Ligand and pH influence on manganese-mediated peracetic acid epoxidation of terminal olefins

Article abstract of DOI:10.1021/ol048846l

(Chemical Equation Presented) Nineteen Mn^II complexes were screened for the catalytic epoxidation of terminal olefins using peracetic acid. Few of these complexes are efficient catalysts at pH < 2, but many are effective at 1 mol % catalyst loading at pH 4. With 0.1 mol % loading, four complexes epoxidize 1-octene in ～80% yield in 5 min. The relative reactivity of the catalysts toward different olefins was probed using a multicomponent intermolecular competition reaction.

Full text of DOI:10.1021/ol048846l

ORGANIC
LETTERS
2
004
Vol. 6, No. 18
119-3122
Ligand and pH Influence on
Manganese-Mediated Peracetic Acid
Epoxidation of Terminal Olefins
3
Andrew Murphy, Allyson Pace, and T. Daniel P. Stack*
Chemistry Department, Stanford UniVersity, Stanford, California 94305
stack@stanford.edu
Received June 17, 2004
ABSTRACT
II
Nineteen Mn complexes were screened for the catalytic epoxidation of terminal olefins using peracetic acid. Few of these complexes are
efficient catalysts at pH < 2, but many are effective at 1 mol % catalyst loading at pH 4. With 0.1 mol % loading, four complexes epoxidize
1
-octene in ∼80% yield in 5 min. The relative reactivity of the catalysts toward different olefins was probed using a multicomponent intermolecular
competition reaction.
III
Metal-catalyzed oxygenations of organic substrates are a
widely used and studied class of reactions. The development
reactivity of [(Fe (phen)
2
)
2
2 2 4 4
(µ-O)(H O) ](ClO ) was found
1
to be significantly dependent on the pH of the reaction
solution, requiring a pH e 2 for greatest activity. By
contrast, [Mn (R,R-mcp)(CF
was effectively invariant to the pH under the conditions
examined previously. To elucidate the pH effect on the
reactivity of manganese catalysts and determine the role of
the ligand on catalytic activity, we have screened a wide
array of [Mn (L)(CF
3
of catalytic epoxidation agents that are rapid, selective,
scalable, and inexpensive with a wide substrate scope remains
an important goal. Terminal olefins are a particularly
challenging class of substrate to epoxidize,2 yet the
resulting 1,2-epoxides are extremely versatile starting materi-
II
3 3 2
SO ) ] exhibited activity that
-9
1
0
als for synthesizing more complicated molecules.
We recently reported a monomeric manganous complex,
II
3
SO
3
)
2
] complexes for their ability to
II
[
[
Mn (R,R-mcp)(CF
3
SO
3
)
2
], and a dimeric ferric complex,
epoxidize terminal olefins with commercial peracetic acid
(PAA , 1% H SO
, pH ∼ 1) or peracetic acid prepared with
strongly acidic resins (PAA Ligands were
III
(Fe (phen)
2
)
2
(µ-O)(H
2
O) ](ClO , that are capable of
2
4
)
4
C
2
4
1
1,12
rapidly catalyzing the epoxidation of terminal olefins using
R
, pH ∼ 4).
2,3
peracetic acid with 400-1000 turnovers within 5 min. The
chosen to highlight the influence of the following properties
on the catalytic activity of the Mn complex: oxidative
II
(
1) Sheldon, R. A.; Kochi, J. K. Metal Catalyzed Oxidations of Organic
Compounds; Academic Press: New York, 1981.
2) Murphy, A.; Dubois, G.; Stack, T. D. P. J. Am. Chem. Soc. 2003,
25, 5250.
stability, coordination mode, and thermodynamic stability
(Figure 1). We have found that most manganous complexes
(
1
of neutral polyamine ligands show a significant increase in
epoxidation reactivity under less acidic conditions with
peracetic acid if the ligation of the manganese center is
appropriate.
(
(
3) Dubois, G.; Murphy, A.; Stack, T. D. P. Org. Lett. 2003, 5, 2469.
4) White, M. C.; Doyle, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2001,
1
23, 7194.
(5) Coperet, C.; Adolfsson, H.; Sharpless, K. B. Chem. Commun. 1997,
1
565.
The complexes were initially screened for their ability to
epoxidize 1-octene at 1 mol % loading in 5 min using
(
6) Sato, K.; Aoki, M.; Ogawa, M.; Hashimoto, T.; Noyori, R. J. Org.
Chem. 1996, 61, 8310.
7) Devos, D. E.; Sels, B. F.; Reynaers, M.; Rao, Y. V. S.; Jacobs, P. A.
Tetrahedron Lett. 1998, 39, 3221.
(
(
(
(
8) Zuwei, X.; Ning, Z.; Yu, S.; Kunlan, L. Science 2001, 292, 1139.
9) Lane, B. S.; Burgess, K. Chem. ReV. 2003, 103, 2457.
10) Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Hansen,
(11) Hawkinson, A. T.; Schmitz, W. R. (E.I. du Pont de Nemours &
Co.). US 2910504, 1959.
(12) PAAR is generated by stirring 50% H2O2 with 10 equiv of CH3-
CO2H and Amberlite IR-120 resin for 12 h at 25 °C. These solutions
typically have <1% residual H2O2.
K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc.
002, 20, 1307.
2
1
0.1021/ol048846l CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/31/2004

