Macrocyclic ligand design. Interaction of a series of successively N-benzylated derivatives of 1,4,8,11-tetraazacyclotetradecane (cyclam) with copper(II) and nickel(II)

Article abstract of DOI:10.1039/b211394e

The interaction of a series of N-benzylated cyclam ligand derivatives incorporating from one to four benzyl groups with nickel(II) and copper(II) is reported. The isolation of a selection of 1 : 1 (metal : ligand) complexes of both metals has been carried out and visible spectral, EPR, electrochemical and X-ray determinations have been employed to probe the nature of the respective complexes in both the solid state and in solution. Emphasis has been given to investigating the effect of the successive N-benzylation along this ligand series on the properties of the above complexes. The UV-VIS spectra obtained for each series of metal complexes indicate that the sequential introduction of N-benzyl substituents produces a change to lower ligand fields of the corresponding macrocyclic tetramines. Cyclic voltammograms of both the nickel(II) (low-spin) and copper(II) complexes both yield evidence for M(III)/M(II) as well as M(II)/M(I) couples in acetonitrile. The Royal Society of Chemistry 2003.

Full text of DOI:10.1039/b211394e

View Article Online / Journal Homepage / Table of Contents for this issue
Macrocyclic ligand design. Interaction of a series of successively
N-benzylated derivatives of 1,4,8,11-tetraazacyclotetradecane
(cyclam) with copper(II) and nickel(II)
Ying Dong,a Geoffrey A. Lawrance,b Leonard F. Lindoy*a and Peter Turnera
a Centre for Heavy Metals Research, School of Chemistry, The University of Sydney, N.S.W.
2006, Australia
b Discipline of Chemistry, School of Environmental and Life Sciences,
The University of Newcastle, Callaghan 2308, Australia
Received 20th November 2002, Accepted 17th February 2003
First published as an Advance Article on the web 18th March 2003
The interaction of a series of N-benzylated cyclam ligand derivatives incorporating from one to four benzyl groups
with nickel() and copper() is reported. The isolation of a selection of 1 : 1 (metal : ligand) complexes of both
metals has been carried out and visible spectral, EPR, electrochemical and X-ray determinations have been employed
to probe the nature of the respective complexes in both the solid state and in solution. Emphasis has been given
to investigating the effect of the successive N-benzylation along this ligand series on the properties of the above
complexes. The UV–VIS spectra obtained for each series of metal complexes indicate that the sequential introduction
of N-benzyl substituents produces a change to lower ligand fields of the corresponding macrocyclic tetramines.
Cyclic voltammograms of both the nickel() (low-spin) and copper() complexes both yield evidence for M()/M()
as well as M()/M() couples in acetonitrile.
Table 1 Conductance values for nickel() and copper() complexes of
Introduction
1–6
The modification of tetra-aza macrocyclic ligands to control
Complex
Colour
Λa/S cm2 molϪ1
and tune the redox properties of coordinated metal centres has
been the subject of continuing interest.1 Variations can be
introduced by altering the size of the macrocyclic ring or by
placing substituents on the nitrogen donors and/or the ring
framework. Of these approaches, N-substitution of the donor
atoms has the potential to generate the biggest steric and elec-
tronic effects, due to the closeness of the site of substitution to
introduced metal ions. For example, since they raise the steric
bulk of the molecule their presence may tend to inhibit the
coordination of additional donors in the coordination sphere.
Thus, apart from an anticipated effect on the redox properties
of metal complexes, structural and spectroscopic effects arising
from N-substitution are anticipated.
[Ni(1)](ClO4)2
Orange
Orange
Orange
Orange
Orange
Purple
Blue
Blue
Blue
Green
Purple
126
125
–
122
126
89
73
85
86
106
96
[Ni(2)](ClO4)2
[Ni(3)](ClO4)2
[Ni(4)](ClO4)2
[Ni(6)](ClO4)2
[Cu(1)(NO3)2]ؒ0.5MeOH
[Cu(2)(NO3)2]ؒMeOH
[Cu(3)(NO3)]NO3
[Cu(4)(NO3)]NO3
[Cu(5)(H2O)](NO3)2ؒEtOHؒH2O
[Cu(6)(NO3)]NO3
a Conductance at 25 ЊC in methanol at ∼10Ϫ3 mol dmϪ3; expected range
for a 1 : 1 electrolyte in methanol is 80–115 S cm2 molϪ1 while that for a
Previous work by us has been concerned with the synthesis
2 : 1 electrolyte is 160–220 S cm2 molϪ1
.
of
a series of N-benzylated cyclam ligand derivatives
incorporating from one to four benzyl groups.2 These repre-
sent excellent candidates for a comparative study of the
effects of such substitution on the metal binding properties
towards selected metal ions. In the present study we report
the results of an investigation of the interaction of 1–5 with
nickel() and copper(), with emphasis on the effect of the
successive benzylation on the structural, spectral and electro-
chemical properties of these complexes. The results are also
compared with the behaviour of the unsubstituted parent
macrocycle 6 (cyclam).
EMX EPR spectrometer at 9.464 GHz (X-band). Atomic
absorption spectroscopy data were obtained on a SpectrAAؒ
800 spectrometer. X-Ray diffraction data were obtained on a
Bruker SMART 1000 CCD diffractometer.
Cyclic voltammetry studies were performed using a con-
ventional three-electrode configuration with iR compensation
and a BAS Model 100B electrochemical system, controlled by a
computer using the BAS software. The working electrodes
employed were glassy carbon or platinum discs. The reference
electrode was Ag/AgCl and was separated from the working
and Pt wire auxiliary electrodes by a glass sleeve fitted with a
Vycor frit. Solutions were prepared in HPLC-grade acetonitrile
or acetone, 0.1 mol dmϪ3 in n-Bu4N(ClO4), and were purged
with argon gas. Complex concentrations in the millimolar range
were used throughout. For the ferrocenium/ferrocene reference
couple, E1/2 was at ϩ0.52 V (acetonitrile) and ϩ0.50 V (acetone)
under these conditions. Scans were measured over the range
10–5000 mV sϪ1, with reported results being mainly those
recorded at 100 mV sϪ1. Aspects of the electrochemical results
are illustrated in Figs. 1–5.
Experimental
Where available, all commercial reagents and solvents were of
analytical or HPLC grade. The preparation of ligands 1–5 has
been reported previously2 1,4,8,11-Tetraazacyclotetradecane
(cyclam; 6) was prepared by the published procedure.3
Solid state UV–VIS spectra were measured on a CARY 1E
UV–VISible spectrophotometer (samples spread on filter
paper) while solution UV–VIS spectra (at (5 4) × 10Ϫ3 mol
dmϪ3) were obtained on a CARY 5E UV–VIS-NIR spectro-
photometer on solutions. EPR spectra were obtained on
samples as dimethylformamide glasses at 77 K using a Bruker
Physical, spectral and electrochemical data for the complexes
are summarised in Tables 1–4.
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 2 0 0 3
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1567

View Article Online
Table 2 UV–VIS spectrophotometric data for the nickel() and copper() complexes of cyclam and 1–5
Complex
λmax a/nm
λmax b/nm (ε/MϪ1 cmϪ1
)
λmax c/nm (ε/MϪ1 cmϪ1
)
[Cu(6)](NO3)2
[Cu(1)](NO3)2
[Cu(2)](NO3)2
[Cu(3)](NO3)2
[Cu(4)](NO3)2
[Cu(5)](NO3)2
[Ni(6)](ClO4)2
[Ni(1)](ClO4)2
[Ni(2)](ClO4)2
[Ni(3)](ClO4)2
[Ni(4)](ClO4)2
508
524
539
543
560
629
459
473
478
471
488
505 (88)
511 (58)
520 (90)
545 (126)
550 (240)
583 (223)
699 (285)
510 (112)
525 (129)
544 (301)
562 (235)
545 (420)
322 (sh), 461 (13), 651 (vw), 935 (vw)
332 (sh), 480 (15), 704 (vw), 995 (vw)
341 (sh), 515 (10), 751 (vw), 974 (vw)
334 (sh), 476 (24), 798 (vw), 1016 (vw)
348 (sh), 517 (10), 820 (vw), 1030 (vw)
a Solid state UV–VIS spectra. b Spectra were recorded in methanol. Spectra for the nickel() complexes in this solvent were not obtained due to their
poor solubility. c Spectra were obtained in acetonitrile.
