Four-coordinate trispyrazolylboratomanganese and -iron complexes with a pyrazolato Co-ligand: Syntheses and properties as oxidation catalysts

Article abstract of DOI:10.1002/chem.201100343

A series of complexes of the type [(Tp^R1,R2)M(X)] (Tp=trispyrazolylborato) with R¹/R² combinations Me/tBu, Ph/Me, iPr/iPr, Me/Me and for M=Mn or Fe coordinating [Pz^Me,tBu] ^- (Pz=pyrazolato) or Cl

Full text of DOI:10.1002/chem.201100343

DOI: 10.1002/chem.201100343
Four-Coordinate Trispyrazolylboratomanganese and -iron Complexes with a
Pyrazolato Co-ligand: Syntheses and Properties as Oxidation Catalysts
Thomas Tietz, Christian Limberg,* Reinhard Stçßer, and Burkhard Ziemer[a]
Dedicated to Professor Hans-Joachim Freund on the occasion of his 60th birthday
Abstract: A series of complexes of the
type [(TpR1,R2)M(X)] (Tp=trispyrazo-
lylborato) with R1/R2 combinations
with cyclohexene oxide occasionally
formed as a side product. The ratios of
the different oxidation products varied
with the reaction conditions: in the
case of a peroxide/alkene ratio of 4:1,
considerably more ketone than alcohol
was obtained and cyclohexene oxide
formation was almost negligible,
whereas a ratio of 1:10 led to a signifi-
cant increase of the alcohol proportion
and to the formation of at least small
amounts of the epoxide. Pre-treatment
of O2, so that it may be argued that
tBuOOH and O2 both lead to the same
active species. The results of EPR spec-
troscopy and ESI-MS suggest that the
initial product of the reaction of
Me/tBu, Ph/Me, iPr/iPr, Me/Me and for
ꢀ
M=Mn or Fe coordinating [PzMe,tBu
]
(Pz=pyrazolato) or Clꢀ as co-ligand X
has been synthesised. Although the
chloride complexes were very unreac-
tive and stable in air, the pyrazolato
series was far more reactive in contact
with oxidants like O2 and tBuOOH.
ACTHNGUTERNNUG
[(TpMe,Me)Mn(PzMe,tBu)] with O2 con-
tains a MnIII(O)2MnIV core. Prolonged
exposure to O2 leads to a different di-
nuclear complex containing three O-
bridges and resulting in different
TOFs/product distributions. Analogous
findings were made for other com-
plexes and formation of these overoxi-
dised products may explain the devia-
tion of the catalytic performances if
the reactions are carried out in an O2
atmosphere.
ACHTUNGTRENNUNG
The [(TpR1,R2)M(PzMe,tBu)] complexes
proved to be active pre-catalysts for
the oxidation of cyclohexene with
tBuOOH, reaching turnover frequen-
cies (TOFs) ranging between moderate
and good in comparison to other man-
ganese catalysts. Cyclohexene-3-one
and cyclohexene-3-ol were always
found to represent the main products,
ACTHNGUTERNNUG
of the dissolved [(TpR1,R2)M(PzMe,tBu)]
pre-catalysts with O2 led to product
distributions and TOFs that were very
similar to those found in the absence
Keywords: catalysis · manganese ·
oxidation · peroxides · pyrazolates
Introduction
methane monooxygenase, acetylacetone dioxygenase, cate-
chol dioxygenases, and a-ketoglutarate-dependent iron en-
zymes.[3,4] In recent years, TpR1,R2 metal oxo research has
also been extended to the element manganese, with a focus
on dioxygen and peroxide reactivity.[5–8]
In the work presented herein, we have systematically
studied the behaviour of new, well-defined, electron-rich
TpR1,R2 manganese and iron complexes in contact with oxi-
dants as well as the reactivity of the oxidised complexes
with respect to organic substrates.
A significant amount of research has been carried out in the
past that exploited the favourable properties of the versatile
hydrotrispyrazolylborato
ligands
(TpR1,R2 [1] in transition-metal chemistry.
)
In the last decade these facially coordi-
nating tripodal N3 donors have also
been increasingly employed for the
complexation of metal oxo moieties in
the area of bioinorganic chemistry:
TpR1,R2 ligands were, for instance, chosen to model histidine-
rich coordination environments in certain oxygenating iron
and copper enzymes,[2] such as the tyrosinases, the soluble
Results and Discussion
Complex synthesis and characterisation: A big advantage of
the TpR1,R2 ligands is that the residues R1 and R2 can be
broadly varied. In the course of our studies we found that
the syntheses of the M’TpR1,R2 salts (M’=Na, K) needed as
ligand precursor compounds can be significantly improved
by the employment of a microwave apparatus. The reaction
time can be reduced from (in some cases) three days to one
hour, while yields can be increased from around 50 to
around 75%. The alkali metal salts (M’TpR1,R2 with M’=Na,
[a] Dipl.-Chem. T. Tietz, Prof. Dr. C. Limberg, Prof. Dr. R. Stçßer,
Dr. B. Ziemer
Department of Chemistry
Humboldt-Universitꢀt zu Berlin
Brook-Taylor-Strasse 2, 12489 Berlin (Germany)
Fax : (+49)030-2093-6966
E-mail: Christian.limberg@chemie.hu-berlin.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100343.
10010
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020

FULL PAPER
K and R1 =methyl, R2 =phenyl or tert-butyl, respectively, or
R1 =R2 =isopropyl) thus obtained were then reacted with
anhydrous MnCl2 in MeOH/CH2Cl2 or with anhydrous
FeCl2 in THF (Scheme 1) to yield the precursor complexes
binding of an additional ligand would be required, they
prefer a tetrahedral coordination sphere.
The compounds 1b,[10] 1c[11] and 2c[12] had been synthes-
ised and characterised earlier, and very recently during the
preparation of this manuscript 2a also appeared in the liter-
ature.[13] Derivatives [(TpMe,Me)Mn(Cl)] and [(TpMe,Me)Fe(Cl)]
could not be obtained, because the formation of the corre-
sponding homoleptic complexes [(TpMe,Me)2M] is favoured.
Compounds 1a–c and 2a–c are very unreactive. They can
be crystallised in contact with air, and they are also inert to-
wards water for longer periods of time. To increase the reac-
tivity of the [(TpR1,R2)M]+ complex metal fragments, the
Scheme 1. Synthesis of [(TpR1,R2)MII(Cl)] complexes 1a: R1 =methyl,
R2 =tert-butyl, M=Mn; 1b: R1 =methyl, R2 =phenyl, M=Mn; 1c: R1 =
isopropyl, R2 =isopropyl, M=Mn; 2a: R1 =methyl, R2 =tert-butyl, M=
Fe; 2b: R1 =methyl, R2 =phenyl, M=Fe; 2c: R1 =isopropyl, R2 =isopro-
pyl, M=Fe.
chlorido ligand in 1a–c and 2a–c was replaced by a pyrazo-
ꢀ
lato ligand [PzMe,tBu
]
by reacting the compounds with the
corresponding sodium salt (Scheme 2). This was supposed to
[(TpR1,R2)M(Cl)] (M=Mn (1a–c), Fe (2a–c)), crystals of
which could be obtained by recrystallisation of the products
from toluene. As an example, the molecular structure of
[(TpMe,tBu)Mn(Cl)] (1a) is shown in Figure 1. The metal ion
ACTHNUTRGNEUNG
Scheme 2. Synthesis of [(TpR1,R2)MII(PzMe,tBu)] complexes 3a: R1 =methyl,
R2 =tert-butyl, M=Mn; 3b: R1 =methyl, R2 =phenyl, M=Mn; 3c: R1 =
isopropyl, R2 =isopropyl, M=Mn; 4a: R1 =methyl, R2 =tert-butyl, M=
Fe; 4b: R1 =methyl, R2 =phenyl, M=Fe; 4c: R1 =isopropyl, R2 =isopro-
pyl, M=Fe.
increase the electron density at the metal centre, and
beyond that the more basic pyrazolato ligand (as compared
to the chlorido ligand) should allow for the introduction of
other (less basic) ligands in reactions with the corresponding
acids.
