A comparative investigation of supramolecular structures involving copper(II) complexes of imidazolinylalkanimidamides

Article abstract of DOI:10.1039/b207031f

The syntheses of three new copper(II) complexes based on the bidentate ligand N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)ethanimidamide are reported. In the solid state these products assemble into elaborate supramolecular structures featuring hydrogen bonds within, and between, complexes as well as weak aryl-aryl and alkyl-aryl interactions between complexes; different coordination geometries and tautomeric forms are also observed. Dimethyl sulfoxide molecules show a variety of interactions within the structures, including coordination to the copper, hydrogen bonding, as well as weak alkyl-aryl interactions with the ligands.

Full text of DOI:10.1039/b207031f

View Article Online / Journal Homepage / Table of Contents for this issue
A comparative investigation of supramolecular structures involving
copper(II) complexes of imidazolinylalkanimidamides
Michael M. Bishop,a,b Leonard F. Lindoy,*a Daniel J. Millerb and Peter Turnera
a Centre for Heavy Metals Research, School of Chemistry, University of Sydney, NSW, 2006,
Australia
b Sydney Grammar School, College Street, Darlinghurst, NSW, 2010, Australia
Received 18th July 2002, Accepted 18th September 2002
First published as an Advance Article on the web 24th October 2002
The syntheses of three new copper() complexes based on the bidentate ligand N-(4-oxo-5,5-diphenyl-4,5-dihydro-
1H-imidazol-2-yl)ethanimidamide are reported. In the solid state these products assemble into elaborate
supramolecular structures featuring hydrogen bonds within, and between, complexes as well as weak aryl–aryl
and alkyl–aryl interactions between complexes; different coordination geometries and tautomeric forms are also
observed. Dimethyl sulfoxide molecules show a variety of interactions within the structures, including coordination
to the copper, hydrogen bonding, as well as weak alkyl–aryl interactions with the ligands.
Weakly polar interactions have long been of interest in bio-
Introduction
chemistry and their occurrence in the interior of globular
proteins has received a good deal of attention. One reason for
this interest is that, even though they are weak, they may con-
tribute from around Ϫ4 to Ϫ10.5 kJ molϪ1 to the corresponding
free energy of stabilisation and they can frequently determine
the adoption of a particular structure. For example, a change
in the stabilisation energy of just Ϫ4 to Ϫ8 kJ molϪ1 is enough
to convert a mesophilic protein into a thermophilic one.5
Aromatic interactions, offset face-to-face stacking and edge-to-
face interactions, also fall into the category of weak inter-
actions and have been investigated over many years;6–8 π–π
interactions may vary considerably in geometry and the cluster-
ing of aromatic rings may make a significant contribution to
overall stability.9 Their role in molecular recognition, and more
generally in supramolecular chemistry, has now been widely
examined.10,11 An electrostatic model for such interactions has
been described.12
In earlier papers we have described the incorporation of a
number of transition metal complexes into supramolecular
arrays using complementary hydrogen bonding motifs.1–3 In
these arrays the complexes occur in three-component subunits,
or in chains, which then interact further by hydrogen bonding
to form more elaborate structures.
In this paper we report an extension of these studies involv-
ing the synthesis and crystallographic analysis of solids
incorporating complexes of the bidentate ligands N-(4-oxo-
5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)ethanimidamide, 1,
and N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)prop-
animidamide, 2.
Aromatic rings have now been well documented to act as
hydrogen bond acceptors.10 This was predicted on the basis of a
simple electrostatic model12 and the possibility that aromatic
rings might interact with alkyl groups in a similar way was
raised by Levitt and Perutz13 who estimated a likely interaction
energy of about 4 kJ molϪ1
.
Our interest in the above alkanimidamido ligands stems
from the fact that they possess a self-complementary doublet
hydrogen bonding motif, which gives them the potential
to form chains in the solid state. They differ from the
2-guanidinobenzimidazole ligand previously used by us1,2 in
two important respects. Namely, once chains of bis(ligand)
complexes form there is no longer the possibility of other
hydrogen bonds occurring between neighbouring complexes
since all other hydrogen bond donor or acceptor groups have
either been removed or are utilised in forming intra-complex
hydrogen bonds. In addition, the presence of pendant phenyl
groups adds a three-dimensional character to the ligands, which
opens up the possibility of forming local environments within
the crystal that involve weakly polar interactions. Although
such interactions may be weak, experience shows that they
should not be ignored in crystal engineering since, they may be
responsible for the difference between two otherwise similar
structures. Further, they may control molecular aggregation
(and thus, which of the possible multiple polymorphic
forms will be observed) or they may even be the only stabilising
interactions apart from van der Waals’ forces.4
The strengths of aromatic–alkyl (and aromatic–aryl) inter-
actions in particular small molecules have been investigated
experimentally by means of a ‘molecular torsion balance’; in
these studies it was demonstrated that cyclohexyl, phenyl and
tert-butyl groups all showed similar affinities for the face of an
aromatic ring, affecting conformational energies by about 2 kJ
molϪ1 at 298 K.14
Experimental
2-Amino-5,5-diphenylimidazolin-4-one, 3
To sodium hydroxide (0.834 g, 21 mmol) in absolute ethanol
(50 mL) was added guanidine hydrochloride (2.097 g, 22 mmol)
in absolute ethanol (50 mL) and the mixture was heated under
reflux. After 2 h a white precipitate of NaCl was filtered from
the hot solution. The filtrate was reheated to boiling and a solu-
tion of benzil (2.190 g, 10.4 mmol) in absolute ethanol (50 mL)
was added dropwise over a period of 2 h. Heating was con-
tinued for 1 h after which the microcrystalline product that
formed was filtered off, washed with cold ethanol (2 × 5 mL)
4128
J. Chem. Soc., Dalton Trans., 2002, 4128–4133
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b207031f

View Article Online
followed by diethyl ether (2 × 5 mL), and air dried. The white
product was recrystallised from ethanol. Yield 1.539g (30%),
mp 360–363 ЊC. IR 3353, 1659, 1495, 1268, 754, 698 cmϪ1
(Found: C, 71.93; H, 5.27; N, 16.91. Calc. for C15H13N3O: C,
71.70; H, 5.21; N, 16.72%).
