Chromophore containing bipyridyl ligands. Part 1: Supramolecular solid-state structure of Ag(I) complexes

Article abstract of DOI:10.1039/b500698h

The solid-state structures of a series of azine or azo chromophore containing bipyridyl ligand complexes of Ag(i) salts have been determined by single-crystal X-ray diffraction. The supramolecular structures are dominated by one-dimensional chains formed through pyridyl-Ag-pyridyl bonding, but the packing of these chains through non-covalent intermolecular interactions is unpredictable. Ag...anion interactions are shown to be important, especially for nitrate and perchlorate species, but these may be supported or replaced by Ag...Ag, Ag...solvent, Ag...azine or Ag...π contacts. The molecular structures of the ligands show little alteration on complex formation, except for the AgNO₃ complex of N,N′-bis-pyridin-4-ylmethylenehydrazine where the normally planar azine ligand adopts a twisted geometry. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2005.

Full text of DOI:10.1039/b500698h

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
Chromophore containing bipyridyl ligands. Part 1: supramolecular
solid-state structure of Ag(I) complexes
Alan R. Kennedy,* Karen G. Brown, Duncan Graham, Jennifer B. Kirkhouse, Madeleine
Kittner, Claire Major, Callum J. McHugh, Paul Murdoch and W. Ewen Smith
Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow, UK G1 1XL
Received (in Durham, UK) 17th January 2005, Accepted 12th April 2005
First published as an Advance Article on the web 5th May 2005
The solid-state structures of a series of azine or azo chromophore containing bipyridyl ligand complexes of
Ag(I) salts have been determined by single-crystal X-ray diffraction. The supramolecular structures are
dominated by one-dimensional chains formed through pyridyl–Ag–pyridyl bonding, but the packing of these
chains through non-covalent intermolecular interactions is unpredictable. Agꢀ ꢀ ꢀanion interactions are shown
to be important, especially for nitrate and perchlorate species, but these may be supported or replaced by
Agꢀ ꢀ ꢀAg, Agꢀ ꢀ ꢀsolvent, Agꢀ ꢀ ꢀazine or Agꢀ ꢀ ꢀ p contacts. The molecular structures of the ligands show little
alteration on complex formation, except for the AgNO3 complex of N,N0-bis-pyridin-4-ylmethylene-
hydrazine where the normally planar azine ligand adopts a twisted geometry.
Our methodology for detecting RDX relies upon the ana-
Introduction
lytes 1 to 4 adhering to the surface of aggregated silver
nanoparticles. Whilst previous work concentrated on simply
detecting these species, it must also be true that the SERS
data collected contains information on the nature of the sur-
face species. To get true SERRS, ligand 5, which absorbs
light at higher wavelength and closer to the resonance condi-
tions of the lasers used, was also prepared and utilised. The
structures of the Ag(I) complexes of these ligands are of
relevance to this work, as knowledge of these will allow
solid-state Raman spectra to be correlated with specific ligand
coordination modes and geometries. This in turn will be used
to interpret SER(R)S spectra from complexed silver nanopar-
ticles, thus giving an insight into ligand binding at silver
nanoparticles.
Surface enhanced Raman scattering (SERS) and surface en-
hanced resonance Raman scattering (SERRS) have recently
been shown to be excellent techniques for the detection and
characterisation of a wide range of analytes at high dilutions,
including the high explosives Semtex1 and TNT,2 drugs of abuse3
and biopolymers such as DNA and proteins. 4 The detection of
hidden high explosives by determining the presence of their
vapour in the atmosphere is a major objective of our work, so
a significant result was the controlled reduction of RDX (1,3,5-
trinitro-1,3,5-triazacyclohexane, a major component of Semtex)
to hydrazine and its subsequent trapping with a number of
different carbonyl functionalities.1 This gave a number of azines,
including compounds 1 to 4 (Scheme 1), which could be
successfully detected at low concentrations using SERS.
1 to 5 are obviously also of the type of bipyridyl ligands
commonly used as linear ‘‘spacers’’ or ‘‘rods’’ in the construc-
tion of supramolecular coordination polymers.5–7 Such net-
works are of general interest, as the intriguing range of
structures which they display often leads to interesting physical
(especially electronic and magnetic) properties—and it is
hoped that studying their structures will eventually make
their self-assembly controllable or at least predictable. The
typically 2-coordinate, linear coordination of Ag(I) complexes
of these ligands have attracted particular interest. This is
both because their labile nature and relative simplicity (as
compared to transition metals with higher coordination num-
bers) makes the growth of single-crystals suitable for X-ray
diffraction studies easier and because it is rationalised that
complex 2D or 3D networks can be viewed as being formed
from aggregations of simpler 1D chains. Therefore, it can be
argued that an understanding of what governs the behavior of
Ag(I) 1D chains is a prerequisite to understanding the behavior
of their higher dimensional equivalents.5,6 In this paper we
report the solid-state structures of a series of complexes formed
by the reaction of Ag(I) salts with ligands 1 to 5. A second
paper will deal with SER(R)S spectra obtained from the
ligands in aqueous silver colloid solution. In combination with
the structures herein and solid-state Raman data, this will
provide rare evidence for the nature of the surface of silver
nanoparticles and for the surface modification of these nano-
particles by dyes.