Table 3 EPR spectroscopic parameters for the copper() complexes of
1–6
Table 5 Bond lengths (Å) and angles (Њ) for [Cu(1)(NO3)2]ؒ0.5MeOH
and [Cu(2)(NO3)2]ؒMeOH
Complex
g||
g⊥
A||/G
Cu(1)(NO3)2
Cu(2)(NO3)2
[Cu(1)](NO3)2
[Cu(2)](NO3)2
[Cu(3)](NO3)2
[Cu(4)](NO3)2
[Cu(5)](NO3)2
[Cu(6)](NO3)2
2.178
2.179
2.182
2.198
2.225
2.173
2.044
2.055
2.050
2.049
2.035
2.048
196
194
182
176
155
200
Cu(1)–O(1)
Cu(1)–O(4)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1)–N(3)
Cu(1)–N(4)
2.804(6)
2.476(4)
2.061(4)
1.997(5)
2.027(5)
2.010(5)
2.7325(15)
2.5663(13)
2.0708(12)
2.0719(12)
2.0207(13)
2.0272(13)
N(1)–Cu(1)–N(2)
N(1)–Cu(1)–N(3)
N(1)–Cu(1)–N(4)
N(2)–Cu(1)–N(3)
N(2)–Cu(1)–N(4)
N(3)–Cu(1)–N(4)
N(1)–Cu(1)–O(1)
N(1)–Cu(1)–O(4)
N(2)–Cu(1)–O(1)
N(2)–Cu(1)–O(4)
N(3)–Cu(1)–O(1)
N(3)–Cu(1)–O(4)
N(4)–Cu(1)–O(1)
N(4)–Cu(1)–O(4)
O(1)–Cu(1)–O(4)
93.77(18)
177.93(17)
86.40(18)
86.2(2)
179.07(19)
93.6(2)
98.05(16)
91.70(16)
92.1(2)
87.74(18)
79.88(18)
90.37(18)
88.8(2)
96.51(5)
175.66(5)
86.55(5)
86.18(5)
175.21(5)
90.55(5)
98.37(4)
91.15(4)
92.67(4)
87.18(4)
84.86(5)
85.57(5)
90.52(4)
89.08(4)
170.42(4)
X-Ray structure determinations
Full sphere data were collected at 150(2) K, with the exception
of the Cu(1) and Cu(3) complexes for which data were collected
at 294(2) K. Data were collected with ω scans to 56Њ 2θ using a
Bruker SMART 1000 CCD diffractometer employing graphite
monochromated Mo-Kα radiation generated from a sealed
tube (0.71073 Å). A recollection of the reflections in the first 50
CCD frames at the end of the collection showed no significant
change in intensities. In general a Gaussian absorption correc-
tion was applied to the data with XPREP,4,5 as was a further
empirical correction determined with SADABS.6 An additonal
empircial correction was not used for the Ni(3) complex data.
The data integration and reduction were undertaken with
SAINT and XPREP,4 and subsequent computations were
carried out with the teXsan,6 WinGX7 and XTAL8 graphical
user interfaces. The structures were solved by direct methods
using SIR97,9 and extended and refined with SHELXL-97.10 In
general, the non-hydrogen atoms were modelled with aniso-
tropic thermal parameters and a riding atom model was used
for the hydrogen atom. The non-hydrogen methanol sites were
modelled with isotropic displacement parameters. ORTEP11
depictions of the molecules with 20% displacement ellipsoids
are provided provided in Figs. 6–10, and pertinent geometry
details are listed in Tables 5–8.
91.34(19)
170.24(16)
Crystal and structure refinement data
The refinement residuals are defined as
R1 = Σ||Fo| Ϫ |Fo||/Σ|Fo| for Fo > 2σ(Fo) and wR2 = (Σw(Fo2 Ϫ
Fc2)2/Σ(wFc2)2)1/2 all reflections where w = 1/[σ2(Fo ) ϩ (AP)2 ϩ
2
2
BP], P = (Fo2 ϩ 2Fo )/3] and A and B are as listed below.
[CuL(NO3)2]ؒ0.5MeOH (L = 1): C17.5H32CuN6O6.5, M =
494.03, monoclinic, space group C2/c (#15), a = 16.0210(17),
b = 19.181(2), c = 15.6723(16) Å, β = 104.095(2)Њ, V = 4671.1(8)
Å3, Dc (Z = 8) = 1.405 g cmϪ3, crystal size 0.370 × 0.247 ×
0.094 mm, colour red, habit prism, µ(Mo-Kα) = 0.981
Table 4 Observed M()/() and M()/() couples from cyclic voltammograms of copper() and nickel() complexes of ligands shown [glassy
carbon working electrode, 100 mV sϪ1 scan rate, acetonitrile/0.1 M Bu4N(ClO4), 1 mM complex, E1/2 (or Ep for irreversible couples) vs. Ag/AgCl]
Complex
M()/(), E1/2/V (∆E/mV)
M()/(), E1/2/V (∆E/mV)
[Cu(6)](NO3)2
[Cu(1)](NO3)2
[Cu(2)](NO3)2
[Cu(3)](NO3)2
[Cu(4)](NO3)2
[Cu(5)](NO3)2
[Ni(6)](ClO4)2
[Ni(1)](ClO4)2
[Ni(2)](ClO4)2
[Ni(3)](ClO4)2
[Ni(4)](ClO4)2
ϩ1.377 (irrev.)
ϩ1.442 (irrev.)
ϩ1.495 (irrev.)
ϩ1.540 (irrev.)
ϩ1.570 (irrev.)
ϩ1.630 (irrev.)
ϩ1.100 (75), ϩ0.583 (70)
ϩ1.074 (205), ϩ0.734 (100)
ϩ1.248 (215), ϩ0.925 (70)
ϩ1.402 (80), ϩ1.015a
ϩ1.52a, ϩ1.19a
Ϫ0.853 (85)
Ϫ0.640 (110)
Ϫ0.430 (230)
Ϫ0.406 (120)
Ϫ0.175 (75)
ϩ0.042 (105)
Ϫ1.447 (225)
Ϫ1.310 (110)
Ϫ1.217 (200)
Ϫ1.011 (60)
Ϫ0.917 (105)
a Reversibility limited or masked by adsorption phenomena.
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1568

View Article Online
Table 6 Bond lengths (Å) and angles (Њ) for [Ni(3)](ClO4)2
[CuL(H2O)](NO3)2ؒ2MeOH (L = 5): C40H58CuN6O9, M =
830.46, monoclinic, space group C2/c (#15), a = 24.200(12),
b = 9.471(5), c = 18.229(9) Å, β = 100.129(9)Њ, V = 4113(4) Å3,
Dc = 1.341 (Z = 4) g cmϪ3, crystal size 0.216 × 0.183 × 0.117
Ni(1)–N(1)
Ni(1)–N(2)
1.9701(15)
1.9441(16)
mm, colour blue, habit prismatc, µ(Mo-Kα) = 0.592 mmϪ1
,
N(1)–Ni(1)–N(2)
92.95(7)
T (Gaussian)min,max = 0.863, 0.936, 2θmax = 56.58, hkl range
Ϫ30 to 30, Ϫ12 to 12, Ϫ23 to 24, N = 21311, Nind = 4881
(Rmerge = 0.0351), Nobs = 4183 (I > 2σ(I )), Nvar = 260, residuals
R1(F) = 0.0479, wR2(F 2) = 0.1420, A = 0.0936, B = 3.68,
Table 7 Bond lengths (Å) and angles (Њ) for [Cu(3)(NO3)]NO3
Cu(1)–O(1)
Cu(1)–N(1)
Cu(1)–N(2)
2.6571(8)
2.1085(15)
1.9898(13)
GoF(all) = 1.072, ∆ρmin,max = Ϫ0.646, 0.872 eϪ ÅϪ3
.
Individual features. The asymmetric unit contains half of the
complex molecule centred on a two-fold axis passing through
Cu(1) and O(1). The asymmetric unit also contains a methanol
solvate molecule and a nitrate counter ion. The water hydrogen
was located and modelled with an isotropic displacement
parameter and a distance restraint.
