Synthesis and X-ray crystal structures of iron(II) and manganese(II) complexes of unsubstituted and benzyl substituted cross-bridged tetraazamacrocycles

Article abstract of DOI:10.1016/S0020-1693(02)01427-5

The Mn²⁺ and Fe²⁺ complexes of the cross-bridged tetraazamacrocyclic ligands, 4,11-dibenzyl-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (1), 4,10-dibenzyl-1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (2), 1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (3), and 1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (4) provide new compounds of these elements for fundamental studies and applications. These unsubstituted and benzyl substituted derivatives were prepared for comparison of their structures and properties with the known catalytically active dimethyl cross-bridged ligand complexes, which are especially notable for their exceptional kinetic stabilities and redox activity. The X-ray crystal structures of five complexes demonstrate that the ligands enforce a distorted octahedral geometry on the metals with two cis sites occupied by labile ligands. The Fe²⁺ complexes of the unsubstitued ligands form μ-oxo dimers upon exposure to air, which have also been structurally characterized. Cyclic voltammetry of the monomeric complexes shows reversible redox processes for the M³⁺/M²⁺ couples, which are sensitive to solvent, ring size, and ring substitution.

Full text of DOI:10.1016/S0020-1693(02)01427-5

Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
www.elsevier.com/locate/ica
Synthesis and X-ray crystal structures of iron(II) and manganese(II)
complexes of unsubstituted and benzyl substituted cross-bridged
tetraazamacrocycles
Timothy J. Hubin a,*, James M. McCormick a,1, Simon R. Collinson a,2,
Nathaniel W. Alcock b, Howard J. Clase b, Daryle H. Busch a
a Chemistry Department, The University of Kansas, Lawrence, KS 66045, USA
b Chemistry Department, The University of Warwick, Coventry CV4 7AL, UK
Received 20 June 2002; accepted 12 August 2002
Abstract
The Mn2ꢀ and Fe2ꢀ complexes of the cross-bridged tetraazamacrocyclic ligands, 4,11-dibenzyl-1,4,8,11-tetraazabicy-
clo[6.6.2]hexadecane (1), 4,10-dibenzyl-1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (2), 1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (3),
and 1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (4) provide new compounds of these elements for fundamental studies and
applications. These unsubstituted and benzyl substituted derivatives were prepared for comparison of their structures and
properties with the known catalytically active dimethyl cross-bridged ligand complexes, which are especially notable for their
exceptional kinetic stabilities and redox activity. The X-ray crystal structures of five complexes demonstrate that the ligands enforce
a distorted octahedral geometry on the metals with two cis sites occupied by labile ligands. The Fe2ꢀ complexes of the unsubstitued
ligands form m-oxo dimers upon exposure to air, which have also been structurally characterized. Cyclic voltammetry of the
monomeric complexes shows reversible redox processes for the M3ꢀ/M2ꢀ couples, which are sensitive to solvent, ring size, and ring
substitution.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Tetraazamacrocycle; Cross-bridged macrocycle; Macrobicycle; Iron complexes; Manganese complexes
1. Introduction
genase [3], methane monooxygenase [4], and lipoxygen-
ase [5] display the power and selectivity found in natural
iron-based systems, while Mn catalase [6], mitochon-
drial superoxide dismutase [7] and Photosystem II [8]
similarly illustrate the effectiveness of manganese deri-
vatives. Manganese and iron complexes of N,N?,N??-
trimethyl-1,4,7-triazacyclononane, Me3[9]aneN3, pro-
vides examples in which some solubility is maintained
and the manganese complex is a useful oxidation
catalyst [9]. However, this and other significant catalysts
having in common nitrogen donors and vacant coordi-
nation sites, tend to form dimers in which higher valent
metal ions are present. Achieving resistance towards
oxidative hydrolysis, but doing so with monomeric
complexes, is the primary goal of the present work.
The principles of modern coordination chemistry [10]
should allow us to design ligands whose complexes are
resistant to oxidative hydrolysis while still having
Functional catalysts based on manganese and iron
complexes of common ligands such as polyamines and
polyethers often have limited utility, especially in aque-
ous media, because of the thermodynamic sink repre-
sented by the mineral forms of these elements [1]. Yet, in
nature, manganese and iron are ubiquitous in redox
related functions. Cytochrome P450 [2], catechol dioxy-
* Corresponding author. Present address: Department of Natural
Science, McPherson College, 1600 E. Euclid, P.O. Box 1402,
McPherson, KS 67460, USA. Tel.: ꢀ
503-212 5746.
E-mail address: hubint@mcpherson.edu (T.J. Hubin).
/
1-620-241 0731x1795; fax: ꢀ1-
/
1
Present address: Division of Science, Truman State, University,
Kirksville, MO 63501, USA.
2
Present address: Department of Inorganic Chemistry, University
of Nottingham, University Park, Nottingham NG7 2RD, UK.
0020-1693/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 0 2 ) 0 1 4 2 7 - 5

T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ
/86
77
available sites for direct binding of the metal ion to
either or both of a terminal oxidant and/or substrate.
We envisioned kinetically stable, mononuclear manga-
nese and iron complexes that were not coordinatively
saturated by their stabilizing ligand and were capable of
supporting higher valent metal ions as well. Specific
organic molecules have been identified that should
possess the sought-after ligand properties, ethylene
cross-bridged tetraazamacrocycles (Fig. 1, Structures
new complexes and compare them to the methyl-
substituted parent ligand complexes.
2. Experimental
N,N-Bis(aminopropyl)ethylenediamine (98%) was
purchased from Acros Organics. Glyoxal (40 wt.% in
water), benzyl bromide (99%), and sodium borohydride
(98%) were purchased from Aldrich Chemical Co.
Cyclen was generously supplied by Dow Chemical Co.
All solvents were of reagent grade and were dried, when
necessary, by accepted procedures [17].
Elemental analyses were performed by the Analytical
Service of the University of Kansas or Desert Analytics.
Mass spectra were measured by the Analytical Service of
the University of Kansas on a VG ZAB HS spectro-
meter equipped with a xenon gun. Matrices used include
1ꢁ6) [11]. These tetradentate ligands are as topologically
/
constrained as the macrobicyclic cryptates, and there-
fore their complexes should have the kinetic and
thermodynamic stabilities associated with this con-
strained topology. But, in contrast to the prototypical
hexadentate cryptate, these ligands should leave labile
coordination sites on the metal ion for further reaction.
Although previously synthesized in other laboratories
[11], these molecules have not been fully exploited as
transition metal ligands. We suspected that the great
rigidity of these short cross-bridged ligands should
impart enormous kinetic stability to their complexes
[10], and we have recently shown this to be the case for
copper(II) [12] and manganese(II) [13]. This kinetic
stability confers great promise in such applications as
homogeneous catalysis, where complex stability is a
problem.
We have described elsewhere [13] the synthesis and
characterization of Fe2ꢀ and Mn2ꢀ complexes of the
methyl-substituted ligands (5 and 6, Fig. 1), and their
one-electron oxidized analogues [14]. These compounds
have exciting properties such as remarkable stabilities in
harsh aqueous conditions and easily accessible higher
oxidation states that suggest useful applications as
aqueous oxidation catalysts and making them of sig-
nificant technological interest [15]. Below, we describe
the synthesis, X-ray crystal structures and electrochemi-
cal behavior of similar complexes where the ligands have
been slightly altered to probe the effect on the metal
center. Ligands having benzyl groups or simple hydro-
gens (unsubstituted ligands) as the two nitrogen sub-
stituents [16] were combined with the metal ions of
interest. We here present the characterization of these
NBA (nitrobenzyl alcohol) and TG/G (thioglycerolꢁ
/
glycerol). NMR spectra were obtained on a Bruker
AM-500 spectrometer.
2.1. Synthesis
The
ligands
4,11-dibenzyl-1,4,8,11-tetraazabicy-
clo[6.6.2]hexadecane (1), and 1,4,8,11-tetraazabicy-
clo[6.6.2]hexadecane (3), were prepared according to
literature procedures [16] from cyclam, which was made
from N,Nꢂ
modified literature procedure [18]. 4,10-Dibenzyl-
1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (2), and
/
-bis(aminopropyl)ethylenediamine using a
1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (4), were pre-
pared in an analogous manner from cyclen following the
Weisman synthesis [16].
