Iron-catalyzed epoxidation of aromatic olefins and 1,3-dienes

Article abstract of DOI:10.1002/adsc.201000091

The combination of iron(III) chloride, pyridine-2,6-dicarboxylic acid and formamidine ligands allows for the epoxidation of styrenes and conjugated dienes in excellent chemoselectivity and yields.

Full text of DOI:10.1002/adsc.201000091

UPDATES
DOI: 10.1002/adsc.201000091
Iron-Catalyzed Epoxidation of Aromatic Olefins and 1,3-Dienes
Kristin Schrçder,a Stephan Enthaler,b Benoꢀt Join,a Kathrin Junge,a
and Matthias Bellera,*
a
Leibniz-Institut fꢁr Katalyse e.V. an der Universitꢂt Rostock, Albert-Einstein-Str. 29a, 18059 Rostock, Germany
Fax : (+49)-381-1281-5000; phone: (+49)-381-1281-0; e-mail: matthias.beller@catalysis.de
Technische Universitꢂt Berlin, Department of Chemistry, Cluster of Excellence “Unifying Concepts in Catalysis”, Straße
b
des 17. Juni 135, 10623 Berlin, Germany
Received: February 3, 2010; Revised: May 22, 2010; Published online: July 7, 2010
Abstract: The combination of iron
(III) chloride,
combination of imidazole ligands together with iron-
pyridine-2,6-dicarboxylic acid and formamidine li-
gands allows for the epoxidation of styrenes and
conjugated dienes in excellent chemoselectivity and
yields.
ACHTUNGTREN(NUGN III) chloride provided a convenient epoxidation cata-
lyst.[11] Unfortunately, in iron-catalyzed epoxidation of
styrenes in the presence of substituted imidazoles as
ligand, with or without pyridine-2,6-dicarboxylic acid
(H2-pydic) as a co-ligand, displayed only moderate
yields and selectivity.[12]
Keywords: epoxidation; hydrogen peroxide; iron;
mechanistic studies; olefins
Based on our formal ligand concept as shown in
Scheme 1 we became attracted by the possibility to
test formamidine ligands instead of imidazole to im-
Introduction
Oxiranes constitute important building blocks in or-
ganic chemistry.[1] Thus, the development of sustaina-
ble, efficient, and selective catalysts for epoxidation
reactions continues to be an important goal in oxida-
tion chemistry. In this respect, the choice of benign
oxidants such as hydrogen peroxide is a key issue.
Scheme 1. Evolution of formamidine ligands from histidine.
prove the epoxidation of aromatic olefins. Notably,
Known molecular-defined transition metal com- the formamidine motif allows easy ligand tuning by
plexes which make use of hydrogen peroxide in cata- substitution of the “amine” part and variation of sub-
lytic epoxidation reactions, are often based on ruthe- stituents on the “imine” nitrogen (Scheme 1). In pre-
nium,[2] rhenium,[3] manganese,[4] and more recently liminary experiments formamidines were successfully
iron.[5] Obviously, the use of iron-based catalysts applied in the epoxidation of trans-stilbene.[13] Herein,
offers significant advantages compared to precious we report an improved protocol for the highly selec-
metals. In addition to its ubiquitous availability, iron tive epoxidation of styrenes and 1,3-dienes applying
is involved in manifold enzyme-catalyzed processes. iron catalysts containing formamidine ligands.
Here, frequently stabilization of the metal centre
takes place via coordination to the imidazole part of
the amino acid histidine, which participates in numer- Results
ous biological structures.[6] For instance, it acts as
ligand and base for non-heme iron enzymes, and is an At the start of our work the reaction of styrene with
emerging motif in the so-called 2-His-1-Carboxylate hydrogen peroxide as terminal oxidant was investigat-
facial triad.[7] As known from several crystal struc- ed in detail. Due to the sensitivity of the resulting sty-
tures of the active sides of enzymes, only the “imine”- rene oxide towards acids, this epoxidation is still a
nitrogen of histidine coordinates to the metal centre. challenging benchmark reaction. In general, catalytic
Therefore, it should be possible to remove the a- tests were run at room temperature under an air at-
amino acid chain and still retain catalytic activity.[8]
Indeed, imidazoles and pyrrazoles have been and various formamidine ligands.
mosphere in the presence of FeCl3·6H2O, H2-pydic
widely employed as additives in epoxidation sys-
As shown in Table 1 N,N-dimethyl-N’-arylformami-
tems.[9,10] More recently, we demonstrated that the dines gave yields of epoxides in the range of 59–78%
Adv. Synth. Catal. 2010, 352, 1771 – 1778
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1771

Kristin Schrçder et al.