[(Dimethyl sulfoxide)bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-
imidazol-2-yl)ethanimidamido)copper(II)], 4
Anhydrous copper() chloride (0.048 g, 0.36 mmol) was
dissolved in acetonitrile (50 mL) and 1 (0.208 g, 0.83 mmol) was
added. The mixture was refluxed for 2 h after which the solution
had become light green; a few drops of a saturated KOH solu-
tion in ethanol were added until a pink precipitate began to
form. When precipitation was complete the solid was filtered
off, washed with acetonitrile (2 × 5 mL) followed by diethyl
ether (2 × 5 mL), and air-dried. The product was obtained as a
pink powder (0.028 g). This pink solid was dissolved in DMSO,
a little ethanol was added and the mixture set aside until
crystals formed. IR 3245, 3064, 1702, 1597, 1215, 695 cmϪ1
(Found: C, 59.25; H, 5.03; N 15.35. Calc. for C36H36CuN8O3S:
C, 59.68; H, 5.02; N, 15.47%).
Fig. 1 An ORTEP23 depiction of complex 4 with 20% displacement
ellipsoids.
Table 1 Pertinent geometry details for 4
Cu(1)–N(4)
Cu(1)–N(5)
Cu(1)–O(3B)
O(1)–C(1)
N(1)–C(1)
N(2)–C(3)
N(3)–C(3)
N(4)–C(4)
N(5)–C(20)
N(6)–C(19)
N(7)–C(21)
1.919(2)
2.034(2)
2.342(18)
1.225(3)
1.367(3)
1.349(3)
1.330(3)
1.283(3)
1.379(3)
1.454(3)
1.368(3)
Cu(1)–N(8)
Cu(1)–N(1)
Cu(1)–O(3A)
O(2)–C(18)
N(1)–C(3)
N(2)–C(2)
N(3)–C(4)
N(5)–C(18)
N(6)–C(20)
N(7)–C(20)
N(8)–C(21)
1.928(2)
2.049(2)
2.409(3)
1.225(3)
1.376(3)
1.459(3)
1.366(3)
1.375(3)
1.339(3)
1.335(3)
1.283(3)
[Bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)prop-
animidamido)copper(II)]ؒ0.5H2O, 5ؒ0.5H2O
Anhydrous copper() bromide (0.269 g, 1.2 mmol) was dis-
solved in propionitrile (70 mL) and solid 1 (0.564 g, 2.1 mmol)
was added slowly to the solution. The resulting mixture was
refluxed for an hour and left to stand at room temperature
overnight after which a solution of KOH (0.135 g, 2.4 mmol) in
methanol (5 mL) was added. The mixture was filtered and
allowed to stand. The product formed as red crystals; yield
0.120 g (Found: C, 63.10; H, 5.06; N, 16.49. Calc. for
C36H35CuN8O2.5: C, 63.29; H, 5.13; N, 16.41%).
