Redox-inactive metal ions promoted the catalytic reactivity of non-heme manganese complexes towards oxygen atom transfer

Article abstract of DOI:10.1039/c4dt03993a

Redox-inactive metal ions can modulate the reactivity of redox-active metal ions in a variety of biological and chemical oxidations. Many synthetic models have been developed to help address the elusive roles of these redox-inactive metal ions. Using a non-heme manganese(ii) complex as the model, the influence of redox-inactive metal ions as a Lewis acid on its catalytic efficiency in oxygen atom transfer was investigated. In the absence of redox-inactive metal ions, the manganese(ii) catalyst is very sluggish, for example, in cyclooctene epoxidation, providing only 9.9% conversion with 4.1% yield of epoxide. However, addition of 2 equiv. of Al³⁺ to the manganese(ii) catalyst sharply improves the epoxidation, providing up to 97.8% conversion with 91.4% yield of epoxide. EPR studies of the manganese(ii) catalyst in the presence of an oxidant reveal a 16-line hyperfine structure centered at g = 2.0, clearly indicating the formation of a mixed valent di-μ-oxo-bridged diamond core, Mn^III-(μ-O)₂-Mn^IV. The presence of a Lewis acid like Al³⁺ causes the dissociation of this diamond Mn^III-(μ-O)₂-Mn^IV core to form monomeric manganese(iv) species which is responsible for improved epoxidation efficiency. This promotional effect has also been observed in other manganese complexes bearing various non-heme ligands. The findings presented here have provided a promising strategy to explore the catalytic reactivity of some di-μ-oxo-bridged complexes by adding non-redox metal ions to in situ dissociate those dimeric cores and may also provide clues to understand the mechanism of methane monooxygenase which has a similar diiron diamond core as the intermediate.

Full text of DOI:10.1039/c4dt03993a

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: C. Choe, L. yang,
Z. Lv, W. Mo, Z. Chen, G. Li and G. Yin, Dalton Trans., 2015, DOI: 10.1039/C4DT03993A.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/dalton

Page 1 of 12
Dalton Transactions
View Article Online
DOI: 10.1039/C4DT03993A
Journal Name
RSCPublishing
ARTICLE
Redox-inactive Metal Ions Promoted Catalytic
Reactivity of Non-heme Manganese Complexes
towards Oxygen Atom Transfer
Cite this: DOI: 10.1039/x0xx00000x
Cholho Choe, Ling Yang, Zhanao Lv, Wanling Mo, Zhuqi Chen*, Guangxin Li* and
Guochuan Yin*
Received 00th January 2012,
Accepted 00th January 2012
DOI: 10.1039/x0xx00000x
Redoxꢀinactive metal ions can modulate the reactivity of the redoxꢀactive metal ions in a
variety of biological and chemical oxidations. Many synthetic models have been developed to
help addressing the elusive roles of these redoxꢀinactive metal ions. Using a nonꢀheme
manganese(II) complex as model, the influence of redoxꢀinactive metal ions as Lewis acid on
its catalytic efficiency in oxygen atom transfer was investigated. In the absence of redoxꢀ
inactive metal ions, manganese(II) catalyst is very sluggish, for example, in cyclooctene
epoxidation, providing only 9.9% conversion with 4.1% yield of epoxide. However, adding 2
equiv. of Al3+ to the manganese(II) catalyst sharply improves the epoxidation up to 97.8%
conversion with 91.4% yield of epoxide. EPR studies of the manganese(II) catalyst in the
presence of oxidant reveal a 16ꢀline hyperfine structure centered at g=2.0, clearly indicating
the formation of a mixed valent diꢀꢁꢀoxoꢀbridged diamond core, MnIIIꢀ(ꢁꢀO)2ꢀMnIV. The
presence of Lewis acid like Al3+ causes the dissociation of this diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core to form monomeric manganese(IV) species which is responsible for improved epoxidation
efficiency. This promotional effect has also been observed in other manganese complexes
bearing various nonꢀheme ligands. The findings presented here have provided a promising
strategy to explore the catalytic reactivity of some diꢀꢁꢀoxoꢀbridged complexes by adding nonꢀ
redox metal ions to in situ dissociate those dimeric cores, and it may also provide clues to
understand the mechanism of methane monooxygenase which has a similar diiron diamond
core as the intermediate.
www.rsc.org/
communities. Different from the complicated biological and
heterogeneous systems, homogeneous system can offer simpler
Introduction
Redoxꢀinactive metal ions are critical components in many models to help understanding how these redoxꢀinactive metal
biological redox enzymes, and the oxygenꢀevolving complex ions participate in the oxidation events and affect the reactivity
(OEC) of photosystem II (PSII) in cyanobacteria and plants is properties of redoxꢀactive metal ions. Using different synthetic
the most wellꢀknown example in which redoxꢀinactive Ca2+ models, it has been found that binding of redoxꢀinactive metal
acts as an essential cofactor in Mn4CaO5 cluster.1 Also in the ions to the high valent metalꢀoxo moieties can modulate their
copperꢀzinc superoxide dismutases (CuZnSOD1), zinc has been oxidative reactivity, and this binding was even evidenced with
proposed to play the significant role on the thermostability and the Xꢀray crystal structure of Sc3+ꢀbound [(TMC)FeIV(O)]2+
activity of the metalloenzyme.2,3 To clarify related oxidation complex determined by Fukuzumi and Nam.5 It should be noted
mechanisms, many inorganic models bearing synthetic ligands that the oxidation state assignment of the iron in this species
have been explored, however, the exact roles of these redoxꢀ has recently been challenged.6 Nam and Fukuzumi reported that
inactive metal ions still remain elusive. In addition to biological Lewis acid like Sc3+ can substantially accelerate the electron
oxidations, redoxꢀinactive metal ions are also frequently transfer rate from a series of oneꢀelectron reductants to a
employed to improve the stability and/or to modulate the nonheme oxoiron(IV), FeIV(N4Py)(O),7 and similar acceleration
reactivity of transition metal catalysts in heterogeneous
oxidations.4 Thereof, addressing the functional roles of these demethylation of N,Nꢀdimethylaniline.8 In particular, Goldberg
redoxꢀinactive metal ions is of great importance and has found that Zn2+ may interact with the MnV
O group in their
attracted much attention in both biological and chemical (corrolazine)MnV(O) complex, and cause the valence
effect was also observed in oxidative dimerization and Nꢀ
≡
This journal is © The Royal Society of Chemistry 2013
J. Name., 2013, 00, 1-3 | 1

Dalton Transactions
Page 2 of 12
VJieowuArrtnicalelONnlainme e
ARTICLE
DOI: 10.1039/C4DT03993A
tautomerization of the (corrolazine)MnV(O) to generate the MnIV=O, the catalytic olefin epoxidation was not improved by
Mn(IV) oxo corrolazine π radical intermediates, thus accelerate adding Lewis acid. Recently, we communicated the first
its rate in both hydrogen abstraction and electron transfer.9 For example of that mixed valent diꢀꢁꢀoxoꢀbridged diamond MnIIIꢀ
BF3 as Lewis acid, Lau also even found that it can dramatically (ꢁꢀO)2ꢀMnIV core, which is sluggish in oxygen atom transfer,
improve the oxidation of alkanes and arylalkanes by KMnO4 can be dissociated by redoxꢀinactive metal ions, leading to
through BF3 binding to the MnVII=O group.10 Notably, Borovik sharp improvement in olefin epoxidation.22 Herein, we report
observed that Ca2+ can promote dioxygen activation by their this remarkable promotional effect with manganese catalysts
manganese(II)
heterobimetallic CaIIꢀ(ꢁꢀOH)ꢀMnIII complex was confirmed by been discussed in details. We hope these findings can provide a
Xꢀray di
raction.11 The dioxygen activation triggered by redoxꢀ new strategy for exploring the catalytic functions of those redox
inactive metal ions was also observed by Que, Nam and metal complexes which generally form sluggish dimers under
Fukuzumi with Fe(TMC) complex.12,13
oxidative conditions. We also hope it may provide new clues to
complexes,
and
the
formation
of bearing different ligands, and the epoxidation mechanism has
ff
Above discoveries clearly reveal that the presence of Lewis understand the mechanism of MMO (Methane Monooxygenase)
acid can modulate the redox behaviors of the active metal ions which has a similar diamond [Fe2(ꢁꢀO)2] core.
at high oxidation state through its bridge or ligation with the
metal oxo functional groups. However, the examples to
demonstrate Lewisꢀacidꢀaccelerated oxidations are still limited,
N
N
and in most cases, only accelerated electron transfer rates were
N
N
N
observed. Although Sc3+ accelerated sulfide oxygenation with
FeIV(N4Py)(O) was even observed by Nam and Fukuzumi, the
oxidation proceeds by switching of mechanism from direct
oxygen transfer to metalꢀionꢀcoupled electron transfer.14
Similarly, binding Sc3+ to a mononuclear nonꢀheme Mn(IV)
oxo complex also accelerates sulfoxidation, however, the rate
of hydrogen atom abstraction was obviously decelerated, which
was rationalized as the steric hindrance raised by Sc3+ binding
to the Mn(IV) oxo moiety.15,16 Collins also found that redoxꢀ
inactive metal ions does not improve olefin epoxidation by their
(TAML)MnV(O) analogue but accelerate triphenylphosphine
oxygenation.17 Due to the fact that examples reporting Lewisꢀ
acidꢀaccelerated oxidation are mostly limited to electron
transfer, one may concern that to which extent the Lewis acid
could improve the efficiency of traditional oxidations.