Table 2. Epoxidation of styrene in the presence of 2.[a]
UPDATES
Table 1. Epoxidation of styrene: Variation of formamidine
ligands.[a]
No H2-pydic FeCl3·6H2O 2
[mol%] [mol%]
Conv.[b] Sel.[c] TON[d]
[%]
[mol%] [%]
1
2
3
4
5
6
7
8
–
5
5
–
–
5
2.5
1
0.5
1
5
-
5
–
5
5
2.5
1
0.5
2
–
–
–
12
12
12
6
2.4
1.2
5
11 (4) 32
1
0
0
0
0
0
0
0
0
0
0
2 (1)
23
94 (93) >99 19
97 (87) 90
71 (65) 91
35
65
0
9
0
0
0
0
10
11
12
13
0
1
1
1
1
1
1
3
2
1
55 (49) 89
50 (43) 88
49
43
0
No Ligand R1
Conv.[b] Yield[b] Sel.[c]
[%]
[%]
[%]
5 (0)
0
[a]
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1f
1g
1h
2
C6H5
75
71
98
>99
76
62
59
78
74
62
19
76
75
93
69
2
83
84
83
74
81
33
87
82
>99
78
37
Reaction conditions: 0.5 mmol styrene, 5 mol%
FeCl3·6H2O and 12mol% formamidine ligand, 5mol%
H2-pydic, tert-amyl alcohol (9 mL), 0.44 mmol dodecane,
(100 mL, internal standard) were added in sequence at
room temperature in air. To this mixture a solution of
30% H2O2 (170 mL, 1.5 mmol) in tert-amyl alcohol
(830 mL) was added over a period of 1 h at room temper-
ature by a syringe pump.
Conversion and yield were determined by GC analysis
by comparison with authentic samples; for reproducibili-
ty the reactions were at least repeated twice. Yields are
given in brackets.
4-MeC6H4
4-t-BuC6H4
4-CF3C6H4
3,5-F2C6H3
3,4,5-F3C6H2
2,4,6-Me3C6H2 88
2,6-i-Pr2C6H3
Cy
58
92
94
88
9
[b]
9
10
11
3a
3b
C6H5
C6H5
[a]
Reaction conditions: 0.5 mmol styrene, 5 mol%
FeCl3·6H2O and 12 mol% formamidine ligand, 5 mol%
H2-pydic, tert-amyl alcohol (9 mL), 0.44 mmol dodecane,
(100 mL, internal standard) were added in sequence at
room temperature in air. To this mixture a solution of
30% H2O2 (170 mL, 1.5 mmol) in tert-amyl alcohol
(830 mL) was added over a period of 1 h at room temper-
ature by a syringe pump.
Conversion and yield were determined by GC analysis
and by comparison with authentic samples; to check for
reproducibility the reactions were at least repeated twice.
Selectivity (Sel.) refers to the chemoselectivity of epox-
ide from olefin.
[c]
[d]
Selectivity (Sel.) refers to the chemoselectivity of epox-
ide from olefin.
TON=turnover number=mole of product per mole of
catalyst.
1:1:2.5. For example, increasing the amount of iron
compared to H2-pydic inhibited the reaction com-
pletely (Table 2, entry 10). This is also true for a de-
crease of the formamidine ligand compared to
FeCl3·6H2O or H2-pydic (Table 2, entry 13).
Next, we were interested in the limitations of this
reaction. Hence, we added further amounts of sub-
strate and oxidant after 1 h reaction time to achieve
[b]
[c]
(Table 1, entries 1–5, 7, 8). Only, ligand 1f with three the maximal TON for our system. The yield of sty-
fluoro substituents led to a diminished yield of 19% rene oxide is improved up to 97% (39 TON) after the
(Table 1, entry 6). To our delight ligand 2 with an N’- addition of an extra portion substrate (0.5 mmol) and
cyclohexyl group induced both improved conversion oxidant (3 equiv.). Further addition was unsuccessful
and excellent chemoselectivity. On the other hand and the substrate remained intact in the reaction solu-
changing methyl to phenyl substituents at the tion despite an excess of oxidant.
“amine” part of formamidines 3 decreased the yield
and the selectivity for styrene oxide.
Then, the regeneration of the catalyst system was
explored. Therefore ligand 2 (6 mol%), H2-pydic
Thus, the catalytic performance of ligand 2 was (6 mol%), styrene (0.5 mmol), and the oxidant
more closely examined. It is clear that all three com- (3 equiv. added over 1 h via syringe pump) were
ponents of the catalyst system are necessary to obtain added at the end of the reaction to investigate the re-
significant activity (Table 2, entries 1–5). Optimization maining catalytic activity. Unfortunately, the addition
of the catalyst loading resulted in a minimized cata- did not result in any further yield of the epoxide. In-
lyst concentration of 2.5 mol% FeCl3·6H2O (Table 2, stead degeneration of the already formed epoxide
entry 7). In order to obtain full conversion it is impor- took place. In contrast, the addition of FeCl3·6H2O
tant to maintain the iron:H2-pydic:ligand ratio of (2.5 mol%), styrene (0.5 mmol), and oxidant
1772
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1771 – 1778

Iron-Catalyzed Epoxidation of Aromatic Olefins and 1,3-Dienes
Table 3. Fe-catalyzed epoxidation: Scope and limitations.[a]
(3 equiv., 1 h) led to an increase of 26% in yield after
2 h compared to the result after 1 h. Another extra
portion did not enhance the yield any further.
After that, the optimized conditions (2.5 mol%
loading of iron catalyst with a 1:1:2.5 ratio of iron/H2-
pydic/formamidine) were applied in the epoxidation
of different substituted styrenes and 1,3-dienes
(Table 3).