Scheme 1
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 5
826
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2

View Article Online
Reaction of 1 with Ag(SbF6)
Experimental
0.249 g (0.78 mmol) of AgSbF6 was dissolved in 30 ml of
acetonitrile and then added to a 30 ml solution of 0.33 g (1.56
mmol) of 1 in CH2Cl2. After stirring for 10 minutes a light
yellow precipitate was isolated by filtration. 0.213 g, 49% yield
based on [Ag1][SbF6] ꢀ CH3CN. Mp 218–225 1C (decomp.).
N,N0-Bis-pyridin-4-ylmethylene-hydrazine, 1
To a solution of pyridine-4-carbaldehyde (4.494 g, 4 ml, 40
mmol) in ethanol (15 ml) was added dropwise hydrazine
hydrate (1.032 g, 1 ml, 21 mmol) with vigorous stirring. After
twenty minutes the pale precipitate was collected by filtration,
washed with cold methanol and dried to afford 1 as yellow
microcrystals (3.150 g, 75%). Mp 178 1C. lmax (MeOH)/nm
286 (e ¼ 36 702). 1H NMR (CDCl3) d 7.69 (4H, d, J ¼ 4.44 Hz,
l
max (solid-state) 284 nm. Major IR peaks (Nujol, cmꢂ1); 1625,
1320, 1245, 1070, 950, 815, 650. Anal. of oven dried sample.
Calcd for C12H10AgSbF6N4: C, 26.02; H, 1.82; N, 10.12.
Found C, 26.04; H, 1.67; N, 10.02%.
ArH), 8.56 (2H, s, CH N), 8.75 (4H, d, J ¼ 4.44 Hz, ArH).
Q
m/z (EI) 210.090 76 [C12H10N4 (M)1 o 1.0 ppm]. Anal. Calcd
for C12H10N4: C, 68.57; H, 4.76; N, 26.66. Found C, 68.56;
H, 4.57; N, 26.53%.
Reaction of 1 with Ag(NO3)
0.132 g (0.78 mmol) of AgNO3 was dissolved in 30 ml of
acetonitrile and then added to a 30 ml solution of 0.33 g (1.56
mmol) of 1 in CH2Cl2. After stirring for 10 minutes a light
yellow precipitate was isolated by filtration. 0.231 g, 77% yield
based on [Ag1][NO3]. Mp 232 1C (decomp.). lmax (solid-state)
275 nm. Major IR peaks (Nujol, cmꢂ1); 1615, 1320, 1310, 1220,
820, 680. Anal. Calcd for C12H10AgO3N5: C, 37.92; H, 2.65; N,
18.43. Found C, 37.32; H, 2.59; N, 18.09%.
N,N0-Bis-pyridin-3-ylmethylene-hydrazine, 2
As per the procedure for azine 1 using pyridine-3-carbaldehyde
(4.494 g, 4 ml, 42 mmol) to afford 2 as yellow microcrystals
1
(2.688 g, 64%). lmax (MeOH)/nm 297 (e ¼ 33 010). H NMR
(DMSO) d 7.55 (2H, dd, J ¼ 4.80, 4.80 Hz, ArH), 8.27 (2H, d,
J ¼ 7.88 Hz, ArH), 8.70 (2H, d, J ¼ 4.64 Hz, ArH), 8.79 (2H, s,
CH N), 9.02 (2H, s, ArH). m/z (EI) 210.091 84 [C H N
Q
12 10
4
Reaction of 2 with Ag(BF4)
(M)1 o 6.2 ppm]. Anal. Calcd for C12H10N4: C, 68.57; H,
4.76; N, 26.66. Found C, 68.48; H, 4.71; N, 26.32%.
A solution of Ag(BF4) (0.156 g, 0.80 mmol) in 30 ml of
acetonitrile was added to 30 ml of a CH2Cl2 solution of 2
(0.329 g, 1.57 mmol). After stirring for 10 minutes a pale yellow
precipitate of [Ag2][BF4] was collected by filtration. 0.084 g,
26% yield based on [Ag2][BF4] ꢀ CH3CN. Mp 193 1C (de-
comp.). Major IR peaks (KBr, cmꢂ1); 3095, 2990, 1620,
1420, 1312, 1050, 805, 702.
N,N0-Bis-pyridin-2-ylmethylene-hydrazine, 3
As per the procedure for azine 1 using pyridine-2-carbaldehyde
(4.494 g, 4 ml, 42 mmol) to afford 3 as yellow microcrystals
1
(3.906 g, 93%). lmax (MeOH)/nm 298 (e ¼ 308 20). H NMR
(DMSO) d 7.52 (2H, t, J ¼ 4.96, 6.28 Hz, ArH), 7.95 (2H, t,
J ¼ 6.40, 7.64 Hz, ArH), 8.11 (2H, d, J ¼ 7.84 Hz, ArH), 8.57
Reaction of 3 with Ag(NO3)
(2H, s, CH N), 8.72 (2H, d, J ¼ 4.72 Hz, ArH). m/z (EI)
Q
210.090 74 [C12H10N4 (M)1 o 0.9 ppm]. Anal. Calcd for
C12H10N4: C, 68.57; H, 4.76; N, 26.66. Found C, 68.41; H,
4.71; N, 26.19%.