2.1.1. Mn(1)(CF3SO3)2
In an inert atmosphere glovebox, manganese(II)
trifluoromethanesulfonate (approximately 1 mmol) was
prepared in situ from manganese metal and trifluor-
omethanesulfonic acid in MeCN, according to a litera-
ture procedure [19], and any excess acid was neutralized
by the addition of 1 ml of Et3N. The ligand 1 (0.406 g, 1
mmol) was dissolved in 6 ml of dry Et3N in a 25 ml
Erlenmeyer flask. The Mn2ꢀ solution was added to the
stirring ligand solution causing an immediate precipita-
tion of a white powder. The reaction was allowed to stir
at 50 8C for 5 more hours, before the precipitate was
collected by filtration and washed sparingly with
MeCN. Drying in vacuo yielded 0.546 g (72%) of the
analytically
pure
product.
Anal.
Calc.
for
C28H38F6MnN4O6S2: C, 44.27; H, 5.04; N, 7.38. Found:
C, 44.50; H, 5.15; N, 7.48%. The FABꢀ mass spectrum
in MeCN (NBA matrix) exhibited peaks at m/zꢂ610
/
(Mn(1)(CF3SO3)ꢀ) and 759 (Mn(1)(CF3SO3)2ꢀ). Crys-
tals suitable for X-ray diffraction were grown from
warm Py.
Fig. 1. Cross-bridged ligands [11,15] used in this work.

78
T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
2.1.2. MLCl2 (Mꢂ
/
Mn2ꢀ, Fe2ꢀ; Lꢂ
The ligand (1 mmol) was dissolved in 10 ml of dry
Et3N in a 25 ml Erlenmeyer flask in an inert atmosphere
/
1, 2)
Yield:
51ꢁ
/
77%.
Fe(4)Cl2:
Anal.
Calc.
for
C10H22Cl2FeN4: C, 36.95; H, 6.82; N, 17.24. Found:
C, 37.17; H, 6.60; N, 17.13%. Mn(4)Cl2: Anal. Calc. for
C10H22Cl2MnN4: C, 37.05; H, 6.84; N, 17.28. Found: C,
36.94; H, 6.74; N, 17.05%. FABꢀ mass spectra in water
(NBA matrix) exhibited peaks at m/zꢂ
MLClꢀ and
MLCl2 for both complexes.
glovebox. A suspension of 1 mmol of M(py)2Cl2 (Mꢂ
/
Fe2ꢀ, Mn2ꢀ), synthesized by literature procedures [20],
in 10 ml of MeCN, was added to the stirring ligand
solution. The reaction mixture was stirred for 5 h at
50 8C, during which time the metal salt dissolved and
pale brown (Fe2ꢀ) or white (Mn2ꢀ) finely divided solids
precipitated. After cooling, the reaction mixtures were
filtered and the solid products were washed sparingly
/
ꢀ
2.1.5. [Cl(3)Fe(m-O)Fe(3)Cl]Cl2×
/
2H2O
The DMF filtrate from the preparation of Fe(3)Cl2
(see above) was removed from the glovebox. Water (3
ml) was added to this solution and air was bubbled
with MeCN and dried in vacuo. Yield: 58ꢁ
Fe(1)Cl2×0.5H2O: Anal. Calc. for C26H39Cl2FeN4O0.5
C, 57.58; H, 7.25; N, 10.33. Found: C, 57.70; H, 7.20; N,
10.18%. 0.5H2O: Anal. Calc. for
Mn(1)Cl2×
C26H39Cl2MnN4O0.5: C, 57.67; H, 7.26; N, 10.35.
Found: C, 57.86; H, 7.38; N, 10.28%. Fe(2)Cl2×
/86%.
/
:
through it overnight. A deep redꢁbrown solution
/
resulted. The solvent was then removed and the brown
residue washed with ether (100 ml) to remove any
remaining DMF. The remaining brown powder was
dried in vacuo. Yield: 801 mg (48% based on ligand).
Anal. Calc. for C24H56Cl4Fe2N8O3: C, 38.02; H, 7.44; N,
14.78. Found: C, 38.06; H, 7.09; N, 14.62%. The FABꢀ
mass spectrum in MeOH (NBA) exhibited peaks at m/
/
/
0.5H2O: Anal. Calc. for C24H35FeCl2N4O0.5: C, 56.11;
H, 6.86; N, 10.91. Found: C, 56.00; H, 6.70; N, 10.71%.
Mn(2)Cl2: Anal. Calc. for C24H34Cl2MnN4: C, 57.15; H,
6.79; N, 11.11. Found: C, 56.99; H, 6.54; N, 10.91%.
zꢂ
317 (Fe(3)Clꢀ), 352 (Fe(3)Cl2ꢀ), and 650
/
FABꢀ mass spectra in MeOH (NBA matrix) exhibited
(Cl(3)Fe(m-O)Fe(3)Clꢀ). X-ray quality crystals were
grown by the slow evaporation of an isopropanol
solution and contained three waters of crystallization.
ꢀ
peaks at m/zꢂ
MLClꢀ and MLCl2 for all complexes.
/
X-ray quality crystals of Fe(2)Cl2 and Mn(2)Cl2 were
grown by the slow evaporation of 3:1 MeOHꢁPy
solutions.
/
2.1.6. [Cl(4)Fe(m-O)Fe(4)Cl][PF6]2
The DMF filtrate from the preparation of Fe(4)Cl2
(see above) was removed from the glovebox. Water (3
ml) was added to this solution and air was bubbled
2.1.3. M(3)Cl2 (Mꢂ )
Mn2ꢀ, Fe2ꢀ
/
The ligand 3 (1.00 g, 4.42 mmol) was dissolved in 40
ml of dry DMF in a 50 ml Erlenmeyer flask in an inert
atmosphere glovebox. Anhydrous MCl2 (4.42 mmol,
through it overnight. A deep redꢁbrown solution
/
resulted. The solvent was then removed and the brown
residue washed with ether (100 ml) to remove any
remaining DMF. The remaining brown powder was
dissolved in 30 ml of dry MeOH. One gram of
ammonium hexafluorophosphate in 15 ml of dry
MeOH was added to the above solution causing a
brown precipitate to form. The solid was collected by
filtration, washed with MeOH and ether and vacuum
dried. Yield: 991 mg (43%). Anal. Calc. for
C20H44Cl2F12Fe2N8OP2: C, 27.14; H, 5.01; N, 12.66.
Found: C, 27.29; H, 4.81; N, 12.38%. The FABꢀ mass
Mꢂ
Fe2ꢀ, Mn2ꢀ) was added to the stirring ligand
/
solution. The reaction was stirred for 16 h at room
temperature (r.t.), during which time the metal salt
dissolved and pale brown (Fe2ꢀ) or white (Mn2ꢀ) finely
divided solids precipitated. The cooled reaction mixtures
were then filtered and the solid products were washed
with ether and dried in vacuo. Yield: 46ꢁ
/
63%. Fe(3)Cl2×
/
0.5DMF: Anal. Calc. for C13.5H29.5Cl2FeN4.5O0.5: C,
41.61; H, 7.63; N, 16.18. Found: C, 41.85; H, 7.28; N,
16.10%. Mn(3)Cl2: Anal. Calc. for C12H26Cl2MnN4: C,
40.92; H, 7.44; N, 15.91. Found: C, 41.08; H, 7.13; N,
spectrum in DMF (NBA) exhibited a peak at m/zꢂ593
/
15.96%. FABꢀ mass spectra in water (NBA matrix)
(Cl(4)Fe(m-O)Fe(4)Clꢀ). X-ray quality crystals of the
crude dichloride salt prior to ammonium hexafluoro-
phosphate addition, were grown by the slow evapora-
tion of a DMF solution and contained two waters of
crystallization.
MLClꢀ and MLCl2 for both
ꢀ
exhibited peaks at m/zꢂ
/
complexes.