Figure 1. Ligands used in this study.
PAA
C
.13,14 The presence of strong acid in solution appears
provide >75% yield of 1,2-epoxyoctane within 5 min using
II
2+
II
to attenuate the catalytic reactivity for the majority of
complexes due to decomposition of the complex into
protonated ligand and free Mn (Table 1). Simple Mn salts
2 equiv of oxidant at 25 °C: [Mn (Me
3
tacn)] , [Mn (R,R-
2
+
II
2+
II
2+
mcp)] , [Mn (bpma)] , and [Mn (bipy) ] . Of all com-
2
II
II
II
2+
plexes tested, [Mn (bipy) ] shows the highest turnover
2
are ineffective catalysts for terminal olefin epoxidation with
frequencies (>1000 turnovers/min) (Table 1).
The bipy ligand has an optimal combination of oxidative
stability and binding proclivity. The X-ray structure of [Mn -
II
PAA
C
under these conditions, although Mn SO
4
has been
shown to be an effective epoxidation catalyst for electron-
II
1
5
rich olefins with peroxycarbonate under basic conditions.
2 3 3 2
(bipy) (CF SO ) ] shows both triflates bound in a cis fashion,
II
2+
II
2+
II
Only [Mn (R,R-mcp)] , [Mn (bisp)] , and [Mn (Me
3
-
II
similar to the coordination geometry of [Mn (R,R-mcp)
2+
tacn)] (Table 1, entries 1, 9, 14) show significant catalytic
activity with PAA . These highly predisposed, neutral
2,18
II
(
3 3 2
CF SO ) ]. Mn complexes with a similar arrangement
C
of labile ligands generally provide potent catalysts for
terminal olefin oxidation, but this coordination geometry does
not ensure efficient reactivity under all conditions. Relative
to the octahedral complexes with cis labile sites, manganese
complexes with tetradentate ligands that bind within the
equatorial plane, creating two trans labile sites, show limited
ligands allow the manganese center to accommodate at least
two exogenous ligands in a cis fashion, and they create
thermodynamically and kinetically stable complexes, which
II
are more resistant to metal decomplexation. Both [Mn (R,R-
2
+
II
2+
mcp)] and [Mn (Me
3
-tacn)] have been previously
reported to be capable of epoxidation at less than 1 mol %
4 2
reactivity toward terminal olefins (Me -cyclam and H -cdb,
2
,16
loadings.
When PAA 11
Table 1, entries 11 and 13). The reactivity differences are
highlighted when the conversion of 1-octene at 15 s is
compared among the catalysts (Table 1). Complexes with
cis labile sites generally catalyze the reaction faster than
complexes with only trans labile sites. Anionic ligation also
seems to attenuate the epoxidation rate (Table 1, entries 12
and 13). The limited reactivity of Mn-salen and Mn-
porphyrin complexes toward terminal olefins is well docu-
R
is used as the oxidant, 14 of the 19
complexes provide >90% conversion with >80% selectivity
for 1,2-epoxyoctane at 1 mol % (Table 1).17 If catalyst
loading is reduced to 0.1 mol %, significant differentiation
among these complexes is observed. Only four complexes
(13) Complexes with acetate anions have been found to have reactivity
identical to that of the triflate complexes.
1
9,20
(
(
(
(
14) Catalysis stops within 15 min for most catalysts.
15) Lane, B. S.; Burgess, K. J. Am. Chem. Soc. 2001, 123, 2933.
16) Devos, D.; Bein, T. Chem. Commun. 1996, 917.
mented.
17) Highest induction observed with ligands 1-5, 12, 13, and 17 under
(18) Smith, J. A.; Galan-Mascaros, J. R.; Clerac, R.; Sun, J. S.; Xiang,
O. Y.; Dunbar, K. R. Polyhedron 2001, 20, 1727.
these conditions is e15%.
3120
Org. Lett., Vol. 6, No. 18, 2004