Moreover, most of these reported examples are based on
stoichiometric oxidations, while the examples on Lewisꢀacidꢀ
accelerated catalytic oxidation are still seldom, which is more
analogous to the biological and chemical oxidations catalyzed
by enzymes and redox catalysts.
N
N
N
N
Bn-TPEN
TPA
N
N
O
O
N
N
N
N
N
N
BPMEN
Phen
Phendio
Scheme1 Molecular Structures of ligands studied in this work.
Experimental
Iodobenzene
trifluoromethanesulfonate
trifluoromethanesulfonate
diacetate
(PhI(OAc)2),
(NaOTf),
sodium
magnesium
and scandium
(Mg(OTf)2),
trifluoromethanesulfonate (Sc(OTf)3) came from Aldrich.
In our works to investigate the oxidative relationships of
active metal oxo and hydroxo moieties, we observed that
increasing the net charge of the active metal species by binding
Brönsted acid (Mn+=O vs. M(n+1)+ꢀOH) can accelerate its
electron transfer rate because of the increased redox potential.18
This finding resembles the acceleration effect caused by
binding Lewis acid to the active Mn+=O moieties as described
above. Inspired by these, we found that the presence of redoxꢀ
inactive metal ions like Al3+ can accelerate sulfide oxidation
catalyzed by a manganese complex having crossꢀbridged
cyclam ligand, and benzene hydroxylation with dioxygen by
PdII(bpym) catalyst (bpym: 2,2’ꢀpyrimidine).19,20 Moreover, the
presence of Ca2+ and Clꢀ can synergistically affect the redox
potentials of the manganese complexes, thus modulate their
reactivity in oxidations,21 which may provide new clues to
understand their roles in oxygen evolution in Photosystem II.
However, due to that the active functional group in the above
mentioned manganese catalyst is a MnIVꢀOH moiety rather than
Barium
trifluoromethanesulfonate
(Ba(OTf)2),
(Ca(OTf)2),
(Zn(OTf)2),
(Al(OTf)3),
(Y(OTf)3) and
calcium
zinc
aluminum
yttrium
trifluoromethanesulfonate
trifluoromethanesulfonate
trifluoromethanesulfonate
trifluoromethanesulfonate
ytterbium
trifluoromethanesulfonate (Yb(OTf)3) came from Aladdin.
Olefins as well as epoxides were purchased from either Aldrich
or Alfa Aesar. H218O (90% 18O atom) came from Acros. Other
organic compounds and organic solvent came from Shaoyuan
or Shanghai Dibai. Tris(pyridinꢀ2ꢀylmethyl)amine) (TPA) and
NꢀbenzylꢀN,N’,N’’ꢀtris(2ꢀpyridylmethyl)ꢀ1,2ꢀdiaminoethane
(BnꢀTPEN) with their manganese complexes, Mn(TPA)Cl2 and
Mn(BnꢀTPEN)Cl2, were synthesized according to literatures.23ꢀ
25
N,N’ꢀdimethylꢀN,N’ꢀbis(2ꢀpyridylmethyl)ꢀ1,2ꢀethanediamine
(BPMEN) and its complex, Mn(BPMEN)Cl2, were synthesized
according to our previous publication.22 Phenantroline (Phen)
was purchased from Aladdin. Phendio (1,10ꢀphenanthrolineꢀ
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 3 of 12
Journal Name
Dalton Transactions
View ArticAle ROnTliInCeLE
DOI: 10.1039/C4DT03993A
5.6ꢀdione) was synthesized according to the literature.26 The initialize the reaction. The reaction mixture was stirred in an
structures of these ligands have been illustrated in Scheme 1. iceꢀwater bath at 273 K and the product analysis was performed
1HꢀNMR data of ligands: TPA (CD3CN): δ 8.48 (d, 3H), 7.69 (t, by HPLC using the internal standard method. Control
3H), 7.60 (d, 3H), 7.17 (t, 3H), 3.81 (s, 6H).; BnꢀTPEN (CDCl3): δ experiments using the manganese(II) complexes or Lewis acid
8.48(m, 2H), 7.58 (m, 3H), 7.45 (t, 3H), 7.19ꢀ7.32 (m, 5H), 7.12 (m, alone as catalyst were carried out in parallel. Reactions were
3H), 3.77 (s, 4H), 3.72 (s, 2H), 3.59 (s, 2H), 2.74 (m, 4H).; Phendio performed at least in triplicate, and the average data were used
(CDCl3): δ 7.56(m, 2H), 8.41(m, 2H), 9.06 (m, 2H).; BPMEN in discussion.
(CDCl3): δ 2.25 (s, 3H), 2.65 (s, 2H), 3.15 (s, 2H), 7.4 (m, 3H), 8.45
(d, H)
Lewis-acid-promoted catalytic epoxidation by the manganese(II)
complexes in the presence of 18O-water In a mixture solution of
UVꢀVis spectra were collected on Analytik jena, specord 0.4 mL of acetonitrile and 0.1 mL of CH2Cl2 containing 0.1 M
205. GCꢀMS analysis was performed on Agilent7890A/5975C. cisꢀstilbene, 2 mM manganese(II) complexes, and 4 mM Lewis
FTꢀIR spectra were collected on Bruker VERTEX70. acid, 0.1 mL of H218O (90% 18O enrichment as received) were
Electrochemical studies were performed on a CS Corrtest added and stirred for 2 minutes, then 0.1 mmol of PhI(OAc)2
electrochemical workstation equipped with glassy carbon as were added to initialize the reaction. The reaction mixture was
both working and counter electrodes and saturated calomel as stirred in an iceꢀwater bath at 273 K and the product analysis
reference electrode. The redox potentials were measured under was performed by GCꢀMS using the procedure as those in
argon with 0.1 M tetrabutylammonium perchlorate as the normal epoxidation. The 18O enrichments are calculated based
supporting electrolyte. EPR experiments were conducted at 130 on the peak abundances of 16Oꢀ and 18Oꢀepoxide in GCꢀMS
K on Bruker A200, with Center Field of 3352.488 G, graphs as shown in supporting information.
Frequency of 9.395 GHz, Power of 19.44 mW, Modulation
Amplitude of 2.00 G and Receiver Gain of 1.00 × 103. 1HꢀNMR
spectra were collected on Bruker AV600(Avance III NMR
Results and discussion
spectrometer, 1Hꢀfrequency: 600.25 MHz)
Adding redoxꢀinactive metal ion as Lewis acid to promote
olefin epoxidation with manganese catalysts was performed in
Lewis-acid-promoted catalytic epoxidation by the manganese(II) acetonitrile/CH2Cl2 (4:1, v/v) using PhI(OAc)2 as oxidant.
complexes In a mixture solution of 4 mL of acetonitrile and 1 Mn(TPA)Cl2 (TPA = tris(pyridinꢀ2ꢀylmethyl)amine) was first
mL of CH2Cl2 containing 0.05 M olefin, 1 mM manganese(II) tested as catalyst which was synthesized according to the
complexes and 2 mM Lewis acid, 0.5 mmol of PhI(OAc)2 were literature.23 As shown in Table 1, without the addition of redoxꢀ
added to initialize the reaction. The reaction mixture was stirred inactive metal ions, Mn(TPA)Cl2 is very sluggish in catalyzing
in an iceꢀwater bath at 273 K, and the product analysis was cyclooctene epoxidation. After 3.5 h reaction at 273 K, only
performed by GC using the internal standard method (methyl 9.9% of cyclooctene was converted with 4.1% yield of 1.2ꢀ
benene or nitrobenzene as the internal standard). Control epoxyoctane formation. When 2 equiv. of Al(OTf)3 were added
experiments using the manganese(II) complexes or Lewis acid to the manganese catalyst in reaction solution, 97.8%
alone as catalyst were carried out in parallel. Reactions were conversion with 91.4% yield of epoxide could be achieved
performed at least in triplicate, and the average data were used under the same conditions. In control experiment, Al(OTf)3
in discussion.
alone as catalyst showed only 3.7% conversion and 1.7% yield.