[NiL](ClO4)2 (L = 3): C24H36Cl2N4NiO8, M = 638.18, mono-
clinic, space group P21/c (#14), a = 9.0053(10), b = 17.8845(19),
c = 8.6964(9) Å, β = 104.909(2)Њ, V = 1353.5(3) Å3, Dc = 1.566 g
cmϪ3, Z = 2, crystal size 0.352 × 0.337 × 0.206 mm, colour
orange, habit primatic, temperature = 294(2) K, µ(Mo-Kα) =
0.970 mmϪ1, T (Gaussian)mim,max = 0.698, 0.834, 2θmax = 56.54,
N(1)–Cu(1)–O(1)
N(1)–Cu(1)–N(2)a
N(1)–Cu(1)–N(2)
N(2)–Cu(1)–O(1)
94.56(4)
86.20(5)
93.80(5)
86.94(4)
a 1 Ϫ x, Ϫy, 1 Ϫ z.
Table 8 Bond lengths (Å) and angles (Њ) [Cu(5)(H2O)](NO3)2ؒ2MeOH
Cu(1)–O(1)
Cu(1)–N(1)
Cu(1)–N(2)
2.226(2)
2.145(2)
2.0841(18)
hkl range Ϫ11 to 11, Ϫ23 to 23, Ϫ11 to 11, N = 13290, Nind
=
3219 (Rmerge = 0.0241), Nobs = 2876 (I > 2σ(I )), Nvar = 178,
residuals R1(F) = 0.0349, wR2(F 2) = 0.1008, A = 0.05, B = 0.5,
N(1)–Cu(1)–N(1)a
N(2)–Cu(1)–N(2)a
N(1)–Cu(1)–N(2)
N(1)–Cu(1)–N(2)a
N(1)–Cu(1)–O(1)
N(2)–Cu(1)–O(1)
177.08(8)
150.76(9)
85.67(6)
93.59(7)
104.62(5)
91.46(4)
GoF(all) = 1.202, ∆ρ mim,max = Ϫ0.387, 0.670 eϪ ÅϪ3
.
Individual features. The asymmetric unit contains the half of
the complex molecule residing on an inversion centre, and a
perchlorate counter ion.
CCDC reference numbers 198031–198035.
See http://www.rsc.org/suppdata/dt/b2/b211394e/ for crystal-
lographic data in CIF or other electronic format.
a 1 Ϫ x, y, 1/2 Ϫ z.
mmϪ1,T (Gausian)min,max = 0.732, 0.914, 2θmax = 56.58, hkl range
Ϫ20 to 20, Ϫ25 to 24, Ϫ20 to 20, N = 24546, Nind = 5563 (Rmerge
= 0.0412), Nobs = 2672 (I > 2σ(I )), Nvar = 280, residuals R1(F) =
0.0672, wR2(F 2) = 0.2226, A = 0.07, B = 10.0, GoF(all) = 1.235,
Copper(II) solvent extraction
The solvent extraction experiments were carried out in screw
top glass vials containing a chloroform phase (5 cm3) and an
aqueous phase (5 cm3). The aqueous phase consisted of an
∆ρmin,max = Ϫ0.596, 0.707 eϪ ÅϪ3
.
unbuffered copper() nitrate solution (1 × 10Ϫ3 mol dmϪ3
)
Individual features. The asymmetric unit contains a complex
molecule with weakly coordinated nitrate counter ions and a
methanol solvate molecule. The occupancy of the methanol
sites was refined and then fixed at 0.5. The large N(5) nitrate
ellipsoids evident in Fig. 6 may be the result of disorder arising
from the partial occupancy of the methanol molecule, with
which it interacts.
and was maintained at a pH of 5.00 0.05 or 5.80 0.05 by
careful addition using microsyringe of dilute sodium
a
hydroxide or nitric acid during the course of the respective
extraction experiments. The chloroform phase (which had
been presaturated with distilled water) contained a ligand
chosen from 1–5 at a concentration of 1 × 10Ϫ3 mol dmϪ3
,
with a four-fold amount of palmitic acid also present. Con-
trol experiments contained only the corresponding concen-
tration of palmitic acid in the organic phase and under the
conditions employed no extraction of copper() was observed
in this case. All experiments involved shaking the extraction
vial for 24 h at 25 ЊC. The fraction of metal extracted was
calculated as: [moles of metal ion in the organic phase]/[total
number of moles of the metal in both phases]. The experi-
ments were performed in duplicate with the relative error
being approximately 10%.
[CuL(NO3)2]ؒMeOH (L = 2): C25H40CuN6O7, M = 600.17,
monoclinic, space group P21/c (#14), a = 13.316(3), b =
18.311(4), c = 12.303(3) Å, β = 113.132(4)Њ, V = 2758.6(11) Å3,
Dc (Z = 4) = 1.445 g cmϪ3, crystal size 0.331 × 0.287 × 0.177
mm, colour purple, habit prism, µ(Mo-Kα) 0.846 mmϪ1
,
T (Gaussian)min,max = 0.744, 0.878, 2θmax = 56.50, hkl range Ϫ17
to 16, Ϫ24 to 24, Ϫ16 to 16, N = 28625, Nind = 6560 (Rmerge
=
0.0263), Nobs = 5778 (I > 2σ(I )), Nvar = 354, residuals R1(F) =
0.0304, wR2(F 2) = 0.0869, A = 0.052, B = 1.08, GoF(all) =
1.031, ∆ρmin,max = Ϫ0.566, 1.077 eϪ ÅϪ3
.
Individual features. The asymmetric unit contains the com-
plex molecule, two nitrate counter ions and a methanol solvate
molecule.
Metal complex synthesis
All complexes were dried over P4O10 in a vacuum before
microanalysis.
[CuL(NO3)2] (L = 3): C24H36CuN6O6, M = 568.13, mono-
clinic, space group P21/c (#14), a = 9.9521(19), b = 11.851(2), c =
10.868(2) Å, β = 97.436(3)Њ, V = 1271.1(4) Å3, Dc (Z = 2) = 1.484
g cmϪ3, crystal size 0.265 × 0.173 × 0.088 mm, colour purple,
[Cu(1)(NO3)2]ؒ0.5MeOH. Copper() nitrate–hydrate (2/5)
(0.24 g, 1 mmol) in methanol (5 cm3) was added to a solution of
1 (0.29 g, 1 mmol) in methanol (15 cm3). The solution was
stirred and heated for 0.5 h and the solvent was removed under
reduced pressure. The residue was recrystallized from a mixture
of dichloromethane and methanol to yield red–purple crystals
(0.30 g, 60%). Single crystals suitable for X-ray diffraction were
obtained from dichloromethane–methanol (1 : 1) (Found: C,
42.68; H, 6.70; N, 16.89. C17.5H32CuN6O6.5 requires C, 42.55; H,
6.53; N, 17.01%).
habit tabular, µ(Mo-Kα) = 0.911 mmϪ1 T (Gaussian)min,max
=
0.794, 0.924, 2θmax = 56.64, hkl range Ϫ12 to 13, Ϫ15 to 15, Ϫ14
to 14, N = 11469, Nind = 3008 (Rmerge = 0.0249), Nobs = 2523 (I >
2σ(I )), Nvar = 169, residuals R1(F) = 0.0292, wR2(F 2) = 0.0835,
A = 0.0481, B = 0.473, GoF(all) = 0.985, ∆ρmim,max = Ϫ0.193,
0.317 eϪ ÅϪ3
.
Individual features. The asymmetric unit conatins half a
complex located on an inversion centre.
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1569

View Article Online
[Ni(1)](ClO4)2. Nickel() perchlorate hexahydrate (0.18 g,
0.5 mmol) in methanol (5 cm3) was added to a solution of 1
(0.15 g, 0.5 mmol) in methanol (15 cm3). The solution was
stirred and heated for 0.5 h and the orange solution was left to
stand at room temperature overnight. The golden yellow
crystals that formed were filtered off and recrystallized from
methanol (0.15 g, 65%) (Found: C, 37.31; H, 5.63; N, 10.03.