N(4)–Cu(1)–N(8)
N(4)–Cu(1)–N(5)
N(8)–Cu(1)–N(5)
N(4)–Cu(1)–N(1)
N(8)–Cu(1)–N(1)
173.37(10)
93.18(9)
86.35(9)
85.76(9)
94.14(9)
N(5)–Cu(1)–N(1)
N(4)–Cu(1)–O(3A)
N(8)–Cu(1)–O(3A)
N(5)–Cu(1)–O(3A)
N(1)–Cu(1)–O(3A)
175.07(8)
98.44(12)
88.16(11)
88.87(10)
96.06(11)
[Bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)prop-
animidamido)copper(II)]ؒ1.75DMSO, 6ؒ1.75DMSO
DMSO solvate molecule exhibits sp3 inversion disorder, with an
occupancy of 0.8 for the S(1A) containing conformer, and a
complementary occupancy of 0.2 for the S(1B) conformer. The
occupancies were refined and then fixed. In general the non-
hydrogen atoms were modelled with anisotropic thermal
parameters and a riding atom model was used for the hydrogen
atoms. The partially occupied non-hydrogen sites were
modelled with isotropic displacement parameters, except for
S(1A) and O(3A) which were treated anisotropically. The
nitrogen bound hydrogen atoms were located and modelled
with isotropic displacement parameters; distance restraints
were required for the N(3) and N(7) hydrogens. Formula
C36H36CuN8O3S, M 724.33, monoclinic, space group C2/c
(no. 15), a 11.3898(15), b 14.843(2), c 40.196(5) Å, β 95.298(2)Њ,
V 6766.4(16) Å3, Dc 1.422 g cmϪ3, Z 8, crystal size 0.213 by
0.137 by 0.022 mm, colour purple, habit plate, temperature
The copper complex 5 (0.020 g) was dissolved in dimethyl
sulfoxide (3 mL) and the solution was warmed to 50 ЊC. Dark
purple crystals containing solvent formed on allowing the
solution to stand. The crystals turned to powder immediately if
washed with alcohol and lost solvent slowly on removal from
the growth solution (Found: C, 58.61; H, 5.55; N, 13.80. Calc.
for C39.5H44.5CuN8O3.75S1.75: C, 58.50; H, 5.49; N, 13.82%).
X-Ray structure determinations
Single crystal diffraction data were collected using ω scans to
56Њ 2θ, with a Bruker SMART 1000 CCD diffractometer
employing graphite monochromated MoKα radiation gener-
ated from a sealed tube. Crystals were cooled with an Oxford
Cryosystems Cryostream. In each case there was no significant
change in the intensities of a set of reflections recollected at the
end of the data collection. The data integration and reduction
were undertaken with SAINT and XPREP,15 and subsequent
computations were carried out with the teXsan,16 WinGX17
and XTAL18 graphical user interfaces. A Gaussian absorption
correction15,19 was applied to the data, and a subsequent
empirical correction determined with SADABS20 was also
applied to the data for 4. The structure for 4 was obtained using
SHELXS9721 and the structures of 5 and 6 were determined
with SIR97.22 The structures were extended and refined with
SHELXL97.21 ORTEP23 depictions of the molecules with 20%
displacement ellipsoids are provided in Figs. 1, 4 and 6; pertin-
ent geometry details are given in Tables 1, 3 and 4.
294(2) K, λ(MoKα) 0.71073 Å, µ(MoKα) 0.757 mmϪ1
,
T (Gaussian and SADABS)min,max 0.896, 0.983, 2θmax 56.62, hkl
range Ϫ15 15, 0 19, Ϫ51 52, N 35758, Nind 8107(Rmerge 0.0590),
Nobs 5341(I > 2σ(I)), Nvar 470, residuals R1(F) 0.0492, wR2(F 2)
0.1201, GoF(all) 1.261, ∆ρmin,max Ϫ0.389, 0.537 e ÅϪ3. R1 =
2
Σ||Fo| Ϫ |Fc||/Σ|Fo| for Fo > 2σ(Fo); wR2 = (Σw(Fo Ϫ Fc2)2/
2
Σ(wFc2)2)1/2 all reflections, w = 1/[σ2 (Fo ) ϩ (0.04P)2 ϩ 2.0P] and
P = (Fo2 ϩ 2Fc2)/3.
[Bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)-
propanimidamido)copper(II)], 5. The asymmetric unit contains
three crystallographically independent complex molecules, two
of which are centred on inversion sites. The non-hydrogen
atoms were modelled with anisotropic displacement param-
eters, and in general a riding atom model was used for the
hydrogen atoms. The H(3N), H(4N), H(6N), H(8N), H(10N)
and H(14N) hydrogen sites were located and modelled with
isotropic displacement parameters. Formula C36H34CuN8O2,
[(Dimethyl sulfoxide)bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-
1H-imidazol-2-yl)ethanimidamido)copper(II)], 4. The asym-
metric unit contains the complex molecule with a weakly
coordinated and slightly disordered DMSO molecule. The
J. Chem. Soc., Dalton Trans., 2002, 4128–4133
4129

View Article Online
M 674.25, monoclinic, space group P21/n (no. 14), a 13.184(3),
b 27.498(7), c 19.121(5) Å, β 108.662(4)Њ, V 6567(3) Å3, Dc 1.364
g cmϪ3, Z 8, crystal size 0.366 by 0.333 by 0.149 mm, colour
purple, habit prism, temperature 150(2) K, λ(MoKα) 0.71073
Å, µ(MoKα) 0.711 mmϪ1, T (Gaussian)min,max 0.788, 0.905,
2θmax 56.60, hkl range Ϫ16 16, Ϫ36 36, Ϫ25 25, N 57558, Nind
15233(Rmerge 0.0368), Nobs 10799(I > 2σ(I)), Nvar 878, residuals
and follows another closely related condensation,28 but it is
noted that it differs from further reports of the same condens-
ation.29 The template synthesis of 4 involved the addition of
the amine functionality across a nitrile triple bond. Although
vigorous conditions are often employed for such reactions, they
have also been reported to occur under mild conditions in the
presence of a copper() catalyst.30 Given the present reaction
conditions, the assistance of copper() in catalysing the reaction
cannot be ruled out.
