Nucleophilic reactivity of a series of peroxomanganese(iii) complexes supported by tetradentate aminopyridyl ligands

Article abstract of DOI:10.1039/c0dt01570a

Peroxomanganese(iii) adducts have been postulated as important intermediates in manganese-containing enzymes and small molecule oxidation catalysts. Synthetic peroxomanganese(iii) complexes are known to be nucleophilic and facilitate aldehyde deformylation, offering a convenient way to compare relative reactivities of complexes supported by different ligands. In this work, tetradentate dipyridyldiazacycloalkane ligands with systematically perturbed steric and electronic properties were used to generate a series of manganese(ii) and peroxomanganese(iii) complexes. X-Ray crystal structures of five manganese(ii) complexes all show the ligands bound to give trans complexes. Treatment of these Mn^II precursors with H₂O₂ and Et₃N in MeCN at -40°C results in the formation of peroxomanganese(iii) complexes that differ only in the identity of the pyridine ring substituent and/or the number of carbons in the diazacycloalkane backbone. To determine the effects of small ligand perturbations on the reactivity of the peroxo group, the more thermally stable peroxomanganese(iii) complexes were reacted with cyclohexanecarboxaldehyde. For these complexes, the rate of deformylation does not correlate with the expected nucleophilicity of the peroxomanganese(iii) unit, as the inclusion of methyl substituents on the pyridines affords slower deformylation rates. It is proposed that adding methyl-substituents to the pyridines, or increasing the number of carbons on the diazacycloalkane, sterically hinders nucleophilic attack of the peroxo ligand on the carbonyl carbon of the aldehyde.

Full text of DOI:10.1039/c0dt01570a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 1707
www.rsc.org/dalton
PAPER
Nucleophilic reactivity of a series of peroxomanganese(III) complexes
supported by tetradentate aminopyridyl ligands†
Robert A. Geiger, Swarup Chattopadhyay, Victor W. Day and Timothy A. Jackson*
Received 11th November 2010, Accepted 3rd December 2010
DOI: 10.1039/c0dt01570a
Peroxomanganese(III) adducts have been postulated as important intermediates in
manganese-containing enzymes and small molecule oxidation catalysts. Synthetic peroxomanganese(III)
complexes are known to be nucleophilic and facilitate aldehyde deformylation, offering a convenient
way to compare relative reactivities of complexes supported by different ligands. In this work,
tetradentate dipyridyldiazacycloalkane ligands with systematically perturbed steric and electronic
properties were used to generate a series of manganese(II) and peroxomanganese(III) complexes. X-Ray
crystal structures of five manganese(II) complexes all show the ligands bound to give trans complexes.
Treatment of these MnII precursors with H2O2 and Et3N in MeCN at -40 ◦C results in the formation of
peroxomanganese(III) complexes that differ only in the identity of the pyridine ring substituent and/or
the number of carbons in the diazacycloalkane backbone. To determine the effects of small ligand
perturbations on the reactivity of the peroxo group, the more thermally stable peroxomanganese(III)
complexes were reacted with cyclohexanecarboxaldehyde. For these complexes, the rate of
deformylation does not correlate with the expected nucleophilicity of the peroxomanganese(III) unit, as
the inclusion of methyl substituents on the pyridines affords slower deformylation rates. It is proposed
that adding methyl-substituents to the pyridines, or increasing the number of carbons on the
diazacycloalkane, sterically hinders nucleophilic attack of the peroxo ligand on the carbonyl carbon of
the aldehyde.
tris(pyrazolyl)borate chelates, have been use to stabilize peroxo-
Introduction
manganese(III) intermediates.13–21 Of the structurally characterized
2
Peroxoiron(III) and peroxomanganese(III) complexes play impor-
complexes, all exhibit side-on (h ) peroxo ligands with Mn–O
13,15,16,21,22
tant roles in biological systems and in catalytic oxidation reactions.
A ferric-peroxo species is implicated as a key intermediate in
certain cytochrome P450-catalyzed reactions, including the arom-
atization of androgen to estrogen by cytochrome P450 aromatase1
and oxidation of L-arginine to citrulline by the P450 mono-
oxygenase nitric oxide synthase.2 Peroxomanganese(III) complexes
are potential intermediates in manganese containing enzymes,
including manganese superoxide dismutase, manganese catalase,
and manganese ribonucleotide reductase.3–7 While the mechanisms
of these systems remain the subject of much current work, the
importance of peroxo-level intermediates has been outlined in
several reports.8–11 An example is the manganese superoxide dis-
mutase enzyme that forms a product-inhibited complex proposed
to contain a peroxomanganese(III) adduct.8,9,12
˚
distances in the range of 1.85 to 1.88 A,
although it has
been proposed that this distance can be modulated by the inclusion
of electron-donating groups trans16 or cis20 to the peroxo unit. A
convenient way to compare the relative reactivities of these diverse
complexes, and thus define different steric and electronic influences
on reactivity, is by using a common reaction. Because aldehyde de-
formylation appears to be characteristic of both MnIII-peroxo15,16
and FeIII-peroxo23,24 complexes, it is ideal for this purpose. Indeed,
peroxomanganese(III) adducts assembled using very different sup-
porting ligands (Scheme 1) deformylate cyclohexanecarboxalde-
hyde (CCA).15–17 For MnIII–O2 adducts supported by the 14-TMC
and 13-TMC ligands (14-TMC = 1,4,8,11-tetramethyl-1,4,8,11-
tetraazacyclotetradecane and 13-TMC = 1,4,7,10-tetramethyl-
1,4,7,10-tetraazacyclotridecane), nearly identical second-order
rate constants were reported (k2 = 2.4 and 1.2 M-1 min-1
respectively at 10 ◦C). Cyclohexene, the expected product
of CCA deformylation (Scheme 1), was observed in both
cases in ~55% yield.15,16 Using a peroxomanganese(III) com-
plex supported by the anionic H3bupa ligand (H3bupa is
the dianion of bis[(N¢-tert-butylurealy)-N-ethyl]-(6-pivalamido-
2-pyridylmethyl)amine), Borovik et al. reported the oxidative
deformylation of CCA to give cyclohexanone in 40% yield.17
A variety of different ligands sets, ranging from neutral,
tetraamine macrocylic ligands to anionic, facially-coordinating
Department of Chemistry, University of Kansas, 1251 Wescoe Hall Drive,
Lawrence, KS 66045, USA. E-mail: taj@ku.edu; Fax: +1 (785) 864-5396;
Tel: +1 (785) 864-3968
† Electronic supplementary information (ESI) available. CCDC reference
numbers 801091–801095. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt01570a
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 1707–1715 | 1707
©

View Article Online
Scheme 1
The identity of the ligand trans to the peroxo group has been
shown to have a dramatic effect on the rate of CCA deformylation.