C17H30Cl2NiN4O8 requires C, 37.26; H, 5.53; N, 10.22%).
dichloromethane–methanol (1 : 1) containing a few drops of
hexane to produce needle-like crystals (0.29 g, 50%) (Found:
C, 30.85; H, 6.11; N, 21.55. C10H24CuN6O6 requires C, 30.97;
H, 6.24; N, 21.67%).
CAUTION: Perchlorate-containing complexes are poten-
tially explosive and appropriate precautions should be in place
for their preparation, handling and storage.
[Cu(2)(NO3)2]ؒMeOH. A similar procedure to that des-
cribed for [Cu(1)(NO3)2]ؒ0.5MeOH was employed. The crude
product was recrystallized from methanol to yield blue crystals
(yield 62%). Single crystals suitable for X-ray diffraction were
obtained from dichloromethane–methanol (Found: C, 49.88;
H, 6.59; N, 14.08. C25H40CuN6O7 requires C, 50.03, H, 6.72;
N, 14.00%).
Results and discussion
Isolation of selected complexes
The syntheses of a range of 1 : 1 (metal : ligand) complexes by
mixing hot methanol or ethanol solutions of the appropriate
copper() or nickel() nitrate or perchlorate salt and ligand
were generally straight forward, leading to analytically pure
compounds in reasonable yields. The complexes obtained
together with their conductance values in methanol are listed in
Table 1. Under the conditions employed, no solid nickel()
species of 5 were isolated, presumably reflecting weak inter-
action of this ion with this sterically hindered system.
Each of the copper() complexes incorporating nitrate
anions yielded a conductance value in methanol which approx-
imated that expected for a 1 : 1 electrolyte; thus indicating that
at least one of the anionic groups is on average coordinated to
(or associated with) the cation under the conditions employed.
[Ni(2)](ClO4)2. A similar procedure to that described for
[Ni(1)](ClO4)2 was employed. The crude product was purified
by recrystallization from methanol to give orange crystals (yield
50%) (Found: C, 45.32; H, 5.80; N, 8.67. C24H36Cl2NiN4O8
requires C, 45.17; H, 5.69; N, 8.78%).
[Cu(3)(NO3)]NO3. Using
a similar procedure to that
described for [Cu(1)(NO3)2]ؒ0.5MeOH, blue–purple crystals
were produced after recrystallization of the initial product from
dichloromethane–methanol (yield 65%). Single crystals suitable
for X-ray diffraction were obtained from the same solvent
mixture (Found: C, 50.61; H, 6.28; N, 14.73. C24H36CuN6O6
requires C, 50.74; H, 6.39; N, 14.79%).
Ϫ
In contrast, the nickel() complexes with ClO4 anions all
yielded values which fall between those expected for a 1 : 1 and
a 2 : 1 electrolyte. The latter behaviour parallels that reported
previously for [Ni(6)](ClO4)2 in methanol4 and indicates that
some measure of cation–anion association also occurs in the
present (nickel) systems in this solvent.
[Ni(3)](ClO4)2. Using a similar procedure to that described
for [Ni(1)](ClO4)2, nickel() perchlorate hexahydrate (0.144 g,
0.395 mmol) and 3 yielded orange crystals (yield 76%) after
recrystallization of the initial product from acetonitrile–water.
Single crystals suitable for X-ray diffraction were obtained from
the same solvent mixture (Found: C, 45.05; H, 5.58; N, 8.69.
C24H36Cl2NiN4O8 requires C, 45.17; H, 5.69; N, 8.78%).
UV–VIS spectral properties
The UV–VIS absorption maxima of the complexes [CuL]-
(NO3)2 (L = 1–6) in methanol and acetonitrile and [NiL](ClO4)2
(L = 1–4 and 6) in acetonitrile together with maxima for the
corresponding solid-reflectance spectra of both metal complex
types are listed in Table 2. Both the solid state and solution
spectra of the copper complexes show a broad envelope of
bands in the visible region which is consistent with the presence
[Cu(4)](NO3)]NO3. A similar procedure to that described for
[Cu(1)(NO3)2]ؒ0.5MeOH yielded the product as blue crystals
after recrystallization from absolute ethanol (yield 53%)
(Found: C, 56.45; H, 6.32; N, 12.85. C31H42CuN6O6 requires
C, 56.56; H, 6.43; N, 12.77%).
of
a copper()–N4 chromophore (in which solvent or
anion may occupy axial positions);13,14 however, as commonly
observed, the featureless nature of the spectra result in them
being of little use diagnostically for assignment of detailed
coordination geometries. It is noted that in the case of the
copper complex of the tetra-benzylated derivative 5, large shifts
in the λmax value occur for the copper complex between the
solid state and the spectra obtained in methanol or acetonitrile,
presumably reflecting differential solvent interaction with the
bound metal ion.
[Ni(4)](ClO4)2. A similar procedure to that described for
[Ni(1)](ClO4)2 yielded a product which was recrystallized from
methanol to give golden yellow crystals (yield 50%) (Found:
C, 50.95; H, 5.59; N, 7.55. C31H42Cl2NiN4O8 requires C, 51.13;
H, 5.81; N, 7.69%).
[Cu(5)(H2O)](NO3)2ؒ2MeOH. Copper() nitrate–hydrate
(2/5) (0.093 g, 0.4 mmol) in absolute ethanol (5 cm3) was added
to a solution of 5 (0.22 g, 0.4 mmol) in absolute ethanol
(50 cm3). The mixture was heated for 10 min and then the
green–blue solution was stirred at room temperature overnight.
The volume of the solvent was reduced to 10 cm3 and the blue–
green powder that formed was filtered off. This product was
recrystallised from methanol to produce blue–green single
crystals (0.15 g, 45%) (Found: C, 57.71; H, 6.95; N, 10.05.
C40H58CuN6O9 requires C, 57.85; H, 7.04; N, 10.12%).
The solid state visible spectra of the orange nickel complexes
of (1–4) each shows a single band with a maximum in the range
473–488 nm, consistent with the adoption of a square planar
(low spin) geometry in each case. The value obtained for the
corresponding nickel complex of cyclam, [Ni(6)]2ϩ, is 459 nm.
This cation12,13 has been well documented to be square planar
in both the solid state14,15 and (in the absence of coordinating
anions) also in aqueous solution;16 in the latter solvent the
absorption maximum in the visible region occurs at 450 nm
(ε = 45 MϪ1 cmϪ1).
[Cu(6)(NO3)]NO3. Copper() nitrate–hydrate (2/5) (0.35 g,
1.5 mmol) in absolute ethanol (5 cm3) was added to a solution
of 6 (cyclam) (0.3 g, 1.5 mmol) in chloroform (50 cm3). The
solution was heated for 0.5 h and the solvent was evaporated on
a rotary evaporator. The residue was dissolved in dichloro-
methane–methanol (1 : 1) (25 cm3) and ether vapour was
allowed to diffuse into the solution over 3 days. The red–purple
crystalline product that formed was recrystallized from
The influence of the incremental addition of N-benzyl sub-
stituents onto the cyclam framework on the electronic structure
of both the respective copper and nickel complexes is especially
worthy of note. The spectra obtained for each series of metal
complexes confirm the expected result that the sequential intro-
duction of N-benzyl substituents tends to produce a change in
the ligand field properties towards lower fields. For example, for
the square-planar nickel complexes of 1–4 and 6, the tendency
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1570

View Article Online
towards a shift to higher wavelengths as the number of
N-benzyl substituents is increased is evident from the data
presented in Table 2. Since this absorption is expected to be
due primarily to the promotion of an electron from the dxy to
dx Ϫy orbital, the variation in the absorption maximum will
represent the effect of changes in the donor properties on the
energy difference between these orbitals. Of course, there is the
possibility that acetonitrile or dimethyl sulfoxide molecules may
interact in one or both of the axial positions and thus influence
the separation between the above orbitals (although the singlet
ground state is certainly maintained in the solid state and
essentialy maintained in solution—see below).
It is noted that a ‘yellow-to-blue’ interconversion was
observed within the present series of [NiL]2ϩ complexes dis-
solved in acetonitrile as evidenced by the presence of weak
absorptions (typical of high-spin, distorted octahedral species)
that were present alongside the main low-spin, square planar
absorption that fell in the range 461–517 nm (Table 2).