Apparently, Lewis acid alone has nearly no catalytic activity for
Lewis-acid-promoted catalytic epoxidation of cis- and trans- olefin epoxidation in our case. Similarly, the addition of 2 equiv.
stilbene by the manganese(II) complexes In a mixture solution of of Sc3+ improved olefin epoxidation with 97.4% conversion and
4 mL of acetonitrile and 1 mL of CH2Cl2 containing 0.05 M 91.0% yield, while Sc3+ alone gave only 4.3% conversion with
olefin, 1 mM manganese(II) complexes and 2 mM Lewis acid, 2.0% yield. In complimentary experiments using the mixture of
0.5 mmol of PhI(OAc)2 were added to initialize the reaction. Al(OTf)3 and TPA ligand as catalyst, it also demonstrated very
The reaction mixture was stirred in an iceꢀwater bath at 273 K sluggish activity, providing only 4.4% of conversion with 2.3%
and the product analysis was performed by HPLC using the yield of epoxide, thus excluded the possibility of that the
internal standard method (mesitylene as the internal standard). synergistic effect originates from the potentially generated
Control experiments using the manganese(II) complexes or complex of Al3+ with ligand. Therefore, this greatly improved
Lewis acid alone as catalyst were carried out in parallel. catalytic activity of the manganese complex by adding redoxꢀ
Reactions were performed at least in triplicate, and the average inactive metal ions strongly supports that the synergistic effect
data were used in discussion.
occurs between the manganese catalyst and the Lewis acid in
epoxidation.
Lewis-acid-promoted catalytic epoxidation by the manganese(II)
This synergistic effect has also been observed by adding other
complexes in the presence of water In a mixture solution of 0.4 redoxꢀinactive metal ions. For examples, adding Y3+ or Yb3+
mL of acetonitrile, 0.1 mL of CH2Cl2 and various amount of achieved 98.8% or 90.1% conversion of substrate, and the yield
H2O containing 0.1 M olefin, 2 mM manganese(II) complexes, of epoxide was 84.4% or 81.2%, respectively. As shown in
and 4 mM Lewis acid, 0.1 mmol of PhI(OAc)2 were added to Table S1, the amount of redoxꢀinactive metal ions can
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 3

Dalton Transactions
Page 4 of 12
VJieowuArrtnicalelONnlainme e
ARTICLE
DOI: 10.1039/C4DT03993A
significantly affect the yield of epoxide. For example, adding destroyed through their protonation. To further rule out the
0.5 equiv. of Al3+ achieved 38.3% conversion with 21.9% yield, influence of water/proton, epoxidation of cyclooctene was
while adding 1.0 or 2.0 equiv. of Al3+ led to 77.2% or 91.4% conducted using dehydrated solvent with Schlenk system in
yield, respectively. However, adding
4
equiv. of Al3+ which the influence of even trace amount of water can be ruled
demonstrated a decreased yield of 51.8%, while the conversion out. In this case, epoxidation using Mn(TPA)Cl2 as catalyst
remained 98.7%. This decreased selectivity of epoxide can be with Al(OTf)3 as Lewis acid still showed efficient activity with
attributed to the Lewis acid catalyzed ringꢀopening of epoxide. 100 % conversion of substrate and 90.8 % yield of epoxide.
As evidence, 1,2ꢀcyclooctanediol product has been detected by Moreover, using protonic acid instead of Lewis acid (Al(OTf)3)
GCꢀMS (Figure S1).
revealed very poor activity. For examples, 23.7% conversion
with 5.9% yield was obtained by adding acetic acid, while for
hydrochloric acid, the conversion was 28.1%, and the yield of
epoxide was 10.5 % (Table S2). These data clearly indicate that
the promotional effect by adding Lewis acid is not attributed to
the protonation of trace amount of water.
Table 1. Catalytic epoxidation of cyclooctene to 1,2ꢀepoxycyclooctane by
Mn(TPA)Cl2 catalyst in the presence of redoxꢀinactive metal ions as Lewis
acid
Additives
Mn(II) + L.A.
Only L.A.
Conv. (%) Yield (%) Conv. (%) Yield (%)
ꢀ
9.9(0.2)
33.3(0.2)
46.2(1.2)
54.9(0.1)
60.1(0.4)
36.8(1.1)
66.6(0.8)
98.8(0.8)
90.1(0.6)
97.8(1.0)
97.4(0.4)
4.1(0.1)
24.7(0.6)
41.7(0.1)
48.8(0.7)
45.6(1.0)
34.1(0.2)
63.3(0.2)
84.4(0.5)
81.2(0.2)
91.4(0.3)
91.0(0.4)
2.4(0.1)
3.7(0.2)
4.5(0.3)
3.6(0.3)
3.9(0.1)
3.5(0.3)
5.0(0.4)
4.2(0.1)
4.2(0.3)
3.7(0.2)
4.3(0.3)
1.5(0.1)
0.8(0.1)
1.7(0.2)
2.0(0.1)
2.4(0.1)
1.5(0.3)
0.8(0.2)
1.5(0.1)
0.7(0.2)
1.7(0.1)
2.0(0.1)
Na+ [a]
Mg2+
Ba2+
Ca2+
Cu2+
Zn2+
Y3+
100
Mn(II)+Al3+
Mn(II)+Sc3+
80
Mn(II)+Zn2+
60
Yb3+
Al3+
Sc3+
Mn(II)+Mg2+
40
Mn(II)+Na+
20
Conditions: acetonitrile/CH2Cl2 (4:1, v/v)
5 mL, cyclooctene 0.05 M,
[a]
Mn(TPA)Cl2 1 mM, Lewis acid 2 mM, PhI(OAc)2 0.1 M, 273 K, 3.5 h. 6 mM
NaOTf was used in the reaction solution. Data in parentheses represent the
deviations.
Mn(II)
No catalyst
0
0
50
100
150
200
250
Reaction time / min
Meanwhile, redoxꢀinactive metal ions with positive charge of
2+, such as Mg2+, Ba2+, Ca2+ and Zn2+, shows relatively low
promotional effect. For example, the addition of 2 equiv. of
Figure 1. Lewis acid promoted epoxidation kinetics by Mn(TPA)Cl2 catalyst.
Conditions: acetonitrile/CH2Cl2 (4:1, v/v) mL, cyclooctene 0.05 M,
Mn(TPA)Cl2 1 mM, Lewis acid 2 mM (6 mM in the case of NaOTf), PhI(OAc)2
5
Ca2+ or Zn2+ provided the conversion of 60.1% or 66.6% with 0.1 M, 273 K, 3.5 h.
epoxide yield of 45.6% or 63.3%, respectively. Similarly, Ca2+
or Zn2+ alone provided only 3.9% or 5.0% conversion with 2.4%
The catalytic kinetics further confirms the aboveꢀmentioned
or 0.8% yield in control experiments, respectively. Adding 6
equiv. of NaOTf just slightly improved cyclooctene oxidation
to 33.3% conversion with 24.7% yield of epoxide, much lower
than that by adding Al(OTf)3, revealing that the promotional
effect cannot be attributed to the presence of OTfꢀ anion,
because the concentration of OTfꢀ under this condition is
identical to that of adding 2 equiv. of Al(OTf)3. In addition,
NaOTf alone under the same conditions gave only 3.7%
conversion and 0.8% yield. It is worth to emphasize that, in
control experiments, all of these nonꢀredox metal ions alone as
catalyst demonstrate very poor catalytic activity for epoxidation
as shown in Table 1. Notably, above improved catalytic
activity is generally dependent on Lewis acid strength, that is,
metal ions with higher positive charge demonstrate better
efficiency in promoting olefin epoxidation with manganese
catalyst.
promotional effect of Lewis acid on the Mn(II) complexes
mediated epoxidation (Figure 1). In the absence of Lewis acid,
catalytic epoxidation of cyclooctene is apparently sluggish.
Along the series of charge 1+, 2+, and 3+ of added redoxꢀ
inactive metal ions, the formation of epoxide speeds up. These
results are well consistent with the data in Table 1, supporting
that adding redoxꢀinactive metal ions with higher positive
charge demonstrates better efficiency in promoting the
epoxidation efficiency of the Mn(TPA)Cl2 catalyst.
This synergistic oxygen atom transfer by adding redoxꢀ
inactive metal ions was also observed in epoxidation of other
cycloꢀolefins and terminal linear olefins. In all of the
investigated substrates, Mn(TPA)Cl2 catalyst or Lewis acid
alone showed sluggish activity for epoxidation. For example in
Table 2, when cyclohexene was used as substrate, Mn(TPA)Cl2
catalyst alone demonstrated sluggish activity in epoxidation,
giving only 18.3 % conversion with 6.7 % yield of epoxide in
3.5 h. Under the identical conditions, addition of 2 equiv. of
Al3+ generated high oxidation activity with 91.3% conversion
and 69.0% yield of epoxide. In the case of norbornene, the
One may argue that this synergistic oxygen transfer
originates from rapid reaction of Lewis acid with trace amounts
of water to form metalꢀoxide/hydroxides, and liberate proton.