The addition of anions to [MnIII(O2)(13-TMC)]+ leads to the for-
mation of neutral, seven-coordinate MnIII–O2 adducts, formulated
as [MnIII(O2)(13-TMC)(X)] (where X- = CN-, NCS-, CF3CO2 ,
-
-
and N3 ) that showed dramatically enhanced rates for CCA
deformylation.16 In particular [MnIII(O2)(13-TMC)(N3)] displayed
a second-order rate constant 104 larger than that of [MnIII(O2)(13-
TMC)]+. This remarkably large rate enhancement was attributed
to the strong electron donating properties of the axial azide ligand.
It was postulated that the trans N3- ligand favors a peroxo binding
mode intermediate between side-on and end-on, increasing the
nucleophilicity of one oxygen atom of the peroxo ligand.
We have recently reported a series of MnIII–O2 complexes
supported by tetradentate dipyridyldiazacycloalkane ligands:
L7py2H, L7py25-Br, L7py26-Me, and L7py2
(L7py2 = 1,4-bis(2-
6-MeO
H
pyridylmethyl)-1,4-diazepane; see Scheme 2).20 It was shown
that by changing the substituent on the pyridine ring, the
spectroscopic properties and stability of these peroxoman-
ganese(III) complexes were perturbed. For [MnIII(O2)(L7py2H)]+
and [MnIII(O2)(L7py26-Me)]+, these spectroscopic changes were
linked to the geometric and electronic structure of the MnIII–O2
unit. Specifically we proposed that the inclusion of the electron-
donating methyl-substituents lead to an elongation in one of
the Mn–O bonds, affording a more asymmetric, end-on binding
mode. In accordance with the reactivity studies described above,
this should render the peroxo in [MnIII(O2)(L7py26-Me)]+ more
nucleophilic than that of [MnIII(O2)(L7py2H)]+. To evaluate this
prediction and further explore steric and electronic influences
on reactivity, in this current work we performed systematic
ligand modifications by altering (i) the nature and location of
pyridine ring substituents, (ii) the size of the diazacycloalkane
backbone, and (iii) the type of N-ligand from pyridine to quinoline
(Scheme 2). The structures of the manganese(II) precursors were
solved by X-ray diffraction, and their geometries report the effects
of these ligand perturbations. In cases where these new ligands
supported stable peroxomanganese(III) adducts, these species were
Scheme 2
reacted with CCA to determine the effect of ligand modifications
on the rate of CCA deformylation. Unlike previous observations,
CCA deformylation by the MnIII–O2 units in these complexes
appears to be largely influence by the sterics of the supporting
ligand.
Results and discussion
To complement our previously reported peroxomanganese(III)
complexes generated using the L7py2H, L7py26-Me, and L7py2
6-MeO
H
ligands,20 we assembled five new MnII complexes using the L6py2
,
L7py24-Me, L7q2, L8py2H, and L8py2
ligands (Scheme 2). With
6-Me
these new ligands, our initial objective was to assess the effects of
four variables on the stability and reactivity of the MnIII–O2 unit.
1708 | Dalton Trans., 2011, 40, 1707–1715
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
◦
˚
Table 1 Selected bond lengths (A) and angles ( ) for manganese(II) complexes
Complex
Mn–N(1) Mn–N(2) Mn–N(3) Mn–N(4) Mn–X(1) Mn–X(2) Mn–X(3)a N3–Mn–N4 C1–N1–Mn–N2–C11
[MnII(L6py2H)(NCMe)3](ClO4)2 2.313(2) 2.307(2) 2.339(2)
2.360(3) 2.286(3)a 2.314(2)a 2.370(3)
2.357(3) 2.268(2)a 2.311(2)a 2.478(2)
62.82(8)
67.40(1)
69.50(7)
71.13(6)
70.36(9)
70.77(6)
78.08(5)
76.27(4)
-57.0(3)
-66.3(6)
-48.1(2)
-44.5(2)
-43.2(2)
-42.1(2)
-68.7(1)
-62.2(1)
b
[MnII(L7py2H)(NCMe)3](ClO4)2 2.306(9) 2.308(3) 2.369(3)
[MnII(L7py26-Me)(ClO4)2]b
[MnII(L7py24-Me)(OTf)2]
[MnII(L7py26-MeO)(ClO4)2]b
[MnII(L7q2)(ClO4)2]
2.283(2) 2.284(2) 2.30692) 2.285(2) 2.212(2)c 2.280(2)a NA
2.216(2) 2.213(2) 2.288(2)
2.176(2) 2.176(2) 2.308(2)
2.226(2) 2.254(2) 2.274(2)
2.246(1) 2.257(1) 2.292(1)
2.273(1) 2.273(1) 2.290(1)
2.277(2) 2.198(2)d 2.235(2)d NA
2.308(2) 2.278(2)c 2.278(2)c NA
2.263(2) 2.209(1)c 2.282(1)c NA
2.320(1) 2.244(1)c 2.266(1)c NA
2.290(1) 2.205(1)d 2.205(1)d NA
[MnII(L8py2H)(ClO4)2]
[MnII(L8py26-Me)(OTf)2]
a X = N(CMe). b From reference 20. c X = O(ClO3)-. d X = OTf.