ACTHNUTRGNEUNG
The compounds of the type [(TpR1,R2)MII(PzMe,tBu)] (M=
Mn (3a–c), Fe (4a–c)) obtained by this procedure were iso-
Figure 1. Molecular structure of [(TpMe,tBu)Mn(Cl)] (1a). Hydrogen atoms
(except for the one belonging to the BH group) are omitted for clarity.
lated and fully characterised. In the case of the complex
ACTHNUTRGNEUNG
[(TpMe,tBu)Mn(PzMe,tBu)] (3a) crystals suitable for X-ray anal-
ꢀ
ꢀ
Selected bond lengths [ꢂ] and angles [8]: Cl Mn 2.2863(7), Mn N1
2.1468(10); N1-Mn-N1 91.45(4), N1-Mn-Cl 124.24(3), N2-B-N2
110.08(10).
ysis were obtained, and the molecular structure of 3a is
shown in Figure 2. The manganese centre is surrounded by
the four binding nitrogen atoms in a distorted tetrahedral
fashion. Due to the pyrazolato ligand, the C3 symmetry
ꢀ
is surrounded by the chlorido ligand and the three coordi-
nating N atoms of the TpMe,tBu ligand in a distorted tetrahe-
dral fashion, which is reasonable for a d5 high-spin configu-
ration. In comparison to the ideal bond angles within a tet-
rahedron, the Cl-Mn-N angles are enlarged (Cl-Mn-N1:
124.24(3)8), whereas the N-Mn-N angles are reduced (N1-
Mn-N1’: 91.44(4)8). It is worth pointing out that the synthe-
ses of 1a–c in the presence of stoichiometric amounts of 3-
methyl-5-tert-butylpyrazole (HPzMe,tBu; Pz=pyrazolato) did
not lead to the isolation of a pyrazole adduct, whereas the
analogous chromium(II) compound [(TpMe,tBu)Cr(Cl)] coor-
dinates HPzMe,tBu to give [(TpMe,tBu)Cr(Cl)(HPzMe,tBu)],[9] pre-
sumably as this results in ligand field stabilisation in the
case of the d4 chromium atom. As this is not possible for a
high-spin d5 manganese ion and as the Lewis acidity of the
manganese centres in 1a–c apparently is not so high that the
found for 1a–c is annulled, which results in differing Mn N
bond lengths and N-Mn-N bond angles.
As the corresponding chloride precursor complexes were
not available for the case of R1 =R2 =methyl (see above), a
different synthetic strategy was developed for the prepara-
ACTHNUTRGNEUNG
tion of [(TpMe,Me)MII(PzMe,tBu)]: MnCl2 or FeCl2, respectively,
were reacted with one equivalent of KTpMe,Me and one
equivalent of NaPzMe,tBu in THF in a one-pot procedure
(Scheme 3). Due to the low solubility of the undesired
[(TpR1,R2)2M] byproducts and the starting materials in THF,
we were able to isolate the complexes [(TpMe,Me)Mn-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(PzMe,tBu)] (3d) and [(TpMe,Me)Fe(PzMe,tBu)] (4d) in pure form.
All attempts to crystallise these compounds failed, but in
the presence of excess HPzMe,tBu crystals of the double-pyra-
ACTHNUTRGNEUNG
zole adduct of 3d [(TpMe,Me)Mn(PzMe,tBu)(HPzMe,tBu)2]
(3d(HPzMe,tBu)2) were obtained. The result of the X-ray crys-
Chem. Eur. J. 2011, 17, 10010 – 10020
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10011

C. Limberg et al.
ꢀ
ed pyrazole ligands are significantly longer than the Mn N
bond between the manganese centre and the pyrazolato
ligand.
EPR spectra of the manganese complexes 3a–d at 77 K
exhibit an intense and broad signal typical for mononuclear
manganese(II) complexes (for the EPR spectrum of com-
pound 3d, see Figure S1 in the Supporting Information).
The iron compounds 4a–d are EPR silent at 77 K, which
can be explained by a possible large zero field splitting of
the high-spin d6 centres. Crystals of 4a–d, suitable for molec-
ular structure analysis by means of X-ray diffraction, could
not be obtained. However, a Mçssbauer spectrum recorded
for 4a at room temperature showed a doublet in line with
the expectations (see Figure S2 in the Supporting Informa-
tion). The value for the isomeric shift of 0.823 mmsꢀ1 is typi-
cal for a four-coordinate high-spin FeII centre and resembles
the shift values characterising the signals of comparable
complexes (see Table 1), which, however, due to distortions
Figure 2. Molecular structure of [(TpMe,tBu)Mn
atoms (except for the one belonging to the BH group) are omitted for
clarity. Selected bond lengths [ꢂ] and angles [8]: Mn N1 2.1571(14), Mn
(PzMe,tBu)] (3a). Hydrogen
ACHTUNGTRENNUNG
ꢀ
ꢀ
ꢀ
ꢀ
N2 2.1872(15), Mn N3 2.1845(15), Mn N7 2.0535(16); N1-Mn-N2
88.70(5), N1-Mn-N3 84.98(6), N2-Mn-N3 97.86(5), N1-Mn-N7 128.72(6),
N2-Mn-N7 114.97(6), N3-Mn-N7 130.85(6).
Table 1. Comparison of the Mçssbauer parameters of 4a with those of
similar compounds.
Compound
d
DQ
Ref.
4a
0.823
0.98
1.09
1.10
0.72
0.870
3.69
3.28
3.41
0.71
[15]
[16]
[16]
[14]
[(Ph)2B
[(tha)2FeII(Cl)2][a]
[(thu)2FeII(Cl)2][b]
(NMe4)2A[FeCl4]
(PzH,Ph)2FeII(Cl)
ACHUTGTNRNEUNG AHCUTNGTREN(NUGN thf)]
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
G
CHTUNGTRENNUNG
[a] tha=thioacetamide. [b] thu=thiocarbamide.
ACTHNUTRGNEUNG
Scheme 3. Synthesis of the complexes [(TpMe,Me)MII(PzMe,tBu)] (3d: M=
Mn; 4d: M=Fe).
of the tetrahedral symmetry, often show much larger quad-
rupole splittings. The small quadrupole splitting found for
4a, which is more comparable to that of the halide complex
tal structure analysis is shown in Figure 3. The manganese
atom is located in a distorted octahedral coordination
sphere, and its bonds to the nitrogen atoms of the coordinat-
(NMe4)2ACHTUNGTRNEUNG
[FeCl4],[14] therefore indicates a quite symmetric
ligand sphere, similar to that of the manganese analogues
displayed in Figure 2.
Reactivity studies: As anticipated (see above), the com-
ACTHNUTRGNEUNG
plexes of the type [(TpR1,R2)M(PzMe,tBu)] (3a–4d) were found
to be much more reactive towards oxidants than the corre-
sponding chloride complexes 1a–2c. However, the products
obtained in reactions of 3a–4d with O2 in situ were not
found to be powerful oxidants: only very reactive organic
substrates such as tetramethylethene were oxygenated to
give the corresponding epoxides. Hence, the iron and man-
ganese complexes 3a–4d were tested with respect to their
potential to catalyse the oxidation of hydrocarbons with
tBuOOH. Cyclohexene was chosen as the substrate as it is
often used representatively for activity tests.[17–24] Two possi-
ble pathways are known for the reaction of complex
metal(II) fragments with tBuOOH,[21,25] which differ from
ꢀ
each other in the way the O O bond of the alkyl hydroper-
oxide is cleaved. Homolytic bond cleavage results in a
ACTHNUTRGNEUNG
Figure 3. Molecular structure of [(TpMe,Me)Mn(PzMe,tBu)(HPzMe,tBu)2]
(3d(HPzMe,tBu)2). Hydrogen atoms (except for those belonging to BH and
metalACTHNURTGNEU(GN III) hydroxido complex and an alkoxo radical that is
NH groups) are omitted for clarity. Selected bond lengths [ꢂ] and
able to abstract a hydrogen atom from the substrate, thus
triggering the radical chain reaction described by Equa-
tions (1)–(10).[26,27] These usually lead to a-hydroxyalkenes,
ꢀ
ꢀ
ꢀ
angles [8]: Mn N3 2.289(4), Mn N9 2.197(4), Mn N11 2.343(4); N1-Mn-
N3 83.61(14), N1-Mn-N11 90.03(14), N3-Mn-N9 177.52(15), N2-B1-N4
109.2(4).
10012
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020

Trispyrazolylborato Manganese and Iron Complexes with a Pyrazolato Co-ligand
FULL PAPER
a,b-unsaturated carbonyl compounds and minor amounts of
epoxide (often epoxides are missing completely in the prod-
uct palettes).
pathway, as epoxides could hardly be detected among the
products.
When the reactions were carried out under the set of re-
action conditions (1), almost no epoxide but only the prod-
ucts originating from an a oxidation of cyclohexene were
obtained (Table 2 showing turnover numbers (TONs), and
Figure 4 and Figure 5 showing turnover frequencies
(TOFs)).