II
] Complexesa
Table 1. Epoxidation Reactivity of [Mn L(CF
3 3
SO )
2
time
catalyst loading
oxidant
5 min
PAACb
1%
5 min
15 sd
PAARc
PAARc
1%
conversion
0.1%
conversion
1%
0.1%
epoxide
27
ligand
R,R-mcp
conversion
epoxide
epoxide
epoxide
epoxide
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
98
0
18
6
18
4
25
5
97
3
5
6
4
80
2
0
12
1
2
94
0
15
2
11
2
24
3
91
2
2
6
2
96
28
91
60
98
95
87
94
92
96
11
51
20
95
94
88
95
88
99
4
92
22
88
55
96
92
87
94
84
92
11
49
15
91
91
83
92
86
95
2
96
2
36
6
33
15
53
49
23
6
1
6
1
85
50
82
15
10
99
4
92
2
31
4
28
12
49
43
18
4
1
5
1
82
46
77
11
9
75
8
R,R-pcp
R,R-6-Me-6′-H-mcp
R,R-6-Me-mcp
R,R-hcp
meso-hcp
mep
uns-mep
bisp
tmpa
Me4cyclam
R,R-H1pcdbe
R,R-H2cdbe
Me3tacn
terpy
19
10
88
61
83
89
19
23
1
2
1
1
1
1
1
1
1
1
1
1
2
50
3
71
1
0
9
1
44
84
83
74
57
94
0
8
43
2
bpma
R,R-tmcp
tmep
bipyf,g
2
0
94
2
93
0
none
0
a
II
Performed with 1-octene (0.5 M), [Mn L(CF3SO3)2] (5 mM), n-nonane (50 mM), 2 equiv of CH3CO3H, 25 °C, 5 min. Conversion and epoxide yields
b
c
determined relative to an internal standard. Results are average of at least three runs. 32% CH3CO3H, 1% H2SO4 in CH3CO2H/H2O (PAAC). 9-10%
CH3CO3H in CH3CO2H (PAAR). 1-Octene (0.5 M), [Mn L(CF3SO3)2] (5 mM), n-nonane (50 mM), 2 equiv of CH3CO3H (PAAR), 25 °C. Reactions were
quenched with Et3N after 15 s. [Mn LOAc] complexes, generated from Mn OAc2‚4H2O and O2. Performed with 2 equiv of bipy/Mn
shows reactivity nearly identical to that of bipy under these conditions.
d
II
e
III
II
f
II g
Phenanthroline
+
II
2+
Ostensibly small perturbations of the R,R-mcp framework
impact dramatically the efficiency of the manganese catalysts.
While most complexes examined are capable of complete
hcp)]2 and [Mn (mep)] , Table 1, entries 6 and 7).
Although a wide variety of bispyridyl-diamine ligands are
II
2+
easily synthesized, [Mn (R,R-mcp)] is by far the most
efficient catalyst within this family of complexes found to
date.
R
conversion of 1-octene at 1% loading using PAA , substitu-
tions that reduce the oxidative or kinetic stability of the
complexes limit the overall turnover numbers of the catalyst.
The introduction of additional benzylic hydrogens via
The ligand structure also influences the relative rates of
reactivity of the various manganese catalysts toward elec-
tronically and sterically distinct olefins. A characteristic
epoxidation reactivity profile can be measured for each
manganese complex using an intermolecular competition
reaction in which a mixture of four distinct olefin substrates
are oxidized under limiting oxidant conditions (Table 2). The
relative ratio of the epoxides formed probes indirectly the
nature of the metal-based oxidant and allows each catalyst’s
electronic and steric preferences for substrate to be compared.
The starting concentrations of the four substrates were biased
to ensure that a measurable conversion of each substrate
occurs under these conditions; the relative starting concentra-
tions of cyclooctene/dihydrocarvone/1-octene/trans-methyl-
cinnamate were 1:10:10:20. The ratio of epoxides of these
substrates reflect neither the absolute rate of epoxidation for
each catalyst nor their catalytic efficiencies.
6-methyl substituents on the pyridyl rings provides not only
a potential point of ligand oxidation but also reduces the
2
1
thermodynamic stability of the formed metal complex.
A
II
stepwise reduction in activity is seen with [Mn (R,R-
2
+
II
2+
II
mcp)] , [Mn (R,R-6-Me-6′-H-mcp)] , and [Mn (R,R-6-
Me-mcp)]2 (Table 1, entries 1, 3, and 4). A similar
reduction in epoxidation efficiency at 0.1% loading is
observed when the backbone methyl groups are replaced with
hydrogen atoms (i.e., [Mn (R,R-hcp)] , Table 1, entry 5).
Less thermodynamically stable complexes formed from
diamines other than trans-1,2-cyclohexyl-diamine are also
+
II
2+
II
not as effective at 0.1 mol % catalyst (i.e., [Mn (meso-
(
19) Srinivasan, K.; Michaud, P.; Kochi, J. K. J. Am. Chem. Soc. 1986,
08, 2309.
20) Collman, J. P.; Brauman, J. I.; Meunier, B.; Raybuck, S. A.;
Kodadek, T. Proc. Natl. Acad Sci. U.S.A. 1984, 81, 3245.
21) Chen, K.; Costas, M.; Que, L. J. Chem. Soc., Dalton Trans. 2002,
72.
1
(
The discrimination between cis-cyclooctene and trans-
methylcinnamate is a combination of electronic and steric
(
6
Org. Lett., Vol. 6, No. 18, 2004
3121