These variables are (i) nature of the pyridine substituent (L7py2
In all cases, the inclusion of pyridine substituents, or sub-
stitution of pyridines for quinolines, results in six-coordinate
complexes. On the basis of their small N3–Mn–N4 bite angles of
~70◦ (Table 1), the MnII complexes supported by the L7q2 ligand
and the substituted L7py2R ligands (R = 6-Me, 6-MeO, and 4-Me)
should be able to accommodate a fifth equatorial ligand. For the
L7py26-Me, L7py26-MeO, and L7q2 complexes, steric crowding of the
equatorial plane hinders the binding of a seventh ligand. For the
[MnII(L7py24-Me)(OTf)2] complex, the lack of heptacoordination is
tentatively attributed to the shorter Mn–N(pyridine) bonds rela-
H
6-Me
H
versus L7py2
and L8py2 versus L8py26-Me), (ii) position of the
pyridine substituent (L7py2
versus L7py24-Me), (iii) size of the
6-Me
diazacycloalkane backbone (L6py2 versus L7py2 and L8py2), and
(iv) identity of the N-donor ligand (L7py2 versus L7q2).
Structural properties of manganese(II) complexes
Fig. 1 displays ORTEP diagrams for the five manganese(II) com-
plexes [MnII(L6py2H)(NCMe)3](ClO4)2, [MnII(L7py24-Me)(OTf)2],
[MnII(L7q2)(ClO4)2], [MnII(L8py2H)(ClO4)2], and [MnII(L8py26-Me)-
(OTf)2]. Selected bond distances are given in Table 1. For each
of the five complexes, the linear, tetradentate ligand is bound
in the trans conformation, creating two axial coordination sites
tive to [MnII(L7py2H)(NCMe)3]2+ (2.22 versus 2.31 A, respectively;
˚
Table 1), which are due to the electron-donating 4-Me groups.
These shorter distances for [MnII(L7py24-Me)(OTf)2] further crowd
the equatorial plane and lead to similar pyridine ring canting when
occupied by solvent (L6py2H) or anionic ligands (ClO4 for L7q2
-
compared with the L7py2
and L7py2
analogues that must
6-Me
6-MeO
and L8py2H, and OTf- for L7py2
and L8py26-Me). In the case of
4-Me
cant to avoid steric clashing between the 6-substituents. Alterna-
tively, the lower coordination number of [MnII(L7py24-Me)(OTf)2]
could be attributed to crystallization of the triflate-coordinated
complex, as opposed to the solvent-coordinated analogue, which
would arise from the relative solubilities of these two forms under
the crystallization conditions. The bulk of the triflate ligands
would then disfavor heptacoordination. Space-filling models high-
lighting the general steric effects observed in the crystal structures
are including in Supplementary Information (Fig. S1, ESI†).
[MnII(L6py2H)(NCMe)3](ClO4)2, a seventh ligand is bound in the
equatorial plane, giving a near pentagonal bipyramidal geometry
around the Mn(II) center. For all the complexes, the MnII–ligand
˚
bond lengths range from ~2.1 to 2.5 A, indicative of high-spin
manganese(II) centers.
The differences in geometry of the Mn(II) complexes supported
by the L6py2H, L7py2H, and L8py2H ligands are directly related to the
variation in diazacycloalkane ring size. The [MnII(L8py2H)(ClO4)2]
complex is six-coordinate, while the corresponding L7py2 and
H
L6py2 complexes display seven-coordinate Mn(II) centers. Al-
H
Formation of peroxomanganese(III) complexes
though heptacoordinate Mn(II) complexes are not unusual,25–28
they are far less common than six-coordinate analogues. We
attribute the structural differences among these ligands to the
variation in N3–Mn–N4 bite angle as a function of the diaza-
cycloalkane ring size (although the identity of the X ligand,
The absorption spectra of acetonitrile solutions of the five
new manganese(II) complexes, [MnII(L6py2H)(NCMe)3]2+,
[MnII(L7py24-Me)(OTf)2], [MnII(L7q2)(ClO4)2], [MnII(L8py2H)-
(ClO4)2], and [MnII(L8py26-Me)(OTf)2], are featureless at energies
below 33 000 cm-1, indicative of high spin manganese(II)
centers. Treatment of the [MnII(L7py24-Me)]2+, [MnII(L7q2)]2+,
and [MnII(L8py2H)]2+ complexes with 5 equivalents H2O2 and
0.5 equivalents triethylamine at -40 ◦C results in the formation of
new absorption features in the visible region (Fig. 2 and Table 2).
Mass spectrometry experiments on these colored solutions
reveal major ion peaks consistent with [MnIII(O2)(L7py24-Me)]+,
[MnIII(O2)(L7q2)]+, and [MnIII(O2)(L8py2H)]+. Thus, these MnII
centers form metastable MnIII–O2 adducts when reacted with
H2O2 in the presence of base at low temperature (Scheme 3),
consistent with a previous report on similar complexes.20
While [MnII(L8py26-Me)]2+ reacts with 5 equivalents H2O2 and
0.5 equivalents triethylamine at -40 ◦C to form a colored
-
NCMe or ClO4 , will also be important). When bound to the
MnII center, the L6py2 ligand creates a N3–Mn–N4 bite angle
H
of 62.8◦, which is smaller than those of the corresponding
L7py2 and L8py2 complexes (67.4◦ and 78.1◦, respectively).