2.1.4. M(4)Cl2 (Mꢂ )
Mn2ꢀ, Fe2ꢀ
The ligand 4 (1.00 g, 5 mmol) was dissolved in 40 ml
of dry DMF in a 50 ml Erlenmeyer flask in an inert
/
2.2. Physical methods
atmosphere glovebox. Anhydrous MCl2 (5 mmol, Mꢂ
/
Fe2ꢀ, Mn2ꢀ) was added to the stirring ligand solution.
The reaction was stirred for 16 h at r.t., during which
time the metal salt dissolved. Pale pink (Fe2ꢀ) or white
(Mn2ꢀ) finely divided solids precipitated. The cooled
reaction mixtures were then filtered and the solid
product was washed with ether and dried in vacuo.
Electrochemical experiments were performed on a
Princeton Applied Research Model 175 programmer
and Model 173 potentiostat in dry CH3CN using a
homemade cell in an inert atmosphere dry box under
N2. Either a button Pt or a button glassy carbon
electrode was used as the working electrode in conjunc-

T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ
/86
79
tion with a Pt-wire counter electrode and a Ag-wire
pseudo-reference electrode. Tetrabutylammonium hexa-
fluorophosphate (0.1 M) was the supporting electrolyte
in all cases. The measured potentials were referenced to
3. Results and discussion
3.1. Preparation of metal complexes
SHE using ferrocene (ꢀ
standard.
/
0.400 V vs. SHE) as an internal
As noted in a previous communication [12], only
Cu2ꢀ and Ni2ꢀ, the thermodynamically strongest
binding (first row, divalent) transition metal ions [25],
had been complexed to ethylene cross-bridged tetraaza-
macrocycles prior to our work [16]. Our early attempts
at the complexation of 5 and 6 with Fe2ꢀ and Mn2ꢀ
salts in protic solvents or in aprotic solvents using
hydrated metal salts were entirely unsuccessful. We
therefore soon concluded that Cu2ꢀ and Ni2ꢀ are
much better at competing with protons for the binding
cavity in protic solvents than are other divalent transi-
tion metal ions. This is especially true of the ions of
primary interest here, the relatively hard oxophilic ions
of manganese and iron. We attribute this behavior to the
proton-sponge nature of the ligands [11], which has been
confirmed by titration. We found that in an aqueous
medium the diprotonated form of free ligand 5 behaves
as a monoprotic weak acid with a pKa1 of 9.58(3);
stoichiometrically, two Hꢀ are bound, demonstrating
that pKa2 must be substantially greater than 13 [13]. The
conclusion is clear, that these ligands bind at least one
proton very strongly indeed, and that only the most
strongly binding metal ions can compete with that
proton for the cavity it occupies.
The obstacle, once identified, was overcome by
minimizing the activity of protons in the reaction
system. The complexes of a broad array of transition
metal ions have been prepared with these ligands
through the use of anhydrous metal reagents with
strictly deprotonated ligands in rigorously dry, aprotic
solvents under a dry, inert atmosphere [12]. The Fe2ꢀ
and Mn2ꢀ complexes with parent ligands 5 and 6 were
prepared from the corresponding anhydrous M(py)2Cl2
starting material in CH3CN. Unfortunately, complexa-
tion of 1 and 2 via the same reaction was not successful.
Possible explanations include decreased solubility of the
ligands in acetonitrile, increased steric bulk (the benzyl
groups) near the ligand cavity, and/or increased ligand
basicity (not measured). Several sets of reaction condi-
tions were investigated until success was achieved in 1:1
2.3. Crystal structure analysis
X-ray data were collected with a Siemens SMART
three-circle system with CCD area detector using
graphite monochromated Mo Ka radiation (lꢂ
/
˚
0.71073 A) [21]. The crystals were held at the specified
temperature with the Oxford Cryosystem Cooler [22].
Absorption corrections were applied by the C-scan
method, and none of the crystals showed any decay
during data collection. Complete crystal data are in
Table 1. The structures were solved by direct methods
using SHELXS [23] (TREF) with additional light atoms
found by Fourier methods. Hydrogen atoms were added
at calculated positions and refined using a riding model.
Anisotropic displacement parameters were used for all
non-H atoms, while H-atoms were given isotropic
displacement parameters equal to 1.2 (or 1.5 for methyl
hydrogen atoms) times the equivalent isotropic displace-
ment parameter for the atom to which the H-atom is
attached. Refinement used SHELXL-96 [24]. Selected
bond lengths and angles for the five complexes are in
Table 2.
2.3.1. Specific structures
Mn(1)(CF3SO3)2: crystal character: yellow diamond
plates. Systematic absences indicated space group Pbcn.
Mn(2)Cl2: crystal character: colorless squares. Systema-
tic absences indicated space group Cc or C2/c; the latter
was chosen and shown to be correct by satisfactory
refinement. Fe(2)Cl2: crystal character: pale brown
plates. Systematic absences indicated space group Cc
or C2/c; the latter was chosen and shown to be correct
by satisfactory refinement. [Cl(3)Fe(m-O)Fe(3)Cl]Cl2×
/
3H2O: crystal character: thin brown plates. Systematic
absences indicated space group P21/c. Three lattice
water molecules were located. Although one ligand is
very precisely located, it appears that the other (attached
to Fe(1)) is partly occupied by the opposite diastereo-
mer. The resulting disorder is visible in the elongated
ellipsoids of the ligand atoms, corresponding to the
superposition of the two species, and required the
inclusion of two additional low occupancy sites
triethylamineꢁacetonitrile solution, from either triflate
/
and dichlorobis-pyridine metal starting materials.
Triethylamine is generally used as a mild base in
complexation reactions, but was found to be a good
solvent for the ligands, which are not readily soluble in
many of the aprotic solvents otherwise necessary for
complexation of these ligands. Complexation of the
unsubstituted ligands 3 and 4 proved less challenging,
and followed protocols [13] established with ligands 1
and 2, except for the substitution of DMF as the solvent.
The air oxidation of the complexes with 3 and 4 have
been studied. Experimentally, the Fe2ꢀ complexes of
(C(12a) and C(17a)). [Cl(4)Fe(m-O)Fe(4)Cl]Cl2×
/2H2O:
crystal character: deep red thin plates. Systematic
absences indicated space group P21/c. Two lattice water
molecules were located.