The crystal structure of 4 is depicted in Figs. 1–3, and selected
geometry details are provided in Table 1. The copper() ion
has a square pyramidal coordination geometry, with minor
disorder associated with sp3 inversion at the sulfur atom. The
axial ligand Cu(1)–O(3A) bond length (2.409(3) Å) is similar to
that found in [Cu(NH3)4(OH2)]SO4.31 The base of the pyramid
is formed by the donor nitrogen atoms of the two bidentate
ligands of type 1. These bidentate ligands are connected via two
N–H ؒ ؒ ؒ O hydrogen bonds resulting in the formation of a
16-membered pseudo-macrocyclic ring (Fig. 2 and Table 2(a)).
R1(F) 0.0375, wR2(F 2) 0.0869, GoF(all) 1.270, ∆ρmin,
max
Ϫ0.434, 0.665 e ÅϪ3. R1 = Σ||Fo| Ϫ |Fo||/Σ|Fo| for Fo > 2σ(Fo);
wR2 = (Σw(Fo Ϫ Fc2)2/Σ(wFo )2)1/2 all reflections, w = 1/[σ2
2
2
2
(Fo )ϩ (0.03P)2 ϩ 7.2571P] where P = (Fo2 ϩ 2Fc2)/3.
[Bis(N-(4-oxo-5,5-diphenyl-4,5-dihydro-1H-imidazol-2-yl)-
propanimidamidocopper(II)]ؒ2DMSO, 6ؒ2DMSO. The asym-
metric unit contains a complex molecule together with two
dimethyl sulfoxide solvate molecules. One of the DMSO mole-
cules is disordered about two sites, and on one of those sites the
molecule has two partially occupied sulfur sites related by sp3
inversion. The disordered DMSO site populations were refined
and then fixed. The disordered DMSO molecule has hydrogen
bond interactions with hydrogens bound to N(2) and N(3), and
these hydrogen sites have corresponding occupancies of 0.25
and 0.75, respectively. The existence of the hydrogen at N(2) is
inferred from the location of the DMSO oxygen O(5). The
ordered DMSO molecule forms a hydrogen bond with H(6N).
In general, the non-hydrogen atoms were modelled with
anisotropic displacement parameters, and a riding atom model
was used for the hydrogen atoms. Isotropic displacement
parameters were used for the disordered S(3) sites, and the
disordered carbon and oxygen sites. The N(3), N(4) N(6) and
N(8) hydrogen atoms were located and modelled with isotropic
displacement parameters. Formula C40H46CuN8O4S2,
M
¯
830.51, triclinic, space group P1 (no. 2), a 11.592(5), b
16.662(5), c 10.435(5) Å, α 93.233(5), β 94.704(5), γ 99.777(5)Њ,
V 1974.4(14) Å3, Dc 1.397 g cmϪ3, Z 2, crystal size 0.290 by
0.058 by 0.058 mm, colour red, habit columnar, temperature
Fig. 2 A PLATON37 depiction of the hydrogen bond linked chains in
complex 4.
150(2) K, λ(MoKα) 0.71069 Å, µ(MoKα) 0.711 mmϪ1
,
T (Gaussian)min, max 0.829, 0.965, 2θmax 56.58, hkl range Ϫ14 14,
Ϫ21 21, Ϫ13 13, N 19578, Nind 9136(Rmerge 0.0446), Nobs 5806-
(I > 2σ(I)), Nvar 539, residuals R1(F) 0.0467, wR2(F 2) 0.0600,
GoF(all) 1.192, ∆ρmin,
Ϫ0.613, 0.943 e ÅϪ3. R1 = Σ||Fo| Ϫ
max
|Fo||/Σ|Fo| for Fo > 2σ(Fo); wR2 = (Σw(Fo2 Ϫ Fc2)2/Σ(wFo )2)1/2 all
2
reflections, w = 1/[σ2(Fo )].
2
CCDC reference numbers 190103–190105.
See http://www.rsc.org/suppdata/dt/b2/b207031f/ for crystal-
lographic data in CIF or other electronic format.
Fig. 3 A PLATON37 view of the unit cell of 4 projected down the a-
axis. Slightly undulating chains or ribbons of hydrogen bond linked
complexes run across the (0,4,0) planes at an angle of approximately 30Њ
to the c-axis. The amplitude of the undulation is approximately 0.03 Å.