Jørgensen17 first reported that the nickel() complex with the
open-chain tetramine trien (1,4,7,10-tetraazadecane) exists in
solution as an equilibrium mixture of yellow, diamagnetic
square species [Ni(trien)]2ϩ and blue, high-spin, octahedral
species of type [Ni(trien)(H2O)2]2ϩ. This behaviour was sub-
sequently found to be common for the nickel() complexes with
larger open-chain tetramines, the value of the equilibrium
constant for the blue-to-yellow conversion being quite depend-
ent upon ligand structure.18 For the complexes of the macro-
cyclic analogues of the linear tetramines in dilute solution and
at room temperature, an equilibrium mixture of high-spin and
low-spin species was found to be present only in the case of the
14-membered macrocycles cyclam and isocyclam, whereas for
ligands of lower and greater ring size a complete predominance
of the blue octahedral complex was found to occur.19
consistent with the planar ligand field becoming weaker while
the axial ligand field becomes stronger,26 reflected in the energy
separation of the dxy and dx Ϫy orbitals.
2
2
2
2
Electrochemistry
Both the copper() and nickel() series of macrocyclic com-
plexes display waves in the cyclic voltammogram associated
with M()/() and M()/() processes. In both series, E1/2 (or Ep,
for irreversible cases) values shift to more positive potentials as
N-benzyl substitution is progressed. For copper, the Cu()/()
couple spans ∼0.9 V from [Cu(cyclam)]2ϩ to tetrabenzylated
[Cu(5)]2ϩ, whereas the Cu()/() couple spans ∼0.3 V. Spans for
the Ni()/() couple are ∼0.55 V and for Ni()/() ∼0.35 V,
for the series from [Ni(cyclam)]2ϩ to [Ni(4)]2ϩ. Results for
acetonitrile at a glassy carbon working electrode are summar-
ized in Table 4; detailed data determined at both C and Pt
electrodes in both acetonitrile and acetone solvents have also
been obtained. The size of the effects is consistent with the
assignment of the waves as metal-centred processes, where the
consequence of N-alkylation of the coordinated amine donors
is sensed. Previously, two series of nickel() complexes of
(N-methyl)cyclams have been examined,27 the two series differ-
ing in the stereochemistry at the coordinated nitrogens. The two
series, trans-III and trans-I forms, showed very little differences
in the redox potentials for each N-methylated geometric pair,
but an overall span of ∼0.55 V from unsubstituted cyclam to
tetramethylcyclam complexes was observed.
The observation of consistent anodic shifts with increasing
N-alkylation for both nickel and copper is illustrated in Fig. 1.
The introduction of substituents on the nitrogens has several
outcomes, which can all contribute to the observed shift. The
substituents raise steric bulk, and the substituents may interact
sterically with any donor entering the axial sites. These substi-
tutents will also give rise to inductive effects. It is notable that
substitution on remote sites on the framework of macrobicyclic
hexaamines shifts the potential of the Co()/() couple over a
range of ∼0.5 V, although alkyl substituents alone do not cause
a marked effect in those examples.28 N-Substitution also leads
to longer M–N distances for the tertiary compared with the
secondary amine donors, the weaker interaction of the tertiary
amine with the metal centre stabilizing antibonding orbitals on
the metal, making electron removal less favourable and electron
addition more favourable. The macrocycle ‘hole’ size will also
play a role, since the metal ions in the various oxidation states
are of different sizes. Although hole size effects will show up
more clearly as the overall macrocyclic framework size changes
(such as when methylene groups are inserted in, or deleted
from, the macrocyclic ring),29 more subtle cavity size effects
arising from N-alkylation may be present in the present series.
It is anticipated that such a combination of effects contribute to
the observed trends.
EPR studies
The X-band EPR spectra of [Cu(cyclam)]2ϩ and [Cu(1)]2ϩ
–
[Cu(5)]2ϩ in dimethylformamide glasses were recorded at 77 K
with no major differences being observed between the members
of this series, although a reasonably consistent small increase
in g|| and a matching decrease in A|| is evident on increasing
the degree of N-alkylation. The parameters obtained for the
complexes are shown in Table 3.
The spectra exhibit four equally spaced lines in the parallel
region, as expected from the coupling of the unpaired electron
density to the copper() nucleus (I = 3/2), although no super-
hyperfine coupling to the nitrogen donors was observed (nor
was it anticipated based on a prior study).20 These spectra, with
g|| > g⊥, are typical of axially symmetric d9 copper() complexes
2
2
in a ground-state doublet with the unpaired electron in a dx Ϫy
orbital.21 The g and A|| values of [Cu(1)]2ϩ–[Cu(5)]2ϩ are
close to that for [Cu(cyclam)]2ϩ 14,22–24 suggesting an essentially
square-planar coordination geometry.21 Moreover, the g|| values
(for [Cu(1)]2ϩ–[Cu(5)]2ϩ) are slightly higher than that for
[Cu(cyclam)]2ϩ. The latter is consistent with a trend to higher g||
values upon N-alkylation of tetraazamacrocycles, perhaps
resulting from true square-planar coordination becoming
disfavoured.20,25
The hyperfine coupling constant, A||, is a more reliable indi-
cator of distortion from square-planar towards pyramidal or
tetrahedral coordination in copper() complexes, with lower
values implying distortion. The present values in the frozen
solution, ranging from 200 G for [Cu(cyclam)]2ϩ to 155 G for
[Cu(5)]2ϩ, are consistent with the result of X-ray diffrac-
tion experiments (see later), with tetrahedral distortion being
greatest in [Cu(5)]2ϩ
.
Comparison of the EPR and electronic spectroscopy param-
eters shows that the trend of a slight increase in g|| and a
decrease in A|| from [Cu(cyclam)]2ϩ to [Cu(5)]2ϩ is matched by a
red shift of λmax (as already discussed). Both observations are
Fig. 1 Variation of redox potentials for NiII–I (ꢀ), CuII–I (᭿), NiII–III
(᭹) and CuII–III (᭹) couples with the number of N-benzyl groups.
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1571

View Article Online
The Cu()/() couples usually display quasi-reversible
electrochemical behaviour, with separation of anodic (Epa) and
cathodic (Epc) maxima in the cyclic voltammograms ranging
from 75 to 250 mV and peak current ratios (Ipa/Ipc) approaching
unity at 100 mV sϪ1 scan rate (Fig. 2), with typical increas-
ing peak-to-peak separation (∆E) as the scan rate increases.
Peak current varies linearly with (scan rate)1/2, in accordance
with a diffusion controlled process; this is consistently found
in the present study wherever quasi-reversibility is observed.
The couples exhibit both solvent- and electrode-dependent
behaviour. For example, [Cu(3)]2ϩ at glassy carbon displays
E1/2 = Ϫ0.406 V (∆E = 123 mV) in acetonitrile and Ϫ0.437 V
(∆E = 160 mV) in acetone, whereas at the Pt electrode E1/2 (∆E)
are Ϫ0.465 V (195 mV) and Ϫ0.490 V (177 mV) in acetonitrile
and acetone, respectively.
The large N-benzyl pendants introduced may act to limit dis-
sociation processes by their bulk, thus enhancing the revers-
ibility of couples.
The copper()/() couple, which occurs at very positive
potential, is irreversible. This is consistent with observations for
copper tetraaza macrocycles generally;29 reversibility has been
reported only in very strong perchloric acid solution.31 The
irreversibility may reflect a major change in coordination
geometry from the d8 to the d7 system, discussed above, but may
equally well relate to increased lability of the copper() leading
to a rapid following dissociation reaction in an EC mechanism,
or a capacity for the powerful copper() oxidant to oxidize the
coordinated saturated polyamine ligand. Both a parallel posi-
tive shift in potential for the M()/() and putative M()/()
processes, and comparable wave-heights consistent with two
one-electron processes, support the assignment of this latter
wave as a metal-centred oxidation. Another wave seen for only
some samples at potentials more anodic than this process as a
shoulder on the edge of the solvent limit is most likely related to
traces of water in the solvent.