Radical chain initiation
LMII þ tBuOOH ! LMIII þ tBuO þ OHꢀ
ð1Þ
ð2Þ
C
LMIII þ tBuOOH ! LMII þ tBuOO þ Hþ
C
Table 2. Catalytic oxidation of cyclohexene with tBuOOH.[a]
Pre-
catalyst
Temperature/
solvent
Overall
TON
CyEp
TON
CyOH
TON
CyO
TON
Radical chain
3a
3a
3a
3a
3b
3c
3d
4a
4a
4a
4a
4b
4c
4d
408C/MeCN
ꢀ408C/MeCN
208C/CH2Cl2
ꢀ408C/CH2Cl2
408C/MeCN
408C/MeCN
408C/MeCN
408C/MeCN
ꢀ408C/MeCN
208C/CH2Cl2
ꢀ408C/CH2Cl2
408C/MeCN
408C/MeCN
408C/MeCN
16.9
1.2
22.5
2.7
24.8
23.9
13.3
4.6
6.4
5.4
7.8
5.0
0
0
0
0
0
0
0
0
0
0
0
0
0.2
0
2.2
0.9
3.5
1.5
0.8
2.4
0.8
1.6
1.0
2.7
2.3
0.5
1.2
3.8
14.7
0.3
19.0
1.2
24.0
21.5
12.5
3.1
5.4
2.7
5.6
4.6
tBuO þ tBuOOH ! tBuOH þ tBuOO
ð3Þ
ð4Þ
ð5Þ
ð6Þ
C
C
2 tBuOO ! 2 tBuO þ O2
C
C
tBuO þ RH ! tBuOH þ R
C
C
R þ O2 ! ROO
C
C
Radical chain termination
7.2
26.9
5.8
23.1
ROO þ ROO ! R0 ¼ O þ ROH þ O2
ð7Þ
ð8Þ
C
C
[a] Catalyst concentration: 1.3 mol% with respect to alkene; ratio of per-
oxide/alkene=4:1; reaction time: 2 h. CyEp: cyclohexene oxide, CyOH:
cyclohexene-3-ol, CyO: cyclohexene-3-one.
ROO þ tBuOO ! R0 ¼ O þ tBuOH þ O2
C
C
R þ ROO ! ROOR
ð9Þ
C
C
R þ tBuOO ! tBuOOR
ð10Þ
C
C
Heterolytic bond cleavage results in a metal(IV) oxido
complex[21,27,28] that oxidises the organic substrate, typically
leading to high yields in epoxide and hardly any a-oxidised
alkenes (though Nam and co-workers proved the existence
of a third principal pathway:[25] a porphinatoiron(IV) oxido
complex was shown to abstract an H atom from a second
equivalent of alkyl peroxide, thereby initiating a radical
chain reaction).
The catalytic oxidations with compounds 3a–4d were car-
ried out under two basically different sets of reaction condi-
tions:
Figure 4. Turnover frequencies (TOFs) reached in the catalytic oxidation
of cyclohexene with tBuOOH; ratio of peroxide/alkene=4:1; catalyst
concentration: 1.3 mol% with respect to cyclohexene; reaction time: 2 h;
temperature: 408C; solvent: MeCN.
1) The concentration of oxidant used was higher than the
concentration of the substrate ([tBuOOH]/[cyclohex-
ene]=4:1), and the concentration of the pre-catalyst em-
ployed was comparatively high (0.325 mol% relating to
tBuOOH, 1.300 mol% relating to cyclohexene).
2) The concentration of oxidant used was lower than the
concentration of the substrate ([tBuOOH]/[cyclohex-
ene]=1:10), and the concentration of the pre-catalyst
employed was comparatively low (0.160 mol% relating
to tBuOOH, 0.016 mol% relating to cyclohexene).
The data suggest that there is no significant influence of
the substituents at the TpR1,R2 ligands on the product distri-
butions and only a small influence on the overall TONs of
the oxidised products (only in the iron series one complex
(4d) sticks out). Both observations can be rationalised if a
radical chain reaction is responsible for the product forma-
tion. If the less polar solvent CH2Cl2 is employed instead of
acetonitrile, TONs increase, which is plausible, too (and in
accordance with literature results[18–20,22]): Coordination of
MeCN to the metal centre slows down the radical chain ini-
tiation reactions [Eqs. (1), (2)], whereas CH2Cl2 is a non-co-
ordinating solvent and thus allows for free access of the oxi-
Irrespective of the reaction conditions, the iron and man-
ganese complexes generated by oxidation of 3a–4d present-
ed herein apparently mediate the radical chain reaction
Chem. Eur. J. 2011, 17, 10010 – 10020
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10013

C. Limberg et al.
Figure 5. TOFs reached in the catalytic oxidation of cyclohexene with
tBuOOH; ratio of peroxide/alkene=4:1; catalyst concentration:
1.3 mol% with respect to cyclohexene; reaction time: 2 h; DCM: di-
chloromethane.
Figure 6. TOFs of the catalytic oxidation of cyclohexene with tBuOOH;
catalysts used: 3a (left columns), 4a (right columns); ratio of peroxide/
alkene=1:10; catalyst concentration: 0.16 mol% with respect to the per-
oxide; reaction time: 2 h.
dant to the metal centre. When tBuOOH is added to the
prepared acetonitrile solutions containing the pre-catalysts
4a–c at 408C, immediate gas formation (O2) is observed.
Probably the iron catalysts formed are much more reactive
than the manganese catalysts, and in combination with a
high oxidant concentration Equations (3) and (4) become
very dominant at 408C, thereby resulting in the decomposi-
tion of a significant part of the oxidant, which is converted
into O2. Consistently, slowing down the reaction by cooling
to ꢀ408C prevented the formation of huge amounts of di-
oxygen and resulted in high TONs compared, for instance,
to those observed for the corresponding manganese complex
3a. In line with this hypothesis, reduction of the ratio of
tBuOOH to cyclohexene led to comparatively high TONs
for the same catalyst (see below).
tions (1): larger amounts of cyclohexene-3-ol were obtained.
Apparently, at low catalyst and high substrate concentra-
tions the reactions of the radical chain become more domi-
nant that lead to the formation of the alcohol, and its fur-
ther oxidation is suppressed.
To elucidate how far the pyrazolato co-ligands are impor-
tant for the catalytic activity, complexes 1a–c were also
tested. Notably, the activity was far lower, partly only slight-
ly above stoichiometric turnover.
The catalytically active species: Because of the high sensitiv-
ity of 3a–d and 4a–d towards oxidants, it is reasonable to
suggest the initial formation of an oxidised complex, which
then represents the active oxidation catalyst. Bubbling of O2
through the reaction mixture prior to the addition of
tBuOOH did not influence the yields of the oxidised prod-
ucts (Table 3, Figure 6), although the complexes are sensi-
tive to O2, so it stands to reason that identical oxidised com-
plexes are generated by the addition of either the alkyl per-
oxide or O2 (of course, in principle the O2 product could
also represent an intermediate on the way to the species
formed with tBuOOH).
When the reactions were carried out with the set of reac-
tion conditions (2), again hardly any epoxide but only the
products of the a oxidation of cyclohexene were obtained
(Table 3 and Figure 6). However, the product distribution
differed from the one observed under the reaction condi-
Table 3. Catalytic oxidation of cyclohexene with tBuOOH.[a]
We assume that these oxidised complexes contain more
than one metal centre, because Kitajima and Moro-oka
Pre-cata-
lyst
Temperature/
solvent
Overall CyEp CyOH CyO
showed that oxidation of the complex [(TpiPr,iPr)MnII
ACHTUNGTRENNUNG
TON
TON TON TON
3a
3a
3a
208C/CH2Cl2
408C/MeCN
408C/MeCN/O2 pre-treat-
ment
408C/MeCN/O2 as co-oxi-
dant
208C/CH2Cl2
408C/MeCN
408C/MeCN/O2 pre-treat-
ment
408C/MeCN/O2 as co-oxi-
dant
42.9
43.9
44.9
3.9
0
0
27.6
19.3
13.3
11.4
24.6
31.6
G
A
mononuclear complex with a peroxido ligand, which is coor-
dinated side-on to the manganese centre, could only be iso-
lated when free pyrazole (HPziPr,iPr) was present in solution
during the oxidation reaction of I, so that it could saturate
the coordination sphere of the manganese atom of the initial
3a
53.1
3.2
49.4
0.6
4a
4a
4a
88.8
106.3
114.4
4.8
3.3
4.8
38.7
35.6
47.8
45.4
67.4
61.8
product
formed.[5]
The
resulting
complex
[(TpiPr,iPr)MnIII(HPziPr,iPr)(O2)] is stabilised by a hydrogen
bond between the NH unit of the pyrazole ligand and the
oxygen atoms of the peroxido ligand; nevertheless it decom-
poses rapidly at room temperature. Mononuclear iron per-
4a
61.5
2.6
49.5
9.4
[a] Catalyst concentration: 0.16 mol% with respect to the peroxide; ratio
of peroxide/alkene=1:10; reaction time: 2 h. For definitions, see Table 2.
10014
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020

Trispyrazolylborato Manganese and Iron Complexes with a Pyrazolato Co-ligand
FULL PAPER
oxide complexes of the type [Fe(TpiPr,tBu
[Fe(TpiPr,tBu
)ACHTUNGTRENNUNG(OOtBu)] could only be obtained by employing
)ACHTNGUTERN(UNG OOH)] (II) and
the strongly sterically demanding TpiPr,tBu ligand at low tem-
peratures (ꢀ808C), as shown by the work of Moro-oka and
Que and their co-workers.[4]
The presence of larger amounts of dioxygen during the
oxidation reaction seems to somewhat inhibit the formation
of the ketone, whereas more cyclohexenol is formed
(Table 3). One could argue that, following Equations (7)
and (8), the presence of excessive dioxygen may avert the
formation of the ketone, but this should generally be the
case and there are differing reports in the literature.[22,29]
Hence, we assumed that an excess of dioxygen might change
the catalytic centres and performed further investigations,
presented below.