80
T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
Table 1
Crystal data and structural refinement details
Mn(1)(CF3SO3)2
Mn(2)Cl2
Fe(2)Cl2
[Cl(3)Fe(m-
O)Fe(3)Cl]Cl2×
[Cl(4)Fe(m-O)Fe(4)Cl]Cl2×
/
/3H2O
2H2O
Empirical
formula
C28H38F6MnN4O6S2
759.68
C24H34Cl2MnN4
C24H34Cl2MnN4
C24H58Cl4Fe2N8O4
C20H44Cl4Fe2N8O3
Formula weight
Temperature (K) 180(2)
504.39
293(2)
505.30
180(2)
776.28
180(2)
698.13
180(2)
Crystal system
Space group
orthorhombic
Pbcn
monoclinic
C2/c
monoclinic
C2/c
monoclinic
P21/c
monoclinic
P21/c
˚
a (A)
16.004(2)
9.71120(10)
21.3157(2)
90
8.97080(10)
13.3513(2)
20.02770(10)
91.4400
2398.00(5)
4
8.9453(4)
13.3191(6)
19.9448(9)
91.7100(10)
2375.2(2)
4
14.3930(10)
22.6960(10)
10.3900(10)
92.971(5)
3389.5(4)
8
10.6863(3)
17.6254(2)
16.4448(2)
108.6220(10)
2935.22(10)
4
˚
b (A)
˚
c (A)
b (8)
3
V (A )
˚
3312.76(5)
4
1.523
Z
rcalc (g cmꢃ3
)
1.397
7.93
1.413
8.79
1.308
10.46
1.580
13.91
Absolute coeffi-
cient (cmꢃ1
6.05
)
Crystal size (mm) 0.80ꢄ
/
0.45ꢄ
/
0.20
0.39ꢄ
28.52
115
175
255
/
0.36ꢄ
/
0.18
0.30ꢄ
28.56
115
175
235
/
0.26ꢄ
/
0.10
0.55ꢄ
25.00
125
275
135
/
0.10ꢄ
/
0.02
0.2ꢄ
24.00
145
235
215
/
0.2ꢄ0.04
/
Maximum u (8)
Index ranges
28.57
205
12 5
275l 5
ꢃ
/
/
h 5
k 5
23
/
21,
ꢃ
/
/
h 5
k 5
l 526
/
5,
ꢃ
/
/
h 5
k 5
l 525
/
11,
ꢃ
/
/
h 5
k 5
l 513
/
18,
ꢃ
/
/
h 5
k 5
l 519
/
10,
ꢃ
/
/
/9,
ꢃ
/
/
/
16,
ꢃ
/
/
/17,
ꢃ
/
/
/
29,
ꢃ
/
/
/
23,
ꢃ/
/
/
ꢃ/
/
/
ꢃ/
/
/
ꢃ/
/
/
ꢃ/
/
/
Reflections
collected
18 842
7040
2766
2260
7022
2797
2168
16 722
13 227
Independent
reflections
Observed
3971
5918
4607
3117
3804
2848
a
reflections
Refinement
method
full-matrix on F2
full-matrix on F2
full-matrix on F2
full-matrix on F2
full-matrix on F2
Data/restraints/
parameters
3971/13/251
2766/0/141
2797/0/141
5918/9/419
4607/0/334
Final R indices
R1ꢂ
/
0.0471
0.1175
R1ꢂ
/
0.0294
R1ꢂ
/
0.0339
0.0820
R1ꢂ
/
0.0608
R1ꢂ
/
0.0619
[I ꢀ2s(I)]
/
R indices
(all data)
wR2ꢂ
/
wR2ꢂ
/0.0756
wR2ꢂ
/
wR2ꢂ
/0.1471
wR2ꢂ
/
0.1290
b,c
Weight para-
meters a, b
Goodness-of-fit
on F2
0.0430, 3.2000
1.086
0.0401, 0.2776
1.035
0.0431, 0.0000
0.985
0.0560, 10.5000
0.997
0.0370, 8.2000
1.014
Largest peak/
ꢃ3
0.504 and ꢃ
/
0.436
0.234 and ꢃ
/
0.304
0.313 and ꢃ
/
0.328
0.663 and ꢃ
/
0.555
0.847 and ꢃ0.873
/
˚
hole (e A
)
both of these ligands immediately turn from pale yellow
to deep redꢁbrown in solution when exposed to air. This
have been structurally characterized (vide infra), show-
ing that these ligand complexes do indeed form m-oxo
Fe3ꢀ dimers [27] on oxidation by air.
/
behavior does not occur with the Fe2ꢀ complexes of
ligands 1, 2, 5, or 6; their solutions only become
somewhat darker yellow in color under the same
conditions. One possible reaction for 3 and 4 is ligand
oxidation, as secondary amines are notoriously easy to
oxidize to imines in transition metal complexes [26].
Another possibility is metal ion oxidation and formation
of oxide or hydroxide bridged dimers. Cache† molecu-
lar modeling studies indicated that dimerization of the
latter complexes should be sterically inhibited by the N-
alkyl groups, which would need to occupy the same
space as the second ligand in a typical m-oxo dimer.
However, no such steric problems exist for the unsub-
stituted ligands 3 and 4. The products of these reactions
The manganese(II) complexes with the methyl-sub-
stituted ligand 5 has been extensively studied under
conditions favorable for dimerization, and no such
reaction has yet been observed. In contrast, the Mn2ꢀ
complex of 4 immediately turns from colorless to deep
green upon exposure of a DMF solution to air. In
addition to its telling color, the solution has the
characteristic 16-line EPR signal, associated with a
Mn3ꢀ/Mn4ꢀ oxo-bridged dimer [28]. Suprisingly, a
solution of the manganese(II) complex of 3 turns a
pale red color under the same conditions and lacks the
16-line EPR. Obviously, a different reaction is taking
place, although neither of the products from the air

T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ
/86
81
Table 2
˚
Selected bond lengths (A) and angles (8)
For Mn(1)(CF3SO3)2
Bond lengths
Mn(1)ꢀ
Mn(1)ꢀ
Mn(1)ꢀ
S(1)ꢀ
/
O(1)#1
2.153(2)
2.266(2)
2.295(2)
1.424(2)
S(1)ꢀ
S(1)ꢀ
S(1)ꢀ
/
O(3)
1.428(2)
1.452(2)
1.852(12)
/N(4)
/O(1)
/N(1)
/C16Aa
/
O(2)
Bond angles
O(1)#1ꢀMn(1)ꢀ
O(1)ꢀMn(1)ꢀ
O(1)ꢀMn(1)ꢀ
N(4)ꢀMn(1)ꢀ
O(1)ꢀMn(1)ꢀ
/
/
O(1)
94.93(9)
93.11(9)
170.40(9)
79.40(14)
92.71(7)
N(4)ꢀ
O(1)ꢀ
N(4)ꢀ
N(1)#1ꢀ
/
Mn(1)ꢀ
Mn(1)ꢀ
Mn(1)ꢀ
Mn(1)ꢀ
/
N(1)#1
N(1)
N(1)
87.58(8)
97.26(7)
81.06(9)
/
/
N(4)
/
/
/
/
N(4)#1
/
/
/
/
N(4)#1
/
/N(1)
165.25(10)
/
/
N(1)#1
For Mn(2)Cl2
Bond lengths
Mn(1)ꢀ
/
N(4)
2.3051(13)
2.3311(12)
Mn(1)ꢀ/Cl(1)
2.4276(4)
Mn(1)ꢀ
/
N(1)
Bond angles
N(4)ꢀMn(1)ꢀ
N(4)ꢀMn(1)ꢀ
N(4)#1ꢀ
N(1)#1ꢀ
N(4)ꢀMn(1)ꢀ
/
/
N(4)#1
74.56(8)
77.69(4)
74.06(4)
144.30(6)
94.13(4)
N(4)#1ꢀ
N(1)#1ꢀ
N(1)ꢀMn(1)ꢀ
Cl(1)ꢀMn(1)ꢀ
/
Mn(1)ꢀ
Mn(1)ꢀ
/
Cl(1)
168.60(4)
102.55(3)
100.84(3)
97.21(3)
/
/N(1)#1
/
/Cl(1)
/
Mn(1)ꢀ
Mn(1)ꢀ
Cl(1)
/N(1)#1
/
/Cl(1)
/
/N(1)
/
/
Cl(1)#1
/
/
For Fe(2)Cl2
Bond lengths
Fe(1)ꢀ
/
N(4)
2.234(2)
2.2757(14)
Fe(1)ꢀ
/
Cl(1)
2.4004(5)
Fe(1)ꢀ
/
N(1)
Bond angles
N(4)#1ꢀFe(1)ꢀ
N(4)ꢀFe(1)ꢀ
N(4)ꢀFe(1)ꢀ
N(1)#1ꢀ
N(4)ꢀFe(1)ꢀ
/
/
N(4)
76.92(9)
79.21(5)
75.63(6)
147.71(8)
94.35(5)
N(1)ꢀ
N(4)ꢀ
N(1)ꢀ
Cl(1)ꢀ
/
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
/
Cl(1)
99.18(4)
171.26(5)
102.61(4)
94.39(3)
/
/
N(1)#1
/
/Cl(1)#1
/
/
N(1)
/
/Cl(1)#1
/
Fe(1)ꢀ
/N(1)
/
/Cl(1)#1
/
/
Cl(1)
For [Cl(3)Fe(m-O)Fe(3)Cl]Cl2×
/3H2O
Bond lengths
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