Results and discussion
The synthesis of 3 and the subsequent template synthesis
of 4 are outlined in Scheme 1. The formation of 3 involves a
benzilic acid rearrangement24 that parallels the well-known
rearrangement of this type that occurs when urea reacts with
benzil.25,26 Our result corresponds to that reported elsewhere,27
Adjacent complexes are linked by hydrogen bonds between
opposing amine acceptor and imidazole nitrogen donor atoms
to form centrosymmetric planar R22 (8) hydrogen bonded rings
(Fig. 2); the graph set notation32 describes an eight membered
ring, with two donors (subscript) and two acceptors (super-
script). There are then slightly undulating chains or ribbons
of complexes running across the (0,4,0) planes at an angle of
approximately 30Њ to the c-axis (Fig. 3); there is no hydrogen
bonding between the chains. The axial ligand of a complex in
the (0,0.25,0) plane is accommodated in pockets defined by the
phenyl residues of the complex and the phenyl residues of two
complexes in the (0,0.75,0) plane. The disposition of the axial
ligand is such that one dimethyl sulfoxide (DMSO) methyl
group (C36) is approximately 3.74 Å from the centroid of the
C(6)–C(11) residue within the same complex, while the other
(C(35)) is approximately 3.38 Å from the centroid of the C(12)–
C(17) phenyl ring of a complex in a neighbouring chain. The
simple electrostatic model13 for the methyl–phenyl interaction
predicts that the interaction energy lies below room tempera-
ture thermal energy (RT) when the carbon atom of the methyl
Scheme 1
4130
J. Chem. Soc., Dalton Trans., 2002, 4128–4133

View Article Online
Table 2 Hydrogen bond geometries within the structures of 4–6
D–H ؒ ؒ ؒ A
D–H/Å
H ؒ ؒ ؒ A/Å
D ؒ ؒ ؒ A/Å
DHA D–H ؒ ؒ ؒ A/Њ
(a) 4
N(4)–H(4N) ؒ ؒ ؒ O(2)
N(2)–H(2N) ؒ ؒ ؒ N(3a)
N(8)–H(8N) ؒ ؒ ؒ O(1)
N(6)–H(6N) ؒ ؒ ؒ N(7b)
0.95(3)
0.834(17)
0.85(3)
1.89(3)
2.279(18)
2.07(3)
2.378(3)
3.110(3)
2.803(3)
3.039(3)
146(3)
175(3)
145(2)
173(3)
0.822(17)
2.221(18)
(b) 5
N(4)–H(4N) ؒ ؒ ؒ O(2)
N(2)–H(3N) ؒ ؒ ؒ N(7c)
N(8)–H(8N) ؒ ؒ ؒ O(1)
N(6)–H(6N) ؒ ؒ ؒ N(3d)
N(10)–H(10N) ؒ ؒ ؒ N(15)
N(12)–H(12N) ؒ ؒ ؒ O(3e)
N(14)–H(14N) ؒ ؒ ؒ N(11)
N(16)–H(16N) ؒ ؒ ؒ O(4f)
0.79(2)
0.83(2)
0.801(14)
0.83(2)
0.789(18)
0.88
2.12(2)
2.14(2)
2.306(19)
2.09(2)
2.141(19)
2.01
2.800(2)
2.971(2)
2.866(2)
2.908(2)
2.927(2)
2.787(2)
2.916(2)
2.807(2)
144(2)
178(2)
127.6(19)
170.6(19)
174.5(18)
146.1
171.7(19)
145.4
0.800(18)
0.88
2.122(19)
2.04
(c) 6
N(2)–H(2N) ؒ ؒ ؒ O(5)
N(3)–H(3N) ؒ ؒ ؒ O(4)
N(4)–H(4N) ؒ ؒ ؒ O(2)
N(6)–H(6N) ؒ ؒ ؒ O(3)
N(8)–H(8N) ؒ ؒ ؒ O(1)
0.88
2.02
2.794(9)
2.803(3)
2.757(3)
2.807(3)
2.807(3)
146.8
0.88(3)
0.81(2)
0.881(18)
0.82(2)
1.93(3)
2.01(2)
1.940(19)
2.05(2)
172(3)
153(2)
167.6(19)
152(2)
Symmetry codes: a Ϫx ϩ 1/2, Ϫy ϩ 1/2, Ϫz. b 1 Ϫ x, y, Ϫz ϩ 1/2. c x Ϫ 1/2, 1/2 Ϫ y, z Ϫ 1/2. d 1/2 ϩ x, 1/2 Ϫ y, 1/2 ϩ z. e Ϫx, 1 Ϫy, Ϫz. f 1 Ϫ x, 1 Ϫ y,
1 Ϫ z.
Table 3 Pertinent geometry details for 5
group lies between 3.6 and 4.4 Å from the centroid of the
aromatic ring (methyl C to aromatic C distances 3.9 and 4.6 Å).
A search of the Cambridge Structure Database33 for
DMSO methyl to phenyl contacts yielded 48 examples of
such intermolecular contacts from 79 structures that contain
both DMSO and phenyl groups. (The search was restricted to
C(methyl) to aromatic centroid distances that lie between 3.0
and 4.5 Å, with the angle between the C(methyl) to centroid
vector and the normal to the aromatic plane being restricted to
lie between 0 and 30Њ.) Forty of the forty-eight examples had
distances in the range 3.5–4.0 Å with the shortest contact being
3.47 Å.34
The phenyl groups project above and below the plane of the
macrocycle and form a herringbone offset edge-to-face pattern.
The angles between the rings and the offsets between centroids
are consistent with attractive interactions in this crystal.6,9
Complexes 5 and 6 utilise the bidentate ligand 2. The crystal
structure of 5 contains three crystallographically independent
complexes for which numbered depictions are provided in
Fig. 4(a–c) and geometric details are listed in Table 3.