The Ni()/() couple shifts anodically with N-substitution
and consistently displays quasi-reversible behaviour (Fig. 3),
albeit complicated in many cases by adsorption–desorption
phenomena. However, for the Ni()/() process it is noted
that two sets of waves that are quasi-reversible in character
are observed, with one usually dominant in terms of peak
currents. The relative size of these pairs of waves varies con-
siderably with the nature of the ligand, solvent and electrode.
However, their summed wave-heights (albeit taken from the
cyclic voltammograms rather than obtained using stirred
voltammetry) are essentially equivalent to that of the simple
Ni()/() process (Fig. 4), which suggests that they represent
components of the same overall one-electron process. The
fact that the shift in each on N-alkylation follows the same
trend supports this. The two processes could simply reflect
adsorbed and free species, which are oxidized at different
potentials according to:
Fig. 2 Cyclic voltammogram for [Cu(4)]2ϩ. Experimental conditions:
glassy carbon working electrode, 100 mV sϪ1 scan rate, 1 mmol dmϪ3
solution in acetonitrile, 0.1 mol dmϪ3 Bu4N(ClO4) electrolyte, potentials
vs. Ag/AgCl.
In some cases, a shift to effectively irreversible behaviour is
observed, particularly at the Pt electrode. This type of elec-
trode-dependent behaviour has also been reported in a study of
a tetraalkylated-cyclam copper() complex.30 The observed
variation in behaviour at each electrode results from different
heterogeneous electron transfer rates at the electrode surfaces,
with the relatively fast electrochemical technique of cyclic
voltammetry highlighting these differences. Further, the more
reversible behaviour on carbon in this series possibly also
reflects a greater compatibility of the complex with the surface
via the benzene rings π-stacking with a pseudo-aromatic array
of surface carbon atoms or else through the ionic complex ion-
pairing to carboxylate groups arising from oxidized carbon sites
on the surface.
The solvent shifts, although modest, contrast with very small
solvent-dependent shifts for ferrocenium/ferrocene, suggesting
solvation effects that are not simply the result of junction
potential differences. Solvent effects may reflect outer-sphere
solvation or even a role for solvent in the inner coordination
sphere; acetonitrile is a better ligand than acetone, and can
readily occupy coordination sites in expanded coordination
number species. In particular, both reduction and oxidation of
d9 copper() remove or diminish the significant Jahn–Teller
elongation of axial sites, enhancing the opportunity for solvent
to bind in these sites whenever higher coordination number
geometries are stabilized [although this outcome is more likely
for copper() than copper()].
Fig. 3 Cyclic voltammogram for [Ni(2)]2ϩ complex. Experimental
conditions: glassy carbon working electrode, 500 mV sϪ1 scan rate,
1
mmol dmϪ3 solutions in acetone, 0.1 mol dmϪ3 Bu4N(ClO4)
electrolyte, potentials vs. Ag/AgCl.
(1)
The electrochemical reversibility displayed implies a capacity
on the timescale relevant for the complexes to adapt to geo-
metrical changes as the oxidation state changes. For the Cu()/
() couple, square planar and tetrahedral geometries may be
anticipated for copper() and copper(), respectively, and the
ligand must be flexible enough to accommodate this transform-
ation, as well as retain the metal ion in the macrocyclic cavity.
Such processes appear in some of the Ni()/() couples, where
the relative size of component shoulders and peaks vary some-
what with scan speed, supporting the assignment. In the Ni()/
() process the shifts are substantial (as much as ∼0.6 V), and
are less influenced by scan rate. More properly, they likely
reflect the presence of low-spin to high-spin equilibria at the
electrode surface:
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1572

View Article Online
in the present study, no clear candidates for ligand-centred
processes exist. A disproportionation reaction may also be
entertained; however, given the limited information available,
further speculation appears unwarranted.
It is noted that an approximately linear relationship occurs
between the position of the d–d transition maximum (νmax) and
E1/2 (V) for both reduction and oxidation processes for each of
the copper and nickel complexes (the situation for copper is
illustrated in Fig. 5). Reasons for correlations of maxima
in electronic spectra with E o are not immediately obvious,
although they have been reported before.27 However, they are by
no means universal; for example, an extensive series of simple
[Co(NH3)5X]nϩ complexes show no such relationship.36
Electrode potential reflects differences in the energies of the
Mnϩ/M(n ϩ 1)ϩ ground states, whereas electronic transitions arise
form different ground and excited states of Mnϩ. Nevertheless,
trends in the common ground state energy may influence both
properties in the same sense. To complicate the outcome, the
broad absorption envelope for copper() is the sum of several
transitions under the severely tetragonally distorted symmetry,
and hence provides a fairly modest measure of ligand field
strength. For nickel(), the spin equilibria in solution discussed
above complicate the situation, as shown, for example, by the
distinctly different spectra for [Ni(4)]2ϩ in solution and the solid
state. Nevertheless, the existence of a relationship is in accord-
ance with observations for other macrocyclic systems, even
though exceptions occur.
Fig. 4 Cyclic voltammogram for [Ni(1)]2ϩ. Experimental conditions:
glassy carbon working electrode, 100 mV sϪ1 scan rate, 1 mmol dmϪ3
solutions in acetonitrile, 0.1 mol dmϪ3 Bu4N(ClO4) electrolyte,
potentials vs. Ag/AgCl.
(2)
The bis(solvent) complex would be expected to be oxidized at a
different potential and show substantial solvent effects, as is
observed. For example, with [Ni(1)]2ϩ, the positions of the pairs
of waves at glassy carbon in acetonitrile (1.074, 0.734 V) differ
clearly from those in acetone (1.394, 0.708 V). It is tempting to
assign the more anodic wave, where shifts are greatest and the
peak currents are larger, to the six-coordinate high spin form.
The Ni(cyclam)]ϩ complex in acetonitrile is known to exist as a
78 : 22 mixture of 6- : 4-coordination species,32 supporting this
proposal. Further, as mentioned already, spin equilibria have
been established in solution for the compounds in this study.
The complex patterns for particular Ni()/() couples may
reflect similar behaviour masked by adsorption, with the
absence of two processes in some cases simply relating to
coincidence of reduction potentials.
Irreversible waves are observed at more negative potentials
following the M()/() couples for both metals. These may be
associated with formal reduction to M(0). The observation of
irreversible processes near Ϫ2.0 V for tetraazamacrocyclic
nickel complexes, assigned to such a process, have been
reported.33 Alternatively, as remaining amine protons become
more acidic with N-substitution, abstraction of a H atom from
the weaker NH bond is possible, allowing a process of the
type:34
Fig. 5 The relationship between νmax (solid state absorption visible
spectra) vs. redox potentials (CuII–I) for the copper() complexes of 1–6.
Crystal structures
ORTEP11 depictions of [Cu(1)(ONO2)2]ؒ0.5MeOH, [Cu(2)-
(ONO2)2]ؒMeOH, [Cu(3)(ONO2)2], [Ni(3)](ClO4)2 and [Cu(5)-
(H2O)](NO3)2ؒ2MeOH are provided in Figs. 6–10 and relevent
geometry details are listed in Tables 5–8. The copper() com-
plex of 1, isolated as its nitrate salt, is formulated as [Cu(1)-
(ONO2)2]ؒ0.5MeOH. The 14-membered macrocyclic ring
adopts the stable trans-III configuration37 (Fig. 6), with the
copper() situated at the centre of the macrocyclic cavity with
the ligand maintained in a planar arrangement. The Cu–N(1)
bond length of 2.061(4) Å (to NCH2Ph) is slightly longer than
that of Cu–N(2) 1.997(5) Å, Cu–N(3) 2.027(4) Å and Cu–N(4)
2.010(5) Å, presumably as a consequence of alkylation on the
N(1) atom. The coordination sphere is completed by oxygen
donors from two nitrato ligands, with a Cu(1)–O(1) bond
length of 2.804(6), and a Cu(1)–O(4) bond length of 2.476(4) Å
(Fig. 6, Table 5). The resulting geometry is a tetragonally dis-
torted octahedron as is commonly found for copper() com-
plexes.38 There is a hydrogen bonding interaction between the
methanol solvate molecule and the N(5) nitrate ion, and hydro-
gen bond interactions between the amine nitrogen atoms and
the nitrate counter ions of neighbouring complex molecules
(see CIF file).