/
O(1)
1.831(4)
2.145(5)
2.157(5)
2.180(5)
2.283(6)
2.368(2)
O(1)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
/
Fe(2)
1.820(4)
2.157(5)
2.152(5)
2.207(5)
2.287(5)
2.350(2)
/
N(111)
N(14)
N(11)
N(18)
Cl(1)
/
N(211)
N(24)
N(28)
N(21)
Cl(2)
/
/
/
/
/
/
/
/
Bond angles
O(1)ꢀFe(1)ꢀ
O(1)ꢀFe(1)ꢀ
N(111)ꢀ
O(1)ꢀFe(1)ꢀ
N(111)ꢀFe(1)ꢀ
N(14)ꢀFe(1)ꢀN(11)
O(1)ꢀFe(1)ꢀN(18)
N(111)ꢀFe(1)ꢀN(18)
N(14)ꢀFe(1)ꢀ
N(11)ꢀFe(1)ꢀ
O(1)ꢀFe(1)ꢀ
N(111)ꢀFe(1)ꢀ
N(14)ꢀFe(1)ꢀ
N(11)ꢀFe(1)ꢀ
N(18)ꢀFe(1)ꢀ
Fe(2)ꢀO(1)ꢀ
/
/
N(111)
95.2(2)
98.6(2)
O(1)ꢀ
O(1)ꢀ
N(211)ꢀ
O(1)ꢀFe(2)ꢀ
N(211)ꢀFe(2)ꢀ
N(24)ꢀFe(2)ꢀN(28)
O(1)ꢀFe(2)ꢀN(21)
N(211)ꢀFe(2)ꢀN(21)
N(24)ꢀFe(2)ꢀ
N(28)ꢀFe(2)ꢀ
O(1)ꢀFe(2)ꢀ
N(211)ꢀFe(2)ꢀ
N(24)ꢀFe(2)ꢀ
N(28)ꢀFe(2)ꢀ
N(21)ꢀFe(2)ꢀ
/
Fe(2)ꢀ
/
N(211)
94.9(2)
101.5(2)
159.6(2)
94.9(2)
/
/N(14)
/
Fe(2)ꢀ
/N(24)
/
Fe(1)ꢀ
N(11)
N(11)
/
N(14)
163.6(2)
95.4(2)
88.3(2)
/
Fe(2)ꢀ
N(28)
N(28)
/N(24)
/
/
/
/
/
/
/
/
81.6(2)
/
/
81.8(2)
/
/
85.1(2)
/
/
171.8(2)
79.3(2)
/
/
173.9(2)
82.7(2)
/
/
/
/
/
/
N(18)
N(18)
85.9(2)
78.5(2)
/
/N(21)
79.7(2)
79.2(2)
/
/
/
/N(21)
/
/Cl(1)
98.50(12)
93.96(14)
92.6(2)
/
/
Cl(2)
98.42(13)
97.56(14)
91.83(13)
166.63(13)
87.42(13)
/
/
Cl(1)
/
/Cl(2)
/
/
Cl(1)
Cl(1)
Cl(1)
/
/
Cl(2)
Cl(2)
Cl(2)
/
/
165.7(2)
88.03(14)
144.1(2)
/
/
/
/
/
/
/
/Fe(1)

82
T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
Table 2 (Continued)
For [Cl(4)Fe(m-O)Fe(4)Cl]Cl2×2H2O
/
Bond lengths
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
Fe(1)ꢀ
/
O(1)
1.818(4)
2.149(6)
2.160(5)
2.173(5)
2.257(5)
2.349(2)
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
Fe(2)ꢀ
/
O(1)
1.807(4)
2.174(5)
2.188(5)
2.198(5)
2.266(5)
2.354(2)
/
N(14)
N(110)
N(11)
N(17)
Cl(11)
/
N(210)
N(24)
N(21)
N(27)
Cl(21)
/
/
/
/
/
/
/
/
Bond angles
O(1)ꢀFe(1)ꢀ
O(1)ꢀFe(1)ꢀ
N(14)ꢀ
O(1)ꢀFe(1)ꢀ
N(14)ꢀFe(1)ꢀ
N(110)ꢀFe(1)ꢀ
O(1)ꢀFe(1)ꢀN(17)
N(14)ꢀFe(1)ꢀN(17)
N(110)ꢀFe(1)ꢀN(17)
N(11)ꢀFe(1)ꢀN(17)
O(1)ꢀFe(1)ꢀCl(11)
N(14)ꢀFe(1)ꢀCl(11)
N(110)ꢀFe(1)ꢀCl(11)
N(11)ꢀFe(1)ꢀ
N(17)ꢀFe(1)ꢀ
O(1)ꢀFe(2)ꢀ
/
/
N(14)
101.2(2)
103.1(2)
148.3(2)
96.3(2)
O(1)ꢀ
N(210)ꢀ
O(1)ꢀFe(2)ꢀ
N(210)ꢀFe(2)ꢀ
N(24)ꢀFe(2)ꢀN(21)
O(1)ꢀFe(2)ꢀN(27)
N(210)ꢀFe(2)ꢀN(27)
N(24)ꢀFe(2)ꢀ
N(21)ꢀFe(2)ꢀ
O(1)ꢀFe(2)ꢀ
N(210)ꢀFe(2)ꢀ
N(24)ꢀFe(2)ꢀ
N(21)ꢀFe(2)ꢀ
N(27)ꢀFe(2)ꢀ
Fe(2)ꢀO(1)ꢀ
/
Fe(2)ꢀ
Fe(2)ꢀ
N(21)
N(21)
/
N(24)
103.2(2)
146.8(2)
94.9(2)
/
/N(110)
/
/N(24)
/
Fe(1)ꢀ
N(11)
N(11)
N(11)
/
N(110)
/
/
/
/
/
/
79.9(2)
/
/
78.7(2)
/
/
76.1(2)
/
/
78.7(2)
174.0(2)
77.2(2)
/
/
172.1(2)
74.6(2)
78.1(2)
/
/
/
/
/
/
/
/N(27)
/
/
76.4(2)
/
/N(27)
77.7(2)
/
/
77.7(2)
98.02(14)
99.1(2)
/
/Cl(21)
97.82(14)
98.6(2)
/
/
/
/Cl(21)
/
/
/
/
Cl(21)
Cl(21)
Cl(21)
99.83(14)
167.23(14)
89.62(14)
178.9(3)
/
/
97.3(2)
/
/
/
/
Cl(11)
Cl(11)
165.6(2)
87.95(14)
101.4(2)
/
/
/
/
/
/Fe(1)
/
/N(210)
oxidation of these Mn2ꢀ precursors has been satisfac-
torily characterized. The cause of this difference in
behavior is unlikely to be steric, since the iron complex
of this ligand can dimerize on air oxidation.
14-membered ring in 1 engulfs the metal ion more fully
than does the 12-membered ring in 2. The NaxꢀMꢀNax
bond angle for Mn(1)(CF3SO3)2 is 165.25(10)8, while for
Mn(2)Cl2, it is only 144.30(6)8. Likewise, the NeqꢀMꢀ
/
/
/
/
Neq angles are 79.40(14) and 74.56(8)8 for
Mn(1)(CF3SO3)2 and Mn(2)Cl2, respectively. The smal-
ler angles for the smaller macrobicycle demostrate how
the metal ion is less engulfed in the ligand cavity.
Surprisingly, the triflate complex of 1 contains a
Mn2ꢀ ion deeper in the ligand cleft than in the related
dichloro complex of 5 [13]. In this structure, Mn(5)Cl2,
3.1.1. Crystal structures
X-ray crystal structures for three monomeric com-
plexes all with dibenzyl ligands, are shown in Fig. 2.
These structures confirm the expected geometric proper-
ties of the cross-bridged ligands, which are constrained
by the short cross-bridge to folded conformations. In all
three cases, the macrobicycle occupies two axial and two
cis equatorial sites of distorted octahedra, while the
remaining cis equatorial sites are occupied by either two
triflate ligands or two chloride ligands. In the chemistry
of the unbridged parent macrocycles, with few excep-
tions [27], 1,4,8,11-tetraazacycotetradecane, or cyclam,
binds metal ions in a square planar fashion [28],
occupying all four equatorial sites of octahedral com-
plexes, which locates the two labile ligand binding sites
at the trans, axial positions. Cyclen is too small to fit
completely around the metal ion, and generally folds to
occupy two cis equatorial and the two axial sites of
octahedral metal ions [29], a strict parallel to the
coordination reported here for 1 and 2. The short
cross-bridge of 1 forces the cyclam ring system to behave
more like cyclen. The consequences that two cis labile
sites have on the reactivity of these complexes are
currently being explored [13,15].
the Naxꢀ
/
Mꢀ
/
Nax and Neqꢀ
/
Mꢀ/Neq bond angles are
158.0(2) and 75.6(2)8, respectively, several degrees
smaller than in Mn(1)(CF3SO3)2. It might be expected
that the more bulky triflate ligands in the complex of 1
would draw the Mn2ꢀ ion further out of the ligand cleft
than the chloride atoms in the complex of 5. Since the
benzyl groups are extended axially, it appears that they
leave sufficient equatorial space to accommodate the
two triflate ligands without costly steric interactions. It
then becomes logical that the smaller triflate oxygen
atoms bound to Mn2ꢀ allow this ion to move deeper
into the ligand cavity than the larger chloro ligands.