observed here correlate well with the previously observed
1
6,22
cis-olefin preference under different oxidation conditions.
Table 2. Intermolecular Competition Studies of Catalysts
II
2+
While [Mn (R,R-mcp)] shows very little preference for
cis-olefins over trans-olefins, there remains a significant
2
preference for cis-cyclooctene over trans-methylcinnamate
due to the different electronic properties of the two substrates,
the former being more electron rich than the latter. Syntheti-
II
2+
cally, [Mn (R,R-mcp)]
is a relatively indiscriminate
oxidant, which shows good regioselectivity with a diolefin
only when large electronic differences exist between the two
corrected relative ratio
of epoxides formed
2
double bonds.
ligand
CO:DH:O:MCa
II
Screening a series of Mn complexes for terminal olefin
epoxidation allows the identification of several Mn catalyst
Me3tacn (14)
bisp (9)
bpma (16)
1500:20:10:1
1400:20:10:1
1100:30:20:1
1100:20:5:1
1000:20:5:1
900:20:10:1
900:20:10:1
900:15:10:1
700:20:10:1
700:10:5:1
700:15:10:1
700:15:10:1
650:15:10:1
600:20:10:1
550:15:5:1
II
that are capable of high activity using PAA
R
as the oxidant.
II
2+
The most active catalyst, [Mn (bipy)
2
] , combines high
Me4cyclam (11)
b
activity with a simple, oxidatively robust ligand. This simple
complex shows that high activity and stability can be
achieved without resorting to more extreme strategies such
R,R-H2cdb (13)
b
R,R-H1pcdb (12)
R,R-tmcp (17)
R,R-6-Me-mcp (4)
meso-hcp (6)
terpy (15)
23
24
as very electron-deficient or complicated ligands. Further
optimization and exploration of substrate scope and mech-
anism is in progress.
tmep (18)
R,R-6-Me-6′-H-mcp (3)
uns-mep (8)
bipy (19)
R,R-pcp (2)
tmpa (10)
Acknowledgment. We are grateful to the National
Institutes of Health (Grant GM50730) for financial support
of this work. We thank UCSF Mass Spectrometry Facility
for performing the high-resolution mass spectrometry.
500:10:5:1
300:10:5:1
R,R-mcp (1)
mep (7)
200:10:5:1
Supporting Information Available: Experimental pro-
a
cedures for the synthesis of novel ligands, catalysts, and
Cyclooctene (CO, 0.025 M), (+)-dihydrocarvone (DH, 0.25 M),
-octene (O, 0.25 M), trans-methylcinnamate (MC, 0.5 M), nitrobenzene
1
(
peracetic acid solutions and reaction conditions and proce-
II
0.0125 M), [Mn (L)(CF3SO3)2] (2.5 mM), and 9% CH3CO3H (0.15 M) in
II
dures for isolated yields using [Mn (bipy)
2 3 3 2
(CF SO ) ]. This
CH3CN, 25 °C. The epoxide yields are determined relative to the internal
standard. The values are the average of at least three runs. The relative
ratios of epoxide are corrected for different initial concentrations.
material is available free of charge via the Internet at
http://pubs.acs.org.
b
III
II
[
Mn LOAc] complexes, generated from Mn OAc2‚4H2O and O2.
OL048846L
III
+
II
effects. Some catalysts, such as [Mn (R,R-cdb)] and [Mn -
(
22) Chang, S.; Lee, N. H.; Jacobsen, E. N. J. Org. Chem. 1993, 3, 6939.
2
+
(Me
3
tacn)] , are cis selective because of the steric con-
straints of the approach of trans-olefins toward the activated
(23) Banfi, S.; Cavazzini, M.; Coppa, F.; Barkanova, S. V.; Kaliya, O.
L. J. Chem. Soc., Perkin Trans. 2 1997, 1577.
(
24) Merlau, M. L.; Mejia, M. D. P.; Nguyen, S. T.; Hupp, J. T. Angew.
metal species. The relative ratios of epoxide products
Chem., Int. Ed. 2001, 40, 4239.
3122
Org. Lett., Vol. 6, No. 18, 2004