No Substrate
1
Conv.[b] [%] Yield[b] [%] Sel.[c] [%]
97
87
94
90
94
For all styrene derivatives good to very good con-
version with excellent selectivity is achieved (Table 3,
entries 1–7)! In the case of 3-chlorostyrene the cata-
lyst loading was increased to 5 mol% in order to
ensure full conversion (Table 3, entry 6). To our de-
light also conjugated cis,cis-1,3-cyclooctadiene and
1,2-dihydronaphthalene resulted in full conversion
and excellent chemoselectivity towards the mono-oxy-
genated product. However, non-conjugated cis,cis-1,5-
cyclooctadiene showed diminished yield comparable
to cyclooctene (Table 3, entries 10 and 11). Beside
styrenes and 1,3-dienes, the relatively stable substrate
trans-stilbene showed diminished yields compared to
former H2-pydic systems resulting in 63% conversion
and 44% yield. Nevertheless, the iron/H2-pydic/forma-
midine catalyst system is the most efficient iron catalyst
for terminal aromatic olefins.
The formation of side products, such as allylic oxi-
dation products, was examined through GC-MS anal-
ysis for the epoxidation of cyclooctene and 1,5-cyclo-
octadiene. Only in the case of 1,5-cyclooctadiene
were the allylic oxidation products, 1,5-cyclooctadien-
4-one (~5–10% yield) and 1,5-cyclooctadien-4-ol (~3–
5% yield), identified. Therefore a-pinene was chosen
as an additional probe.[14] Indeed, formation of the al-
lylic products verbenol (4,6,6-trimethylbicyclo-
2
3
4
5
6
7
8
9
>99
>99
>99
>99
81 (>99)[f]
95
92
95
92
95
>99 (83)[d,e] >99
79 (94)[f]
92
98 (94)[f]
97
82
98
>99
88
82 (60)[d]
86
10
11
57
27
48
34
18
24
9
70
50
12
[a]
A
and
verbenone
in
(4,6,6-
small
Reaction conditions: 0.5 mmol styrene, 2.5 mol%
FeCl3·6H2O and 6 mol% formamidine ligand, 2.5 mol%
H2-pydic, tert-amyl alcohol (9 mL), 0.44 mmol dodecane,
(100 mL, internal standard) were added in sequence at
room temperature in air. To this mixture a solution of
30% H2O2 (170 mL, 1.5 mmol) in tert-amyl alcohol
(830 mL) was added over a period of 1 h at room temper-
ature by a syringe pump.
Conversion and yield were determined by GC analysis
by comparison with authentic samples.
Selectivity (Sel.) refers to the chemoselectivity of epox-
ide from olefin.
Isolated yields.
Traces of the ring opening product 2-p-tolylacetaldehyde
were detected.
The catalyst loading was increased to 5 mol%
FeCl3·6H2O, 12 mol% formamidine ligand and 5 mol%
H2-pydic to achieve complete conversion.
ACHTUNGTRENNUNG
amounts beside the corresponding epoxide was ob-
served.
After exploring the scope of this novel iron catalyst
system, we were interested in mechanistic aspects. Ini-
tially, the role of water was examined. Water incorpo-
ration is of special mechanistic interest because tauto-
meric iron oxo species can exchange the oxygen of
the hydroperoxide with the oxygen of water.[15] In
general, the used tert-amyl alcohol (stirring at air and
room temperature) had a water content of 0.3–
0.5 wt% as determined by Carl–Fischer titration. Ad-
dition of >10 equiv. of water to the reaction mixture
(50% H2O2 or the urea-H2O2 adduct as oxidant) led
to a significant induction period. However, after 24 h
of stirring a yield of 64% or 31% is detected for 50%
H2O2 or for urea-H2O2 adduct, respectively. Hence,
[b]
[c]
[d]
[e]
[f]
an increased amount of water slows down the reac- system were carried out. Here, logACTHNUTRGNE(UNG kX/kH) was corre-
tion, but does not deactivate the catalyst system.
Next, investigations on the reaction rate of compet-
itive epoxidation reactions of p-substituted styrenes in
the presence of the iron-H2-pydic-formamidine
lated with the Hammett s+ value.
Adv. Synth. Catal. 2010, 352, 1771 – 1778
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1773

Kristin Schrçder et al.
UPDATES
Table 4. Competitive epoxidation of different p-substituted
styrenes.[a]
[b]
.[b]
No
X (Y)
log krel
smb
sJJ
1
2
3
4
5
H
0
0
0.15
0.22
0
Me (H)
Cl (F)
F (Me)
CF3 (Cl)
0.462
0.179
0.240
ꢀ0.264
ꢀ0.29
0.11
ꢀ0.24
0.49
Figure 1. Hammett correlation for the epoxidation of differ-
ent styrenes: y=ꢀ0.9976x+0.3187 with R2 =0.96.
ꢀ0.02
ꢀ0.01
[a]
Reaction conditions: 2.5 mmol of each substrate, 5 mol%
FeCl3·6H2O and 12 mol% formamidine ligand, 5 mol%
H2-pydic, tert-amyl alcohol (9 mL), 0.44 mmol dodecane,
(100 mL, internal standard) were added in sequence at
room temperature in air. To this mixture a solution of
30% H2O2 (57 mL, 0.5 mmol) in tert-amyl alcohol
(443 mL) was added over a period of 1 h at room temper-
ature by a syringe pump.