To obtain further information about the nature of the cat-
alytically active species in our system, the products of the
reactions of 3a–d and 4a–d with dioxygen were inspected
more closely. In some cases over a longer period of time
crystals could be grown, which, however, were very sensitive
after isolation from the mother liquor so that, except for the
case of 3d, all attempts to perform an X-ray crystal struc-
ture determination failed. After oxidation of compound 3d
crystals were obtained that could be analysed, but still the
solution did not reach a quality that would allow for a de-
tailed discussion of bond lengths and angles. Nevertheless,
the identity of the oxidation product 5 as a dinuclear
oxygen-bridged manganese complex, as shown in Figure 7, is
Figure 8. a) EPR spectrum of a solution of 3d in MeCN after treatment
with O2 at 77 K and b) its simulation; gx =2.0014, gy =2.0030, gz =1.9865;
A
A
N
,
A
E
,
A
E
,
;
[MnIV]=75.3ꢃ10ꢀ4, A
yA
S
2
Figure 7. Heavy-atom framework of the molecular structure of 5.
beyond doubt. To determine the nature of the bridging li-
gands in this oxidised complex, which does not become ob-
vious from the X-ray data, resonance Raman spectra were
recorded. Because of occurring self-heating processes or un-
favourable excitation frequencies, the spectra did not reveal
any reliable information. The results of investigations by in-
frared, mass and EPR spectroscopy suggested that the crys-
talline samples isolated after the oxidation of 3a–4d with O2
do not correspond to the catalytically active species which
are formed originally from the pre-catalysts 3a–4d under
the reaction conditions used and which are responsible for
the TONs in Table 2. The spectra, recorded shortly after the
addition of O2 to solutions of the pre-catalysts, differ from
those of the crystalline oxidation products of 3a–4d (Fig-
ures 8 and 9; for IR spectra, see the Supporting Information
Figure S3).
Figure 9. High-resolution mass spectra of a solution of 3d in MeCN
shortly after addition of 16O2 (black) or 18O2 (blue) and of a solution of 5
in MeCN.
Nevertheless, the catalytically active species generated ini-
tially appear to represent oxygen-bridged dinuclear metal
complexes as well, because the EPR spectrum of a solution
Chem. Eur. J. 2011, 17, 10010 – 10020
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10015

C. Limberg et al.
of 3d measured shortly after the addition of O2 shows a
16 line signal, typical for dinuclear MnIII–MnIV com-
plexes[30,31] (Figure 8). By contrast, compound 5 is EPR-
silent. The high-resolution electrospray ionisation/atmos-
pheric-pressure chemical ionisation (ESI/APCI) mass spec-
trum (Figure 9, top) measured shortly after the addition of
O2 to a solution of 3d in MeCN supports the formation of a
tions (2). The distributions of the oxidation products ob-
tained are very different from those found in the case of 3a
or 4a (Table 4), which indicates again that 3aOx–3cOx, 5
Table 4. Catalytic oxidation of cyclohexene with tBuOOH.[a]
A
Overall CyEp CyOH CyO
TON
TON TON TON
dinuclear oxido-bridged manganeseACHTUNTRGNEUNG(III/IV) complex, as sug-
3a
408C/MeCN/O2 as co-oxi- 53.1
3.2
49.4
0.6
gested by the EPR spectra and additionally points to two
bridging oxido ligands. A signal with a maximum at m/z
736.2674 can be assigned to the ion [(TpMe,Me)MnIII(O)2MnIV-
dant
3aOx
4a
408C/MeCN
408C/MeCN/O2 as co-oxi- 61.5
45.2
8.5
2.6
30.7
49.5
6.0
9.4
ACHTUNGTRENNUNG
(TpMe,Me)]+; further signals can be assigned to Na+ and pyra-
dant
4aOx
5
408C/MeCN
408C/MeCN
89.4
168.0
13.7
12.2
58.6
77.7
17.2
78.2
zole adducts of this ion. This interpretation is corroborated
by the fact that on usage of 18O2 for the same investigation,
the corresponding peaks appeared shifted by 4 mass units as
expected for the incorporation of two O atoms (blue signals
in Figure 9, top).[32] The mass spectrum of 5 (Figure 9,
bottom) strongly differs from the spectrum of 3d shortly
after addition of O2. It is in agreement with the molecular
structure shown in Figure 7 and suggests the constitution
[a] Catalyst concentration: 0.1–0.14 mol% with respect to the peroxide;
ratio of peroxide/alkene=1:10; reaction time: 2 h. TONs of the dinuclear
complexes are calculated per metal centre for better comparison. For
definitions, see Table 2
and 4aOx–4dOx are not the active catalysts formed initially
during the catalytic oxidations with tBuOOH, but secondary
products originating from the latter after prolonged expo-
sure to O2. The fact that the catalytic results obtained on
employment of 3aOx, 4aOx and 5 resemble those obtained
for 3a or 4a, respectively, in the presence of O2 as a co-oxi-
dant can be explained by assuming that the formation of
3aOx–3cOx, 5 and 4aOx–4dOx under catalytic conditions
(408C/O2) is much faster than under those that had led to
their isolation, so that 3aOx and 4aOx significantly contrib-
ute to the catalytic activity of the systems 3a/tBuOOH/O2
and 4a/tBuOOH/O2, respectively.
[(TpMe,Me)Mn(O)3Mn
(TpMe,Me)], as it is dominated by a peak
ACHTUNGTRENNUNG
at m/z 751.2692 that deviates by one mass unit from the mo-
ACTHNUTRGNEUNG
lecular ion [(TpMe,Me)Mn(O)3Mn(TpMe,Me)]+ and may be ex-
plained by the abstraction of Hꢀ from 5 in the mass spec-
trometer. Besides, only one further signal can be found at
m/z 737.2896 corresponding to [(TpMe,Me)MnIII(O)2MnIII-
ACHTUNGTRENNUNG
(TpMe,Me)]+H+. The conclusions drawn from these measure-
ments are summarised in Scheme 4.
To further prove the hypothesis that the isolated oxidation
products of 3a–4d—analogues of 5—are not the active cata-
lysts generated from the pre-catalysts 3a–4d, the final oxi-
dation products of 3a (3aOx) and 4a (4aOx) as well as 5
were also investigated as catalysts for the oxidation of cyclo-
hexene with tBuOOH under the set of reaction condi-
Conclusion
ACTHNUTRGNEUNG
A new class of metal complexes, [(TpR1,R2)M(PzMe,tBu)] (M=
Mn, Fe), which show a very high reactivity towards oxidants,
has been synthesised and characterised. The complexes
proved to be catalytically active in the oxidation of cyclo-
hexene with tBuOOH to give mainly cyclohexene-3-one and
cyclohexene-3-ol and minor amounts of cyclohexene oxide,
respectively, and reached TOFs between average and good
relative to other manganese complexes reported in the liter-
ature (Table 5). Catalysis is based on a radical-type mecha-
nism, and hence there is only a small influence of the sub-
stituents at the Tp ligand on the TONs, which can be ex-
plained by the effect the different steric demands of the sub-
stituents have on the approach of the oxidant to the metal
centre. The significant influence of the solvent can be ex-
plained along the same lines because, contrary to dichloro-
methane, acetonitrile can coordinate to the metal centre.
Additionally, the different stabilities of the radicals formed
during the reaction in the different solvents could be a
reason for the different catalytic activities. The results of
EPR spectroscopy as well as ESI-MS studies indicate that
the active catalysts formed in situ consist of dinuclear com-
plexes containing two bridging oxido ligands. Exposure of
Scheme 4. Proposed behaviour of the complexes 3a–d and the catalytical-
ly active species in contact with O2.
10016
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020

Trispyrazolylborato Manganese and Iron Complexes with a Pyrazolato Co-ligand
FULL PAPER
Table 5. Comparison of overall TOFs found for selected manganese com-
plexes that catalyse the oxidation of cyclohexene with hydroperoxides.
X-ray structure determination: The crystals were mounted on a glass
fibre and then transferred into the cold nitrogen gas stream of a Stoe
IPDS2T diffractometer equipped with MoKa radiation (l=0.71073 ꢂ).
The structures were solved by direct methods (SIR 2004)[35] and refined
versus F2 (SHELXL-97)[36] with anisotropic temperature factors for all
non-hydrogen atoms (Table 1). All hydrogen atoms were added geomet-
rically and refined by using a riding model. CCDC-811467 (1a), 811468
(3d) and 811469 (3a) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif.
Catalyst
Oxidant/conditions
Overall Ref.