30 ml of an acetonitrile solution of Ag(NO3) (0.136 g, 0.80
mmol) was added to a solution of 3 (0.331 g, 1.58 mmol) in
30 ml of CH2Cl2. After 10 minutes of stirring a bright yellow
precipitate of [{Ag3}2][NO3]2 was collected by filtration (0.250
g, 82%). Mp 232 1C (decomp.). Major IR peaks (KBr, cmꢂ1);
3053, 2955, 1634, 1583, 1470, 1400, 1290, 1219, 1030, 738, 702.
Anal. Calcd for C24H20Ag2O6N10: C, 37.92; H, 2.65; N, 18.43.
Found C, 37.94; H, 2.46; N, 18.13%.
N,N0-Bis-(1-pyridin-4-yl-ethylidene)-hydrazine, 4
To a solution of 4-acetylpyridine (1.089 g, 9.0 mmol) in ethanol
(20 ml) was added hrdrazine hydrate (2.30 g, 4.5 mmol) and
acetic acid (2 drops). The resulting solution was heated to
reflux for two hours, cooled and then neutrilised by dropwise
addition of NaHCO3 (1 M, pH 8). Ethanol was removed
in vacuo to afford 4 as an off white powder. After washing
with cold ethanol and drying 0.536 g (50%) were recovered.
Reaction of 5 with Ag(NO3)
Orange crystals of [Ag5][NO3] were grown at the interface
between an acetonitrile solution of AgNO3 (0.20 g, 1.2 mmol)
layered onto a CH2Cl2 solution of azo ligand 5 (0.22 g, 1.2
mmol). Mp 286 1C (decomp.). Major IR peaks (KBr, cmꢂ1);
3457, 2427, 1644, 1588, 1480, 1378, 1224, 1050, 840.
1
lmax (MeOH)/nm 269 (e ¼ 31 105). H NMR (CDCl3) d 2.28
(6H, s, 2 ꢁ CH3), 7.75 (4H, d, J ¼ 4.56 Hz, ArH), 8.70 (4H, d,
J ¼ 5.28 Hz, ArH). m/z (EI) 238.120 74 [C14H14N4 (M)1
o
4.6 ppm]. Anal. Calcd for C14H14N4: C, 70.58; H, 5.88; N,
23.52. Found C, 70.35; H, 5.72; N, 22.99%.
Reactions of 1, 2 and 4 with Ag(ClO4)
Due to the safety hazards associated with perchlorate salts no
attempt was made to complete these reactions in bulk. In each
case a few yellow crystals of the Ag(I) complexes were grown at
the interface between 10 ml of an acetonitrile solution of
AgClO4 (0.02 g, 0.01 mmol) layered onto a 10 ml CH2Cl2
solution of azine ligand (0.02 mmol).
trans-4,40-Azobis(pyridine), 5
5 was synthesised as the dihydrate in 40% yield using the
8
method of Clarke
NaOCl).
(treatment of 4-aminopyridine with
Reaction of 1 with Ag(BF4)
Crystallography
0.15 g (0.78 mmol) of AgBF4 was dissolved in 30 ml of
acetonitrile and then added to a 30 ml solution of 0.33 g
(1.56 mmol) of 1 in CH2Cl2. After stirring for 10 minutes a
light yellow precipitate was isolated by filtration, 0.184 g, 58%
yield based on [Ag1][BF4] ꢀ CH3CN. Mp 200–205 1C (de-
comp.). Major IR peaks (Nujol, cmꢂ1); 1610, 1310, 1290,
1025, 820, 698. Further analysis was limited as the powder
degraded over a period of 5 to 7 days.
Measurements on 1A, 1B, 2A, 2B, 3A, 4A and 5A were made
on a Nonius Kappa CCD diffractometer and those on 1C and
1D were made with a Rigaku AFC7S diffractometer. Both used
graphite monochromated Mo-Ka radiation (l ¼ 0.7107 A).
Data were corrected for absorption by multi-scan or psi-scan
methods respectively, depending on the data collection instru-
mentation.9 All structures were refined to convergence on F2
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2
827

View Article Online
Table 1 Crystallographic parameters and refinement data for Ag complexes
1A 1B 1C 1D 2A
2B
3A
4A
5A
[Ag1][BF4] ꢀ [Ag1][ClO4] ꢀ [Ag1][SbF6] ꢀ
[Ag1][NO3] [Ag2][BF4] ꢀ [Ag2][ClO4] ꢀ [{Ag3}2][NO3]2 [Ag4][ClO4] ꢀ [Ag5][NO3]
MeCN
445.97
MeCN
458.61
Monoclinic Monoclinic
MeCN ꢀ xH2O
MeCN
445.97
Monoclinic Triclinic
MeCN
458.61
0.25[4]
505.18
Triclinic
Fw
626.44a
Triclinic
380.12
760.24
Triclinic
354.09
Monoclinic
C2/c
System
Space
group
a/A
b/A
c/A
Monoclinic
P21/n
ꢀ
P1
ꢀ
P1
ꢀ
P1
ꢀ
P1
P21/m
P21/m
P21/a
6.0846(2)
15.6681(4)
8.6321(3)
90
6.0509(2)
15.8105(4)
8.6688(2)
90
10.496(2)
11.904(2)
8.6632(12)
97.956(13)
110.611(9)
92.935(13)
997.5(2)
2
6.7441(12)
13.452(2)
15.028(2)
90
7.5292(2)
9.7601(3)
11.8669(4)
75.444(1)
74.672(1)
83.659(2)
813.13(1)
2
9.3202(4)
14.2960(6)
13.2368(4)
90
8.0933(2)
8.8628(3)
9.3988(4)
89.868(2)
82.849(2)
72.614(4)
637.93(4)
1
10.4948(3)
13.9768(4)
14.3519(4)
64.474(2)
88.167(2)
77.080(1)
1846.64
4
25.0213(5)
9.3011(2)
15.9564(4)
90
a/1
b/1
96.589(1)
90
97.244(1)
90
97.568(13)
90
110.178(2)
90
109.683(1)
90
g/1
V/A3
Z
817.50(4)
2
150
822.71(4)
2
123
1351.5(4)
4
123
1655.45(11)
4
123
3496.49(13)
12
123
T/K
123
4841
123
15331
123
17254
2893
123
15793
No. refl. 8971
3719
1935
3284
3019
32485
3781
0.048
13990
4183
Unique
Rint
R
1878
4589
0.0370
3634
0.0429
8459
0.0382
0.0712
0.0305
0.0787
0.0289
0.0253
0.0653
0.0354
0.0347
0.0862
0.0520
0.0254
0.0577
0.0483
0.0311
0.0645
0.0307
0.0804
0.0296
0.0567
0.0300
0.0686
0.0500
0.1117
wR2
a
Based on 1.75 H2O molecules per asymmetric unit.