Next, the oxo and hydroxide bridged complexes is simply
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 5 of 12
Journal Name
Dalton Transactions
View ArticAle ROnTliInCeLE
DOI: 10.1039/C4DT03993A
conversion can be promoted from 11.1% to 84.3%, and the different from that of using Mn(TPA)Cl2 alone. In the case of
yield increased from 5.1% to 47.4% by adding Al3+. transꢀstilbene, the combination of Mn(TPA)Cl2 with Al3+
Conversion of styrene was also promoted from 9.1% to 71.9%, provided 64.7% yield of transꢀepoxide with 2.8% yield of
and the yield of epoxystyrene increased to 45.1%. For these benzaldehyde, while Mn(TPA)Cl2 alone gave only 19.7% yield
terminal linear olefins such as 1ꢀhexene and 1ꢀdodecene, the of transꢀepoxide with 3.8% yield of benzaldehyde.
improved epoxidation efficiency with Mn(TPA)Cl2 catalyst was Remarkably, there was no cisꢀstilbene epoxide formation.
also observed by adding redoxꢀinactive metal ions. However,
Table 3. Stilbene epoxidations by Mn(TPA)Cl2 catalyst promoted by
the required reaction time for electronꢀdeficient linear olefins is
different ratio of Al(OTf)3
much longer (8 h) than that of cycloꢀolefins or electronꢀrich
olefins (3.5 h), which leads to a lower selectivity because of
Lewisꢀacidꢀcatalyzed ringꢀopening of epoxide.
Mn(II):Al3+
Substrate
Product
1:0
1:0.5
1:1
1:2
cisꢀstilbene
cisꢀepoxide
5.9
(0.1)
3.5
(0.3)
3.3
39.6
(0.7)
2.8
(0.1)
3.4
41.5
(2.0)
3.6
(0.1)
2.0
50.7
(0.2)
5.8
(0.4)
6.1
Table 2. Al(OTf)3 promoted olefin epoxidations by Mn(TPA)Cl2 catalyst
transꢀepoxide
benzaldehyde
Mn(II)
Substrate
Product
Conv. %
Yield%
: Al3+
(0.2)
(0.3)
(0.3)
(0.6)
transꢀstilbene cisꢀepoxide
transꢀepoxide
0:0
0:2
1:0
1:2
2.4(0.1)
3.7(0.2)
9.9(0.2)
97.8(1.0)
1.5(0.1)
1.7(0.1)
4.1(0.1)
91.4(0.3)
ꢀ
ꢀ
ꢀ
ꢀ
19.7
(0.7)
3.8
43.5
(1.7)
4.3
44.6
(0.7)
3.6
64.7
(1.9)
2.8
1,2ꢀ
cyclooctene
epoxycyclooctane
benzaldehyde
0:0
0:2
1:0
1:2
3.8(0.6)
6.0(0.3)
18.3(0.6)
91.3(0.9)
1.0(0.3)
2.5(0.5)
6.7(0.4)
69.0(0.2)
(0.4)
(0.6)
(0.2)
(0.4)
Conditions: acetonitrile/CH2Cl2 (4:1, v/v) 5 mL, cyclooctene 0.1 M, Mn(TPA)Cl2
2 mM, PhI(OAc)2 0.1 M, 273 K, 8 h. The data in parentheses represent the
deviations.
1,2ꢀ
cyclohexene
norbornene
epoxycyclohexane
0:0
0:2
1:0
1:2
0:0
0:2
1:0
1:2
0:0
0:2
1:0
1:2
0:0
0:2
1:0
1:2
2.6(0.2)
3.5(0.4)
11.1(0.9)
84.3(0.7)
3.7(0.3)
5.5(0.4)
9.1(0.4)
71.9(0.6)
3.1(0.2)
5.3(0.3)
9.3(0.4)
75.1(1.7)
3.8(0.1)
7.3(0.4)
13.5(1.3)
90.3(0.4)
1.4(0.2)
1.6(0.2)
5.1(0.2)
47.4(0.8)
2.4(0.3)
1.9(0.2)
4.9(0.3)
45.1(0.6)
1.2(0.2)
1.5(0.1)
3.7(0.1)
51.7(0.6)
1.3(0.1)
1.7(0.1)
2.5(0.2)
45.1(1.9)
Interaction between high valent manganese species and
redox-inactive metal ions. In literatures, the promotional
effects of those redoxꢀinactive metal ions are generally
attributed to their linkages to the redox metal ions through the
Mn+=O functional group. Here, the reaction data presented
above have clearly revealed that the synergistic effect exists
between Mn(TPA)Cl2 catalyst and redoxꢀinactive metal ions in
catalyzing olefin epoxidation, which is one of the rare examples
that redoxꢀinactive metal ions as Lewis acid can accelerate
direct oxygen atom transfer.22 The evidence to display their
interaction first comes from EPR studies (Figure 2). In the
absence of Lewis acid, a typical 16ꢀline signal centered at g=2.0
and regularly spaced by about 84 G was observed at 130 K
when Mn(TPA)Cl2 was oxidized by PhI(OAc)2 in
acetonitrile/CH2Cl2 (4:1, v/v), accounting for the coupling of
two nonꢀequivalent 55Mn nuclei (I=5/2).22 This typical 16ꢀline
signal is highly characteristic of the mixed valent diꢀꢁꢀoxoꢀ
bridged diamond core, MnIIIꢀ(ꢁꢀO)2ꢀMnIV, which has been
reported in chemical oxidation or electrochemical process of
various manganese(II) complexes.24,31ꢀ33 The 16ꢀline signal
spreads over a 1260 G range (first peak to last trough), which is
highly consistent with the reported value of similar mixedꢀ
valent species in literature (1240 G).24 On the other hand, this
16ꢀline signal gradually disappeared with the addition of redoxꢀ
inactive metal ions. As shown in Figure 2, at 10 minutes after
adding Al(OTf)3 to the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core, the
16ꢀline hyperfine structure turns more irregular when compared
with its initial EPR signal. Instead, after 30 minutes, a 6ꢀline
signal at g=1.9 with average spaced by 95 G was observed,
while the 16ꢀline hyperfine structure has almost vanished,
indicating a highꢀspin mononuclear manganese(II) species was
generated.25,34,35 This change of EPR spectra clearly indicates
an Al3+ induced dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
epoxynorbornene
styrene
epoxystyrene
1ꢀhexene[a]
1,2ꢀepoxyhexene
1,2ꢀ
epoxydodecene
1ꢀdodecene
Conditions: acetonitrile/CH2Cl2 (4:1, v/v)
Mn(TPA)Cl2 1 mM, Al(OTf)3 2 mM, PhI(OAc)2 0.1 M, 273 K, 3.5 h reaction
time, 8 h.The data in parentheses represent the deviations.
5 mL, cyclooctene 0.05 M,
[a]
Generally, epoxidation of cisꢀstilbene can provide more
mechanistic information than other olefins. In active Mn+=O
moieties mediated cisꢀstilbene epoxidation, the ratio of cis and
transꢀepoxide products is highly dependent on the reaction
pathway as well as the coordination environments of the redox
metal ions.27ꢀ30 In some cases, the yields of cis and trans
ꢀ
epoxide are comparable, while in other cases, it provides
dominant cisꢀepoxide. Here, Mn(TPA)Cl2 alone as catalyst
provided 5.9% cisꢀepoxide and 3.5% transꢀepoxide when using
cisꢀstilbene as substrate, and the ratio of substrate to
Mn(TPA)Cl2 does not affect the ratio of cis/trans in epoxide
product (Table S3). In contrast, adding Al3+ can sharply
improve the yield of cisꢀepoxide, whereas the yield of trans
ꢀ
epoxide remained unchanged. In the presence of 2 equiv. of
Al3+, Mn(TPA)Cl2 catalyst gave 50.7% yield of cisꢀepoxide
with only 5.8% yield of transꢀepoxide (Table 3). Clearly, the
active intermediate for epoxidation in the presence of Al3+ is
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 5

Dalton Transactions
Page 6 of 12
ARTICLE
VJieowuArrtnicalelONnlainme e
DOI: 10.1039/C4DT03993A
mode which is generated by disassembling of FeIIIꢀ(ꢁꢀO)2ꢀ
FeIV.36,37 In literature, it has been reported that O=FeIVꢀOꢀFeIII
can cleave the CꢀH bond in 9,10ꢀdihydroanthracene (DHA)
over 106 fold faster, and transfer oxygen atom to
diphenyl(pentafluorophenyl) phosphine over 103ꢀfold faster
than the corresponding diamond FeIIIꢀ(ꢁꢀO)2ꢀFeIV core.36 The
formation of a terminal iron(IV) oxo in open core has been
proposed as the key factor for the activity enhancement. Here,
similar dissociation of diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core has
been induced by the addition of Lewis acid, which improves
oxygen transfer efficiency as observed in olefin epoxidation.
core happens (Scheme 2, vide infra). Here, the generation of
manganese(II) species is possibly related to the instability of
the activated monomeric manganese(IV) species in the
presence of Al3+. As evidence, a similar 6ꢀline hyperfine
structure was directly observed for Mn(TPA)Cl2 complex, and
this structure remains unchanged by adding Al(OTf)3, which
simultaneously eliminates other influence of Al(OTf)3 on the
EPR signal of manganese complexes. The lack of the direct
EPR signal of Mn(IV)=O/Al3+ adduct is possibly related to its
short life time and/or its weakness compared with the strong
signal of the highꢀspin mononuclear manganese(II) species. It is
worth to mention that, because the EPR signal of the Mn(II)
species is so strong, even though the Mn(II) signal has been
observed, it does not suggest that the Mn(IV) species have been
completely reduced to the corresponding Mn(II) species. On the
other hand, the dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core does not happen by adding 3 equiv. of NaOTf (Figure S2),
which is consistent with its poor synergistic effect in
epoxidation. This result further confirms that the Lewis acid
induced promotional effect is not attributed to the presence of
OTfꢀ anion.