(3)
However, no reverse wave near Ϫ0.45 V, seen for hydrogen-
saturated acetone at the Pt electrode, is observed in the cyclic
volammograms, suggesting that this process is not occurring.
The Ni()/() couple for [Ni(4)]2ϩ in acetonitrile does not shift
appreciably on addition of Et3N as base (Ϫ0.915 vs. Ϫ0.902 V
with base), nor does the first following process (Ϫ1.49 vs
Ϫ1.40 V with base), also not supporting a proton abstraction
mechanism. Instead, for the copper complexes, a sharp oxid-
ation peak with a Lorentzian profile, indicative of an electrode
stripping process, is observed near Ϫ0.2 V; for nickel complexes,
this is absent. These observations suggest irreversible processes
with rapid following chemical reactions, such as would occur
for reduction to M(0). Irreversible waves at very negative poten-
tial for some N-pendants including thioether functionality have
been ascribed to ligand-centred processes involving this group
as they are absent when this group is not present;35 however,
The copper() complex of 2 is similarly formulated as
[Cu(2)(ONO2)2]ؒMeOH, and also has the macrocycle in a trans-
III configuration, with the two benzyl substituents lying on the
same side of the square plane of the macrocycle donors (Fig. 7,
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1573

View Article Online
Fig. 6 ORTEP depiction of [Cu(1)(ONO2)2]ؒ0.5MeOH, with 20%
displacement ellipsoids.
Fig. 8 ORTEP depiction of [Ni(3)](ClO4)2, with 20% displacement
ellipsoids. The molecule is centred on an inversion site.
Fig.
7
ORTEP depiction of [Cu(2)(ONO2)2]ؒMeOH, with 20%
displacement ellipsoids.
Table 5). As expected, the Cu–N bond lengths (to NCH2Ph) of
2.0708(12) and 2.0719(12) Å are longer than the remaining
Cu–N bond lengths (to NH) at 2.0207(13) and 2.0272(13) Å.
The counter ions are weakly coordinated to the metal, with a
2.7325(16) Å Cu(1)–O(1) bond length and a Cu(1)–O(4) length
of 2.5663(14) Å. There is also a hydrogen bond interaction
between the coordinated nitrate atom O(1) and the methanol
O(7) (see CIF file). The overall coordination geometry is a
tetragonally distorted octahedron.
The nickel ion in [Ni(3)](ClO4)2 has a typical square-planar
coordination geometry (Fig. 8), and the cyclam ring once again
adopts the trans-III configuration. The Ni–N bond lengths are
1.9441(16) Å (to NH) and 1.9701(15) Å (to NCH2Ph), respect-
ively (Table 6). Two perchlorate counter ions are located above
and below the coordination plane, but are beyond bonding
distances.
Like the [Ni(3)](ClO4)2 complex, the Cu(3) analogue resides
on an inversion centre (Fig. 9, Table 7); however, the copper
complex is hexacoordiante with two weakly coodinated nitrate
counter ions and so is formulated as [Cu(3)(ONO2)2]. The metal
to axial ligand oxygen distance is 2.6571(8) Å. Again the 14-
membered ring adopts a square-planar conformation with the
amine nitrogens, respectively, adopting a RRSS arrangement
(trans-III). The Cu–N bond lengths are 1.9898(13) Å (to NH)
and 2.1085(13) Å (to NCH2Ph), and the metal to nitrate oxygen
bond distance is 2.6571(8) Å. There is a weak hydrogen bond
between N(2) and O(2) on a neighbouring complex molecule,
with a donor–acceptor separation of 2.948(2) Å (see CIF file).
Fig. 9 ORTEP depiction of [Cu(3)(ONO2)2], with 20% displacement
ellipsoids. The molecule is centred on an inversion site, and there are
hydrogen bonding interactions between the equatorial and axial
ligands.
The copper() complex of the tetrabenzyl derivative 5, with
additional methanol solvate molecules, is formulated as
[Cu(5)(H2O)](NO3)2ؒ2MeOH. A two-fold axis passes through
the metal and coordinated water molecule (Fig. 10, Table 8).
The crystal structure confirms that all four benzyl groups lie on
the same side of the coordinated macrocycle in an RSRS con-
figuration (trans-I isomer).37 There is a pseudo-tetrahedral dis-
tortion at the copper() centre with the N(1)–Cu(1)–N(1) bond
angle being aproximately linear at 177.08(8)Њ in contrast to the
N(2)–Cu(1)–N(2) angle of 150.76(9)Њ. The Cu(1)–N(1) bond
length of 2.145(2) Å is slightly longer than the 2.0841(18) Å
Cu(1)–N(2) bond length. The tetrahedral distortion presum-
ably minimises steric interactions between the benzyl side-
chains and the macrocycle backbone. The five coordination
sphere is completed by one oxygen donor from a water mole-
cule. The resulting geometry approximates square pyramidal.
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1574

View Article Online
Fig. 10 ORTEP depiction of [Cu(5)H2O](NO3]2ؒ2MeOH, with 20% displacement ellipsoids. A two fold axis passes through the axial oxygen and
metals atoms.
There is a moderate hydrogen bond interaction between the
nitrate counter ions and the coordinated water molecule, and
between the nitrate ions and the methanol solvate molecules.
The O(1) ؒ ؒ ؒ O(2) donor to acceptor distance is 2.817(2), and
the O(5) ؒ ؒ ؒ O(4) distance is 2.779(4) Å (see CIF file).
In summary, the crystal structures of the above complexes
show several features in common. First of all, the metal ion is
usually situated in the centre of the macrocyclic cavity and
coordinated by four nitrogen atoms from the cyclam backbone
with, in many instances, oxygens from either nitrato anions or
water molecules occupying axial sites. Secondly, the cyclam ring
has the tendency to adopt the most stable trans-III configur-
ation although the slightly less stable trans-I arrangement is
also observed in the case of [Cu(5)]2ϩ. Finally, benzylation of
nitrogen donors results in longer metal to nitrogen bond lengths
than those involving secondary nitrogens in all structures.
Fig. 11 The effect of increasing N-benzylation of cyclam on the
efficiency of solvent extraction of copper(). Aqueous phase: 5 cm3,
contained copper() nitrate with an initial concentration of 1 × 10Ϫ3
mol dmϪ3 (pH maintained at 5.00 0.05). Chloroform phase: 5 cm3,
contained ligand chosen from 1–5 at 1 × 10Ϫ3 mol dmϪ3 and palmitic
acid at 4 × 10Ϫ3 mol dmϪ3. The extraction vial was shaken for 24 h at
25 ЊC.
Copper(II) solvent extraction studies
A comparative investigation of the effect of incremental
N-benzylation of cyclam on the efficiency of 1–5 to act as
extractants for copper() has been performed.
for the ligands’ amine sites. Besides pH, and as discussed
previously,39 the interplay of two further subtle factors may
dominate extraction behaviour of the present type. First, the
lipophilicity of ligand will play a role in influencing extraction
efficiency. More lipophilic ligands (lipophilicity is expected to
increase with increasing N-benzylation) may aid the extraction
process by inhibiting the bleeding of the amine ligand (or its
metal complex) from the organic to the aqueous phase as
well as enhancing extraction through favorable solvation
effects. Indeed, the appearance of a pink–violet coloration in
the aqueous phase for each of the systems incorporating the
less lipophilic systems 1–3 is attributed to bleeding of macro-
cycle from the organic phase under the conditions employed.
Secondly, the overall thermodynamic affinity of the ligand for
the metal ion of interest will also be important in influencing
the efficiency of extraction behaviour. In this regard both the
electronic and steric nature of the extractant may play
important roles in solvent extraction. For example, when four
N-substituted phenyl groups are present, as occurs in 5, steric
hindrance associated with the four benzyl substituents may be
the cause of the observed somewhat reduced extraction effi-
ciency even though the lipophilicity of this metal–ligand system
might be expected to be the highest of the present series.