using Shelxl-97.10 Further crystallographic parameters and
refinement statistics can be found in Table 1.
CCDC reference numbers 266716–266724.
See http://www.rsc.org/suppdata/nj/b5/b500698h/ for crys-
tallographic data in CIF or other electronic format.
of the powders where available, allows the assumption that
they are chemically similar.
The structures of the silver complexes of 1 and 2 appear on
initial examination to consist of simple 1D polymeric chains
constructed by the ligands bridging near linear, two-coordinate
metal centers (Fig. 1). Closer inspection shows that this
apparent simplicity is misleading. The coordinative unsatura-
tion about the Ag atoms encourages the formation of non-
covalent intermolecular interactions, which link the individual
chains to form higher dimensional networks. In 1A the octa-
hedral Ag coordination is completed by two mutually trans
Agꢀ ꢀ ꢀanion interactions (Agꢀ ꢀ ꢀF 2.828(2) A) and two mutually
trans Agꢀ ꢀ ꢀp contacts (closest Agꢀ ꢀ ꢀC3 is 3.280(2) A). As can
be seen in Fig. 2 this causes the 1D chains to form sheets
through bridging BF4 anions. These sheets lie offset to each
other and form a 3D network via the Agꢀ ꢀ ꢀp interactions. The
importance of the anion in these systems is shown in that
replacing BF4 with the similarly shaped ClO4 gives an essen-
tially isostructural complex (1B) whilst replacing BF4 with
SbF6 fundamentally alters the inter-chain bonding. As well
as being a considerably larger anion, SbF6 also has a lower
charge per F atom than BF4 or ClO4, which leads to it being a
less effective metal binding species. Thus unlike 1A and 1B, the
Results and discussion
Solid-state structure of Ag(I) complexes of 1
The dyes 1 to 5 (Scheme 1) are easily prepared in good yield. Of
these, the azo 5 has widely been reported to form coordination
networks, with a search of the Cambridge Crystallographic
database11 giving 50 structures with coordinated metals,12
whilst the coordination chemistry of 1 to 4 is less well known.13
All the dyes would be classified as ‘‘off-axis rod type’’ ligands
by the method described in ref. 5 as, in the expected trans
geometry, each will display two metal–N vectors rotated by
1801 but offset from each other by the stepping effect of the
azine or azo link. In terms of size and shape, 5 has an obvious
analogue in 1,2-bis{40-pyridyl}ethylene which has itself been
widely studied,14 however we are not aware of any correspond-
ing simple equivalents of the azine ligands. Mixing the ligands
with Ag(I) salts led to rapid deposition of fine powders. Scaling
down the bulk preparative techniques described in the Experi-
mental section and layering acetonitrile solutions of the Ag
salts onto dichloromethane solutions of the ligands gave
crystals suitable for single crystal analysis. This showed the
crystals to be {[Ag1][BF4] ꢀ NCMe}N, hereafter 1A, {[Ag1]
[ClO4] ꢀ NCMe}N 1B, {[Ag1][SbF6] ꢀ xNCMe ꢀ yOH2}N 1C,
{[Ag1][NO3]}N 1D, {[Ag2][BF4] ꢀ NCMe}N 2A, {[Ag2][ClO4] ꢀ
NCMe}N 2B, [(Ag3)2][NO3]2 3A, {[Ag4][ClO4] ꢀ 0.25(4)}N 4A
and {[Ag5][NO3]}N 5A. These single crystal samples have
identical IR and solid-state Raman spectra to their powder
counterparts which, taken together with the microanalysis data
Fig. 2 Part of a 2D sheet in 1A showing the chains linked by bridging
BF4 anions. The 3D structure is obtained by off-setting the adjoining
2D sheets so that Agꢀ ꢀ ꢀp bonds are formed with the pyridyl rings.