Thereof,
a
similar open O=MnIVꢀOꢀAl3+ core can be
alternatively proposed as the dissociation product from the
diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core. If it is correct, the improved
activity in oxidation demonstrated here may provide clues to
understand the reactivity of nonꢀheme diiron enzymes, where
the intermediate Q in MMO is proposed to have a diamond
[Fe2(ꢁꢀO)2] core (vide infra).38,39
MnIII/MnIV
MnII/MnIII
Mn(II)+PhI(OAc) +Al(OTf)
2
3
IV, IV
Mn2III, IV/Mn2
2750
3000
3250
3500
3750
4000
Mn(II)+PhI(OAc)2
（
a
）
MnIII/MnIV
MnII/MnIII
Mn(II)+PhI(OAc)2
Mn(II)
（
b
）
T=10min
Al(OTf)3
PhI(OAc)2
Mn(II)+Al(OTf)3+PhI(OAc)2
（
c
）
T=30min
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Potential / V (vs SCE)
Mn(II)+Al(OTf)3+PhI(OAc)2
Figure 3 Cyclic voltammogram for Mn(TPA)Cl2 in the presence/absence of
PhI(OAc)2 and Al(OTf)3. Conditions: 5 mM Mn(TPA)Cl2 in acetonitrile/CH2Cl2
(4:1,v/v), 5 equiv. of PhI(OAc)2, 2 equiv. of Al(OTf)3.
（
d
）
Mn(II)+Al(OTf)3
The dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core has
also been evidenced by electrochemical studies (Figure 3).
MnII(TPA)Cl2 alone revealed the MnII/MnIII and MnIII/MnIV
（
e
）
Mn(II)
2750
couples of +0.79
V
and +1.57
V
(vs SCE) in
3000
3250
3500
3750
4000
G
acetonitrile/dichloromethane (4:1, v/v), respectively, which are
similar to those potentials of Mn(BPMEN)Cl2 in literature.22
These electrochemical oxidation waves also indicate that TPA
and BPMEN have similar electron donation ability and
coordination structure around manganese(II) ions. In situ
oxidizing of Mn(TPA)Cl2 with PhI(OAc)2 revealed a third
redox couple at +1.12 V (vs SCE), which can be assigned to the
couple of dinuclear core Mn2(III,IV) to Mn2(IV,IV) (it is +1.10
Figure 2. EPR spectra of (a) Mn(TPA)Cl2 plus 5 equiv. of PhI(OAc)2, (b) 10
minutes after adding 1 equiv. of Al3+ into (a), (c) 30 minutes after adding 1 equiv.
of Al3+ into (a), (d) Mn(TPA)Cl2 plus
1 , (e) Mn(TPA)Cl2.
equiv. of Al3+
Conditions: 5 mM Mn(TPA)Cl2 in acetonitrile/CH2Cl2 (4:1, v/v) at 130 K.
Although the accurate structure is still elusive, one plausible
interaction between Mn(IV) species and Al3+ can be described
as formation of the Mn(IV)=O•••Al3+ unit, and similar binding
mode has been suggested in other Lewisꢀacidꢀaccelerated
oxidations by active metal oxo moieties in electron transfer,
oxygenation and hydrogen abstraction.7ꢀ9 Another alternative
structure of the generated active intermediate is the form of
O=MnIVꢀOꢀAl3+. This hypothesis is similar of O=FeIVꢀOꢀFeIII
V
vs SCE for Mn(BPMEN)Cl2).24 In the presence of Al3+, this
dinuclear redox couple almost vanished, indicating an Al3+
induced dissociation of the dinuclear core happens. Moreover,
no electrochemical wave can be monitored when using
Al(OTf)3 or PhI(OAc)2 alone in control experiments, indicating
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 7 of 12
Journal Name
Dalton Transactions
View ArticAle ROnTliInCeLE
DOI: 10.1039/C4DT03993A
the aboveꢀmentioned redox couples were not contributed from while vanishing of this band by adding Al3+ indicates the
Lewis acid or oxidant.
dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core (Fig. 4a).
1000
900
800
700
600
500
2.5
(a)
Mn(TPA)Cl2
Al(OTf)3
2.0
1.5
1.0
0.5
0.0
PhI(OAc)2
(a)
Mn(TPA)Cl2+PhI(OAc)2
Mn(TPA)Cl2+PhI(OAc)2+Al(OTf)3
Mn(II)+PhI(OAc)2+Al(OTf)3
670 cm-1
(b)
(c)
Mn(II)+PhI(OAc)2
400
600
800
1000
Al(OTf)3
Wavelength / nm
(d)
(e)
PhI(OAc)2
1.0
0.8
0.6
0.4
0.2
0.0
(b)
0 min
3 min
30 min
60 min
Mn(II)
(f)
Solvent
1000
900
800
700
600
500
-1
Wavenumber / cm
Figure 4. FTꢀIR spectra of manganese(II) complex in the presence/absence of
PhI(OAc)2 and Al(OTf)3. Conditions: 5 mM Mn(TPA)Cl2 in acetonitrile/CH2Cl2
(4:1, v/v), 5 equiv. of PhI(OAc)2, 2 equiv. of Al(OTf)3.
400
600
800
1000
Wavelength / nm
The FTꢀIR spectra also indicate the dissociation of the Figure 5. (a) UVꢀVis spectra of the manganese complex, Al(OTf)3 and PhI(OAc)2.
dinuclear MnIIIꢀ(ꢁꢀO)2ꢀMnIV core by adding Al3+. Both
Conditions: 1 mM Mn(TPA)Cl2 in acetonitrile/CH2Cl2 (4:1, v/v), 5 equiv. of
PhI(OAc)2, 2 equiv. of Al(OTf)3. (b) UVꢀVis spectra of (a) upon addition of 20
equiv. of cyclooctene.
Mn(TPA)Cl2 and PhI(OAc)2 in acetonitrile/CH2Cl2 (4:1, v/v)
gave a broad absorption band from 650 to 800 cmꢀ1 and another
sharp one around 920 cmꢀ1, which can be attributed to organic
The UVꢀVis spectra provide further clue for the interaction
solvent (Figure 4f). The high valent diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core, generated by oxidation of the manganese(II) complex
with PhI(OAc)2, showed an obvious shoulder band around 670
cmꢀ1 (Figure 4b). In literature, a diamond MnO2Mn core, in
which each manganese ion bears a tetradentate Schiff base
ligand, gives rise to a prominent and characteristic IR
absorption from 600 to 650 cmꢀ1.40 In addition, a characteristic
MnIVꢀO stretching mode of the diamond MnO2Mn core with
pyridylꢀbenzimidazole ligand is also located around 695 cm−1,
which has been believed as a useful reference probe for
structural characterization of Mn containing active sites
involving two oxo bridges.41 Accordingly, in this work, the
shoulder band at 670 cmꢀ1 raised by oxidation of Mn(TPA)Cl2
with PhI(OAc)2 is consistent with those in literatures, indicating
the formation of diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core (Fig. 4b),
of Al3+ with the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core. As shown in
Figure 5, MnII(TPA)Cl2 alone demonstrated a characteristic
absorbance at 257 nm with
ε
value of 2230 Mꢀ1cmꢀ1, which can
be attributed to the absorbance of TPA ligand, while PhI(OAc)2
alone gives no absorption band beyond 400 nm. With the
addition of 5 equiv. of PhI(OAc)2, a broad absorption band
from 400 to 820 nm was observed, which is similar to the
previous reported absorbance from diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core.24 Adding 2 equiv. of Al(OTf)3 generated a distinct UVꢀ
Vis absorbance, which contained a shoulder peak around 440
nm and two weak but nonetheless detectable peaks at 560 nm
and 650 nm. This change in the absorption is consistent with
those from EPR, FTꢀIR and electrochemistry, indicating a
dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core to form the
active Mn(IV)=O/Al3+ adduct for epoxidation. As evidence,
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 7

Dalton Transactions
Page 8 of 12
VJieowuArrtnicalelONnlainme e
ARTICLE
DOI: 10.1039/C4DT03993A
upon adding cyclooctene, the shoulder bands above 500 nm addition of cyclooctene, supporting that this band is a signal of
decayed gradually, and GC analysis confirmed the formation of high valent manganese intermediate, proposed as
epoxide. Similarly, adding Mg2+ or Zn2+ can also induce similar Mn(IV)=O/Al3+ adduct, responsible for epoxidation (vide
spectra changes as that by adding Al3+ (Figure S3), indicating a supra); 3) it has been reported that epoxidation of cisꢀstilbene
dissociation of diamond core has also been initialized by Mg2+ by Lewis acid pathway with PhIO can yield transꢀstilbene as
or Zn2+.