TOFs
[h]
3b
4d
3a
3a
4a
5
tBuOOH, MeCN, 408C[a]
tBuOOH, MeCN, 408C[a]
tBuOOH, CH2Cl2, 208C[a]
tBuOOH, CH2Cl2, 208C[b]
tBuOOH, MeCN, 408C[b]
12.4
13.5
11.3
21.5
53.2
84.5
43.8
0.5
5.0
0.4
tBuOOH, MeCN, 408C[b]
Materials: Solvents were dried using
a Braun Solvent Purification
[c]
[23]
A
E
(tacn)}] O2/tBuOOH, hot C7F16
System. Potassium hydrotris(3,5-dimethylpyrazolyl)-borate and 3,5-diiso-
propylpyrazole were purchased from Aldrich; 3,5-methylphenylpyra-
zole,[37] [(TpiPr,iPr)Mn(Cl)][8] and [(TpPh,Me)Mn(Cl)][10] were synthesised as
described earlier. The syntheses of the alkali metal hydrotrispyrazolylbo-
rate[38] compounds of 3,5-diisopropylpyrazole, 3,5-tert-butylmethylpyra-
zole and 3,5-phenylmethylpyrazole, as well as 3,5-tert-butylmethylpyra-
[22]
[Mn
[Mn
[Mn
[Mn
G
G
tBuOOH, MeCN, 258C[d]
O2/tBuOOH, MeCN, 258C[e]
tBuOOH, MeCN, 258C[d]
[22]
R
[22]
E
G
[22]
N
R
O2/tBuOOH, MeCN, 258C[e] 14.2
[24]
[20]
[18]
[29]
A
E
tBuOOH, CH2Cl2, 258C[f]
tBuOOH, CH2Cl2, 408C[g]
tBuOOH, CH2Cl2, 408C[h]
tBuOOH, cyclohexene,
608C[i]
14.0
0.7
51.0
35.0
[Mn
[Mn
[Mn
ACHTUNGTRENNUNG
zole,[39] [(TpiPr,iPr)Fe(Cl)][12] and [(TpMe,tBu)Fe(Cl)][13]
published procedures.
, corresponded to
G
ACHTUNGTRENNUNG
Alkali metal hydrotrispyrazolylborate, MHBACTHNUTRGNEUNG
(PzR1,R2)3: In a typical ex-
[33]
[33]
[34]
periment the corresponding pyrazole (36.0 mmol, 3 equiv) was combined
with the corresponding alkali metal tetrahydroborate (12.0 mmol,
1 equiv) in a Schlenk flask equipped with an excess pressure valve. The
flask was then heated to 2058C in a microwave apparatus and kept at
this temperature for 60 min under vigorous stirring. After cooling the re-
sulting light brown slurry to approximately 1008C, toluene (ca. 30 mL)
was poured into the flask. The slightly turbid solution was filtered off,
and the solvent was removed at reduced pressure. Excess pyrazole was
removed by means of sublimation (1308C, 1ꢃ10ꢀ2 mbar). Further subli-
mation of the resulting brown solid (2208C, 8ꢃ10ꢀ3 mbar) afforded the
corresponding alkali metal hydrotrispyrazolylborate in pure form (with
yields ranging between 65 and 75%) as a white solid.
[Mn(Cl)
[Mn(Cl)
N
CumylOOH, CH2Cl2,
24.0
192.0
120.3
208C[j]
CumylOOH, CH2Cl2,
208C[k]
[Mn
(bpy)
E
G
tBuOOH, MeCN, 08C[l]
[a] 16 mmol cat., 1.26 mmol substrate, 5 mmol tBuOOH, 2 h. [b] 1.6 mmol
cat., 10 mmol substrate, 1 mmol tBuOOH, 2 h. [c] 3.5 mmol cat., 2 mL
substrate, 72 mmol tBuOOH, excess O2, 3 h. [d] 5 mm cat., 1m substrate,
20 mm tBuOOH, 3 h. [e] Like [d] but additionally 1 atm O2. [f] 0.25 mm
cat., 0.3m substrate, 22 mm tBuOOH, 1.5 h. [g] 0.74 mmol cat., 10 mmol
substrate, 16 mmol tBuOOH, 8 h. [h] 5 mmol cat., 1 mL substrate, 2 mL
tBuOOH, 8 h. [i] 0.1 mmol cat, stepwise per 5 mmol tBuOOH, reaction
under air. [j] Substrate/cumylOOH/cat.=50:5:1, 5 min. [k] Substrate/cu-
mylOOH/cat./HIm=50:5:1:10, 5 min. [l] 2 mmol cat., 8 mmol substrate,
16 mmol tBuOOH, 6 h. Tacn=triazacyclononane, acac=acetylacetonate,
H2Mabenzil=bis(2-mercaptoanil)benzil, pbi=polybenzimidazole, tpp=
tetraphenylporphyrin, HIm=imidazole, bpy=2,2’-bipyridine.
Na/K[HBACHTUNGTRNEUNG
(PztBu,Me)3]: 1H NMR (400 MHz, CDCl3): d=1.23 (s, 9H; tBu),
2.39 (s, 3H; methyl), 5.71 ppm (s, 1H; CH); 13C NMR (100 MHz,
CDCl3): d=13.5 (CH3methyl), 30.9 (CH3tBu), 31.6 (CqtBu), 100.4 (CHarom.),
143.9 (Cqarom.), 160.1 ppm (Cqarom.); IR (KBr): n˜ =2961 (s), 2927 (s), 2904
(m), 2864 (m), 2494 (m), 1648 (m), 1539 (s), 1486 (m), 1460 (m), 1428
(m), 1361 (m), 1345 (m), 1337 (m), 1242 (m), 1183 (s), 1173 (m), 1109
(w), 1067 (s), 1023 (w), 1008 (w), 982 (w), 843 (w), 801 (m), 778 (s), 731
(w), 654 (m), 641 cmꢀ1 (m); ESI/APCI-MS (neg, THF): m/z (%): calcd
for [Tp]ꢀ: 423.3443; found: 423.3374; elemental analysis calcd (%) for
C24H40N6BK: C 62.32, H 8.72, N 18.17; found: C 60.45, H 8.73, N 17.52.
these complexes to O2 over longer periods of time or at ele-
vated temperatures yields dinuclear products containing
three oxido bridges which, when employed as catalysts, lead
to different product distributions (in favour of the alcohol).
Na/K[HB(PziPr,iPr)3]: 1H NMR (400 MHz, CDCl3): d=1.19 (d, J=
6.80 Hz, 12H; CH3iPr), 2.83 (sept, J=6.80 Hz, 2H; CHiPr), 5.79 ppm (s,
1H; CH); 13C NMR (100 MHz, CDCl3): d=22.6 (CHiPr), 23.1 (CH3iPr),
97.8 ppm (CHarom.); IR (KBr): n˜ =3373 (vw), 2962 (s), 2929 (s), 2869 (m),
2464 (w), 1567 (w), 1530 (m), 1459 (m), 1424 (w), 1379 (m), 1367 (m),
1361 (m), 1302 (m), 1186 (m), 1173 (m), 1137 (w), 1105 (w), 1073 (w),
1047 (m), 1004 (m), 991 (w), 792 (s), 786 (s), 721 (w), 663 cmꢀ1 (w); ESI/
APCI-MS (neg, THF): m/z (%): calcd for [Tp]ꢀ: 565.3913; found:
565.3504; elemental analysis calcd (%) for C27H46N6BK: C 64.27, H 9.19,
N 16.64; found: C 64.60, H 9.48, N 16.73.
Experimental Section
Apart from the ligand synthesis all manipulations were carried out in a
glovebox, or else by means of Schlenk-type techniques involving the use
of a dry argon atmosphere. 1H and 13C NMR spectra were recorded on a
Bruker AV 400 NMR spectrometer (1H 400.13 MHz; 13C 100.63 MHz)
with CDCl3 as solvent at 208C. 1H and 13C NMR spectra were calibrated
against the residual proton and natural abundance 13C resonances of the
deuterated solvent (CDCl3 dH =7.26, dC =77.0 ppm; C6D6 dH =7.15, dC =
128.0 ppm). Microanalyses were performed on a Leco CHNS-932 or HE-
KAtech EURO elemental analyser. ESI mass spectra were recorded on
an Agilent Technologies 6210 time-of-flight LC-MS instrument. IR spec-
tra were recorded for samples prepared as KBr pellets with a Digilab Ex-
calibur FTS 4000 FTIR spectrometer. EPR spectra were recorded of
probes within a silica glass tube on an ERS 300 X-band EPR spectrome-
ter at 9.2 GHz. Mçssbauer spectra were recorded at the Bundesanstalt
fꢄr Materialforschung und -prꢄfung in Berlin Adlershof on a constantly
accelerating Mçssbauer spectrometer. The source employed was 57Co in
Pt, and the detector was a proportionality counter at 1950 V, which was
calibrated with Fe2O3.