Indeed, earlier workers found no evidence for the complexation
Clearly, the pH of the aqueous phase in a solvent extraction
system may play a crucial role in influencing the efficiency of
extraction of metal ions when amine-containing ligands are
involved. In a previous study,31 the copper() extraction effi-
ciency of a 1:4 mixture of tetrabenzylcyclam (5) and palmitic
acid under various pH conditions was investigated. This
experiment enabled the plotting of a classical ‘S-curve’; namely
a curve of percent extraction against pH. For pH values less
than about 5, copper() extraction became no longer favoured
reflecting proton competition for the ligand. At pH values
above about 6, competing hydrolysis occurred leading to pre-
cipitation of copper hydroxide. As a consequence of these
results, pH values between 5.00 and 5.80 were chosen for the
aqueous phase for the experiments undertaken in the current
study.
Fig. 11 shows that, at pH 5.00 0.05, the extraction efficiency
increased almost linearly as the number of N-benzyl groups
increased from 1 to 3, but then dropped for tetrabenzylcyclam.
Related behaviour to this was observed when the pH was main-
tained at 5.80 0.05 with the difference that for each ligand
system somewhat enhanced extraction was observed at this
(higher) pH—presumably reflecting lower proton competition
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1575

View Article Online
18 A. Anichini, L. Fabbrizzi, P. Paoletti and R. M. Clay, Inorg. Chim.
Acta, 1977, 24, L21; R. W. Hay, B. Kinsman and C. I. Smith,
Polyhedron, 1995, 14, 1245; L. Fabbrizzi, Comments Inorg. Chem,
1985, 4, 33; A. De Blas, G. De Santis, L. Fabbrizzi, M. Licchelli and
P. Pallavicini, Pure Appl. Chem., 1993, 65, 455.
19 M. Ciampolini, L. Fabbrizzi, M. Licchelli, A. Perotti, F. Pezzini and
A. Poggi, Inorg. Chem., 1986, 25, 4131.
20 A. E. Geota, J. A. K. Howard, D. Maffeo, H. Puschmann, J. A. G.
Williams and D. S. Yufit, J. Chem. Soc., Dalton Trans., 2000, 1873.
21 H. Kurosaki, C. Bucher, E. Espinosa, J.-M. Barbe and G. Guilard,
Inorg. Chim. Acta, 2001, 322, 145.
of either copper() or nickel() by this ligand in a number of
common solvents.40
Acknowledgements
We thank the Australian Research Council for support.
References
1 D. Meyerstein, Coord. Chem. Rev., 1999, 185–186, 141.
2 Y. Dong and L. F. Lindoy, Aust. J. Chem., 2001, 54, 291.
3 E. K. Barefield, F. Wagner, A. W. Herlinger and A. R. Dahl,
Inorg. Synth., 1976, 16, 220.
4 Bruker (1995); SMART, SAINT and XPREP. Area detector control
and data integration and reduction software. Bruker Analytical
X-ray Instruments Inc., Madison, WI, USA.
5 P. Coppens, L. Leiserowitz and D. Rabinovich, Acta Crystallogr.,
1965, 18, 1035.
6 Molecular Structure Corporation, teXsan for Windows: Single
Crystal Structure Analysis Software, MSC, 3200 Research Forest
Drive, The Woodlands, TX 77381, USA, 1997–1998.
22 S. V. Rosokha, Y. D. Lampeka and I. M. Maloshtan, J. Chem. Soc.,
Dalton Trans., 1993, 631.
23 M. Lachkar, G. Guilard, A. Atmani, A. De Cian, Jean Fischer and
Raymond Weiss, Inorg. Chem., 1998, 37, 1575.
24 E. Zeigerson, G. Ginzburg, D. Meyerstein and L. J. Kirschenbaum,
J. Chem. Soc., Dalton Trans., 1980, 1243.
25 F. E. Mabbs and D. Collison, Electron paramagnetic resonance of d
transition metal compounds, Elsevier, Amsterdam, 1992.
26 X. H. Bu, Z. H. Zhang, D. L. An, Y. T. Chen, M. Shionoya and
E. Kimura, Inorg. Chim. Acta, 1996, 249, 125.
27 E. K. Barefield, G. M. Freeman and D. G. van Derveer,
Inorg. Chem., 1986, 25, 552.
7 WinGX L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
8 S. R. Hall, D. J. du Boulay and R. Olthof-Hazekamp, eds. Xtal3.6
System, University of WesternAustralia, 1999.
9 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
C. Giacovazzo, A. Guagliardi, A. G. C. Moliterni, G. Polidori and
R. Spagna, J. Appl. Crystallogr., 1999, 32, 115.
10 G. M. Sheldrick, SHELX97. Programs for Crystal Structure
Analysis, University of Göttingen, Institüt für Anorganische
Chemie der Universität, Tammanstrasse 4, D-3400 Göttingen,
Germany, 1998.
28 A. M. Bond, G. A. Lawrance, P. A. Lay and A. M. Sargeson,
Inorg. Chem., 1983, 22, 2010.
29 E. Kimura, T. Koike, R. Machida, R. Nagai and M. Kodama,
Inorg. Chem., 1984, 23, 4181.
30 R. Bilewicz and M. Pietraszkiewicz, Polyhedron, 1990, 9, 2353.
31 M. Ciampolini, L. Fabbrizzi, A. Perotti, A. Poggi, B. Seghi and
F. Zanobibi, Inorg. Chem., 1987, 26, 3527.
32 N. Sabatini and L. Fabbrizzi, Inorg. Chem., 1979, 18, 438.
33 F. V. Lovecchio, E. S. Gore and D. H. Busch, J. Am. Chem. Soc.,
1974, 96, 3109.
11 C. K. Johnson, ORTEPII. Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
12 B. Bosnich, M. L. Tobe and G. A. Webb, Inorg. Chem., 1965, 4, 1109.
13 R. Buxtrof and T. A. Kaden, Helv. Chim. Acta, 1974, 57, 1035;
M. Hediger and T. A. Kaden, Helv. Chim. Acta, 1983, 66, 86.
14 L. Fabbrizzi, L. Montagna, A. Poggi, T. A. Kaden and L. C.
Siegfried, J. Chem. Soc., Dalton Trans., 1987, 2631.
34 H. A. Boucher, G. A. Lawrance, P. A. Lay, A. M. Sargeson,
A. M. Bond, D. F. Sangster and J. C. Sullivan, J. Am. Chem. Soc.,
1983, 105, 4652.
35 C. L. Schmid, C. Kempf, A. Taubert, M. Neuburger, M. Zehnder
and T. A. Kaden, Helv. Chim. Acta, 1996, 79, 1011.
36 N. J. Curtis, G. A. Lawrance and A. M. Sargeson, Aust. J. Chem.,
1983, 36, 1327.
15 K. R. Adam, M. Antolovich, L. G. Brigden, A. J. Leong, L. F.
Lindoy, P. J. Baillie, D. K. Uppall, M. McPartlin, B. Shah,
D. Proserpio, L. Fabbrizzi and P. A. Tasker, J. Chem. Soc.,
Dalton Trans., 1991, 2493; K. A. Adam, I. M. Atkinson and
L. F. Lindoy, Inorg. Chem., 1997, 36, 480; M. A. Donnelly and
M. Zimmer, Inorg. Chem., 1999, 38, 1650.
16 E. Zeigerson, G. Ginzburg, D. Meyerstein and L. J. Kirschenbaum,
J. Chem. Soc., Dalton. Trans., 1980, 1243.
17 C. K. Jøgensen, Acta Chem. Scand., 1957, 11, 399.
37 L. F. Lindoy, The Chemistry of Macrocyclic Ligand complexes,
Cambridge University Press, Cambridge, 1989.
38 M. H. Meyer, P. Singh, W. E. Hatfield and D. J. Hodgson,
Acta Crystallogr., Sect. B, 1972, 28, 1607.
39 K. R. Adam, I. M. Atkinson, S. Farquhar, A. J. Leong, L. F. Lindoy,
M. S. Mahinay, P. A. Tasker and D. Thorp, Pure Appl. Chem., 1998,
70, 2345.
40 M. R. Oberholzer, M. Neuburger, M. Zehnder and T. A. Kaden,
Helv. Chim. Acta, 1995, 78, 505.
D a l t o n T r a n s . , 2 0 0 3 , 1 5 6 7 – 1 5 7 6
1576