Fig. 1 Part of the linear chain structure of 1A.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2
828

View Article Online
Table 2 Coordination geometry about the Ag atom (A and 1)
1A
1B
1C
1D
2A
2B
4A0
4A00
5A0
5A00
Dative bonds
Ag–N (py)
Ag–N (py)
Ag–N (NCMe)
py–Ag–py
Agꢀ ꢀ ꢀanion
Agꢀ ꢀ ꢀF
2.143(2)
2.143(2)
2.153(2)
2.153(2)
2.168(3)
2.177(2)
2.586(4)
167.9(1)
2.181(3)
2.182(3)
2.147(2)
2.153(2)
2.600(2)
167.03(7)
2.155(2)
2.157(2)
2.702(2)
2.134(4)
2.134(4)
2.143(3)
2.154(4)
2.170(2)
2.177(2)
2.187(2)
2.187(2)
180.0
180.0
173.4(1)
175.56(8) 174.47(13) 174.28(13) 168.27(9)
180.0
2.828(2)
2.828(2)
3.247(4)
3.387(2)
Agꢀ ꢀ ꢀF
Agꢀ ꢀ ꢀOa
2.837(2)
2.837(2)
2.682(3)
2.730(3)
2.747(3)
2.911(2)
3.009(2)
2.813(7)
2.653(7)
2.865(7)
2.764(2)
2.811(2)
2.791(2)
3.119(2)
2.838(2)
2.906(2)
2.838(2)
2.906(2)
Agꢀ ꢀ ꢀOa
2.830(11)
Agꢀ ꢀ ꢀO
Agꢀ ꢀ ꢀp
Agꢀ ꢀ ꢀC or N
(shortest)
Agꢀ ꢀ ꢀC or N
(shortest)
3.280(2)
para-C
3.280(2)
para-C
3.271(2)
para-C
3.271(2)
para-C
3.116(4)
azine-N
3.366(4)
ring
3.328(2)
ortho-C
3.371(4)
meta-C
3.329(4)
meta-C
3.308(2)
ortho-C
3.308(2)
ortho-C
centroid
Agꢀ ꢀ ꢀAg
Agꢀ ꢀ ꢀAg
a
3.1131(4)
3.2719(5)
3.2719(5)
For both geometries of 5A the nitrate anions bind in an Z2 mode. Both 4A and 5A have two crystallographically independent molecules present,
these are denoted above as 4A0, 4A00, 5A0 and 5A00.
structure of 1C is not cross-linked through anion interactions.
Although in 1C the Ag atom can be described as having six
pseudo-octahedral contacts, two of these are very long (Agꢀ ꢀ ꢀF
is 3.247(4) and the shortest Agꢀ ꢀ ꢀC is 3.586(4) A, which are
approximately 0.42 and 0.30 A longer than their respective
equivalents in 1A) leaving four shorter/stronger Ag–N bonds
(see Table 2 for details). Of these stronger bonds, that to the
NCMe solvent is terminal, and so as far as network formation
is concerned each Ag site acts as a T-shaped 3-fold junction
and a ladder structure is formed (Fig. 3). Although the ladder is
drawn so as to emphasise its inter-chain Agꢀ ꢀ ꢀN(azine)
‘‘rungs’’ it should be noted that the ladder is also positioned
so as to allow face-to-face aromatic interactions of the pyridyl
rings (Cꢀ ꢀ ꢀC min. 3.337 A). These are in the more common and
energetically favourable head-to-tail orientation15 and it may
largely be this rather than the Agꢀ ꢀ ꢀN(azine) interaction that
holds the ladders together. The looser interconnectivity of 1C
leaves space for extra, disordered solvent (a mixture of NCMe
and water) and also for the anion to be disordered.
perpendicular to the b axis. Pairs of sheets are joined by
ligating NO3 anions to give a ‘‘double-layer’’ sheet. This
architecture is tightly enough packed to exclude solvent mole-
cules, and it also seems that it is energetically favoured enough
to overcome the azine’s preferred13a,17 flat geometry (the
torsion angle around the central N–N bond, C6N2N3C7, is
138.1(4)1). This is the only complex reported herein where there
is a significant change in the geometry of the ligand upon metal
complexation (Fig. 4). It is possible that this is because the
structural motif imposed by the strongly coordinating nitrate
groups features head-to-head ordering of the pyridyl groups.
As mentioned above this is relatively unfavourable for pyridyl
groups (due to the large electrostatic effect caused by metal
complexation),15 and so the azines twist to give edge-to-face
Overall the Agꢀ ꢀ ꢀanion and Agꢀ ꢀ ꢀp interactions of 1A and
1B are replaced in 1C by Agꢀ ꢀ ꢀsolvent and Agꢀ ꢀ ꢀN(azine)
contacts. All the contacts observed in 1C are potentially
available to 1A and 1B (both have non-bonded NCMe in their
lattices) and so it can be concluded that for this system the
relatively highly coordinating BF4 and ClO4 anions dominate
inter-chain packing, as only in their absence do Agꢀ ꢀ ꢀsolvent
and Agꢀ ꢀ ꢀN(azine) contacts occur. It would thus be expected
that introduction of the more highly coordinating NO3 anion
would increase the domination of the inter-chain interactions
by Agꢀ ꢀ ꢀanion contacts. Examination of the structure of 1D
shows that this is so.16 There are no contacts with Ag apart
from the Ag–N(py) and Agꢀ ꢀ ꢀO3N interactions. In 1D the Ag
coordination is five-coordinate and square-pyramidal with
both NO3 groups and azine ligands acting as bridges between
metals. This gives a structure based around 2D sheets lying
Fig. 4 Detail of the molecular structure of 1D showing the twisted
nature of the dye ligand.