Epoxidation Mechanism. It is generally accepted that
product in addition to cisꢀepoxide,43 however, this
isomerization was not observed here; 4) in using
iodosylbenzene analog as oxidant, Lewisꢀacidꢀassisted oxygen
transfer, in which the valence of metal ion does not change, was
commonly observed with PhIO;43,49,50 here, using PhIO for
cyclooctane epoxidation gave only 27.1% conversion with 23.3%
yield of epoxide (Table S5), much lower than that of using
PhI(OAc)2. Accordingly, the generation of the active
intermediate, Mn(IV)=O/Al3+ adduct, does not proceed by
oxidation of manganese(II) catalyst with potentially in situ
generated PhIO, but with PhI(OAc)2 directly.
In active Mn+=O moieties mediated olefin epoxidation, it has
been reported that there exist at least two distinct reaction
pathways:27,51 1) olefin attacking on the Mn+=O moiety through
one electron oxidation to generate an olefinic C=C π bond
broken radical intermediate, followed by collapse or rotationꢀ
collapse process which in turn provides the mixture of cis and
trans epoxides, known as the radicalꢀtype oxygen atom
transfer;28,52 2) 2+2 cycloaddition of the Mn+=O moiety with
olefin to generate a metallaoxetane intermediate, followed by
its subsequent breakdown to form epoxide, which remains the
stereoꢀstructure of olefin, known as the concerted oxygen atom
transfer pathway.53ꢀ55 It is worth to mention that, in concerted
versatile active Mn+=O species can conduct oxygen atom
transfer reaction like epoxidation, and the key evidence to
support this comes from 18O isotope labelling experiments. In
Groves’ studies, MnV(TMPyP)(O) species can exchange its oxo
functional group with 18Oꢀwater to form MnV(TMPyP)(18O).
Next, it was identified that 18O can be subsequently
incorporated into olefin, forming the 18Oꢀepoxide product, thus
the MnV=O moiety was established as the active intermediate in
olefin epoxidation.42 Here, incorporation of 18O into epoxide
from 18Oꢀwater was also observed in Mn(TPA)Cl2 catalyzed
epoxidation of cisꢀstilbene (Figure S4). Importantly, the
abundance of 18O in epoxide is highly Al3+ꢀdependent. In the
absence of Al3+, the yields of cisꢀ and transꢀepoxide were 5.9%
and 3.5% (Table 3), in which the abundances of 18O were 89.2%
and 81.0%, respectively (Table 4). Along with increasing the
dosage of Al3+, the yield of cisꢀepoxide sharply increased up to
50.7%, whereas the abundance of 18O in cisꢀepoxide decreased
down to 66.1%. This decrease of the 18O enrichment can be
attributed to the fact that with the addition of Al3+, the
generated Mn(IV)=O/Al3+ adduct is highly active for
epoxidation, which reacts with olefin much faster than its
exchange with 18Oꢀwater, otherwise Al3+ has delayed the 18O
exchange between the Mn(IV)=O/Al3+ adduct and 18Oꢀwater oxygen atom transfer process, the exact transition state is still
(
vide infra).
elusive. The threeꢀmember ring transition state, generated by
the oxygen atom of the Mn+=O moiety attacking on the olefinic
double bond, has also been proposed.56
Table 4. Isotopic labelling experiments using H218O for the catalytic
oxidation of cisꢀstilbene
Here, in the absence of Al3+, the sluggish Mn(TPA)Cl2 alone
provides comparable yields of cisꢀ and transꢀepoxide in each
case (Table S3), indicating the occurrence of an olefinic C=C π
bond broken radical intermediate (pathway A in Scheme 2).
The sluggish activity of the manganese(II) catalyst alone is
related to the formation of the dominant diamond core structure
which is believed to be much less active than the corresponding
Mn+=O moieties.39,55 As shown in Table 4, the high enrichment
of 18O in epoxide indicates that, here, the manganese species
responsible for epoxidation can rapidly exchange oxygen with
18Oꢀwater, which eliminates the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core serving epoxidation. As described in literature, a feasible
coordination site on the manganese intermediate is essential for
18O exchange with 18Oꢀwater, in which binding of 18Oꢀwater
followed by protonation on the Mn+=16O moiety and
simultaneous deprotonation on 18O generates the 18Oꢀlabelled
Mn+=18O species.57 In the case of diꢀꢁꢀoxo bridged Mn dimer,
coordination of water at the vacated sites is also essential for
the ꢁꢀoxo exchange happens.58 Apparently, our wellꢀ
characterized diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV core, in which both
manganese ions are coordination saturated in diamond core,
cannot offer feasible oxygen exchange with 18Oꢀwater to
achieve a high enrichment of 18O in epoxide. Accordingly, the
The abundance of 18O (%)
Substrate
Product
1:0[a]
1:0.5[a]
1:1[a]
1:2[a]
18O-labelled cisꢀ
epoxide
18O-labelled transꢀ
89.2
68.6
67.5
66.1
cisꢀstilbene
81.0
65.0
63.9
65.5
epoxide
Conditions: acetonitrile/CH2Cl2 (4:1, v/v) 0.5 mL, cisꢀstilbene 0.1 M,
Mn(TPA)Cl2 2 mM, PhI(OAc)2 0.2 M, H218O 0.05 mL, 273 K, 8 h. [a] The ratio
of Mn(II):Al3+ as catalyst.
One may argue that, in literature, some manganese
complexes and redoxꢀinactive metal ions have been reported to
serve as Lewis acid to assist olefin epoxidation with oxidant in
which the valence of metal ions does not change.43ꢀ48 Here, we
still prefer that it is manganese oxo moiety that delivers the
oxygen atom to olefin to form epoxide due to several reasons: 1)
control experiments using either Mn(TPA)Cl2 or redoxꢀinactive
metal ions alone as catalyst provided very limited olefin
conversion and epoxide formation, supporting none of them can
serve as Lewis acid to achieve efficient epoxidation; 2) in UVꢀ
Vis studies, the weak but nontheless absorption bands beyond
500 nm can be observed by adding Al3+ to the manganese(II)
complex in the presence of PhI(OAc)2, and it decays with the
high enrichment of 18O in epoxide products (89.2 % in cis
ꢀ
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 9 of 12
Journal Name
Dalton Transactions
View Article Online
ARTICLE
DOI: 10.1039/C4DT03993A
epoxide and 81.0 % in transꢀepoxide, Table 4) suggests that thus facilitate the oxygen atom transfer. In the case of 18O
there exist certain active species other than diamond MnIIIꢀ(ꢁꢀ labelling experiments, the decreased 18O enrichment in epoxide
O)2ꢀMnIV core responsible for epoxidation. In literature, the can be attributed to the presence of Al3+ which delays the 18O
equilibrium between the dimeric core and monomeric metal exchange between the MnIV=O•••Al3+ species and 18Oꢀwater
oxo moiety has been reported.32 Accordingly, similar minor (Table 4).
free MnIV=O moiety may exist in the reaction solution when
Alternatively, another form of Mn(IV)=O/Al3+ adduct,
Mn(TPA)Cl2 alone is employed as catalyst, and it is responsible O=MnIVꢀOꢀAl3+, may be generated when the diamond MnIIIꢀ(ꢁꢀ
for epoxidation. Due to its low concentration, the catalytic O)2ꢀMnIV core structure is disassembled in the presence of
efficiency of Mn(TPA)Cl2 is poor.
redoxꢀinactive metal ions and excess oxidant. Similarly, a
metallaoxetane intermediate can also be formed by 2+2
cycloaddition of O=MnIVꢀOꢀAl3+ and olefin, as shown in
Scheme 2. In this case, the dramatically increased epoxidation
efficiency can be attributed to two main factors: (1) generation
of a highly active MnIV=O species by dissociation of sluggish
diꢀꢁꢀoxoꢀbridged species with Al3+; and (2) the linkage of Al3+
through ꢁꢀOꢀbridge with manganese(IV) species, not via
directly bound to the oxygen atom which will be delivered to
olefin, increases the net charge of the manganese(IV)
intermediate, thus improves the oxygen transfer capability of
the MnIV=O moiety in epoxidation. Apparently this hypothesis
is also consistent with the fact that redoxꢀinactive metal ions
with higher positive charge demonstrate better efficiency in
promoting the epoxidation efficiency of the Mn(II) complex
(see Table 1). In this case, 18O exchange is also delayed
because binding of Al3+ to form O=MnIVꢀOꢀAl3+ prevents the
coordination of 18Oꢀwater onto the manganese(IV) ion.