Correlation following Eq. (1) (X=p-CF3, p-Me, p-
Cl, p-F) gave after a linear regression a reaction con-
stant value of 1=ꢀ1.00ꢁ0.14 (Figure 1) and 1=
ꢀ0.85ꢁ0.31 including styrene. Notably, electron-do-
nating substituents at the phenyl moiety increased the
rate constant of the epoxidation. The relatively small
negative Hammett values indicated less electron den-
sity in the transition state in comparison with the sub-
strates. Similar Hammett values are reported for the
epoxidation of olefins applying hydrogen peroxide as
oxidant in the presence of metal peroxo catalysts, for
example, W or Re complexes.[16] Interestingly, related
[b]
See ref.[17]
Table 5. Epoxidation of styrene in the presence of various
radical trapping agents.[a]
Conv. [%][b]
,
Sel. [%][c],
(%)[d]
styrene oxidations involving iron porphyrins like
No Radical scav-
enger (rs)
nrs
[mol%] (%)[d]
+
[Fe(IV)ACHTUNGTRENNUNG
In order to prove the involvement of a spin delocal-
ization resonance effect, which indicates the participa-
tion of a carbon radical centre,[18] we also performed
experiments with radical scavengers (see Table 5). In
1
2
3
4
5
6
7
–
–
100
5
100
2.5
100
5
97
0
90
0
TEMPO
TEMPO
BPN
BPN
Duroquinone
Duroquinone
13 (13)
32 (54)
15 (86)
63 (90)
67 (82)
0 (44)
55 (74)
5 (67)
79 (84)
91 (74)
.
addition, the sJJ and smb substituent constants were
used for the Hammett correlation as developed by
Jiang and Ji [Eq. (2)].[19]
[a]
Reaction conditions: 0.5 mmol styrene, 2.5 mol%
FeCl3·6H2O and 6 mol% formamidine ligand 2,
2.5 mol% H2-pydic, tert-amyl alcohol (9 mL), 0.44 mmol
dodecane, (100 mL, internal standard) were added in se-
quence at room temperature in air. To this mixture, a so-
lution of 30% H2O2 (170 mL, 1.5 mmol) in tert-amyl alco-
hol (830 mL) was added over a period of 1 h at room tem-
perature by a syringe pump.
Conversion and yield were determined by GC analysis
and determined by comparison with authentic samples.
Selectivity refers to the chemoselectivity of epoxide from
olefin.
Here, values of polar (smb) and spin delocalization
.
effects (sJJ ) were established and separated (Table 4,
Figure 2). The negative 1mb value accords with the
Hammett plot trend and shows the electrophilicity of
.
the active iron catalyst. Whereas the positive 1JJ sug-
[b]
[c]
[d]
gests a spin density delocalization at the carbon radi-
cal centre similar to epoxidations performed with
dioxo-Ru-porphyrins as catalysts.[20]
To assess whether the spin polarization or the polar
substitution effect is more important, the magnitude
Yields are obtained after additionally stirring for 3 days.
.
of j1mb/1JJ j was calculated. With a value of 0.86
.
(j1mb/1JJ j<1) it implies a slightly higher significance added several radical scavengers such as TEMPO
of the spin polarization effect in contrast to the polar- (2,2,6,6-tetramethylpiperidine-1-oxyl), BPN (N-tert-
ization effect.
butylphenylnitrone), and duroquinone (2,3,5,6-tetra-
To tackle the problem of possible radical incorpora- methyl-p-benzoquinone) (Table 5). In general, all re-
tion, we performed the epoxidation of styrene with actions led to diminished yields of styrene oxide and
1774
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1771 – 1778

Iron-Catalyzed Epoxidation of Aromatic Olefins and 1,3-Dienes
Figure 2. Dual-parameter Hammett correlation for the ep-
oxidation of different styrenes (Table 4). y=(0.72ꢁ0.07)x +
(0.07ꢁ0.02) with R2 =0.96.
lower chemoselectivity; however only the addition of
1 equiv. TEMPO inhibited the reaction completely.
Apparently, the persistent TEMPO radical blocks
any reactivity and does not favour the oxidation pro-
cess like it is known for alcohol oxidations.[21] BPN
and duroquinone diminished also the conversion and,
in the case of BPN, selectivity, too.
Figure 3. Oxidation of b-pinene with hydrogen peroxide in
the presence of Fe/H2-pydic/2.
Next, b-pinene was used as a so-called indicative
substrate. It is well known that in the case of radical by GC-MS analysis, which were identified by compar-
intermediates rearrangement of the strained four- ison with authentic samples.
membered ring occurs to give limonene-type prod-
ucts.[22] Under our optimized conditions with
a cata- dine ligands under catalytic conditions. Hence, we ex-
lyst loading of 2.5 mol% the iron/H2-pydic/2 system posed 1a to two hours of stirring with FeCl3·6H2O
gave a multitude of products (Figure 3).
and H2-pydic in non-dry tert-amyl alcohol.[23] Then,
Finally, we explored the stability of the formami-
AHCTUNGTRENNUNG
Interestingly, epoxide 4a followed by myrtenol 4c hydrogen peroxide and substrate were added. Beside
and the double hydroxylated product 4d constituted the formation of the epoxide, no change or decompo-
the main products. Using perillyl alcohol 4e as sub- sition of the ligand was observed in both cases. How-
strate selective oxidation to the corresponding perilla ever, investigating the stability of ligand 2, we ob-
aldehyde 4f occurred. The significant amount of rear- served partial decomposition to cyclohexylamine 6
ranged products suggests at least partial participation and dicyclohexylurea 9. In addition, small amounts of
of OH radicals in this oxidation process.