Equatorial angles of 72◦ are expected for idealized pentagonal
bipyramidal complexes, consistent with [MnII(L6py2H)(NCMe)3]2+
and [MnII(L7py2H)(NCMe)3]2+ being seven-coordinate. The bite
angles also offer a rationale for the smaller meridional Mn–
NCMe bond length for [MnII(L6py2H)(NCMe)3]2+ compared with
H
H
[MnII(L7py2H)(NCMe)3]2+ (2.37 versus 2.48 A, respectively; see
˚
Mn–X3 in Table 1). Binding of a fifth equatorial ligand is also
associated with significant canting of the pyridines, as assessed by
the torsional angle resulting from the two planes of the pyridine
rings (Table 1). This canting is nicely illustrated pictorially in the
space-filling model of [MnIII(L6py2H)]2+ shown in Fig. 1F.
intermediate, the fleeting nature of this chromophore (t1/2
=
seconds at -40 ◦C) prevented its further characterization.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 1707–1715 | 1709
©

View Article Online
Fig. 1 ORTEP diagrams of (A) [MnII(L6py2H)(NCMe)3](ClO4)2, (B) [MnII(L7py24-Me)(OTf)2], (C) [MnII(L7q2)(ClO4)2], (D) [MnII(L8py2H)(ClO4)2], and
(E) [MnII(L8py26-Me)(OTf)2] showing 50% probability thermal ellipsoids. (F) Space filling model for [MnII(L6py2H)(NCMe)3]2+ viewed along the Mn–X3
axis. For ORTEP structures, hydrogen atoms and noncoordinating counteranions have been removed for clarity. Significant interatomic distances and
angles are listed in Table 1.
Notably, [MnII(L6py2H)]2+ showed no reaction under these
conditions.
a “bulky” pyridine). With the aid of density functional theory
(DFT) computations, we previously attributed differences in
absorption and, especially, magnetic circular dichroism (MCD)
spectra of [MnIII(O2)(L7py2H)]+ and [MnIII(O2)(L7py26-Me)]+ to
a change from a side-on-bound peroxo ligand with identical
The absorption maxima of [MnIII(O2)(L7py24-Me)]+ and
[MnIII(O2)(L8py2H)]+ are nearly identical to those of
[MnIII(O2)(L7py2H)]+, consisting of a weak feature at ~17 000 cm-1
and a more intense band at ~22 000 cm-1 (Table 2). In contrast,
the absorption spectrum of [MnIII(O2)(L7q2)]+ resembles that of
[MnIII(O2)(L7py26-Me)]+ and [MnIII(O2)(L7py26-MeO)]+ (lmax ª 16 000
and 24 000 cm-1). Among this set of complexes, the common link
between ligand structure and absorption maxima is steric bulk
in the 6-position of the pyridine ring (assuming the quinoline as
˚
Mn–Operoxo distances of 1.87 A in the former complex to a more
˚
end-on peroxo with Mn–Operoxo distances of 1.87 and 1.91 A in the
latter species.20 It was suggested that the increased Lewis acidity
of the manganese center in [MnIII(O2)(L7py26-Me)]+ led to the
longer, asymmetric Mn–Operoxo bond lengths. The data presented
in Table 2 suggests an alternative explanation. The inclusion of
1710 | Dalton Trans., 2011, 40, 1707–1715
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 2 Properties of peroxomanganese(III) complexes
Complex
l
max/cm-1 (e/M-1 cm-1)
t1/2 at 0 ◦
15 min
6 min
C
[MnIII(O2)(L7py2H)]+a
[MnIII(O2)(L7py26-Me)]+a
[MnIII(O2)(L7py26-MeO)]+a
[MnIII(O2)(L7py24-Me)]+
[MnIII(O2)(L7q2)]+
22 470 (280)
16 950 (120)
24 100 (280)
~16 130 (80)
24 000 (250)
17 900 (80)
22 470 (260)
~17 000 (100)
24 050 (305)
~16 500 (70)
21 550 (280)
16 700 (120)
~ seconds
6 min
4 min
[MnIII(O2)(L8py2H)]+
12 min
a From reference 20.
characteristic visible bands at 22 470 and 16 950 cm-1, indicating
the decay of the peroxomanganese(III) complex (Fig. 3A). This
decay followed pseudo-first order behavior when excess CCA was
used. By plotting the range of pseudo-first order rate constants
(kobs) as a function of CCA concentration, we determined a
second-order-rate constant (k2) of 3.11 M-1 min-1 at -40 ◦C (Fig.
3B). Reactions performed with either perchlorate or triflate as
counteranion displayed identical rates. The organic product of
this reaction was determined by GC-MS, and cyclohexanone was
identified in ~50% yield based on the concentration of the peroxo-
manganese(III) complex. No other organic products were detected.
The formation of a ketone product in aldehyde deformylation
has been previously observed for other peroxoiron(III)23,24 and
peroxomanganese(III)17 complexes.
Fig. 2 Top: Formation of [MnIII(O2)(L8py2H)]+ after the addition of
5 equivalents H2O2 and 0.5 equivalents triethylamine of at -40 ◦C in
MeCN. Bottom: Electronic absorption spectra of [MnIII(O2)(L7q2)]+ and
[MnIII(O2)(L7py24-Me)]+ at -40 ◦C in MeCN.
To date the most reactive peroxomanganese(III) complex for
CCA deformylation is [MnIII(O2)(13-TMC)(N3)].16 Using the DH‡
and DS‡ values reported for this reaction, we estimate a k2 value
of ~220 M-1 min-1 at -40 ◦C, which is 102-fold larger than that of
[MnIII(O2)(L7py2H)]+. On the basis of this comparison, as well as
the other reactivity data reported by Nam et al. for the series of
[MnIII(O2)(13-TMC)(X)] complexes,16 this implies that the k2 value
of [MnIII(O2)(L7py2H)]+ is comparable to that of [MnIII(O2)(13-
TMC)(CN)] and 102-fold greater than that of the parent complex,
[MnIII(O2)(13-TMC)]+. Thus, [MnIII(O2)(L7py2H)]+ represents the
fastest MnIII–O2 CCA deformylation agent lacking an anionic
ligand.
bulky groups in the 6-position is expected to cause significant
canting of the pyridine rings, as evidenced in the XRD structures
of the Mn(II) complexes (Fig. 1 and Table 1). This increased
canting causes the 6-substituents to partially block the axial
positions, sterically disfavoring the formation of a symmetric,
side-on peroxo ligand in the L7py26-Me, L7py26-MeO, and L7q2
complexes. Importantly, this model would also account for the
similarities of the absorption maxima of the [MnIII(O2)(L7py2H)]+
and [MnIII(O2)(L7py24-Me)]+ complexes.