Na/K[HBACHTUNGTRNEUNG
(PzPh,Me)3]: 1H NMR (400 MHz, CDCl3): d=2.31 (s, 9H;
methyl), 6.26 (s, 3H; CH), 7.18 (m, 1H; CHarom.), 7.28 (m, 2H; CHarom.),
7.54 ppm (m, 2H; CHarom.); 13C NMR (100 MHz, CDCl3): d=12.7
(CH3methyl), 102.5 (CHarom.), 125.3 (CHPh), 125.5 (CHPh), 128.7 (CHPh),
131.2 (CqPh), 143.1 (Cqarom.), 149.8 ppm (Cqarom.); IR (KBr): n˜ =3200 (br,
m), 2957 (m), 2927 (m), 2858 (m), 2449 (m), 1952 (vw), 1888 (vw), 1827
(vw), 1634 (m), 1603 (m), 1541 (m), 1508 (m), 1494 (w), 1469 (m), 1454
(m), 1429 (w), 1412 (s), 1343 (m), 1219 (m), 1177 (s), 1071 (s), 1026 (m),
961 (m), 914 (w), 841 (w), 797 (m), 763 (s), 696 (s), 661 (w), 639 cmꢀ1
(m); ESI/APCI-MS (neg, THF): m/z (%): calcd for [Tp]ꢀ: 483.2504;
found: 483.2452; elemental analysis calcd (%) for C30H28N6BK: C 68.96,
H 5.40, N 16.08; found: C 68.82 H 5.40 N 16.07.
AHCTUNGTRENNUNG
A
)
3
Chem. Eur. J. 2011, 17, 10010 – 10020
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10017

C. Limberg et al.
solid started to precipitate immediately, and the reaction mixture was
stirred at room temperature for 48 h. Subsequently, the solvent was re-
moved, and the resulting light brown solid was suspended in dichlorome-
thane (15 mL) and acetonitrile (15 mL). After stirring for 30 min, the
white suspension was allowed to settle for an additional 30 min before
being filtered. The filtrate was concentrated to about 10 mL, which re-
sulted in the formation of a light rose coloured precipitate. The reaction
mixture was stored at ꢀ308C for 15 h and subsequently filtered. Drying
of the filter cake in a vacuum gave [(TpR1,R2)Mn(Cl)] (with yields ranging
between 50 and 70%) as a light rose coloured solid.
N
G
A
solution of KHB
A
)
(500 mg,
3
(PztBu,Me)] (480 mg,
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[(TptBu,Me)Mn(Cl)] (1a): IR (KBr): n˜ =2935 (s), 2930 (m), 2906 (m), 2865
(m), 2575 (m), 1537 (s), 1508 (m), 1471 (m), 1427 (m), 1382 (m), 1363 (s),
1336 (m), 1241 (m), 1180 (s), 1124 (w), 1069 (s), 1027 (m), 984 (w), 852
(w), 807 (m), 791 (s), 765 (s), 680 (w), 648 cmꢀ1 (m); ESI/APCI-MS (pos,
THF): m/z (%): calcd for [TpMnCl]H+: 514.2591; found: 514.2633; ele-
mental analysis calcd (%) for C24H40N6BMnCl: C 56.10, H 7.85, N 16.36,
Cl 6.90; found: C 56.10, H 7.85, N 16.33, Cl 7.28.
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
of FeCl2 (2 mmol, 1 equiv) in THF. The FeCl2 immediately started to dis-
solve, and a white solid started to precipitate. The reaction mixture was
stirred at room temperature for 48 h. Subsequently, the solvent was re-
moved and the resulting light brown solid was suspended in dichlorome-
thane (15 mL) and acetonitrile (15 mL). After stirring for 30 min the
ochre suspension was allowed to settle for an additional 30 min before
being filtered. The filtrate was concentrated to about 10 mL, which re-
sulted in the formation of a light ochre coloured precipitate. The reaction
mixture was stored at ꢀ308C for 15 h and subsequently filtered. Drying
of the filter cake in a vacuum gave [(TpR1,R2)Fe(Cl)] (with yields ranging
between 50 and 70%) as a light ochre coloured solid.
(PztBu,Me)] (with yield ranging
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[(TptBu,Me)Fe(Cl)] (2b): IR (KBr): n˜ =3274 (m), 3061 (m), 3042 (m), 2980
(m), 2961 (m), 2927 (m), 2548 (m), 1945 (vw), 1878 (vw), 1801 (vw), 1605
(w), 1566 (m), 1543 (s), 1506 (m), 1495 (m), 1474 (s), 1450 (s), 1436 (s),
1415 (s), 1364 (m), 1341 (m), 1306 (m), 1284 (w), 1268 (w), 1192 (s), 1174
(s), 1069 (s), 1019 (m), 976 (m), 915 (m), 831 (m), 797 (m), 778 (s), 761
(s), 694 (s), 658 (m), 637 cmꢀ1 (m); ESI/APCI-MS (pos, THF): m/z (%):
calcd for [TpFeCl]H+: 575.1662; found: 575.1897; elemental analysis
calcd (%) for C30H28N6BFeCl: C 62.70, H 4.91, N 14.62, Cl 6.17; found: C
62.95, H 5.03, N 14.27, Cl 6.54.
ACHUTNGRENNUG CAHTUNGTRENNUGN
[(TpPh,Me)Fe(PztBu,Me)] (4b): IR (KBr): n˜ =3060 (m), 2961 (m), 2925 (m),
2856 (m), 2542 (m), 1541 (s), 1505 (m), 1472 (m), 1452 (m), 1433 (m),
1416 (s), 1366 (m), 1342 (m), 1298 (w), 1260 (m), 1191 (s), 1172 (s), 1094
(m), 1067 (s), 1029 (s), 977 (m), 793 (s), 778 (s), 762 (s), 692 (s), 636 cmꢀ1
(m); ESI-MS (pos, THF): m/z (%): calcd for [TpFe]+: 539.1895; found:
539.2098; elemental analysis calcd (%) for C38H41N8BFe: C 67.47, H 6.11,
N 16.57; found: C 66.55, H 6.19, N 15.87.
ACTHNUTRGNEUNG
[(TpiPr,iPr)Fe(PztBu,Me)] (4c): IR (KBr): n˜ =2964 (s), 2928 (m), 2869 (m),
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
[(TpR1,R2)Mn(PztBu,Me)]: In a typical experiment a solution of NaPztBu,Me
2540 (w), 1532 (m), 1520 (w), 1464 (m), 1419 (w), 1393 (w), 1379 (m),
1369 (m), 1300 (m), 1184 (s), 1173 (s), 1051 (s), 1026 (w), 1015 (w), 864
(w), 791 (s), 767 (w), 659 cmꢀ1 (m); ESI-MS (pos, THF): m/z (%): calcd
for [TpFe]+: 521.3309; found: 521.2757; elemental analysis calcd (%) for
C35H59N8BFe: C 63.83, H 9.03, N 17.02; found: C 63.27, H 8.92, N 16.35.
(192 mg, 1.2 mmol, 1 equiv) in toluene (10 mL) was added to a solution
of [(TpR1,R2)Mn(Cl)] (1.2 mmol, 1 equiv) in toluene (10 mL). The result-
ing slightly turbid reaction mixture was stirred at room temperature for
48 h before being filtered. The solvent of the filtrate was removed under
reduced pressure and recrystallisation of the resulting light brown solid
A
E
A
solution of KHB
G
)
3
(500 mg,
from acetonitrile at ꢀ308C gave [(TpR1,R2)Mn
AHCTUNGTRENNUNG
(PztBu,Me)] (with yields rang-
1.49 mmol,1 equiv) and NaPztBu,Me (238 mg, 1.49 mmol, 1 equiv) in THF
(10 mL) was slowly added to a stirred suspension of FeCl2 (189 mg,
1.49 mmol, 1 equiv) in THF (5 mL). The resulting suspension was stirred
at room temperature for 48 h before being filtered. The solvent was re-
moved under reduced pressure and the resulting light green solid was
suspended in toluene. After stirring for 30 min, the reaction mixture was
filtered. The solvent was removed under reduced pressure and recrystalli-
sation of the resulting light green solid from acetonitrile at ꢀ308C gave
ing between 70 and 80%) as a white solid.
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
[(TptBu,Me)Mn(PztBu,Me)] (3a): IR (KBr): n˜ =2963 (s), 2929 (s), 2907 (m),
2866 (m), 2547 (m), 1541 (m), 1465 (m), 1431 (m), 1362 (m), 1243 (m),
1185 (s), 1068 (s), 1028 (m), 1019 (m), 985 (m), 807 (m), 796 (m), 787
(m), 771 (m), 648 cmꢀ1 (m); ESI-MS (pos, THF): m/z (%): calcd for
[TpMn]+: 478.2824; found: 478.2855; elemental analysis calcd (%) for
C32H53N8BMn: C 62.44, H 8.68, N 18.20; found: C 59.65, H 8.28, N 16.00.
