with Dr Markus Neuburger are gratefully
acknowledged.
the van der Waals radii.
Ligand 3 again shows a preference for forming a 1-dimen-
sional polymer when treated with ZnI
2
under the same condi-
tions as its reaction with ZnCl . Crystals of [{ZnCl (3)} ] grew
2
2
n
within one month when carefully layered solutions of 3 in
,2-C H Cl /MeOH and ZnCl in MeOH with a dividing layer of
1
6
4
2
2
neat MeOH were left to stand at room temperature. In contrast
to [{ZnI (3)} ]$0.25H O, [{ZnCl (3)} ] crystallizes in the ortho-
rhombic space group Pbca. There is one independent {ZnCl (3)}
unit and the Zn/Zn separations along each chain are
Notes and references
2
n
2
2
n
1 P. J. Stang and B. Olenyuk, Acc. Chem. Res., 1997, 30, 502.
2
2
3
4
S. Leininger, B. Olenyuk and P. J. Stang, Chem. Rev., 2000, 100, 853.
S. R. Seidel and P. J. Stang, Acc. Chem. Res., 2002, 35, 972.
B. H. Northrop, H.-B. Yang and P. J. Stang, Chem. Commun., 2008, 5896.
ꢁ
1
3.207(2) A, i.e. similar to that in the iodido analogue. Each
chain follows the c-axis and is generated by the glide plane along
c. Packing of chains is illustrated in Fig. 7. Pairs of chains col-
oured green (or red) are related by inversion, and pairs of chains
5 M. Yoshizawa, J. K. Klosterman and M. Fujita, Angew. Chem., Int.
Ed., 2009, 48, 3418.
6
7
8
G. F. Swiegers and T. J. Malefetse, Chem. Rev., 2000, 100, 3483.
B. Moulton and M. J. Zaworotko, Chem. Rev., 2001, 101, 1629.
M. Fujita, K. Umemoto, M. Yoshizawa, N. Fujita, T. Kusukawa and
K. Biradha, Chem. Commun., 2001, 509.
coloured green and red are related by a glide along b or a. The
ꢀ
3
-chloropyridyl ring is rotated 38.0 out of the plane of the
ꢀ
9 D. W. Johnson and K. N. Raymond, Supramol. Chem., 2001, 13, 639.
0 B. J. HollidayandC. A. Mirkin, Angew. Chem., Int. Ed., 2001, 40, 2022.
1 G. F. Swiegers and T. J. Malefetse, Coord. Chem. Rev., 2002, 225, 91.
pyridine ring to which it is bonded (compare 42.3 for
1
1
[
{ZnI (3)} ]); the angles between bonded pairs of pyridine rings
2 n
ꢀ
ꢀ
in the tpy unit are 25.1 and 33.4 (compare 14.6 and 30.8 in
{ZnI (3)} ]). Crystal packing is dominated by CH/Nchloropy
hydrogen bonds and weak C–H/Cl contacts (shortest CH/Cl
12 L. F. Lindoy, Coord. Chem. Rev., 2008, 252, 811.
13 E. C. Constable, Coord. Chem. Rev., 2008, 252, 842.
14 E. C. Constable, Chem. Ind., 1994, 56.
[
2
n
33
1
5 See, for example: S. R. Batten, B. F. Hoskins and R. Robson, Angew.
Chem., Int. Ed., 1995, 34, 820; S. R. Batten, B. F. Hoskins and
R. Robson, J. Am. Chem. Soc., 1995, 117, 5385; B. F. Abrahams,
S. R. Batten, H. Hamit, B. F. Hoskins and R. Robson, Chem.
Commun., 1996, 1313; B. F. Abrahams, S. R. Batten, H. Hamit,
B. F. Hoskins and R. Robson, Angew. Chem., Int. Ed., 1996, 35,
ꢁ
¼
2.66–2.83 A compared to the sum of the van der Waals radii of
ꢁ
.00 A) and no efficient p-stacking interactions between
aromatic rings are present.
3
1
3
690; K. Biradha and M. Fujita, Angew. Chem., Int. Ed., 2002, 41,
392; O. Ohmori, M. Kawano and M. Fujita, Angew. Chem., Int.
Ed., 2005, 44, 1962; Y. Inokuma, T. Arai and M. Fujita, Nat.
Chem., 2010, 2, 780; Y. Inokuma, M. Kawano and M. Fujita, Nat.
Chem., 2011, 3, 349.