N-cyclohexylformamide 8 are observed (Scheme 2).
Apparently, N,N’-1,3-dicyclohexylurea 9 is pro-
The formation of perillyl alcohol 4e and the addi-
tional hydroxylated derivative 4d indicate radical in- duced by the reaction of 6 with N-cyclohexylforma-
termediates. In addition, b-pinene epoxide 4a, myrte- mide 8, which is known in the presence of ruthenium
nol (6,6-dimethylbicyclo
nol) 4c and the cleavage product norinone (6,6-
dimethylbicyclo[3.1.1]heptan-2-one) 4b are observed ligand 2 did not result in similar selectivity (Table 6).
[3.1.1]-hept-2-en-2-yl]metha- complexes.[24]
ACHTUNGTRENNUNG
Testing the different decomposition products of our
ACHTUNGTRENNUNG
Scheme 2. Partial decomposition of ligand 2 (products were identified by GC-MS and comparison with authentic samples).
Adv. Synth. Catal. 2010, 352, 1771 – 1778
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1775

Kristin Schrçder et al.
UPDATES
Table 6. Iron-catalyzed epoxidation of styrenes in the pres-
action as shown by the oxidation of b-pinene. Nota-
bly, ligand 2 is of limited stability during catalysis and
decomposition products of 2 gave significant lower
conversion and/or selectivity.
ence of ligand decomposition products.[a]
No Ligand Substrate
Conv. Yield
[%][b] [%][b]
Sel.
,
[%][c],
(%)[d]
(%)[d]
1
2
3
4
5
6
“2”
5
6
8
>99
2
97
1
2
93
0
75
0
0
76
93
0
77
0
0
78
Experimental Section
General
All solvents and chemicals were obtained commercially and
were used as received. “30%” aqueous H2O2 from Merck
was used as received. The peroxide content varied from
30% to 40% as determined by titration. NMR spectra were
9
5+6
98
7
8
9
6
6
6
6
>99
>99
>99
85
85 (>99)
79 (94)
76 (92)
71 (92)
85 (>99)
79 (94)
76 (92)
83 (97)
measured using
a Bruker AMF 200 spectrometer at
200 MHz (1H), 75.5 MHz (13C). All spectra were recorded in
CDCl3 or CD2Cl2 and chemical shifts (d) are reported in
ppm relative to tetramethylsilane referenced to the residual
solvent peaks. Spectra were measured at room temperature
unless otherwise stated. Mass spectra were in general re-
corded on an HP 6890/5973 GC-MS. In each case character-
istic fragments with their relative intensities in percentages
are shown. Infrared spectra were recorded on a Nicolet
Magna-IR-Serie 550 spectrometer using KBr plates. Wave
numbers (n) are reported in cmꢀ1 and relative intensities are
given (w=weak, m=medium, s=strong, sh=shoulder).
10
[a]
Reaction conditions: 0.5 mmol styrene/substrate, 5 mol%
FeCl3·6H2O and 12 mol% ligand, 5 mol% H2-pydic, tert-
amyl alcohol (9 mL), 0.44 mmol dodecane, (100 mL, inter-
nal standard) were added in sequence at room tempera-
ture in air. To this mixture a solution of 30% H2O2
(170 mL, 1.5 mmol) in tert-amyl alcohol (830 mL) was
added over a period of 1 h at room temperature by a sy-
ringe pump.
Conversion and yield were determined by GC analysis
by comparison with authentic samples.
Selectivity (Sel.) refers to the chemoselectivity of epox-
ide from olefin.
Synthesis of N,N-Dimethylformamidines
Dimethylformamide (25.3 mmol) was dissolved in dry dieth-
yl ether (50 mL) and cooled to 108C. Under an atmosphere
of nitrogen phosphorus oxychloride (25.3 mmol) was added
carefully maintaining the reaction temperature under 108C,
while a white precipitate was formed. The mixture was
stirred for two hours at room temperature yielding an in-
soluble oily residue. A white precipitate was obtained after
cooling to 108C and slow addition of the corresponding pri-
mary amine (25.3 mmol). The stirring was continued for
12 h at room temperature. The mixture was transferred to a
separation funnel and carefully treated with water (50 mL)
and an aqueous sodium hydroxide solution until the aque-
ous layer reached a pH >8. The aqueous layer was extract-
ed with diethyl ether (2ꢄ50 mL). The combined organic
layers were washed with water (50 mL) and brine (50 mL).
Drying with Na2SO4 and removal of the solvent yield the
crude product, which was purified by distillation.
[b]
[c]
[d]
Yield and selectivities obtained with ligand 2.
Furthermore, only in the presence of cyclohexylamine
were comparable conversions albeit with lower selec-
tivity obtained (Table 6, entry 3). This is also true for
other styrene substrates (Table 6, entries 7–10). These
results demonstrate the importance of the formami-
dine ligand structure.