Fig. 3 The ladder structure of 1C with H-atoms omitted for clarity.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2
829

View Article Online
Fig. 7 Brick-wall structure of 2A.
supported by short Agꢀ ꢀ ꢀp interactions (with the pyridyl rings
trans to the Agꢀ ꢀ ꢀAg contacts). The coordination shell of Ag is
completed (Table 2) by a terminally bonded NCMe (trans to a
BF4 anion that is itself too far from Ag to be considered as a
ligand). Whilst the structural difference between 1A and 2A can
be rationalised once it is realised that the zig-zag nature of the
2A chains profoundly effect the chain packing, it is harder to
explain why for 1 the BF4 and ClO4 salts are isostructural but
for 2 they are not. 2B differs from 2A as here the polymeric
chains pair through Z2-ClO4, [Agꢀ ꢀ ꢀO–Cl–Oꢀ ꢀ ꢀ]2 bridges (see
Fig. 8) and do not aggregate further. The Agꢀ ꢀ ꢀp and Agꢀ ꢀ ꢀAg
contacts seen in 2A are absent and the Ag to solvent distance is
lengthened.
Fig. 5 Part of the double sheet structure of 1D highlighting the edge-
to-face nature of the ring-to-ring contacts.
aromatic interactions instead of the more common face-to-face
interactions (Fig. 5). Calculations in ref. 17 show that in azines
the N–N bond is rotationally soft and so the energy needed to
twist 1 can be easily regained by maximising inter-chain
interactions.
The AgClO4 complex of 4, a methyl substituted derivative of
1, reverts to the straight chain structural type found for all the
complexes of 1. However, the extended structure of 4A is
similar to 2B in that the pairs of crystallographically indepen-
dent polymeric chains link together by ClO4 bridges between
Ag atoms. Here, however, the bridge involves only one O atom
(at least in the major configuration of the disordered anions),
leading to a much shorter Agꢀ ꢀ ꢀAg contact distance (3.2719(5)
and 3.814(1) A in 4A and 2B, respectively). Despite the short
Agꢀ ꢀ ꢀAg contact the structure is very different from that of 2A,
as in 4A there is no solvent to Ag interaction. Indeed, the
NCMe seems to have been replaced by a discrete molecule of 4
lying orthogonal to the polymeric Ag4 chains, see Fig. 9.
The structure of 5A is also hard to predict. Not only is a
psuedo-polymorphic structure, 5B, formed under slightly dif-
Solid-state structure of Ag(I) complexes of 2, 3, 4 and 5
The difficulties of deciphering the rules that predict 3D packing
are illustrated by extending the study to the structures of Ag(I)
complexes of 2, 3, 4 and 5. Whereas the complexes of 1 form a
series of structures that can be rationalised by understanding
the simple differences between their anions’ bonding character-
istics, attempting to extrapolate this behaviour to the other
ligands fails. Reaction of 2 with AgBF4 or AgClO4 gives 2A
and 2B, the chain structures of which contrast with those of the
linear complexes of 1 as they adopt pronounced zig-zag shapes
(compare Fig. 1 with Fig. 6). Similar effects are seen, amongst
18
others, on going from {[Ag(4,40-pytz)][BF4]}N to {[Ag(3,30-
19
pytz)][BF4]}N
(pytz ¼ 3,6-bis(pyridinyl)-1,2,4,5-tetrazine)
and have been discussed in terms of the change from linear
to off-axis rod ligands.5 That this is not quite accurate is shown
here as both 1 and 2 are of the off-axis type. It would be better
to describe the change from linear chain to zig-zag chain in
terms of the angle between the rod axis and the N–Ag axis.
Only if these are approximately parallel can a straight chain
form.
2A has the expected py–Ag–py polymer chain, but the BF4
anion does not then dominate the inter-chain packing as in 1A.
This may be because the zig-zag nature of the chains leads to
the dye molecules impinging on the space needed for a putative
BF4 bridge. The coordination network is instead completed by
a short Agꢀ ꢀ ꢀAg contact (3.1131(4) A). This leads to a corru-
gated 2D sheet of the brick wall type (Fig. 7) that is internally
Fig. 8 Detail of the structure of 2B illustrating the perchlorate
bridging mode.
Fig. 6 The primary structural feature of 2A is a zig-zag chain.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2
830

View Article Online
Fig. 11 The dimeric molecular structure of 3A with nitrate ions
omitted.