Reactivity of peroxomanganese(III) complexes
The addition of excess CCA to the green acetonitrile solution of
The nucleophilic nature of the aldehyde deformylation re-
action was confirmed by reacting [MnIII(O2)(L7py2H)]+ with
[MnIII(O2)(L7py2H)]+ at -40 ◦C, led to the disappearance of the
Scheme 3
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 1707–1715 | 1711
©

View Article Online
Table 3 Second-order-rate constants for CCA deformylation at -40◦
C
Complex
k2/M-1min-1
[MnIII(O2)(L7py2H)]+
[MnIII(O2)(L7py26-Me)]+
[MnIII(O2)(L7py24-Me)]+
[MnIII(O2)(L8py2H)]+
3.1
0.32
0.40
0.19
Fig. 4 Hammett correlation of the reaction with para-substituted ben-
zaldehydes with [MnIII(O2)(L7py2H)]+. Error bars represent two standard
deviations from three replicate experiments.
constants are less than that for [MnIII(O2)(L7py2H)]+ by an order of
magnitude (Table 3). Because the rate of aldehyde deformylation
is expected to increase as the nucleophilicity of the peroxo ligand
increases,16,23 it is initially surprising thatthe [MnIII(O2)(L7py26-Me)]+
and [MnIII(O2)(L7py24-Me)]+ complexes, which bear pyridine rings
with electron-donating methyl groups, exhibit smaller second-
order-rate constants than [MnIII(O2)(L7py2H)]+. We attribute this
to steric hindrance of CCA-peroxo interactions caused by canting
of the pyridine rings (L7py2
and L7py24-Me) or the presence of
6-Me
the bulkier 8-membered diazacycloalkane backbone (L8py2H). In
support, Fig. 5 shows space-filling models of [MnIII(O2)(L7py2H)]+,
[MnIII(O2)(L7py26-Me)]+ and [MnIII(O2)(L7py24-Me)]+ complexes de-
veloped using density functional theory (DFT) geometry op-
timization. The peroxo ligand in the optimized model of
[MnIII(O2)(L7py2H)]+ is the most exposed, whereas the peroxo
groups in the other complexes are shielded by the canting of the
pyridine rings.
Fig.
3
(A)
Electronic
absorption
spectral
changes
for
[MnIII(O2)(L7py2H)]+ (2.5 mM) upon addition of cyclohexanecar-
boxaldehyde (60 equivalents, 150 mM) at -40 ◦C in MeCN. Insert
shows the time course of the absorbance at 23 000 cm-1. (B) kobs versus
cyclohexanecarboxaldehyde concentration. Error bars represent two
standard deviations determined from 3 or 4 replicate experiments.
para-substituted benzaldehydes (Cl, H, Me, and OMe). Because
the deformylation reaction presumably involves nucleophilic at-
tack of the peroxo on the carbonyl carbon on the aldehyde,
electron withdrawing groups on the substrate should make the
reaction more facile. The kobs values for all four para-substituted
benzaldehydes were determined under invariant conditions of
temperature and concentration. The para-substituents were found
to affect kobs in the manner expected for a nucleophilic reaction (i.e.,
kobs increased with increasing electron-withdrawing properties: Cl
> H > Me > OMe). Indeed, the plot of log kobs versus Hammett
constants (sp) are linear with a r value of 1.25, as shown in Fig.
4. Similar values for r have been observed for other nucleophilic
aldehyde deformylation reactions.16
Conclusion
Because peroxomanganese(III) intermediates play an important
role in manganese-containing enzymes and small molecule oxida-
tion catalysts, it is important to identify the factors affecting the
nucleophilicity of the peroxo ligand. A characteristic reaction of
synthetic peroxomanganese(III) complexes is aldehyde deformy-
lation, which offers a convenient method for comparing relative
reactivities of peroxomanganese(III) complexes supported by dif-
ferent ligands. Herein, we reported the X-ray structures of a series
of manganese(II) complexes with tetradentate dipyridyldiazacy-
cloalkane ligands. Both the size of the diazacycloalkane backbone
and the position and identity of the ring substituents have a
large influence on the geometry of the Mn(II) complex, with both
seven- and six-coordinate complexes observed. Treatment of some
The other three MnIII–O2 complexes [MnIII(O2)(L7py26-Me)]+,
[MnIII(O2)(L7py24-Me)]+, and [MnIII(O2)(L8py2H)]+ were reacted with
CCA in a similar manner, and in each case cyclohexanone
was observed as the organic product. The second-order-rate
1712 | Dalton Trans., 2011, 40, 1707–1715
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 5 Space filling models for (A) [MnIII(O2)(L7py2H)]+, (B) [MnIII(O2)(L7py26-Me)]+, and (C) [MnIII(O2)(L7py24-Me)]+ based on structures derived from
DFT energy minimizations.
of these manganese(II) complexes with H2O2/Et3N at -40 ◦C
resulted in the formation of peroxomanganese(III) adducts. The
thermally stable complexes were treated with CCA to determine
the effects of small ligand perturbations on the reactivity of
the peroxo group. [MnIII(O2)(L7py2H)]+ was observed to have
the largest second-order rate constant (k2), by an order of
magnitude, compared to the other complexes. This result was
formed by Columbia Analytical Services, Tucson AZ. The gas
chromatography-mass spectrometric data were collected on an
Agilent 6890 N Gas Chromatograph interfaced with quadrupole
mass analyzer (Quattro Micro GC, Waters Corporation, Milford
MA). A 5% phenyl, methyl silicone stationary phase (HP-5 MS),
15 metre column with a 0.25¢ ID was used. The carrier gas was
helium and constant flow mode was used to maintain 1.5 ml min-1.