N,N-Dimethyl-N’-cyclohexylformamidine (2): yield: 8%;
white solid, mp 378C; bp 50–528C (1 mbar); 1H NMR
(200 MHz, CDCl3, 258C): d=1.03–1.85 (m, 10H, C6H10),
2.82 (s, 6H, CH3), 7.33 [s, 1H, N=C(H)N]; 13C NMR
(50 MHz, CDCl3, 258C): d=25.5, 25.7, 36.2, 37.2, 64.8, 153.3
[N=C(H)N]; IR (KBr): n=2929 (s), 1651 (s), 1600 (m),
1376 (m), 1325 (m), 1108 (m), 1076 (m), 844 (w) cmꢀ1; MS
(ESI): m/z=155 (25, M+), 73 (100); UV-VIS (EtOH, 258C):
l=276.5 nm.
Conclusions
In summary, we have developed novel bio-inspired
iron catalysts for oxidation reactions. The convenient
and inexpensive combination of FeCl3·6H2O, H2-
pydic, and formamidine ligand 2 allows for the epoxi-
dation of substituted styrenes and conjugated 1,3-
dienes in good to excellent yield and chemoselectivity
(8 examples with selectivities >90%). Improved re-
sults are obtained compared to all previously known
iron catalysts for this type of epoxidations. Although
we assume an electrophilic peroxide mechanism, par-
General Procedure for the Epoxidation of Olefins
In a test tube, FeCl3·6H2O (0.0125 mmol), tert-amyl alcohol
(9 mL), formamidine ligand (0.030 mmol), pyridine-2,6-di-
carboxylic acid (0.0125 mmol), olefin (0.5 mmol) and dodec-
tial radical formation might take place during the re- ane (GC internal standard, 100 mL) were added in sequence
1776
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1771 – 1778

Iron-Catalyzed Epoxidation of Aromatic Olefins and 1,3-Dienes
at room temperature in air. To this stirred mixture, a solu-
tion of 30% hydrogen peroxide (aqueous, 170 mL, 1.5 mmol)
in tert-amyl alcohol (830 mL) was added over a period of 1 h
at room temperature by a syringe pump. In some cases the
reaction mixture was additionally stirred for 5 h to achieve
better results. Conversion and yield were determined by GC
analysis without further manipulations and compared with
authentic samples. In isolation purposes the reaction mix-
ture was quenched with Na2SO3 to destroy the excess of hy-
drogen peroxide and diluted with water. The solution was
extracted with ethyl acetate (3ꢄ10 mL) and dried over an-
hydrous MgSO4. After filtration and solvent removal under
vacuum, the crude product was purified either by distillation
or by silica gel chromatography on a short column (hexane:
ethyl acetate 20:1, 1% Et3N). For isolation in some cases
the reaction mixture was filtered over a short column
(silica) and washed with hexane:Et3N (100:1). Afterwards
the solvent was removed under vacuum, followed by a silica
gel chromatography (hexane: ethyl acetate 20: 1, 1% Et3N).
Data of the synthesized epoxides were compared with litera-
ture known data.[25]
Acknowledgements
We thank the State of Mecklenburg-Western Pommerania, the
Federal Ministry of Education and Research (BMBF) and
the Deutsche Forschungsgemeinschaft (SPP 1118 and Leib-
niz-prize) for financial support. K. S. appreciates the finan-
cial support provided by the Graduiertenkolleg 1213 “Neue
Methoden fꢀr Nachhaltigkeit in Katalyse und Technik” and
the Max-Buchner-Forschungsstiftung (DECHEMA). Dr. S.
E. thanks the Cluster of Excellence “Unifying Concepts in
Catalysis” (sponsored by the Deutsche Forschungsgemein-
schaft and administered by the Technische Universitꢁt
Berlin). Dr. B. J. thanks the Alexander-von-Humboldt-Stif-
tung. Dr. habil. Haijun Jiao is acknowledged for critical dis-
cussion during the preparation of the manuscript and Mr. S.
Peitz for performing the Carl-Fischer titration. Mrs. M.
Heyken, Mrs. K. Mçller and Mr. G. Wienhçfer (LIKAT) are
acknowledged for their valuable support in the laboratory.
p-(Trifluoromethyl)phenyloxirane: 1H NMR (300 MHz,
CDCl3, 258C): d=7.60 (d, J=8.1 Hz, 2H), 7.39 (d, J=
8.3 Hz, 2H), 3.91 (dd, J=2.6 Hz, J=4.1 Hz, 1H), 3.18 (dd,
J=4.1 Hz, J=5.7 Hz, 1H), 2.76 (dd, J=2.6 Hz, J=5.6 Hz,
1H); 13C NMR (100 MHz, CDCl3, 258C): d=142.0, 125.9,
125.6 (q, J=3.7 Hz), 51.9, 51.6; MS (EI): m/z (rel. int.)=187
(19) [Mꢀ1]+, 169 (14), 158 (46), 138 (22), 119 (100), 109
(23), 91 (40), 63 (15).
References
[1] a) L. P. C. Nielsen, E. N. Jacobsen, in: Aziridines and
Epoxides in Organic Synthesis, (Ed.: A. K. Yudin),
Wiley-VCH, Weinheim, 2006, pp 229–269; b) J. F.
Larrow, E. N. Jacobsen, Top. Organomet. Chem. 2004,
6, 123–152; c) Q.-H. Xia, H.-Q. Ge, C.-P. Ye, Z.-M.