Complimentarily, the concerted oxygen transfer from the
O=MnIV functional group in the O=MnIVꢀOꢀAl3+ unit to the
substrate, which is not illustrated in Scheme 2, may happen
without the formation of a metallaoxetane as described in
literature.56
The plausible O=MnIVꢀOꢀAl3+ species demonstrated here
recalls the reported O=FeIVꢀOꢀFeIII species from the geometric
and electronic structural point of view. In literature, the
formation of a terminal FeIV=O unit in open core was also
proposed as the main factor that contributes the enhancement of
reactivity towards CꢀH bond cleavage and oxygen atom transfer
reactions.36 This phenomenon can be extended to help
understanding the catalytic mechanism of MMO which has a
diamond FeIIIꢀ(ꢁꢀO)2ꢀFeIV core as the key active Q intermediate.
If it is correct, the second iron in MMO enzymes may enhance
the hydrogen abstraction capability of the FeIV=O from the
robust methane through the dissociation of diamond FeIIIꢀ(ꢁꢀ
O)2ꢀFeIV core to form more active O=FeIVꢀOꢀFeIII structure as
well as the plausible O=MnIVꢀOꢀAl3+ species here. In view of
this point, this work may provide new clues in understanding
the catalytic mechanism of methane monooxygenase and
identifying the real reactive species in catalytic cycle.
Scheme 2. Proposed mechanism of redoxꢀinactive Al3+ promoted epoxidation by
Mn(TPA)Cl2 catalyst.
In the case of adding Al3+, the yield of cisꢀepoxide (50.7%)
was sharply improved, whereas the yield of transꢀepoxide
remains unchanged (5.8%) (it is worth to mention that again,
Al3+ alone is very sluggish for epoxidation as demonstrated in
Table 1). Clearly, the presence of Al3+ has shifted the
epoxidation from the radicalꢀtype oxygen atom transfer
(pathway A) to the concerted oxygen atom transfer, which leads
to the high stereoselectivity of cisꢀepoxide (pathway B). This
switch in epoxidation mechanism can be attributed to the
formation of a plausible Mn(IV)=O/Al3+ adduct through Lewis
acid induced dissociation of the diamond MnIIIꢀ(ꢁꢀO)2ꢀMnIV
core as evidenced by EPR and others. Then, the
Mn(IV)=O/Al3+ adduct mediates the concerted oxygen atom
transfer, which provides the stereoꢀstructure maintained
epoxide product with sharply improved yield.
In our previous communication,22 the interaction between
Mn(IV) species and Al3+ was described as Mn(IV)=O•••Al3+
for which a similar, though controversial, structure has been
illustrated in literature.7ꢀ9 In this case, a metallaoxetane
intermediate can be generated by 2+2 cycloaddition of
Mn(IV)=O•••Al3+ with olefin as shown in Scheme 2. Because
of the Lewis acidity of Al3+, MnꢀO bond in this metallaoxetane
intermediate is weakened and oxygen atom can be easily
transferred to the substrate, which remarkably increases the
epoxidation efficiency (Table 2). Also this hypothesis has
rationalized the fact that redoxꢀinactive metal ions with higher
positive charge demonstrate better efficiency in promoting the
Mn(II) complex catalyzed epoxidation (Table 1), in which
higher Lewis acidity may lead to a more stretched MnꢀO bond
The stability and reactivity of high valent transition metal
clusters having ꢁꢀoxo bridge have attracted much attention over
the past decades, however, the formation of diꢀꢁꢀoxoꢀbridged
core from the Mn+=O moieties also greatly reduce its oxidative
activity.39,55 Apparently our results have illustrated a novel
strategy to explore the catalytic reactivity of a variety of redox
metal complexes which form inactive clusters under oxidation
conditions. To further testify this strategy, manganese
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 9

Dalton Transactions
Page 10 of 12
ARTICLE
VJieowuArrtnicalelONnlainme e
DOI: 10.1039/C4DT03993A
complexes with different ligands are employed as catalyst for
This work was supported by the National Natural Science
the epoxidation of cyclooctene (Table 5). Due to the formation Foundation of China (No 21303063 and 21273086). The GCꢀ
of the binuclear diamond core structure under oxidative MS analysis was performed in the Analytical and Testing
conditions as reported in literature,59 these manganese Center of Huazhong University of Science and Technology. Mr.
complexes alone are very sluggish in olefin epoxidation as well Hong Yi was appreciated sincerely for EPR analysis in Wuhan
as Mn(TPA)Cl2 complex. For example, Mn(BnꢀTPEN)Cl2 University.
alone as catalyst only gave 18.5% conversion and 8.7% yield in
epoxidation of cyclooctene. Remarkably, the addition of Lewis
Notes and references
acid sharply promoted conversion up to 95.3% with 88.9%
School of Chemistry and Chemical Engineering, Huazhong University of
yield of epoxide. This acceleration has been observed in each
Science and Technology. Key Laboratory for LargeꢀFormat Battery
case, which further supports the effectiveness of our strategy.
Materials and System, Ministry of Education. Luoyu Road 1037, Wuhan
That is, in situ dissociating of these sluggish dimeric cores to
430074, PR China. Eꢀmail: zqchen@hust.edu.cn, ligxabc@163.com,
generate the active intermediates by simply adding redoxꢀ
gyin@hust.edu.cn.
inactive metal ions can provide a convenient protocol to
Electronic Supplementary Information (ESI) available: Experimental
explore their applications in versatile oxidation processes.
details of catalytic epoxidation, GCꢀMS graphs, EPR spectra are available
in supporting information. See DOI: 10.1039/c000000x/
Table 5. Catalytic epoxidation of cyclooctene by Mn(II) complexes bearing
different ligands
1. I. Rivalta, G. W. Brudvig and V. S. Batista, Curr. Opin. Chem. Biol.
2012, 16, 11.
,
Al(OTf)3
mM
Conv.
%
Yield
%
Ligands
2. S. V. Seetharaman, D. D. Winkler, A. B. Taylor, X. H. Cao, L. J.
Whitson, P. A. Doucette, J. S. Valentine, V. Schirf, B. Demeler, M.
0
2
9.9(0.2)
97.8(1.0)
4.1(0.1)
91.4(0.3)
TPA
0
2
18.5(0.8)
95.3(0.3)
8.7(0.3)
88.9(0.1)
C. Carroll, V. C. Culotta and P. J. Hart, Biochemistry, 2010, 49
,
BnꢀTPEN
5714.
0
2
0
2
0
2
11.0(0.8)
76.3(1.1)
14.1(0.6)
86.3(1.7)
10.3(2.0)
99.5(0.5)
5.8(0.5)
67.4(0.3)
6.9(0.5)
72.5(0.8)
2.8(1.0)
79.3(0.4)
Phen[a]
Phendio[a]
BPMEN
3. J. S. Valentine, P. A. Doucette and S. Zittin Potter, Annu. Rev.
Biochem, 2005, 74, 563.
4. R. K. Grasselli, Top Catal, 2002, 21, 79.
5. S. Fukuzumi, Y. Morimoto, H. Kotani, P. Naumov, Y.ꢀM. Lee and
W. Nam, Nat. Chem., 2010, 2, 756.
Conditions: acetonitrile/CH2Cl2 (4:1, v/v) 0.5 mL, cyclooctene 0.05 M, Mn(II)
complex 1 mM, 3.5 h, 273 K. The data in parentheses represent the deviations. [a]
Preꢀmixing MnCl2 with ligand to inꢀsitu generate the catalyst for epoxidation.
6. J. Prakash, G. T. Rohde, K. K. Meier, A. J. Jasniewski, K. M. Van
Heuvelen, E. Münck and L. Que, J. Am. Chem. Soc., 2015, 137, 3478.
7. Y. Morimoto, H. Kotani, J. Park, Y.ꢀM. Lee, W. Nam and S.
Fukuzumi, J. Am. Chem. Soc., 2010, 133, 403.
Conclusions
8. J. Park, Y. Morimoto, Y. M. Lee, Y. You, W. Nam and S. Fukuzumi,
Inorg. Chem., 2011, 50, 11612.
We have illustrated an early example that redoxꢀinactive metal
ions can sharply improve the oxygen atom transfer efficiency of
redox metal catalysts like olefin epoxidation, which represents
another important enzymatic and chemical oxidative process.
Disclosed by various spectroscopies, oxidation of Mn(TPA)Cl2
with PhI(OAc)2 dominantly generated a mixed valent diꢀꢁꢀoxoꢀ
bridged diamond core as MnIIIꢀ(ꢁꢀO)2ꢀMnIV which is very
sluggish for olefin epoxidation. Dissociation of this diamond
core is indicated in the presence of redoxꢀinactive metal ions to
form a plausible MnIV=O•••Al3+ or O=MnIVꢀOꢀAl3+ moiety
which is highly active for epoxidation. Notably, the presence of
Al3+ has shifted the epoxidation mechanism from radicalꢀtype
oxygen atom transfer to concerted oxygen atom transfer. Our
results have illustrated a novel strategy to explore the catalytic
reactivity of a variety of redox metal complexes which form
inactive clusters under oxidative conditions. Moreover, studies
demonstrated here may also provide new clues to understand
the catalytic mechanism of methane monooxygenase.