Injections of 1.0 ml were made into the injector port heated to
240 ◦C and a split ratio of 20 : 1 was used. The GC thermal gradient
was an initial 30 ◦C with a 1 min hold after which the temp◦erature
initially surprising because the electron rich L7py24-Me and L7py2
6-Me
ligands were expected to increase the nucleophilicity of the peroxo.
[MnIII(O2)(L8py2H)]+ was observed to have the slowest k2 value.
The decrease in k2 values was attributed to the steric hindrance
of nucleophilic attack of the peroxo ligand on the carbonyl
carbon of the aldehyde. This is the first observation of steric
hindrance in aldehyde deformylation reaction for a series of MnIII–
O2 complexes.
◦
was increased 15 C min-1 to a final temperature of 300 C and
held for 2 min. Ionization was by electron impact at 70 eV and the
mass analyzer scanned from 45 to 600 u in 0.5 s. The analyzers
were tuned to 0.6 u FWHH and data collected in centroid mode.
Ligand synthesis
1,4-Bis(2-pyridylmethyl)piperazine (L6py2H), 1,4-bis(2-pyridyl-
methyl)-1,4-diazepane (L7py2H) and 1,4-bis(6-methyl-2-pyridyl-
methyl)-1,4-diazepane (L7py26-Me) were prepared as previously
described.30,31 The ligands 1,4-bis(4-methyl-2-pyridylmethyl)-
1,4-diazepane (L7py24-Me) and 1,4-bis(2-quinolinylmethyl)-1,4-
diazepane (L7q2) were respectively prepared by reacting ho-
mopiperazine with 4-methylpicolinaldehyde or quinoline-2-
carboxaldehyde (67% and 63% yield), using similar procedures.
Experimental
Materials
All chemicals were obtained from commercial vendors at ACS
reagent-grade or better and were used without further purification
except for CCA. Fractional distillation at 43 ◦C was preformed
under reduced pressure (10 Torr) to remove acid impurities from
CCA. All solvents were dried by routine techniques under an inert
argon atmoshphere.29 Unless otherwise stated, manipulations of
MnII complexes were carried out under argon using a glovebox or
Schlenk techniques.
1H NMR data (400 MHz) for L7py24-Me, and L7q2 are as follows:
4-Me
L7py2
(CDCl3; d) 8.36 (d, 2H JHH = 4.04), 7.28 (s, 2H), 6.96
(br d, 2H, JHH = 3.88), 3.77 (s, 4H), 2.81 (t, 4H, JHH = 4.84), 2.77
(s, 4H), 2.34 (s, 6H) 1.85 (p, 2H, JHH = 4.56): L7q2 (CDCl3; d)
8.12–8.05 (m, 4H), 7.79–7.65 (m, 6H), 7.51–7.49 (m, 2H), 3.99
Instrumentation
(s, 4H), 2.85 (t, 4H, JHH = 6.00), 2.83 (s, 4H), 1.86 (p, 2H, JHH
=
1H NMR spectra were obtained on a Bruker DRX 400 MHz
NMR spectrometer. Electronic absorption spectra were recorded
on a Cary 50 Bio spectrophotometer (Varian) or an Agilent
8453 spectrometer, each of which is interfaced with a Unisoku
cryostat (USP-203-A). ESI-mass spectrometry experiments were
performed using an LCT Primers MicroMass electrospray-
ionization time-of-flight instrument. Elemental analysis was per-
6.44). 1,5-Bis(2-pyridylmethyl)-1,5-diazacyclooctane (L8py2H) was
prepared by a previously reported substitution reaction, where
1,5-diazacyclooctane dihydrobromide is reacted with 2-picolyl
chloride hydrochloride.30 1,5-Bis((6-methyl-2-pyridyl)methyl)-1,5-
diazacyclooctane (L8py26-Me) was prepared by the same procedure
1
using 2-(bromomethyl)-6-methylpyridine (42% yield). H NMR
data (400 MHz) L8py26-Me are as follows: (CDCl3; d) 7.55–7.51 (m,
Dalton Trans., 2011, 40, 1707–1715 | 1713
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
+
quires: M+, 469.1], [Found M+ {[MnIII(O2)(L7py24-Me)]} : 397.1.
2H), 7.34 (d, 2H, JHH = 7.68), 7.00 (d, 2H, JHH = 7.48), 3.79 (s, 4H),
2.79 (t, 8H, JHH = 5.88), 2.51 (s, 6H), 1.67 (p, 4H, JHH = 5.84).
{[MnIII(O2)(L7py24-Me)]} requires: M+, 397.1], and [Found M+
+
{[MnIII(O2)(L8py2H)]} : 383.1. {[MnIII(O2)(L8py2H)]} requires:
+
+
M+, 383.1]
Preparation of manganese(II) complexes
Caution! While no problems were encountered during this work, care
should always be taken when handling perchlorate salts of metal
complexes because of the possibility of explosion.
Kinetic and reactivity studies
Kinetic analyses of aldehyde deformylation were carried out
by adding excess CCA to 2.5 mM acetonitrile solutions of
[MnIII(O2)(L7py2R)]+ or [MnIII(O2)(L8py2H)]+ and monitoring the
time-dependent decay of the characteristic bands of the MnIII–
O2 intermediates at -40 ◦C in MeCN. Kinetic measurements
were made under pseudo-first-order conditions, which permit-
ted the determination of kobs values by fitting the decrease
of absorption bands at ~23 000 cm-1. Corresponding second-
order-rate constants were obtained using kobs values collected
over a range of CCA concentrations. A 2.5 mM solution of
[MnIII(O2)(L7py2H)](OClO3) and [MnIII(O2)(L7py2H)](OTf) was re-
acted with 120 equivalents of CCA to compare if counter anion
had any role on the pseudo first-order rate. The identity of the
counteranion had no effect on the rate (kobs).