Electronic factors might also contribute; the poorly
complementary triflate ligands do not add enough
electron density, so the metal ion has to delve deeper
into the tetradentate ligand cavity. These postulated
steric effects also explain the larger Clꢀ
/
Mnꢀ
/
Cl angle
Mnꢀ
(98.85(6)8) in Mn(5)Cl2, versus that of the Oꢀ
/
/
O
A more detailed inspection of the three structures
reveals several interesting trends. First, the size of the
ring system dictates the distortion of the octahedra. The
angle (94.93(9)8) in Mn(1)(CF3SO3)2.
A second observation is that the smaller Fe2ꢀ ion is
more fully engulfed by macrobicycle 2 than is the larger

T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ
/86
83
Fig. 3. The crystal packing diagram of Fe(2)Cl2 showing the pꢁp
/
stacking interactions between benzyl groups of neighboring molecules.
benzyl groups extending from the macrobicyclic com-
plexes make classical pꢁp interactions with benzyl
/
groups from neighboring molecules, as demonstrated
in the crystal packing diagram of Fe(2)Cl2 (Fig. 3). The
two benzyl groups are in the slipped arrangement where
one aromatic carbon is directly above the centroid of its
partners aromatic ring [30]. The paired benzyl groups
˚
are 3.32 A apart, also a normal distance for this type of
Fig. 2. Views of the crystal structures of: (a) Mn(1)(CF3SO3)2; (b)
Mn(2)Cl2; and (c) Fe(2)Cl2.
interaction [30]. The strengths of these interactions in
the solid state may contribute to the general low
solubilities of the complexes of ligands 1 and 2.
Mn2ꢀ ion. Naxꢀ
144.30(6)8 for Mn(2)Cl2. The Neqꢀ
76.92(9) and 74.56(8)8 for Fe(2)Cl2 and Mn(2)Cl2,
respectively. In these two complexes, the average Feꢀ
/
Mꢀ
/
Nax is 147.71(8)8 for Fe(2)Cl2,
The crystal structures of the dimeric iron(III) com-
plexes of ligands 3 and 4 are represented in Fig. 4. The
Fe3ꢀ ions of these complexes are again found in pseudo-
octahedral, six-coordinate geometries, similar to most
monomeric complexes of the methyl- and benzyl sub-
stituted ligands [13,14]. Commonly, such dimers have a
coordination number of 5, although six- and seven-
coordinate dimers are known [27]. It is interesting that
the monodentate chloro ligands are maintained in the
present structures, but since the macrobicyclic ligands
are uncharged, the only Coulombic forces influencing
the monoanionic halide are attractive. Also, the folded
ligand conformations help separate the ligands from
each other, easing the steric interactions that might
/
MꢀNeq angles are
/
/
˚
N distance is 2.255 A, while the average Mnꢀ
/
N distance
2ꢀ
˚
is 2.318 A, confirming the size relationship of the Fe
and Mn2ꢀ cations. These effects of macrocycle ring size
and metal ionic radius, influence how deep the metal ion
resides within the ligand cavity and determine how close
the coordination geometry is to a true octahedron; they
have been consistently observed with several different
metal ions and ligand ring sizes [13,14].
The intermolecular interactions of the two ligand 2
complexes in the solid state should also be noted. The

84
T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
˚
˚
type [27], averaging 1.83 A for the dimer of 1 and 1.81 A
for the dimer of 2.
Perhaps most interesting is the degree of linearity in
the Feꢀ
angle is almost completely linear, with a value of
178.9(3)8. In contrast, the FeꢀOꢀFe angle in the ligand
/
Oꢀ/Fe bond angles. For the ligand 2 dimer, this
/
/
1 dimer is only 144.1(2)8. These values approach the
extremes for linearity and non-linearity, respectively, for
m-oxo bridged iron(III) dimers [25]. A possible rationa-
lization of this observation is that the bonding in 4 is
compromised by poor complementarity (the small
ligand size). This dimensional relationship decreases
the Nꢀ
/
Feꢀ/N bond angles, which in turn reduces orbital
overlap.
The trend in the ability of the ligands to enclose the
metal ion is unaffected by absence of N-substitution.
The ligand 3 dimer has an average Naxꢀ
161.608, while this angle averages 147.558 for the ligand
Neq angles average 78.85 and
/
Mꢀ/Nax angle of
4 dimer. The Neqꢀ
/
Mꢀ
/
77.708 for the ligand 3 dimer and the ligand 4 dimer,
respectively.
3.1.2. Electrochemistry
Electrochemical studies were used to compare the
monomeric complexes of the benzyl-substituted ligands
1 and 2 and the unsubstituted ligands 3 and 4 with the
methyl-substituted ligand 5 and 6 complexes, whose
previous characterization gave promise of oxidative
catalytic behavior in aqueous media [13ꢁ15]. The new
/
complexes provide the opportunity to probe the effect of
the N-substituents on the accessibility of higher metal
oxidation states. The cyclic voltammograms of the
complexes were obtained in either acetonitrile or DMF
in an inert atmosphere glovebox. The redox potentials
and peak separations are shown in Table 3. These rigid
ligands stabilize a range of oxidation states for manga-
nese in acetonitrile, from Mn2ꢀ to Mn4ꢀ as shown by
Fig. 4. Views of the crystal structures of: (a) [Cl(3)Fe(m-
O)Fe(3)Cl]Cl2×
/
3H2O; and (b) [Cl(4)Fe(m-O)Fe(4)Cl]Cl2×2H2O.
/
favor lower coordination numbers with more nearly
planar ligands. In the ligands, all of the CꢀN bond
/
lengths are within the accepted values for single bonds,
confirming the absence of ligand oxidation during the
formation of the dimers [31]. In contrast, such ligand
oxidation is common in Fe2ꢀ complexes of secondary
amine-containing macrocycles [26]. As expected, the
bond distances between the donor nitrogens are shorter
for these Fe3ꢀ complexes than for similar Fe2ꢀ com-
plexes. For example, in the Fe2ꢀ complexes of 5 and 6,
Table 3
Redox potentials (vs. SHE) for the complexes with peak separations
Complex
Solvent Redox couple
E1/2 (V) (EaꢃEc) (mV)
/
Mn(5)Cl2
MeCN
MeCN
Mn3ꢀ/Mn2ꢀ
Mn4ꢀ/Mn3ꢀ
Mn3ꢀ/Mn2ꢀ
Mn4ꢀ/Mn3ꢀ
Mn3ꢀ/Mn2ꢀ
Mn3ꢀ/Mn2ꢀ
Mn3ꢀ/Mn2ꢀ
Mn3ꢀ/Mn2ꢀ
Mn3ꢀ/Mn2ꢀ
Fe3ꢀ/Fe2ꢀ
ꢀ
/
0.585
1.343
0.466
61
65
70
ꢀ
/
Mn(6)Cl2
ꢀ
/
ꢀ1.232 102
/
the Feꢀ
/
N(38) bond distances average 2.27 (Lꢂ5) and
/
Mn(5)Cl2
Mn(1)Cl2
Mn(2)Cl2
Mn(3)Cl2
Mn(4)Cl2
Fe(5)Cl2
Fe(6)Cl2
Fe(1)Cl2
Fe(2)Cl2
Fe(3)Cl2
Fe(4)Cl2
DMF
DMF
DMF
DMF
DMF
MeCN
MeCN
DMF
DMF
DMF
DMF
ꢀ
/
0.522
0.577
0.400
0.239
67
72
65
79
˚
2.26 A (Lꢂ
are 2.24 (Lꢂ
amineꢀ
N(28) averages 2.15 A for Lꢂ
/
6), while for 1 and 2, the average Feꢀ
/
N(38)
ꢀ
/
˚
/
1) and 2.22 A (Lꢂ
/
2). The secondary
ꢀ
/
/
Fe3ꢀ bond lengths are somewhat shorter: Feꢀ
/
ꢀ/
ꢀ0.389 280
/
˚
˚
/
1 and 2.17 A for Lꢂ
/2.