The ligand ethyl groups lie approximately in the plane of each
ligand and the terminal methyl groups of these residues have a
syn conformation with respect to the coordinated imino groups.
The two complexes incorporating Cu(2) and Cu(3) are square
planar, with the metal ions residing on crystallographic
inversion centres. The complex containing Cu(1) has a signifi-
cant tetrahedral distortion from square planar geometry, and
the two ligand planes (the least squares planes of the five ligand
atoms in the metallarings) form a dihedral angle of 26.21(7)Њ.
The copper ion in this complex is essentially located in the
ligand least squares plane defined by N(5), C(21), N(7), C(22),
and N(6), but is 0.375(2) Å from the least squares plane defined
by N(1), C(3), N(3), C(4) and N(4).
Cu(1)–(4)
Cu(1)–(1)
N(1)–C(3)
N(2)–C(3)
N(3)–C(3)
N(4)–C(4)
N(5)–C(19)
N(6)–C(20)
N(7)–C(22)
Cu(2)–N(12)
N(9)–C(37)
N(10)–C(38)
N(11)–C(39)
N(12)–C(40)
Cu(3)–N(16)
N(13)–C(55)
N(14)–C(57)
N(15)–C(57)
N(16)–C(58)
1.9257(18)
1.9804(16)
1.372(2)
1.346(2)
1.325(2)
1.290(2)
1.378(2)
1.449(2)
1.362(2)
1.9163(16)
1.388(2)
1.454(2)
1.327(2)
1.292(2)
1.9247(16)
1.378(2)
1.341(2)
1.325(2)
1.293(2)
Cu(1)–N(8)
Cu(1)–N(5)
N(1)–C(1)
N(2)–C(2)
N(3)–C(4)
N(5)–C(21)
N(6)–C(21)
N(7)–C(21)
N(8)–C(22)
Cu(2)–N(9)
N(9)–C(39)
N(10)–C(39)
N(11)–C(40)
1.9323(18)
1.9835(15)
1.378(2)
1.460(2)
1.367(2)
1.378(2)
1.337(2)
1.330(2)
1.291(2)
2.0279(15)
1.378(2)
1.343(2)
1.362(2)
Cu(3)–N(13)
N(13)–C(57)
N(14)–C(56)
N(15)–C(58)
2.0447(15)
1.378(2)
1.451(2)
1.362(2)
N(4)–Cu(1)–N(8)
N(4)–Cu(1)–N(1)
N(8)–Cu(1)–N(1)
N(4)–Cu(1)–N(5)
160.72(8)
87.18(7)
93.59(7)
94.95(7)
N(8)–Cu(1)–N(5)
N(1)–Cu(1)–N(5)
N(9)–Cu(2)–N(12)
N(13)–Cu(3)–N(16)
87.84(7)
169.37(6)
87.66(6)
87.64(6)
hydrogen bond interaction. A related example of a twisted
hydrogen-bonded ring, perhaps reflecting steric interactions,
occurs in the dimeric structure of N,N Ј-bis(2,6-diisopropyl-
phenyl)acetamidine.35
As in the structure of 4, the hydrogen bonds link the com-
plexes in 5 into chains, however this time there are two distinct
chains present (Fig. 5). Linked complexes incorporating Cu(1)
comprise one chain, while the second chain has alternating
Cu(2)- and Cu(3)-containing complexes. The two twisted
crystallographically independent chains are quasi-parallel to
one another, and the pendant phenyl groups that project from
the chains are tightly interlocked.
Red crystals of 5 dissolve in DMSO and the dark purple
crystals of 6 isolated from this solution contain two DMSO
molecules per complex molecule (Fig. 6, Table 4). In contrast to
4, there is no axial DMSO coordinated to the copper, which has
essentially square planar coordination geometry, with the trans
N–Cu–N angles being 179.27(10) and 178.89(9)Њ respectively.
The distortion from square planar coordination does not
prevent the formation of the hydrogen-bonded, pseudo-
macrocycle in 5. As in 4, complementary R22 (8) hydrogen bond
patterns are present between neighbouring complexes; however
the rings are considerably twisted in 5 (Fig. 5). Although the two
hydrogen bonds are linear, they are far from being co-planar. It
is interesting to note that if the methyl group had an anti dis-
position (pointing towards the neighbouring complex instead
of away from it), then steric hindrance would prevent the
J. Chem. Soc., Dalton Trans., 2002, 4128–4133
4131

View Article Online
Fig. 4 ORTEP23 depictions with 20% displacement ellipsoids of the three crystallographically independent complexes in 5; individually (a–c) and
their positions with respect to one another (d).