9. P. Leeladee, R. A. Baglia, K. A. Prokop, R. Latifi, S. P. de Visser and
D. P. Goldberg, J. Am. Chem. Soc., 2012, 134, 10397.
10. W. W. Y. Lam, S.ꢀM. Yiu, J. M. N. Lee, S. K. Y. Yau, H.ꢀK. Kwong,
T.ꢀC. Lau, D. Liu and Z. Lin, J. Am. Chem. Soc., 2006, 128, 2851.
11. Y. J. Park, J. W. Ziller and A. S. Borovik, J. Am. Chem. Soc., 2011,
133, 9258.
12. F. Li, K. M. Van Heuvelen, K. K. Meier, E. Münck and L. Que, J.
Am. Chem. Soc., 2013, 135, 10198.
13. Y. Nishida, Y. M. Lee, W. Nam and S. Fukuzumi, J. Am. Chem. Soc.
,
2014, 136, 8042.
14. J. Park, Y. Morimoto, Y. M. Lee, W. Nam and S. Fukuzumi, J. Am.
Chem. Soc., 2011, 133, 5236.
15. J. Chen, Y. M. Lee, K. M. Davis, X. Wu, M. S. Seo, K. B. Cho, H.
Yoon, Y. J. Park, S. Fukuzumi, Y. N. Pushkar and W. Nam, J. Am.
Chem. Soc., 2013, 135, 6388.
16. H. Yoon, Y.ꢀM. Lee, X. Wu, K.ꢀB. Cho, R. Sarangi, W. Nam and S.
Fukuzumi, J. Am. Chem. Soc., 2013, 135, 9186.
17. C. G. Miller, S. W. GordonꢀWylie, C. P. Horwitz, S. A. Strazisar, D.
K. Peraino, G. R. Clark, S. T. Weintraub and T. J. Collins, J. Am.
Chem. Soc., 1998, 120, 11540.
Acknowledgements
18. G. Yin, Acc. Chem. Res., 2013, 46, 483.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 11 of 12
Journal Name
Dalton Transactions
View ArticAle ROnTliInCeLE
DOI: 10.1039/C4DT03993A
19. L. Dong, Y. Wang, Y. Lv, Z. Chen, F. Mei, H. Xiong and G. Yin, 47. G. Yin, J. M. McCormick, M. Buchalova, A. M. Danby, K. Rodgers,
Inorg. Chem., 2013, 52, 5418.
V. W. Day, K. Smith, C. M. Perkins, D. Kitko, J. D. Carter, W. M.
Scheper and D. H. Busch, Inorg. Chem., 2006, 45, 8052.
48. G. Yin, M. Buchalova, A. M. Danby, C. M. Perkins, D. Kitko, J. D.
Carter, W. M. Scheper and D. H. Busch, J. Am. Chem. Soc., 2005,
127, 17170.
20. H. Guo, Z. Chen, F. Mei, D. Zhu, H. Xiong and G. Yin, Chem.-Asian
J., 2013, 8, 888.
21. Z. Zhang, K. L. Coats, Z. Chen, T. J. Hubin and G. Yin, Inorg.
Chem., 2014, 53, 11937.
22. Z. Chen, L. Yang, C. Choe, Z. Lv and G. Yin, Chem. Commun.
2015, 51, 1874.
,
49. W. Nam and J. S. Valentine, J. Am. Chem. Soc., 1993, 115, 1772.
50. T. Niwa and M. Nakada, J. Am. Chem. Soc., 2012, 134, 13538.
51. N. S. Finney, P. J. Pospisil, S. Chang, M. Palucki, R. G. Konsler, K.
23. B. K. Shin, Y. Kim, M. Kim and J. Han, Polyhedron, 2007, 26, 4557.
24. C. Hureau, G. Blondin, M. F. Charlot, C. Philouze, M. Nierlich, M.
Cesario and E. AnxolabehereꢀMallart, Inorg. Chem., 2005, 44, 3669.
B. Hansen and E. N. Jacobsen, Angew. Chem. Int. Edit., 1997, 36
,
1720.
25. X. Wu, M. S. Seo, K. M. Davis, Y. M. Lee, J. Chen, K. B. Cho, Y. N. 52. M. Palucki, N. S. Finney, P. J. Pospisil, M. L. Guler, T. Ishida and E.
Pushkar and W. Nam, J. Am. Chem. Soc., 2011, 133, 20088.
26. G. Conte, A. J. Bortoluzzi and H. Gallardo, Synthesis, 2006, 3945.
27. R. D. Arasasingham, G. X. He and T. C. Bruice, J. Am. Chem. Soc.
1993, 115, 7985.
N. Jacobsen, J. Am. Chem. Soc., 1998, 120, 948.
53. J. P. Collman, J. I. Brauman, B. Meunier, T. Hayashi, T. Kodadek
and S. A. Raybuck, J. Am. Chem. Soc., 1985, 107, 2000.
54. E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker and L. Deng, J.
Am. Chem. Soc., 1991, 113, 7063.
,
28. W. Adam, K. J. Roschmann, C. R. SahaꢀMöller and D. Seebach, J.
Am. Chem. Soc., 2002, 124, 5068.
55. E. M. McGarrigle and D. G. Gilheany, Chem. Rev., 2005, 105, 1563.
29. R. V. Ottenbacher, D. G. Samsonenko, E. P. Talsi and K. P. 56. L. Kurti, M. M. Blewett and E. J. Corey, Org. Lett., 2009, 11, 4592.
Bryliakov, Acs Catal., 2014,
4
, 1599.
57. J. Bernadou, A.ꢀS. Fabiano, A. Robert and B. Meunier, J. Am. Chem.
Soc., 1994, 116, 9375.
30. S. E. Park, W. J. Song, Y. O. Ryu, M. H. Lim, R. Song, K. M. Kim
and W. Nam, J. Inorg. Biochem., 2005, 99, 424.
58. R. Tagore, R. H. Crabtree and G. W. Brudvig, Inorg. Chem., 2007,
46, 2193.
31. C. Hureau, E. AnxolabehereꢀMallart, G. Blondin, E. Riviere and M.
Nierlich, Eur. J. Inorg. Chem., 2005, 4808.
59. E. M. Mcgarrigle and D. G. Giheany, Chem. Rev., 2005, 105, 1563.
32. C. Hureau, L. Sabater, E. AnxolabehereꢀMallart, M. Nierlich, M. F.
Charlot, F. Gonnet, E. Riviere and G. Blondin, Chem.-Eur J, 2004,
10, 1998.
33. S. D. Hicks, D. Kim, S. L. Xiong, G. A. Medyedey, J. Caruthers, S.
Hong, W. Nam and M. M. AbuꢀOmar, J. Am. Chem. Soc., 2014, 136
,
3680.
34. J. Kliava and J. Purans, J. Magn. Reson., 1980, 40, 33.
35. M. V. Golynskiy, W. A. Gunderson, M. P. Hendrich and S. M.
Cohen, Biochemistry, 2006, 45, 15359.
36. G. Q. Xue, R. De Hont, E. Munck and L. Que, Nat. Chem, 2010, 2,
400.
37. H. Zheng, S. J. Yoo, E. Munck and L. Que, J. Am. Chem. Soc., 2000,
122, 3789.
38. S. Friedle, E. Reisner and S. J. Lippard, Chem. Soc. Rev., 2010, 39
,
2768.
39. G. Q. Xue, D. Wang, R. De Hont, A. T. Fiedler, X. P. Shan, E.
Munckt and L. Que, P. Natl. Acad. Sci. USA, 2007, 104, 20713.
40. L. J. Boucher and C. G. Coe, Inorg. Chem., 1975, 14, 1289.
41. B. C. Dave and R. S. Czernuszewicz, Inorg. Chim. Acta, 1994, 227
,
33.
42. J. T. Groves, J. B. Lee and S. S. Marla, J. Am. Chem. Soc., 1997, 119
,
6269.
43. Y. H. Yang, F. Diederich and J. S. Valentine, J. Am. Chem. Soc.
,
1991, 113, 7195.
44. S. Y. Tosaki, R. Tsuji, T. Ohshima and M. Shibasaki, J. Am. Chem.
Soc., 2005, 127, 2147.
45. S. H. Wang, B. S. Mandimutsira, R. Todd, B. Ramdhanie, J. P. Fox
and D. P. Goldberg, J. Am. Chem. Soc., 2003, 126, 18.
46. J. P. Collman, L. Zeng and J. I. Brauman, Inorg. Chem., 2004, 43
,
2672.
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 11

Dalton Transactions
Page 12 of 12
View Article Online
DOI: 10.1039/C4DT03993A
Table of contents
Oxygenation efficiency of manganese catalyst can be sharply improved by Lewis acid
which causes the dissociation of diamond Mn2(III,IV) core.