ACTHNUTRGNEUNG
[(TpMe,Me)Fe(PztBu,Me)] (459 mg, 0.94 mmol, 63%) as a white solid. IR
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
[(TpPh,Me)Mn(PztBu,Me)] (3b): IR (KBr): n˜ =3059 (m), 2961 (m), 2926 (m),
(KBr): n˜ =3118 (vw), 2961 (s), 2927 (m), 2887 (m), 2515 (m), 1557 (w),
1541 (s), 1521 (m), 1508 (w), 1488(w), 1447 (s), 1418 (s), 1381 (m), 1365
(m), 1350 (w), 1313 (m), 1261 (m), 1236 (w), 1203 (s), 1066 (s), 1044 (s),
1027 (s), 983 (w), 869 m), 806 (s), 778 (s), 719 (w), 695 (m), 677 m), 665
(m), 648 cmꢀ1 (m); elemental analysis calcd (%) for C23H35N8BFe: C
56.35, H 7.20, N 22.86; found: C 55.59, H 7.01, N 21.92.
2856 (m), 2544 (m), 1543 (s), 1505 (m), 1473 (m), 1451 (m), 1433 (m),
1415 (s), 1364 (m), 1341 (m), 1302 (w), 1259 (m), 1192 (s), 1172 (s), 1092
(m), 1067 (s), 1029 (s), 977 (m), 793 (s), 776 (s), 762 (s), 692 (s), 636 cmꢀ1
(m); ESI-MS (pos, THF): m/z (%): calcd for [TpMn]+: 538.1885; found:
538.1880; elemental analysis calcd (%) for C38H41N8BMn: C 67.56, H
6.12, N 16.59; found: C 66.76, H 6.43, N 15.68.
ACHTUNGTRENNUNG
[(TpiPr,iPr)Mn(PztBu,Me)] (3c): IR (KBr): n˜ =2960 (s), 2927 (m), 2912 (m),
Catalytic oxidation of cyclohexene with tBuOOH and 3a–4d as pre-cata-
lysts
2903 (m), 2891 (m), 2866 (m), 2542 (m), 1534 (m), 1516 (w), 1473 (m),
1461 (m), 1428 (w), 1392 (m), 1381 (m), 1364 (m), 1298 (m), 1261 (w),
1170 (s), 1054 (s), 816 (m), 807 (m), 790 (s), 766 (m), 716 (w), 657 cmꢀ1
(m); ESI-MS (pos, THF): m/z (%): calcd for [TpMn]+: 520.3293; found:
520.3180; elemental analysis calcd (%) for C35H59N8BMn: C 63.92, H
9.04, N 17.04; found: C 63.12, H 8.74, N 16.18.
Reaction conditions (1): Molecular sieves (30 mg) and successively cyclo-
hexene (125.0 mL, 1.250 mmol), bromobenzene (62.5 mL, 0.640 mmol) and
an approximately 5.5m solution of tBuOOH in n-decane (0.91 mL,
5 mmol) were added to a solution of the pre-catalyst (16.0 mmol) in
MeCN (5 mL) at 408C. Subsequently, the solution was stirred at 408C for
10018
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020

Trispyrazolylborato Manganese and Iron Complexes with a Pyrazolato Co-ligand
FULL PAPER
2 h before an aliquot (ca. 0.4 mL) was added to a mixture of 1–2m HCl
(2 mL) and diethyl ether (10 mL). After vigorous shaking of the mixture
to transfer the metal components into the aqueous phase, an aliquot of
the ethereal phase was analysed by GC-MS and GC-flame ionisation de-
tection (FID).
Maruno, K. Koizumi, Y. Kitagawa, S. Yamanaka, T. Kawakami, M.
Okumura, K. Yamaguchi, Int. J. Quantum Chem. 2004, 100, 887;
A. J. Skulan, T. C. Brunold, J. Baldwin, L. Saleh, J. M. Bollinger, Jr.,
J. Martin, E. I. Solomon, J. Am. Chem. Soc. 2004, 126, 8842; N. Leh-
nert, K. Fujisawa, E. I. Solomon, Inorg. Chem. 2003, 42, 469; T. C.
Brunold, N. Tamura, N. Kitajima, Y. Moro-oka, E. I. Solomon, J.
Am. Chem. Soc. 1998, 120, 5674; T. Ogihara, S. Hikichi, M. Akita,
Y. Moro-oka, Inorg. Chem. 1998, 37, 2614; R. C. Scarrow, M. G. Tri-
mitsis, C. P. Buck, G. N. Grove, R. A. Cowling, M. J. Nelson, Bio-
chemistry 1994, 33, 15023; N. Kitajima, M. Ito, H. Fukui, Y. Moro-
oka, J. Am. Chem. Soc. 1993, 115, 9335; I. Siewert, C. Limberg,
Chem. Eur. J. 2009, 15, 10316; M. Wagner, C. Limberg, T. Tietz,
Chem. Eur. J. 2009, 15, 5567–5576.
Reaction conditions (2): Molecular sieves (30 mg) and successively cyclo-
hexene (1.00 mL, 10.0 mmol), bromobenzene (20 mL, 0.200 mmol) and an
approximately 5.5m solution of tBuOOH in n-decane (181.0 mL, 1 mmol)
were added to a solution of the pre-catalyst (1.6 mmol) in MeCN (5 mL)
at 408C. Subsequently, the solution was stirred at 408C for 2 h before an
aliquot (ca. 0.4 mL) was added to a mixture of 1–2m HCl (2 mL) and di-
ethyl ether (10 mL). After vigorous shaking of the mixture to transfer
the metal components into the aqueous phase, an aliquot of the ethereal
phase was analysed by GC-MS and GC-FID.
[4] X. Shan, J.-U. Rohde, K. D. Koehntop, Y. Zhou, M. R. Bukowski,
M. Costas, K. Fujisawa, L. Que, Inorg. Chem. 2007, 46, 8410; T. Ogi-
hara, S. Hikichi, M. Akita, T. Uchida, T. Kitagawa, Y. Moro-oka,
Inorg. Chim. Acta 2000, 297, 162.
[5] N. Kitajima, H. Komatsuzaki, S. Hikichi, M. Osawa, Y. Moro-oka, J.
Am. Chem. Soc. 1994, 116, 11596.
[6] J. E. Sheats, R. S. Czernuszewicz, G. C. Dismukes, A. L. Rheingold,
V. Petrouleas, J. Stubbe, W. H. Armstrong, R. H. Beer, S. J. Lippard,
J. Am. Chem. Soc. 1987, 109, 1435.
Applied GC method: Injector: temperature at t=0: 508C; subsequently
heating of the injector at 2008Cminꢀ1 until 3008C was reached then cool-
ing down. Column oven: temperature at t=0: 508C; subsequently hold-
ing temperature at 508C for 5 min; heating of the column oven at
28Cminꢀ1 until 808C was reached; heating of the column oven at
1008Cminꢀ1 until 3008C was reached; subsequently holding temperature
at 3008C for 5.80 min.
[7] N. Kitajima, M. Osawa, M. Tanaka, Y. Moro-oka, J. Am. Chem. Soc.
1991, 113, 8952.
[8] N. Kitajima, U. P. Singh, H. Amagai, M. Osawa, Y. Moro-oka, J.
Am. Chem. Soc. 1991, 113, 7757.
Acknowledgements
[9] K. Qin, C. D. Incarvito, A. L. Rheingold, K. H. Theopold, Angew.
Chem. 2002, 114, 2439; Angew. Chem. Int. Ed. 2002, 41, 2333.
[10] U. P. Singh, P. Tyagi, S. Upreti, Polyhedron 2007, 26, 3625.
[11] N. Kitajima, M. Osawa, N. Tamura, Y. Moro-oka, T. Hirano, M.
Hirobe, T. Nagano, Inorg. Chem. 1993, 32, 1879.
[12] N. Kitajima, H. Fukui, Y. Moro-oka, Y. Mizutani, T. Kitagawa, J.
Am. Chem. Soc. 1990, 112, 6402.
[13] F. A. Jovꢅ, C. Pariya, M. Scoblete, G. P. A. Yap, K. H. Theopold,
Chem. Eur. J. 2011, 17, 1310–1318.
We are grateful to the Fonds der Chemischen Industrie, the BMBF and
the Dr. Otto Rçhm Gedꢀchtnisstiftung for financial support as well as to
Bayer Services GmbH & Co., OHG, BASF AG and Sasol GmbH for the
supply of chemicals. We thank C. Knispel and P. Neubauer for crystal
structure analyses and Dr. M. Menzel for Mçssbauer measurements. Fi-
nally, we acknowledge helpful discussions within the Cluster of Excel-
lence “Unifying Concepts in Catalysis” funded by the DFG.
[14] G. M. Bancroft, J. M. Dubery, J. Chem. Phys. 1969, 50, 2264.
[15] M. Wagner, Dissertation thesis, Humboldt-Universitꢀt zu Berlin
(Berlin), 2009.
[16] T. Birchall, M. F. Morris, Can. J. Chem. 1972, 50, 211.
[17] R. van Gorkum, J. Berding, D. M. Tooke, A. L. Spek, J. Reedijk, E.