Fig. 9 View of 4A down the a axis, showing the relative orientations
of the polymer chains and the free molecules of ligand 4.
favours dimer formation over polymer formation. In 3A the N
coordination about Ag is distorted from T-shaped trigonal due
to the small bite angle of the chelate (N3–Ag1–N4 71.45 (7)1)
and, interestingly, whilst the distance to the azine N is compar-
able to that to the trans pyridyl (2.264(2) and 2.228(2) A,
respectively) it is considerably shorter than the Ag to chelated
pyridyl bond (2.435(2) A). As reported above, only the twisted
ligand geometry of 1D shows any sign that complex forma-
tion alters the dye ligands’ internal geometry. Even for 3A
with its direct azine to Ag bond there is no apparent change in
the azines’ bond lengths or angles (for example N2–N3 is
1.411(3) compared to 1.407(5) to 1.416(6) A found herein for
ferent reaction conditions,20 but 5A itself contains two different
linear-chain geometries, which interconnect through anion
bridges and p interactions. Both chains exhibit the expected
governing linear N–Ag–N motif (N3–Ag1–N4 168.27(9) and
N6–Ag2–N6* 180.01) but whilst both Ag centres also contact
two bidentate nitrate anions only Ag2 makes further connec-
tions through two Agꢀ ꢀ ꢀpyridyl interactions (Ag2ꢀ ꢀ ꢀC1*
3.308(2) A). This is very different from the reported structure
of 5B 20 in which the nitrate unexpectedly plays no role in inter-
chain contacts with only Agꢀ ꢀ ꢀAg and longish pꢀ ꢀ ꢀp contacts
reported. 5A is also unusual (and again differs from 5B) as the
1D chains do not all run parallel to each other,5 eschewing this
simple and effective packing mode to instead cross at a
pronounced angle (Fig. 10). A crystallographic curiosity is that
the structure contains approximately 3% of the monochlori-
nated azo dye as an impurity (as often happens when forming
azos through amine reduction by sodium hypochlorite),8 but
that this impurity is only included in the chain bonding
through Ag2.
complexes of 1 and 2, the C N distances and the angles
Q
subtended by N in 3A also lie within the ranges found for the
other complexes).
Conclusions
From the structural descriptions above it is obvious that
predicting the packing arrangement of py–Ag–py chains is
not trivial. Changing the anion changes the structure and to
some extent this change is predictable when the anion’s metal
bonding characteristics are considered. Thus, for ligand 1 the
relatively highly coordinating NO3 anion dominates the inter-
chain packing in 1D and excludes all other contact types, in 1B
and 1A the similarly sized and shaped ClO4 and BF4 anions
contribute to inter-chain packing whilst, in 1C, the SbF6 anion
makes little contribution and other intermolecular interactions
dominate. In agreement with this trend, NO3 also dominates
the inter-chain bonding in 5A. Indeed it may be the strength of
these interactions that leads to the absence of NCMe from all
three of the NO3 salts examined herein. The other six com-
plexes all contain NCMe—except for 4A where its place is
unexpectedly taken by a ‘‘solvent’’ molecule of 4. The beha-
viour of the ClO4 anions is also consistent, in that all bridge
two metal centers. However, there is much more that does not
fit any obvious pattern. For example, the NO3 and ClO4 anions
all bridge metals but they do not do so in any consistent way
and adopt a variety of bonding modes. Furthermore, the
previously published polymorph of 5A, 5B has non-bonding
NO3 anions and, for ligand 1, the BF4 anion acts in the same
fashion as ClO4, but for ligand 2 the two anions give different
structures. A final unpredictable feature is the non-parallel
packing of py–Ag–py chains found for 5A.
Although some 2-pyridyl species are known to form coordi-
nation polymers with Ag(I),21 the presence of hetero-atoms
between the pyridyl rings in 3 favours chelation. Indeed
reaction with AgNO3 gives the planar, centrosymmetric,
dimer [(Ag3)2][NO3]2, 3A, as shown in Fig. 11. The structurally
similar O3SCF3 and PF6 salts of [(Ag3)2] have been re-
22,23
ported
but the BF4 salt is reported to form a polymeric
network similar to that known for a Cu(I) complex of 3.23
Anion control of these structures is not obvious, however it
may be worth noting the site disorder present in the structure
of the PF6 salt. This can be interpreted as evidence of a
polymeric nature23 and if this was the case then the greater
bonding capabilities of the O containing anions may be what
24
In conclusion, we have synthesised and examined a series of
silver compounds which further our understanding of the
fundamental forces that govern the packing of 1D polymeric
coordination compounds into 2D and 3D networks. Of parti-
cular note is the crucial role played by the anions and the
differences in packing arrangements available to linear as
opposed to zig-zag chains but other, currently unidentified,
factors also contribute and thus make general structure pre-
diction unreliable.
Fig. 10 Part of the crystal structure of 5A viewed along the a axis,
illustrating the non-parallel packing of the polymer chains.
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2
831

View Article Online
12 For examples see: (a) L. Carlucci, G. Ciani and D. M. Proserpio,
New J. Chem., 1998, 1319; (b) L. Carlucci, G. Ciani and D. M.
Proserpio, J. Chem. Soc., Dalton Trans., 1999, 1789; (c) M.
Kondo, M. Shimamura, S. Noro, S. Minakoshi, A. Asami, K.
Seki and S. Kitagawa, Chem. Mater., 2000, 12, 1288; (d) M. A.
Withersby, A. J. Blake, N. R. Champness, P. A. Cooke, P.
Acknowledgements
We thank the Home Office UK for funding CJM, the BBSRC
for the award of a David Philips Fellowship to DG and Avecia
Ltd and the EPSRC for funding JBK.
Hubbersty, A. L. Realf, S. J. Teat and M. Schro
Soc., Dalton Trans., 2000, 3261.
¨
der, J. Chem.