Product analysis was determined using GC-MS and percent
yield was determined by comparing the ratio of the product to an
internal standard and comparing the ratio to a standard curve for
the product. A detailed procedure is a follows: an approximately
10 mM solution of the peroxometal complex (formed by treating
10 mM solution of the manganese(II) precursor with 5 equivalents
H2O2 in the presence of 0.5 equivalents Et3N) was reacted with
120 equivalents of CCA at -40 ◦C. When the reaction was
completed, it was allowed to warm to room temperature, and 1,2-
dichlorobenzene was added as an internal standard. A blank run,
which lacked only the manganese(II) complex, showed formation
of neither cyclohexanone nor cyclohexene.
The [MnII(L7py2R)](ClO4)2 (R = H, 6-Me, and 6-MeO) com-
plexes were prepared by a previously reported metalation
reaction.20 The other metal complexes were synthesized with
excellent yields (>85%) by reacting L6py2H, L7py24-Me, L7q2, and
L8py2R (R = H and 6-Me) ligands with Mn(ClO4)2 or Mn(OTf)2 in
MeCN solution in a 1 : 1 molar ratio. Mn(OTf)2 was prepared by a
previously reported procedure by reacting equimolar amounts of
(CH3)3Si(OTf) and anhydrous MnCl2.15 Details of a representative
preparation for a metal complex are as followed. To a stirred
solution of 277 mg (0.785 mmol) of Mn(OTf)2 in 10 mL of MeCN
was added L7q2 (300 mg, 0.785 mmol) in 10 mL of MeCN. The yel-
low solution was stirred overnight and evaporated under reduced
pressure. The solid thus obtained was dried in vacuo. Recrys-
tallization of the crude solid from MeCN–diethyl ether afforded
nearly colorless crystals of [Mn(L7q2)](OTf)2. (505 mg, 87%)
[Found for [Mn(L7q2)](OTf)2: C27H26F6MnN4O6S2: C, 43.6; H,
+
3.5; N, 7.4%; M+ ({[Mn(L7q2)](OTf)} ) 586.1. C27H26F6MnN4O6S2
requires: C, 44.1; H, 3.6; N, 7.6%; M+, 586.1], [Found for
[Mn(L6py2H)](OTf)2: C18H20F6MnN4O6S2: C, 34.4; H, 3.2; N,
+
8.8%; M+ ({[Mn(L6py2H)](OTf)} ) 472.1. C18H20F6MnN4O6S2
requires: C, 34.8; H, 3.2; N, 9.0%; M+, 472.1], [Found for
[Mn(L7py24-Me)](OTf)2: C21H26F6MnN4O6S2: C, 38.0; H, 3.8; N,
+
8.4%; M+ ({[Mn(L7py24-Me)](OTf)} ) 514.1. C18H20F6MnN4O6S2
requires: C, 38.0; H, 4.0; N, 8.4%; M+, 514.1], [Found for
[Mn(L8py2H)](OTf)2: C20H24F6MnN4O6S2: C, 37.1; H, 3.5; N,
+
8.5%; M+ ({[Mn(L8py2H)](OTf)} ) 500.1. C20H24F6MnN4O6S2
The Hammett study was preformed by reacting a 2.5
mM acetonitrile solution of [MnIII(O2)(L7py2H)]+ with ben-
zaldehyde, 4-methylbenzaldehyde, 4-methoxybenzaldehyde, or 4-
chlorobenzaldehyde under pseudo-first-order conditions, which
permitted the determination of kobs values by fitting the decrease
absorption band at ~23 000 cm-1.
requires: C, 37.0; H, 3.7; N, 8.6%; M+, 500.1], and [Found for
[Mn(L8py26-Me)](OTf)2: C22H28F6MnN4O6S2: C, 39.0; H, 4.0; N,
+
8.2%; M+ ({[Mn(L7py24-Me)](OTf)} ) 514.1. C22H28F6MnN4O6S2
requires: C, 39.0; H, 4.2; N, 8.3%; M+, 514.1].
X-Ray crystallography
Density functional theory calculations
Single
crystals
of
[Mn(L6py2H)(CH3CN)3](ClO4)2,
The ORCA 2.7 software package was used for all DFT
computations.32 Initial models of [MnIII(O2)(L7py2R)]+ (R = H,
4-Me, and 6-Me) were built using the X-ray coordinates of the
corresponding manganese(II) complexes, then adding side-on per-
oxo ligands, and then rotated the pyridines from a canted position
to a planar position. Geometry optimizations for each model of
the metal complexes were converged to the S = 2 spin system. These
calculations employed the Becke-Perdew (BP86) functional33,34
and the SVP (Ahlrichs split valence polarized)35,36 basis with
the SV/C auxiliary basis for all atoms except for manganese,
nitrogen, and oxygen, where the larger TZVP (Ahlrichs triple-z
valence polarized)20 basis in conjunction with the TZV/J auxiliary
basis were used. The resolution of identity (RI) approximation,
developed by Neese,37 was used for all calculations. Solvation
effects associated with acetonitrile (dielectric constant e = 36.6)
were incorporated using COSMO, as implemented in ORCA.38
[Mn(L7py24-Me)(OTf)2], [Mn(L7q2H)(ClO4)2], [Mn(L8py2H)(ClO4)2],
and [Mn(L8py26-Me)(OTf)2] were grown by vapor diffusion of
diethyl ether into an MeCN solution of the complex at room
temperature. Data collection and refinement parameters are
summarized in the ESI.†
In situ preparation of peroxomanganese(III) complexes
The green peroxomanganese(III) intermediates were formed
by treating a 2.5 mM MeCN solution of metal complex
with 5 equivalents H2O2 and 0.5 equivalents triethylamine at
-40 ◦C. The formation of the peroxomanganese(III) complexes
was evident from the appearance of characteristic bands in
the electronic absorption spectra and ESI-mass spectrometry.
+
+
[Found M+ {[MnIII(O2)(L7q2)]} : 469.1. {[MnIII(O2)(L7q2)]} re-
1714 | Dalton Trans., 2011, 40, 1707–1715
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
17 R. L. Shook, W. A. Gunderson, J. Greaves, J. W. Ziller, M. P. Hendrich
and A. S. Borovik, J. Am. Chem. Soc., 2008, 130, 8888–8889.