Bouwman, J. Catal. 2007, 252, 110.
[18] M. Salavati-Niasari, H. Babazadeh-Arani, J. Mol. Catal. 2007, 274,
58.
[19] M. Salavati-Niasari, S. H. Banitaba, J. Mol. Catal. 2003, 201, 43.
[20] M. Salavati-Niasari, F. Farzaneh, M. Ghandi, J. Mol. Catal. 2002,
186, 101.
[21] W. Nam, S.-Y. Oh, M. H. Lim, M.-H. Choi, S.-Y. Han, G.-J. Jhon,
Chem. Commun. 2000, 1787.
[22] T. Matsushita, D. T. Sawyer, A. Sobkowiak, J. Mol. Catal. 1999, 137,
127.
[23] J.-M. Vincent, A. Rabion, V. K. Yachandra, R. H. Fish, Angew.
Chem. 1997, 109, 2438.
[24] T. G. Carrell, S. Cohen, G. C. Dismukes, J. Mol. Catal. 2002, 187, 3.
[25] W. Nam, H. J. Han, S.-Y. Oh, Y. J. Lee, M.-H. Choi, S.-Y. Han, C.
Kim, S. K. Woo, W. Shin, J. Am. Chem. Soc. 2000, 122, 8677.
[26] P. A. MacFaul, K. U. Ingold, D. D. M. Wayner, L. Que, J. Am.
Chem. Soc. 1997, 119, 10594; I. W. C. E. Arends, K. U. Ingold,
D. D. M. Wayner, J. Am. Chem. Soc. 1995, 117, 4710; J. Kim, R. G.
Harrison, C. Kim, L. Que, J. Am. Chem. Soc. 1996, 118, 4373.
[27] T. L. Foster, J. P. Caradonna, J. Am. Chem. Soc. 2003, 125, 3678.
[28] W. Nam, I. Kim, M. H. Lim, H. J. Choi, J. S. Lee, H. G. Jang, Chem.
Eur. J. 2002, 8, 2067.
[29] G. Olason, D. C. Sherrington, React. Funct. Polym. 1999, 42, 163.
[30] K. Wieghardt, U. Bossek, B. Nuber, J. Weiss, J. Bonvoisin, M. Cor-
bella, S. E. Vitols, J. J. Girerd, J. Am. Chem. Soc. 1988, 110, 7398.
[31] S. Romain, C. Baffert, C. Duboc, J.-C. LeprÞtre, A. Deronzier, M.-
N. Collomb, Inorg. Chem. 2009, 48, 3125; R. Hage, B. Krijnen, J. B.
Warnaar, F. Hartl, D. J. Stufkens, T. L. Snoeck, Inorg. Chem. 1995,
34, 4973; J. W. de Boer, P. L. Alsters, A. Meetsma, R. Hage, W. R.
[1] S. Trofimenko, Chem. Rev. 1993, 93, 943; S. Trofimenko, Scorpio-
nates: The Coordination Chemistry of Pyrazolylborate Ligands, Im-
perial College Press, London 1999.
[2] Y. Takano, K. Koizumi, H. Nakamura, Inorg. Chim. Acta 2009, 362,
4578; D. Diaconu, Z. Hu, S. M. Gorun, J. Am. Chem. Soc. 2002, 124,
1564; A. de la Lande, H. Gꢅrard, V. Moliner, G. Izzet, O. Reinaud,
O. Parisel, J. Biol. Inorg. Chem. 2006, 11, 593; C. Diedrich, R. J.
Deeth, Inorg. Chem. 2008, 47, 2494; M. Sundararajan, I. H. Hillier,
N. A. Burton, J. Phys. Chem. B 2007, 111, 5511; K. Fujisawa, A.
Tateda, Y. Miyashita, K.-i. Okamoto, F. Paulat, V. K. K. Praneeth,
A. Merkle, N. Lehnert, J. Am. Chem. Soc. 2008, 130, 1205; C. L.
Foster, X. Liu, C. A. Kilner, M. Thornton-Pett, M. A. Halcrow, J.
Chem. Soc. Dalton Trans. 2000, 4563; D. W. Randall, S. D. George,
P. L. Holland, B. Hedman, K. O. Hodgson, W. B. Tolman, E. I. Solo-
mon, J. Am. Chem. Soc. 2000, 122, 11632; G. Shengli, S. Liu, Y.
Yuanqi, Y. Kaibei, J. Coord. Chem. 1998, 46, 145; M. A. Halcrow,
L. M. L. Chia, J. E. Davies, X. Liu, L. J. Yellowlees, E. J. L. McInnes,
F. E. Mabbs, Chem. Commun. 1998, 2465; C. E. Ruggiero, S. M. Car-
rier, W. E. Antholine, J. W. Whittaker, C. J. Cramer, W. B. Tolman, J.
Am. Chem. Soc. 1993, 115, 11285; M. J. Baldwin, D. E. Root, J. E.
Pate, K. Fujisawa, N. Kitajima, E. I. Solomon, J. Am. Chem. Soc.
1992, 114, 10421.
[3] L. Que, Jr., W. B. Tolman, Angew. Chem. 2002, 114, 1160; Angew.
Chem. Int. Ed. 2002, 41, 1114; L. Que, Jr., W. B. Tolman, Angew.
Chem. 2002, 114, 1900; Angew. Chem. Int. Ed. 2002, 41, 1821; A.
Mukherjee, M. Martinho, E. L. Bominaar, E. Mꢄnck, L. Que, Jr.,
Angew. Chem. 2009, 121, 1812; Angew. Chem. Int. Ed. 2009, 48,
1780; T. Borowski, A. Bassan, P. E. M. Siegbahn, Inorg. Chem. 2004,
43, 3277; I. Siewert, C. Limberg, Angew. Chem. 2008, 120, 8071;
Angew. Chem. Int. Ed. 2008, 47, 7953; W.-G. Han, L. Noodleman,
Inorg. Chim. Acta 2008, 361, 973; T. K. Paine, H. Zheng, L. Que,
Inorg. Chem. 2005, 44, 474; T. Liu, T. Lovell, W.-G. Han, L. Noodle-
man, Inorg. Chem. 2004, 43, 6858; M. Shoji, Y. Nishiyama, Y.
Chem. Eur. J. 2011, 17, 10010 – 10020
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10019

C. Limberg et al.
Browne, B. L. Feringa, Dalton Trans. 2008, 6283; K.-O. Schꢀfer, R.
Bittl, F. Lendzian, V. Barynin, T. Weyhermꢄller, K. Wieghardt, W.
Lubitz, J. Phys. Chem. B 2003, 107, 1242; A. Magnuson, P. Liebisch,
J. Hçgblom, M. F. Anderlund, R. Lomoth, W. Meyer-Klaucke, M.
Haumann, H. Dau, J. Inorg. Biochem. 2006, 100, 1234; J. W.
de Boer, W. R. Browne, J. Brinksma, P. L. Alsters, R. Hage, B. L.
Feringa, Inorg. Chem. 2007, 46, 6353; J. A. Lessa, A. Horn, ꢆ. S.
Bull, M. R. Rocha, M. Benassi, R. R. Catharino, M. N. Eberlin, A.
Casellato, C. J. Noble, G. R. Hanson, G. Schenk, G. C. Silva,
O. A. C. Antunes, C. Fernandes, Inorg. Chem. 2009, 48, 4569.
[32] A further signal at m/z 887.3598 may originate from the product of
an intramolecular ligand oxidation containing two (Tp)B-PzMe-CH2-
O-Mn moieties and a Mn-O-Mn unit. Kitajima et al. have reported
the formation of an analogous product in the reaction of
[(TpiPr,iPr)Mn(OH)2Mn(TpiPr,iPr)] with O2.[7]
[34] J. E. Sarneski, M. Didiuk, H. H. Thorp, R. H. Crabtree, G. W. Brud-
vig, J. W. Faller, G. K. Schulte, Inorg. Chem. 1991, 30, 2833.
[35] SIR 2004: Program for Crystal Structure Determination, Perugia
(Italy) 2005; M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini,
G. L. Cascarano, L. De Caro, C. Giacovazzo, G. Polidori, R. Spanga,
J. Appl. Crystallogr. 2005, 38, 381–388.
[36] SHELXL-97, G. M. Sheldrick, Program for Crystal Structure Re-
finement, University of Gçttingen, Gçttingen (Germany), 1997.
[37] D. T. Puerta, S. M. Cohen, Inorg. Chim. Acta 2002, 337, 459.
[38] S. Trofimenko, J. Am. Chem. Soc. 1967, 89, 3165.
[39] F. W. Swamer, C. R. Hauser, J. Am. Chem. Soc. 1950, 72, 1352;
R. H. Wiley, P. E. Hexner, Org. Synth. 1951, 31, 43.
[33] D. Mansuy, P. Battioni, J.-P. Renaud, J. Chem. Soc. Chem. Commun.
1984, 1255.
Received: January 31, 2011
Published online: July 8, 2011
10020
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10010 – 10020