References
13 (a) D. M. Ciurtin, Y.-B. Dong, M. D. Smith, T. Barclay and H.-C.
zur Loye, Inorg. Chem., 2001, 40, 2825; (b) Y. Diskin-Posner, G.
K. Patra and I. Goldberg, J. Chem. Soc., Dalton Trans., 2001,
2775; (c) Y.-B. Dong, M. D. Smith and H.-C. zur Loye, Inorg.
Chem., 2000, 39, 4927.
14 For example: (a) O.-S. Jung, S. H. Park, K. M. Kim and H. G.
Jang, Inorg. Chem., 1998, 37, 5781; (b) J. Y. Lu and A. M. Babb,
Inorg. Chem., 2001, 40, 3261; (c) S. R. Batten, A. R. Harris,
P. Jensen, K. S. Murray and A. Ziebell, J. Chem. Soc., Dalton
Trans., 2000, 3829.
15 C. Janiak, J. Chem. Soc., Dalton Trans., 2000, 3885.
16 The structure of 1D has been reported by G. Kumar and I.
Goldberg, Cryst. Growth Des., 2003, 3, 321, but all geometric
parameters discussed are based on our own measurements.
17 E. C. Kesslen, W. B. Euler and B. M. Foxman, Chem. Mater.,
1999, 11, 336.
1
C. J. McHugh, W. E. Smith, R. J. Lacey and D. Graham, Chem.
Commun., 2002, 2514.
2
(a) C. J. McHugh, R. Keir, D. Graham and W. E. Smith, Chem.
Commun., 2002, 580; (b) R. Keir, E. Igata, M. Arundell, W. E.
Smith, D. Graham, C. J. McHugh and J. Cooper, Anal. Chem.,
2002, 74, 1503.
3
4
K. Faulds, W. E. Smith, D. Graham and R. J. Lacey, Analyst,
2002, 127, 282.
(a) L. Fruk, A. Grondin, W. E. Smith and D. Graham, Chem.
Commun., 2002, 2100; (b) D. Graham, R. Brown and W. E. Smith,
Chem. Commun., 2001, 1002; (c) D. Graham, B. S. Mallinder,
D. Whitcombe and W. E. Smith, ChemPhysChem, 2001, 2, 746; (d)
D. Graham, B. S. Mallinder, D. Whitcombe, N. D. Watson and
W. E. Smith, Anal. Chem., 2002, 74, 1069.
5
6
7
A. N. Khlobystov, A. J. Blake, N. R. Champness, D. A. Leme-
novskii, A. G. Majouga, N. V. Zyk and M. Schroder, Coord.
Chem. Rev., 2001, 222, 155 and references therein.
A. J. Blake, G. Baum, N. R. Champness, S. S. M. Chung, P. A.
Cooke, D. Fenske, A. N. Khlobystov, A. A. Lemenovskii, W.-S.
18 M. A. Withersby, A. J. Blake, N. R. Champness, P. Hubbersty,
W.-S. Li and M. Schroder, Angew. Chem., Int. Ed. Engl., 1997, 36,
¨
¨
2327.
19 M. A. Withersby, A. J. Blake, N. R. Champness, P. A. Cooke, P.
Li and M. Schroder, J. Chem. Soc., Dalton Trans., 2000, 4285.
¨
(a) A. J. Blake, N. R. Champness, P. Hubbersty, W.-S. Li, M. A.
Hubbersty, W.-S. Li and M. Schroder, Crystal Engineering, 1999,
2, 123.
¨
Withersby and M. Schroder, Coord. Chem. Rev., 1999, 183, 117;
20 C. He, B.-G. Zhang, C. Duan, J. Li and Q.-J. Meng, Eur. J. Inorg.
Chem., 2000, 2549.
21 (a) C. Richardson and P. J. Steel, Inorg. Chem. Commun., 1998, 1,
260; (b) M.-L. Tong, X.-M. Chen, B.-H. Ye and S. W. Ng, Inorg.
Chem., 1998, 37, 5278.
¨
(b) M. J. Zaworotko, Chem. Commun., 2001, 1; (c) M. Fujita,
D. Oguro, M. Miyazawa, H. Oka, K. Yamaguchi and K. Ogura,
Nature, 1995, 378, 469.
8
9
T. Clarke, J. Org. Chem., 1971, 36, 3816.
(a) R. H. Blessing, J. Appl. Crystallogr., 1997, 30, 421; (b) A. C. T.
North, D. C. Phillips and F. S. Mathews, Acta Crysallogr., Sect.
A, 1968, 24, 351.
22 J. Hamblin, A. Jackson, N. W. Alcock and M. J. Hannon,
J. Chem. Soc., Dalton Trans., 2002, 1635.
23 F. Tuna, J. Hamblin, J. Jackson, G. Clarkson, N. W. Alcock and
M. J. Hannon, J. Chem. Soc., Dalton Trans., 2003, 2141.
24 G. Dong, H. Cheng, D. Chun-Ying, Q. Chun-Qi and M. Queng-
Jin, New J. Chem., 2002, 26, 796.
10 G. M. Shedrick, SHELXL-97, a program for crystal structure
refinement, University of Gottingen, Gottingen, Germany, 1997.
11 F. H. Allen, Acta Crystallogr., Sect. B, 2002, 58, 380.
¨
¨
832
N e w J . C h e m . , 2 0 0 5 , 2 9 , 8 2 6 – 8 3 2