Table 4 Pertinent geometry details for 6
Cu(1)–N(8)
Cu(1)–N(5)
N(1)–C(1)
N(2)–C(3)
N(3)–C(4)
N(4)–C(4)
N(5)–C(21)
N(6)–C(20)
N(7)–C(22)
1.921(2)
2.0275(19)
1.364(3)
1.296(3)
1.349(3)
1.287(3)
1.388(3)
1.447(3)
1.357(3)
Cu(1)–N(4)
Cu(1)–N(1)
N(1)–C(3)
N(2)–C(2)
N(3)–C(3)
N(5)–C(19)
N(6)–C(21)
N(7)–C(21)
N(8)–C(22)
1.928(2)
2.0303(19)
1.393(3)
1.464(3)
1.370(3)
1.367(3)
1.342(3)
1.321(3)
1.293(3)
N(1)–Cu(1)–N(8)
N(1)–Cu(1)–N(4)
N(1)–Cu(1)–N(5)
93.23(9)
87.43(9)
178.89(9)
N(4)–Cu(1)–N(8)
N(4)–Cu(1)–N(5)
N(5)–Cu(1)–N(8)
179.27(10)
92.12(9)
87.21(9)
Fig. 5 A PLATON37 depiction of the two independent chains of
complexes in the crystal structure of 5. The twisted chains have
‘interlocking’ pendant phenyl groups.
N–H donors (Fig. 6 and Table 2(c)). The DMSO hydrogen
bond interaction with the ligands prevents the formation of
the R22 (8) synthon36 and thus the concatenation observed in
4 (which was also crystallised from DMSO) and in 5 does not
occur in 6.
Perhaps surprisingly, one of the two ligands in complex 6
exists in two different tautomeric forms within the crystal, and
there is associated DMSO disorder. The predominant ligand
tautomer is protonated at the metallaring amine nitrogen N(3)
(75% occupancy), whereas the second form of this ligand is
protonated at the imidazole nitrogen N(2) (25% occupancy).
The second ligand in the complex is protonated only at the
imidazole nitrogen N(6). That is, the two otherwise identical
ligands in the predominant form of the complex have different
tautomeric forms. The presence of the low population tautomer
was indicated by both residual electron density in the final
crystallographic difference maps, and by the disorder present in
the DMSO on that side of the complex; O(5) would be too close
to N(2) if it were not protonated. It seems not possible to
determine if the tautomerism is responsible for the DMSO
disorder, or if the DMSO disorder is responsible for the
tautomerism.
Fig. 6 An ORTEP23 depiction of complex 6 with 20% displacement
ellipsoids.
Again the two bidentate ligands are connected by hydrogen
bonds to form a sixteen-membered pseudo-macrocycle, with
the DMSO molecules being hydrogen bonded to the ligand
4132
J. Chem. Soc., Dalton Trans., 2002, 4128–4133

View Article Online
Unlike the ordered DMSO molecule, the disordered DMSO
molecule is close to an equivalent symmetry related site and
it may be the proximity of another DMSO molecule that drives
the DMSO disorder and perhaps then the tautomerism. The
presence of the nearby DMSO molecule prevents the dis-
ordered molecule from having a similar orientation, with
respect to the complex, to that of the ordered DMSO molecule.
The phenyl groups in complex 6 are involved in either C–H to
π-facial or in methyl–phenyl interactions (Fig. 7), and the
different steric requirements that may preclude the formation
of the hydrogen bonded rings and which would also change the
electronic properties of the ligand.
Acknowledgements
We thank the Australian Research Council for support.
References
1 M. M. Bishop, L. F. Lindoy, B. Skelton and A. H. White, Supramol.
Chem., 2001, 13, 293.
2 M. M. Bishop, L. F. Lindoy and P. Turner, Supramol. Chem., 2001,
14, 179.
3 M. M. Bishop, L. F. Lindoy, B. Skelton and A. H. White, J. Chem.
Soc., Dalton Trans., 2002, 377.
4 M. C. Etter, J. Phys. Chem., 1991, 95, 4601.
5 S. K. Burley and G. A. Petsko, Adv. Protein Chem., 1988, 39, 125.
6 S. K. Burley and G. A. Petsko, Science, 1985, 229, 23.
7 (a) J. Singh and J. M. Thornton, FEBS Lett., 1985, 191, 2989; (b)
J. Singh and J. M. Thornton, J. Mol. Biol., 1990, 211, 595.
8 C. A. Hunter, J. Singh and J. M. Thornton, J. Mol. Biol., 1991, 218,
837.
Fig. 7 A PLATON37 illustration of the principal intermolecular
hydrogen bonding and weak interactions in 6; the disorder and the
hydrogen atoms have been omitted for clarity. Each phenyl ring is
involved in a C–H to π-facial interaction or in an interaction with a
DMSO methyl group.
9 C. A. Hunter, Chem. Soc. Rev., 1994, 101.
10 L. F. Lindoy and I. M. Atkinson, Self-Assembly in Supramolecular
Chemistry, Royal Society of Chemistry, Cambridge, UK, 2000.
11 I. Dance, Supramolecular Inorganic Chemistry, in G. R. Desiraju
(Editor), The Crystal as a Supramolecular Entity, John Wiley &
Sons, Chichester, UK, 1996.
ordered DMSO molecule, participates in a methyl–phenyl
interaction with the complex to which it is hydrogen bonded.