1
1
1
1
2
6 E. C. Constable, G. Zhang, C. E. Housecroft, M. Neuburger and
J. A. Zampese, CrystEngComm, 2009, 11, 2279.
7 E. C. Constable, G. Zhang, E. Coronado, C. E. Housecroft and
M. Neuburger, CrystEngComm, 2010, 12, 2139.
8 E. C. Constable, G. Zhang, C. E. Housecroft, M. Neuburger and
J. A. Zampese, CrystEngComm, 2010, 12, 2146.
9 E. C. Constable, G. Zhang, C. E. Housecroft, M. Neuburger and
J. A. Zampese, CrystEngComm, 2010, 12, 3733.
0 M. Barqu ꢂı n, J. Cancela, M. J. Gonz ꢂa lez Garmendia, J. Quintanilla
and U. Amador, Polyhedron, 1998, 17, 2373.
21 G. W. V. Cave and C. L. Raston, J. Supramol. Chem., 2002, 2, 317.
22 L. Hou and D. Li, Inorg. Chem. Commun., 2005, 8, 190.
23 B.-C. Wang, Q.-R. Wu, H.-M. Hu, X.-L. Chen, Z.-H. Yang,
Y.-Q. Shangguan, M.-L. Yang and G.-L. Xue, CrystEngComm,
2010, 12, 485.
Fig. 7 Packing of chains of [{ZnCl
2 n
(3)} ].
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 6864–6870 | 6869

View Article Online
2
4 J. Song, B.-C. Wang, H.-M. Hu, L. Gou, Q.-R. Wu, X.-L. Yang,
Y.-Q. Shangguan, F.-X. Dong and G.-L. Xue, Inorg. Chim. Acta,
32 A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2009, 65, 148.
33 T. Steiner, Angew. Chem., Int. Ed., 2002, 41, 48.
34 C. Liu, Y.-B. Ding, X.-H. Shi, D. Zhang, M.-H. Hu, Y.-G. Yin and
D. Li, Cryst. Growth Des., 2009, 9, 1275.
35 For recent examples, see: T. Hasell, X. Wu, J. T. A. Jones, J. Bacsa,
A. Steiner, T. Mitra, A. Trewin, D. J. Adams and A. I. Cooper,
Nat. Chem., 2010, 2, 750; X. Kuang, X. Wu, R. Yu, J. P. Donahue,
J. Huang and C.-Z. Lu, Nat. Chem., 2010, 2, 461; L.-X. Sun, Y. Qi,
Y.-X. Che, S. R. Batten and J.-M. Zheng, Cryst. Growth Des., 2009,
9, 2995.
36 G. R. Desiraju and T. Steiner, The Weak Hydrogen Bond, Oxford
University Press, Oxford, 1999, p. 223.
37 Conquest v. 1.13, CSD version 5.32, November 2010 with updates to
May 2011.
38 F. Kr o€ hnke, Synthesis, 1976, 1.
2011, 366, 134.
2
5 X.-Z. Li, M. Li, Z. Li, J.-Z. Hou, X.-C. Huang and D. Li, Angew.
Chem., Int. Ed., 2008, 47, 6371.
6 J. Heine, J. Schmedt auf der G u€ nne and S. Dehnen, J. Am. Chem.
Soc., 2011, 133, 10018.
7 C. B. Smith, C. L. Raston and A. N. Sobolev, Green Chem., 2005, 7, 650.
8 Stoe & Cie, IPDS Software v. 1.26, Stoe & Cie, Darmstadt, Germany,
2
2
2
1
996.
9 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
4, 112.
2
3
6
0 I. J. Bruno, J. C. Cole, P. R. Edgington, M. K. Kessler, C. F. Macrae,
P. McCabe, J. Pearson and R. Taylor, Acta Crystallogr., Sect. B:
Struct. Sci., 2002, 58, 389.
3
1 C. F. Macrae, I. J. Bruno, J. A. Chisholm, P. R. Edgington,
P. McCabe, E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de
Streek and P. A. Wood, J. Appl. Crystallogr., 2008, 41, 466.
39 Conquest v. 1.13, CSD version 5.32: refcodes AGUPEY, AGUPIC,
AJURIG, CUXDOP, FEPRUO, GAQYUS, LOCTED, NOGFIZ,
NOGFOF.
6
870 | CrystEngComm, 2011, 13, 6864–6870
This journal is ª The Royal Society of Chemistry 2011