ꢀ
/
0.110
0.036
0.157
0.071
0.113
0.055
63
64
85
85
78
89
Fe3ꢀ/Fe2ꢀ
ꢀ
/
Even more affected by the change in metal oxidation
˚
Cl bond distances: 2.43 A for Fe(5)Cl2,
state are the Feꢀ
/
Fe3ꢀ/Fe2ꢀ
ꢀ
/
˚
2.41 A for Fe(6)Cl2, 2.36 A for the dimer of 1, and 2.35
˚
Fe3ꢀ/Fe2ꢀ
ꢀ
/
Fe3ꢀ/Fe2ꢀ
ꢃ
/
˚
A for the dimer of 2 (averages). The Feꢀ
/
O bond
distances are in the range expected for dimers of this
Fe3ꢀ/Fe2ꢀ
ꢃ/

T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ
/86
85
the two reversible oxidations of Mn(5)Cl2 and
Mn(6)Cl2. In contrast, only the Fe3ꢀ/Fe2ꢀ couple is
observed for Fe(5)Cl2 and Fe(6)Cl2 in acetonitrile.
plexes are still relatively rigid regardless of substitution,
and thus their redox potentials are less modified, in the
unsubstituted complexes.
Unfortunately, the complexes of ligands 1ꢁ4 are not
/
sufficiently soluble in acetonitrile to permit electroche-
mical studies in that medium. But, these complexes do
have good solubility in DMF, and their cyclic voltam-
mograms were obtained in this solvent. The complex
Mn(5)Cl2 was examined in both solvents as a test for the
effect of solvent on the redox potentials. This complex
only exhibits one reversible oxidation in DMF, because
the solvent itself is oxidized at potentials corresponding
to the Mn4ꢀ/Mn3ꢀ couple. The Mn3ꢀ/Mn2ꢀ couple is
observed in both solvents with only small changes in
4. Conclusions
FeII and MnII, as noted in Section 1, are enormously
important biological ions. However, their properties
have made them difficult to control in oxygenated
aqueous systems without the protection of the bulky,
hydrophobic proteins provided by nature. The ligands
1ꢁ
6 are designed to produce Mn2ꢀ and Fe2ꢀ complexes
/
that are stable in aqueous solutions, and under other
harsh conditions, to take advantage of their known
catalytic reactivities. Here, we have reported the synth-
esis of Mn2ꢀ and Fe2ꢀ complexes of the cross-bridged
potential or reversibility: ꢀ
/
0.585 V (61 mV) in MeCN
versus ꢀ0.522 V(67 mV) in DMF.
/
Comparison of the 14- and 12-membered ligands
shows a significant ring size effect. The smaller ring
complexes are generally easier to oxidize, which may be
explained simply on the basis of size; the smaller cavity
more greatly stabilizes the smaller oxidized metal ion.
Conversely, the larger ligands favor the larger lower
valent metal ions. Only in the unsubstituted ligands 3
and 4 is this trend reversed. With these ligands, for both
the Fe2ꢀ and Mn2ꢀ complexes, the complex of the
larger ligand is slightly easier to oxidize. Here, the
unsubstituted ligands may be less rigid than those
containing only tertiary N, allowing 3 to more easily
adapt to the small Fe3ꢀ ion than with the substituted
ligands. Ligand 4, because of its small size, should retain
more of its rigidity even when unsubstituted, and may
not adjust to small Fe3ꢀ as well. In spite of the
electrochemistry, the overall ability of the ligand 3 to
engulf Fe3ꢀ is still greater than that of ligand 4 (see
discussion of dimer crystal structures above.)
macrobicyclic ligands 1ꢁ4, overcoming their proton-
sponge nature by reducing the activity of protons in the
reaction mixture, following the lead provided in work
/
with ligands 5ꢁ6. The derivatives of the larger rings
/
display more positive oxidation potentials, which can be
traced to a better size match between the larger ring and
the larger lower valent ions and the smaller rings with
the smaller higher valent ions. The potentials of the
complexes of the present ligand 1ꢁ
greatly changed from those of the previously character-
ized complexes of ligands 5ꢁ6. It is therefore expected
/4 complexes are not
/
that they will not have vastly different aqueous reactiv-
ities. The new complexes also suffer the disadvantage of
being only sparingly soluble in water. These factors
combine to suggest that further studies on the catalytic
behavior of this class of complexes are best reserved for
the parent, methyl-substituted complexes, until other
derivatives of greater solubility and more perturbed
electrochemistry are designed and synthesized.
Going from the methyl- to the benzyl-substituted
derivatives has only a small effect on the redox
potentials. For example the Mn3ꢀ/Mn2ꢀ wave changes
from ꢀ
The Fe3ꢀ/Fe2ꢀ wave is similarly changed from ꢀ
V for Fe(5)Cl2 to ꢀ0.157 V in Fe(1)Cl2. Similar effects
are observed in the complexes of 4 as compared to those
of 6. Comparison of the methyl-substituted and the
unsubstituted ligands shows substantial changes. The
/
0.522 V for Mn(5)Cl2 to ꢀ
/
0.577 V in Mn(1)Cl2.
5. Supporting information available
/
0.110
/
Tables of bond distances and angles, atomic coordi-
nates, anisotropic displacement parameters, and isotro-
pic displacement parameters for Mn(1)(CF3SO3)2,
Mn(2)Cl2, Fe(2)Cl2, [Cl(3)Fe(m-O)Fe(3)Cl]Cl2×
/H2O,
Mn3ꢀ/Mn2ꢀ wave changes from
Mn(5)Cl2 to ꢀ
wave is also substantially changed from ꢀ
Fe(5)Cl2 to ꢃ0.113 V in Fe(3)Cl2. This effect is lessened
in the complexes of 4 and 6, where the Mn3ꢀ/Mn2ꢀ
wave changes only from ꢀ0.466 V for Mn(6)Cl2 to ꢀ
0.389 V in Mn(4)Cl2, and the Fe3ꢀ/Fe2ꢀ potential, only
from ꢀ0.036 V for Fe(6)Cl2 to ꢃ0.055 V in Fe(4)Cl2.
ꢀ
/
0.522
0.239 V in Mn(3)Cl2. The Fe3ꢀ/Fe2ꢀ
0.110 V for
V
for
and [Cl(4)Fe(m-O)Fe(4)Cl]Cl2×
/2H2O are available from
/
the authors upon request.
/
/
Acknowledgements
/
/
The funding of this research by the Procter and
Gamble Company is gratefully acknowledged. T.J.H.
thanks the Madison and Lila Self Graduate Research
Fellowship of the University of Kansas for financial
support. We thank EPSRC and Siemens Analytical
Instruments for grants in support of the diffractometer.
/
/
Again, large changes in the flexibility between the
substituted and unsubstituted versions of the cyclam-
derived ligands might explain the large difference in
their redox potentials. The smaller 12N4 derived com-

86
T.J. Hubin et al. / Inorganica Chimica Acta 346 (2003) 76ꢁ86
/
[13] T.J. Hubin, J.M. McCormick, S.R. Collinson, M. Buchalova,
C.M. Perkins, N.W. Alcock, P.K. Kahol, A. Raghunathan, D.H.
Busch, J. Am. Chem. Soc. 122 (2000) 2512.
The Kansas/Warwick collaboration has been supported
by NATO.
[14] T.J. Hubin, J.M. McCormick, N.W. Alcock, D.H. Busch, Inorg.
Chem. 40 (2001) 435.