Liu, K.-X. Su, Chem. Rev. 2005, 105, 1603–1662.
[2] For recent work on Ru-based epoxidations, see: a) D.
Chatterjee, Coord. Chem. Rev. 2008, 176–198; b) M. K.
Tse, C. Dçbler, S. Bhor, M. Klawonn, W. Mꢂgerlein, H.
Hugl, M. Beller, Angew. Chem. 2004, 116, 5367–5372;
Angew. Chem. Int. Ed. 2004, 43, 5255–5260; c) S. Bhor,
G. Anilkumar, M. K. Tse, M. Klawonn, C. Dçbler, B.
Bitterlich, A. Grotevendt, M. Beller, Org. Lett. 2005, 7,
3393–3396; d) M. K. Tse, M. Klawonn, S. Bhor, C.
Dçbler, G. Anilkumar, H. Hugl, W. Mꢂgerlein, M.
Beller, Org. Lett. 2005, 7, 987–990; e) I. W. C. E.
Arends, T. Kodama, R. A. Sheldon, in: Topics in Or-
ganometallic Chemistry, (Eds.: C. Bruneau, P. H. Dix-
neuf), Springer-Verlag, Berlin, Heidelberg, 2004,
pp 277–320.
[3] a) W. A. Herrmann, R. W. Fischer, D. W. Marz, Angew.
Chem. 1991, 103, 1706–1709; Angew. Chem. Int. Ed.
Engl. 1991, 30, 1638–1641; b) W. A. Herrmann,
A. M. J. Rost, J. K. M. Mitterpleininger, N. Szesni, W.
Sturm, R. W. Fischer, F. E. Kꢁhn, Angew. Chem. Int.
Ed. 2007, 46, 7901–7903; c) M. Zhou, J. Zhao, J. Li, S.
Yue, C. Bao, J. Mink, S. Zang, F. E. Kꢁhn, Chem. Eur.
J. 2007, 13, 158–166; d) F. E. Kꢁhn, A. Scherbaum,
W. A. Herrmann J. Organomet. Chem. 2004, 689, 4149–
4164; e) A. Schrçckeneder, P. Traar, G. Raber, J.
Baumgartner, F. Belaj, N. C. Mçsch-Zanetti, Inorg.
Chem. 2009, 48, 11608–11614.
p-Methylphenyloxirane: 1H NMR (300 MHz, CD2Cl2,
258C): d=7.18–7.13 (m, 4H), 3.79 (dd, J=2.6 Hz, J=
4.1 Hz, 1H), 3.09 (dd, J=4.0 Hz, J=5.5 Hz, 1H), 2.77 (dd,
J=2.6 Hz, J=5.5 Hz, 1H), 2.34 (s, 3H); 13C NMR
(100 MHz, CD2Cl2, 258C): d=138.4, 135.1, 129.5, 125.8,
52.7, 51.3, 21.3; MS (EI): m/z (rel. int.)=134 (30) [M]+, 119
(22), 105 (100), 98 (25), 91 (18), 75 (16), 77 (24).
1,2-Dihydronaphthaleneoxirane: 1H NMR (300 MHz,
CD2Cl2, 258C): d=7.00–7.34 (m, 4H), 3.77 (d, J=4.3 Hz,
1H), 3.65 (m, 1H), 2.70
ACHTUNGTNER(NUNG m, 1H), 2.48 (m, 1H), 2.33 (m,
1H), 1.69
ACHTUNGTRENNUNG
136.7, 132.6, 129.6, 128.5, 128.4, 126.2, 55.2, 52.8, 24.4, 21.8;
MS (EI): m/z (rel. int.)=146 (100) [M]+, 145 (27), 131 (35),
112 (49), 113 (57), 115 (61), 104 (82), 91 (31), 78 (15), 63
(15).
General Procedure for the Competetive Epoxidation
of p-Substituted Styrenes
A test tube was charged in sequence with FeCl3·6H2O
(0.025 mmol), tert-amyl alcohol (9 mL), formamidine ligand
(0.060 mmol), pyridine-2,6-dicarboxylic acid (0.025 mmol)
and dodecane (GC internal standard, 100 mL) at room tem-
perature in air. Then two different p-substituted styrenes
(each 2.5 mmol) were added and the solution was stirred for
five minutes. To this mixture a solution of 30% hydrogen
peroxide (aqueous, 57 mL, 0.5 mmol) in tert-amyl alcohol
(443 mL) was added over a period of 1 h at room tempera-
ture by a syringe pump. Conversion and yield were deter-
mined by GC analysis without further manipulations and
compared with authentic samples. Reactions were carried
out twice to check for reproducibility.
[4] a) E. M. McGarrigle, D. G. Gilheany, Chem. Rev. 2005,
105, 1563–1602; b) K. Muniz-Fernandez, C. Bolm, in:
Transition Metals for Organic Synthesis, 2nd edn.,
(Eds.: M. Beller, C. Bolm), Wiley-VCH, Weinheim,
2004, pp 344–356.
[5] For reviews on iron based catalysts see: a) C. Bolm, J.
Legros, J. Le Paih, L. Zani, Chem. Rev. 2004, 104,
6217–6254; b) B. S. Lane, K. Burgess, Chem. Rev. 2003,
Adv. Synth. Catal. 2010, 352, 1771 – 1778
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1777

Kristin Schrçder et al.