18 S. Groni, G. Blain, R. Guillot, C. Policar and E. Anxolabe´he`re-Mallart,
Inorg. Chem., 2007, 46, 1951–1953.
19 S. Groni, P. Dorlet, G. Blain, S. Bourcier, R. Guillot and E. Anxo-
labe´he`re-Mallart, Inorg. Chem., 2008, 47, 3166–3172.
20 R. A. Geiger, S. Chattopadhyay, V. W. Day and T. A. Jackson, J. Am.
Chem. Soc., 2010, 132, 2821–2831.
21 U. P. Singh, A. K. Sharma, S. Hikichi, H. Komatsuzaki, Y. Moro-oka
and M. Akita, Inorg. Chim. Acta, 2006, 359, 4407–4411.
22 N. Kitajima, H. Fukui and Y. Moro-oka, J. Am. Chem. Soc., 1990, 112,
6402–6403.
Acknowledgements
The authors acknowledge the University of Kansas for start-up
support, the KU New Faculty General Research Fund and the
American Chemical Society Petroleum Research Fund (50287-
DNI3) for support. The authors thank the National Science
Foundation (Grant CHE-00079282) and KU for funds to purchase
the X-ray instrument and computers.
References
23 J. Annaraj, Y. Suh, M. S. Seo, S. O. Kim and W. Nam, Chem. Commun.,
2005, 4529–4531.
1 M. Momenteau and C. A. Reed, Chem. Rev., 1994, 94, 659–698.
2 J. J. Woodward, M. M. Chang, N. I. Martin and M. A. Marletta, J.
Am. Chem. Soc., 2009, 131, 297–305.
3 A. J. Wu, J. E. Penner-Hahn and V. L. Pecoraro, Chem. Rev., 2004, 104,
903–938.
4 A.-F. Miller, Curr. Opin. Chem. Biol., 2004, 8, 162–168.
5 L. E. Grove and T. C. Brunold, Comments Inorg. Chem., 2008, 29,
134–168.
6 J. A. Cotruvo and J. Stubbe, Biochemistry, 2010, 49, 1297–1309.
7 A. Willing, H. Follmann and G. Auling, Eur. J. Biochem., 1988, 170,
603.
24 Y. Goto, S. Wada, I. Morishima and Y. Watanabe, J. Inorg. Biochem.,
1998, 69, 241–247.
25 S. E. Ghachtouli, A. Mohamadou and J.-P. Barbier, Inorg. Chim. Acta,
2005, 358, 3873–3880.
¨
26 S. I. M. Paris, U. A. Laskay, S. Liang, O. Pavlyuk, S. Tschirschwitz,
P. Lo¨nnecke, M. C. McMills, G. P. Jackson, J. L. Petersen, E. Hey-
Hawkins and M. P. Jensen, Inorg. Chim. Acta, 2010, 363, 3390–
3398.
27 S. Wang and T. D. Westmoreland, Inorg. Chem., 2009, 48, 719–727.
28 J. C. Dabrowiak, L. A. Nafie, P. S. Bryan and A. T. Torkelson, Inorg.
Chem., 1977, 16, 540–544.
29 W. L. F. Armarego, and D. D. Perrin, Purification of Laboratory
Chemicals, Butterworth-Heinemann, Oxford, U.K., 1997.
30 J. A. Halfen, J. M. Uhan, D. C. Fox, M. P. Mehn and L. Que, Jr., Inorg.
Chem., 2000, 39, 4913–4920.
8 C. Bull, E. C. Niederhoffer, T. Yoshida and J. A. Fee, J. Am. Chem.
Soc., 1991, 113, 4069–4076.
9 A. S. Hearn, C. K. Tu, H. S. Nick and D. N. Silverman, J. Biol. Chem.,
1999, 274, 24457–24460.
10 J. Hayden and M. Hendrich, JBIC, J. Biol. Inorg. Chem., 2010, 15,
31 R. Mayilmurugan, H. Stoeckli-Evans and M. Palaniandavar, Inorg.
Chem., 2008, 47, 6645–6658.
729–736.
11 N. Cox, H. Ogata, P. Stolle, E. Reijerse, G. Auling and W. Lubitz, J.
Am. Chem. Soc., 2010, 132, 11197–11213.
12 T. A. Jackson, A. Karapetian, A.-F. Miller and T. C. Brunold,
Biochemistry, 2005, 44, 1504–1520.
13 R. B. VanAtta, C. E. Strouse, L. K. Hanson and J. S. Valentine, J. Am.
Chem. Soc., 1987, 109, 1425–1434.
14 N. Kitajima, H. Komatsuzaki, S. Hikichi, M. Osawa and Y. Moro-oka,
J. Am. Chem. Soc., 1994, 116, 11596–11597.
15 M. S. Seo, J. Y. Kim, J. Annaraj, Y. Kim, Y.-M. Lee, S.-J. Kim, J. Kim
and W. Nam, Angew. Chem., Int. Ed., 2007, 46, 377–380.
16 J. Annaraj, J. Cho, Y.-M. Lee, S. Y. Kim, R. Latifi, S. P. de Visser and
W. Nam, Angew. Chem., Int. Ed., 2009, 48, 4150–4153.
32 F. Neese, ORCA - an ab initio, Density Functional and Semiempirical
Program Package, Version 2.7, University of Bonn, 2009.
33 A. D. Becke, J. Chem. Phys., 1986, 84, 4524–4529.
34 J. P. Perdew, Phys. Rev. B, 1986, 33, 8822–8824.
35 A. Scha¨fer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 97, 2571–
2577.
36 G. Scha¨fer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 100, 5829–
5835.
37 F. Neese, J. Comput. Chem., 2003, 24, 1740–1747.
38 S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen and F. Neese,
J. Phys. Chem. A, 2006, 110, 2235–2245.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 1707–1715 | 1715
©