The methyl groups of the disordered DMSO molecule, how-
ever, are oriented so that they do not interact with the phenyl
residues of the ligand to which it is hydrogen bonded. Instead
they show methyl–phenyl interactions with the phenyl group of
a neighbouring complex. Additionally, the two ethyl groups
have different orientations; the terminal methyl group C(24) lies
in an approximately anti conformation, while the C(6) methyl
group, approximates a syn orientation. A nearby neighbouring
phenyl residue prevents C(6) from having an anti-conformation
and the consequential position of C(6) prevents C(24) having a
syn-conformation. A simple calculation suggests that if the
C(5)–C(6) ethyl residue orientation was similar to that of the
C(23)–C(24) residue, then it would be approximately 2.7 Å from
the phenyl C(11) on the neighbouring complex. It would also be
approximately 3.0 Å from O(4) of the disordered DMSO
molecule and so prevent the observed tautomerism.
12 C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112,
5525.
13 M. Levitt and M. F. Perutz, J. Mol. Biol., 1988, 201, 751.
14 S. Paliwal, S. Geib and C. S. Wilcox, J. Am. Chem. Soc., 1994, 116,
4497.
15 SMART, SAINT and XPREP, Area detector control and data
integration and reduction software, Bruker Analytical X-ray
Instruments Inc., Madison, WI, 1995.
16 teXsan for Windows: Single Crystal Structure Analysis Software,
Molecular Structure Corporation, The Woodlands, TX, 1997–1998.
17 WinGX, L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
18 S. R. Hall, D. J. du Boulay, and R. Olthof-Hazekamp (Editors),
Xtal3.6 System, University of Western Australia, 1999.
19 P. Coppens, L. Leiserowitz and D. Rabinovich, Acta Crystallogr.,
1965, 18, 1035.
20 (a) R. H. Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33; (b)
G. M. Sheldrick, SADABS, Empirical absorption correction
program for area detector data, University of Göttingen, Germany,
1996.
21 G. M. Sheldrick, SHELX97, Programs for Crystal Structure
Analysis, University of Göttingen, Germany, 1998.
22 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and
R. Spagna, J. Appl. Crystallogr., 1999, 32, 115.
23 C. K. Johnson, ORTEPII, Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
24 K. Bowden and K. D. Williams, J. Chem. Soc., Perkin Trans. 2, 1994,
77.
25 A. R. Butler and E. Leitch, J. Chem. Soc., Perkin Trans. 2, 1976,
1972.
26 T. Nishimura and K. Kitajima, J. Org. Chem., 1979, 44, 818.
27 L. Call, Monatsh. Chem., 1970, 101, 344.
28 H. W. Schramm, Sci. Pharm., 1991, 59, 123.
29 S. R. Saha and A. Das, Indian J. Chem., Sect. A, 1993, 32, 802.
30 G. Rousselet, P. Capdevielle and M. Maumy, Tetrahedron Lett.,
1993, 34, 6395.
Conclusions
The new ligands reported here possess a number of features
that are capable of being altered synthetically in such a way that
the changes might be expected to affect the supramolecular
assembly of their complexes and, as a consequence, contribute
to a better understanding of rational crystal design.
First, the anionic form of the ligand has self-complementary
doublet hydrogen bonding motifs and the potential for hydrogen
bonding between neighbouring complexes is restricted to these
motifs because the other hydrogen bond donor and acceptor in
the ligand are used in the formation of the pseudo-macrocyclic
rings. Steric or electronic inhibition of the above hydrogen
bonding will clearly likely result in changes in the supramolec-
ular structure adopted.
31 (a) F. Massi, Acta Crystallogr, 1955, 8, 137; (b) B. Morosin, Acta
Crystallogr., Sect. B, 1969, 25, 19.
32 J. Bernstein, R. E. Davis, L. Shimoni and N-L. Chang, Angew.
Chem., Int. Ed. Engl., 1995, 34, 1555.
33 F. H. Allen and O. Kennard, Chem. Des. Automat. News, 1993, 8, 31.
34 E. Bermejo, A. Castineiras, R. Dominguez, R. Carballo,
C. Maichle-Moessmer, J. Straehle and D. X. West, Z. Anorg. Allg.
Chem., 1999, 625, 961.
35 R. T. Boeré, V. Klassen and G. Wolmershaüser, J. Chem. Soc.,
Dalton Trans., 1998, 4147.
36 G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311.
37 (a) A. L. Spek, Acta Crystallogr., Sect A, 1990, 46, C34; (b)
A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
Utrecht University, The Netherlands, 1998.
In addition, there is a structural feature that provides an
alternative means of assembling the complexes into chains
(though in this work only dimers have been observed); namely,
the complementary CH to π-facial interactions between the
pairs of aromatic rings on each ligand. Such an assembly, based
as it is on weak interactions, would not be expected to occur
unless the formation of the self-complementary hydrogen
bonded rings mentioned above did not form, whether for steric
reasons or, for example, because the ligand was in its neutral
form.
Finally, the R groups, Me or Et in the current work, might be
changed, for example, to –OMe, –NMe2 or –NHPh, which have
J. Chem. Soc., Dalton Trans., 2002, 4128–4133
4133