UPDATES
103, 2457–2473; c) M. Costas, M. P. Mehn, M. P.
Jensen, L. Que Jr, Chem. Rev. 2004, 104, 939–986;
d) E. B. Bauer, Current Org. Chem. 2008, 12, 1341–
1369; for recent iron-catalyzed oxidations see: e) M. S.
Chen, M. C. White, Science 2007, 318, 783–787; f) A.
Company, L. Gꢅmez, M. Gꢁell, X. Ribas, J. M. Luis, L.
Que, Jr., M. Costas, J. Am. Chem. Soc. 2007, 129,
15766–15767.
[13] S. Enthaler, K. Schrçder, S. Inuoe, B. Eckhardt, K.
Junge, M. Beller, M. Driess, Eur. J. Org. Chem. 2010,
accepted.
[14] U. Neuenschwander, F. Guignard, I. Hermans, Chem-
SusChem 2010, 3, 75–84.
[15] a) A. Company, L. Gꢅmez, X. Fontrodona, X. Ribas,
M. Costas, Chem. Eur. J. 2008, 14, 5727–5731; b) J.
Bautz, P. Comba, C. Lopez de Laorden, M. Menzel, G.
Rajaraman, Angew. Chem. 2007, 119, 8213–8216;
Angew. Chem. Int. Ed. 2007, 46, 8067–8070.
[6] S. J. Lippard, J. M. Berg, in: Principles of Bioinorganic
Chemistry University Science Books, Mill Valley, CA,
1994.
˘
[16] A. M. Al-Ajlouni. ꢆ. Saglam, T. Diafla, F. E. Kꢁhn, J.
[7] a) E. L. Hegg, L. Que, Jr., Eur. J. Biochem. 1997, 250,
625–629; b) P. C. A. Bruijnincx, G. van Koten, R. J. M.
Klein Gebbink, Chem. Soc. Rev. 2008, 37, 2716–2744.
[8] One of the referees pointed out that this displays a
formal point of view due to the importance of the spe-
cial protein environment in each enzyme.
[9] For iron see: W. Nam, H.-J. Lee, S.-Y. Oh, C. Kim,
H. G. Jang, J. Inorg. Biochem. 2000, 80, 219–225.
[10] For manganese, see: a) J.-P. Renaud, P. Battioni, J. F.
Bartoli, D. Mansay, J. Chem. Soc. Chem. Commun.
1985, 888–889; b) P. Pietikꢂinen, Tetrahedron Lett.
1994, 35, 941–944; c) R. Krishnan, S. Vancheesan, J.
Mol. Catal. A 1999, 142, 377–382.
[11] For work our on iron/imidazole systems, see: a) K.
Schrçder, X. Tong, B. Bitterlich, M. K. Tse, F. G. Gelal-
cha, A. Brꢁckner, M. Beller, Tetrahedron Lett. 2007,
48, 6339–6342; b) K. Schrçder, S. Enthaler, B. Bitter-
lich, T. Schulz, A. Spannenberg, M. K. Tse, K. Junge,
M. Beller, Chem. Eur. J. 2009, 15, 5471–5481.
Mol. Catal. A: Chem. 2008, 187, 159–164.
[17] J. T. Groves, Y. Watanabe, J. Am. Chem. Soc. 1986, 108,
507–508.
[18] F. G. Gelalcha, G. Anilkumar, M. K. Tse, A. Brꢁckner,
M. Beller, Chem. Eur. J. 2008, 14, 7687–7698.
[19] X. K. Jiang, Acc. Chem. Res. 1997, 30, 283–289.
[20] C.-J. Liu, W.-Y. Yu, C.-M. Che, C.-H. Yeung, J. Org.
Chem. 1999, 64, 7365–7374.
[21] R. A. Sheldon, I. W. C. E. Arends, Adv. Synth. Catal.
2004, 346, 1051–1071.
[22] F. Bellesia, L. Forti, F. Gheli, U. M. Pagnoni, Synth.
Commun. 1997, 27, 961–971.
[23] The reaction was carried out under standard condition
with 5 mol% FeCl3·6H2O, 5 mol% H2-pydic and
12 mol% 1a. Afterwards the substrate trans stilbene
and hydrogen peroxide was added.
[24] T. Kondo, S. Kotachi, Y. Tsuji, Y. Watanabe, T. Mitsu-
do, Organometallics 1997, 16, 2562–2570.
[25] a) H. Sasaki, R. Irie, T. Hamada, K. Suzuki, T. Katsuki,
Tetrahedron 1994, 50, 11827–11838; b) M. K. Tse, S.
Bhor, M. Klawonn, G. Anilkumar, H. Jiao, A. Span-
nenberg, C. Dçbler, W. Mꢂgerlein, H. Hugl, M. Beller,
Chem. Eur. J. 2006, 12, 1875–1888.
[12] For recent work on similar iron/pydic systems, see:
a) B. Bitterlich, G. Anilkumar, F. G. Gelalcha, B. Spilk-
er, A. Grotevendt, R. Jackstell, M. K. Tse, M. Beller,
Chem. Asian J. 2007, 2, 521; b) B. Bitterlich, K. Schrçd-
er, M. K. Tse, M. Beller, Eur. J. Org. Chem. 2008, 29,
4867–4870.
1778
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1771 – 1778