[15] (a) D.H. Busch, S.R. Collinson, T.J. Hubin, Catalysts and
Methods for Catalytic Oxidation, WO 98/39098, September 11,
1998;
References
[1] F.A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, 5th
ed., Wiley, New York, 1988.
(b) D.H. Busch, S.R. Collinson, T.J. Hubin, R. Labeque, B.K.
Williams, J.P. Johnston, D.J. Kitko, J.C.T.R. Burkett-St. Laur-
ent, C.M. Perkins, Bleach Compositions, US Patent 6,218,351 B1.
[16] G.R. Weisman, E.H. Wong, D.C. Hill, M.E. Rogers, D.P. Reed,
J.C. Calabrese, J. Chem. Soc., Chem. Commun. (1996) 947.
[17] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of
Laboratory Chemicals, 2nd ed., Pergamon Press, New York,
1980.
[2] P.R. Ortiz de Montellano (Ed.), Cytochrome P450: Structure,
Mechanism, and Biochemistry, Plenum Press, New York, 1986.
[3] (a) H.G. Jang, D.D. Cox, L. Que, Jr., J. Am. Chem. Soc. 113
(1991) 9200;
(b) S. Han, L.D. Eltis, K.N. Timmis, S.W. Muchmore, J.T. Bolin,
Science 270 (1995) 976.
[4] (a) B.J. Waller, J.D. Lipscomb, Chem. Rev. 96 (1996) 2625;
(b) A.C. Rosenzweig, P. Nordlund, P.M. Takahara, C.A.
Frederich, S. Lippard, J. Chem. Biol. 2 (1995) 409;
(c) L. Que, Jr., Y. Dong, Acc. Chem. Res. 29 (1996) 190.
[5] J.C. Boyington, B.J. Gaffney, L.M. Amzel, Science 260 (1993)
1482.
[18] E.K. Barefield, F. Wagner, A.W. Herlinger, A.R. Dahl, Inorg.
Synth. 16 (1976) 220.
[19] P.S. Bryan, J.C. Dabrowiak, Inorg. Chem. 2 (1975) 296.
[20] H.T. Witteveen, B. Nieuwenhuijse, J. Reedijk, J. Inorg. Nucl.
Chem. 36 (1974) 1535.
[21] SMART User’s Manual, Siemens Industrial Automation Inc.,
Madison, WI, USA.
[6] (a) V.L. Pecoraro (Ed.), Manganese Redox Enzymes, VCH, New
York, 1992;
[22] J. Cosier, A.M. Glazer, J. Appl. Crystallogr. 19 (1986) 105.
[23] G.M. Sheldrick, Acta Crystallogr., A 46 (1990) 467.
[24] G.M. Sheldrick, SHELX-96 (beta-test) (including SHELXS and
(b) B.K. Vainshtein, W.R. Melik-Adayamyan, V.V. Barynin,
A.A. Vagin, A.I. Grebenko, Proc. Int. Symp. Biomol. Struct.
Interact., Suppl. J. Biosci. 8 (1985) 471;
SHELXL), University of Gottingen, Gottingen, Germany, 1996.
¨
¨
(c) V.V. Barynin, A.A. Vagin, W.R. Melik-Adayamyan, A.I.
Grebenko, S.V. Khangulov, A.N. Popov, M.E. Andrianova, B.K.
Vainshtein, Dokl. Akad. Nauk, SSSR 228 (1986) 877;
(d) K. Weighardt, Angew. Chem., Int. Ed. Engl. 28 (1989) 1153.
[7] (a) M.L. Ludwig, A.L. Metzger, K.A. Pattridge, W.C. Stallings, J.
Mol. Biol. 219 (1991) 335;
[25] H. Irving, R.J.P.J. Williams, Chem. Soc. (1953) 3192 (Part III).
[26] V.L. Goedken, D.H. Busch, J. Am. Chem. Soc. 94 (1972) 7355.
[27] K.S. Murray, Coord. Chem. Rev. 12 (1974) 1.
[28] (a) P.A. Goodson, D.J. Hodgson, J. Glerup, K. Michelson, H.
Weihe, Inorg. Chem. Acta 197 (1992) 141;
(b) P.A. Goodson, J. Glerup, D.J. Hodgson, K. Michelson, H.
Weihe, Inorg. Chem. 30 (1991) 4909;
(b) G.E.O. Borgstahl, H.E. Parge, M.J. Hickey, W.F. Beyer, Jr.,
R.A. Hallewell, J.A. Tainer, Cell 71 (1992) 107.
[8] (a) D.M. Prosperpio, R. Hoffman, G.C. Dismukes, J. Am. Chem.
Soc. 114 (1992) 4374;
(c) P.A. Goodson, J. Glerup, D.J. Hodgson, K. Michelson, E.
Pederson, Inorg. Chem. 29 (1990) 503;
(d) A.R. Oki, J. Glerup, D.J. Hodgson, Inorg. Chem. 29 (1990)
2435;
(b) T. Ono, T. Noguchi, Y. Inoue, M. Kusunoki, T. Matsushita,
H. Oyanagi, Science 258 (1992) 1335;
(c) U. Bossek, T. Weyhermuller, K. Weighardt, B. Nuber, J.
¨
(e) K.J. Brewer, M. Calvin, R.S. Lumpkin, J.W. Otvos, L.O.
Spreer, Inorg. Chem. 28 (1989) 4446;
Weiss, J. Am. Chem. Soc. 112 (1990) 6387.
[9] (a) K. Wieghardt, U. Bossek, B. Nuber, J. Weiss, J. Bonvoisin, M.
Corbella, S.E. Vitols, J.J. Girerd, J. Am. Chem. Soc. 110 (1988)
7398;
(f) K.S. Hagen, W.H. Armstrong, H. Hope, Inorg. Chem. 27
(1988) 267;
(g) S.R. Cooper, G.C. Dismukes, M.R. Klein, M. Calvin, J. Am.
Chem. Soc. 100 (1978) 7248;
(b) V.C. Quee-Smith, L. DelPizzo, S.H. Jurellar, J.L. Kershner, R.
Hage, Inorg. Chem. 35 (1996) 6461;
(h) P.M. Plaksin, R.C. Stoufer, M. Mathew, G.J. Palenik, J. Am.
Chem. Soc. 94 (1972) 2121.
(c) J.H. Koek, S.W. Russel, L. van der Wolf, R. Hage, J.B.
Warnaar, A.L. Spek, J. Kershner, L. DelPizzo, J. Chem. Soc.,
Dalton. Trans. (1996) 353;
[29] (a) K.J. Brewer, M. Calvin, R.S. Lumpkin, J.W. Otvos, L.O.
Spreer, Inorg. Chem. 28 (1989) 4446;
(b) C.-M. Che, S.-S. Kwong, C.-K. Poon, T.-F. Lai, T.C.W. Mak,
Inorg. Chem. 24 (1985) 1359;
(d) D.E. De Vos, T. Bein, J. Organomet. Chem. 520 (1996) 195.
[10] D.H. Busch, Chem. Rev. 93 (1993) 847.
(c) C.-K. Poon, C.-M. Che, J. Chem. Soc., Dalton. Trans. (1981)
1336;
[11] (a) G.R. Weisman, M.E. Rogers, E.H. Wong, J.P. Jasinski, E.S.
Paight, J. Am. Chem. Soc. 112 (1990) 8604;
(b) A. Bencini, A. Bianchi, C. Bazzicalupi, M. Ciampolini, V.
Fusi, M. Micheloni, N. Nardi, P. Paoli, B. Valtancoli, Supramol.
Chem. 3 (1994) 41.
(d) T.-F. Lai, C.-K. Poon, Inorg. Chem. 15 (1976) 1562.
[30] C.A. Hunter, J.K.M. Sanders, J. Am. Chem. Soc. 112 (1990)
5525.
[12] T.J. Hubin, J.M. McCormick, S.R. Collinson, N.W. Alcock, D.H.
Busch, J. Chem. Soc., Chem. Commun. (1998) 1675.
[31] F.H. Allen, O. Kennard, D.G. Watson, L. Brammer, A.G. Orpen,
R. Taylor, J. Chem. Soc., Perkin Trans. II (1987) S1.