Cis-dihydroxylation of alkenes with oxone catalyzed by iron complexes of a macrocyclic tetraaza ligand and reaction mechanism by ESI-MS spectrometry and DFT calculations

Article abstract of DOI:10.1021/ja100967g

[Fe^III(L-N₄Me₂)Cl₂]⁺ (1, L-N₄Me₂ = N,N′-dimethyl-2,11-diaza[3.3](2,6) pyridinophane) is an active catalyst for cis-dihydroxylation of various types of alkenes with oxone at room temperature using limiting amounts of alkene substrates. In the presence of 0.7 or 3.5 mol % of 1, reactions of electron-rich alkenes, including cyclooctene, styrenes, and linear alkenes, with oxone (2 equiv) for 5 min resulted in up to >99% substrate conversion and afforded cis-diol products in up to 67% yield, with cis-diol/epoxide molar ratio of up to 16.8:1. For electron-deficient alkenes including α,β-unsaturated esters and α,β-unsaturated ketones, their reactions with oxone (2 equiv) catalyzed by 1 (3.5 mol %) for 5 min afforded cis-diols in up to 99% yield with up to >99% substrate conversion. A large-scale cis-dihydroxylation of methyl cinnamate (9.7 g) with oxone (1 equiv) afforded the cis-diol product (8.4 g) in 84% yield with 85% substrate conversion. After catalysis, the L-N₄Me₂ ligand released due to demetalation can be reused to react with newly added Fe(ClO₄)₂?4H₂O to generate an iron catalyst in situ, which could be used to restart the catalytic alkene cis-dihydroxylation. Mechanistic studies by ESI-MS, isotope labeling studies, and DFT calculations on the 1-catalyzed cis-dihydroxylation of dimethyl fumarate with oxone reveal possible involvement of cis-HO-Fe ^V O and/or cis-O Fe^V O species in the reaction; the cis-dihydroxylation reactions involving cis-HO-Fe^V O and cis-O Fe^V O species both proceed by a concerted but highly asynchronous mechanism, with that involving cis-HO-Fe^V O being more favorable due to a smaller activation barrier.

Full text of DOI:10.1021/ja100967g

Published on Web 09/02/2010
cis-Dihydroxylation of Alkenes with Oxone Catalyzed by Iron
Complexes of a Macrocyclic Tetraaza Ligand and Reaction
Mechanism by ESI-MS Spectrometry and DFT Calculations
Toby Wai-Shan Chow, Ella Lai-Ming Wong, Zhen Guo, Yungen Liu,
Jie-Sheng Huang, and Chi-Ming Che*
Department of Chemistry and Open Laboratory of Chemical Biology of the Institute of
Molecular Technology for Drug DiscoVery and Synthesis, The UniVersity of Hong Kong,
Pokfulam Road, Hong Kong
Received February 3, 2010; E-mail: cmche@hku.hk
Abstract: [FeIII(L-N4Me2)Cl2]+ (1, L-N4Me2 ) N,N′-dimethyl-2,11-diaza[3.3](2,6)pyridinophane) is an active
catalyst for cis-dihydroxylation of various types of alkenes with oxone at room temperature using limiting
amounts of alkene substrates. In the presence of 0.7 or 3.5 mol % of 1, reactions of electron-rich alkenes,
including cyclooctene, styrenes, and linear alkenes, with oxone (2 equiv) for 5 min resulted in up to >99%
substrate conversion and afforded cis-diol products in up to 67% yield, with cis-diol/epoxide molar ratio of
up to 16.8:1. For electron-deficient alkenes including R,ꢀ-unsaturated esters and R,ꢀ-unsaturated ketones,
their reactions with oxone (2 equiv) catalyzed by 1 (3.5 mol %) for 5 min afforded cis-diols in up to 99%
yield with up to >99% substrate conversion. A large-scale cis-dihydroxylation of methyl cinnamate (9.7 g)
with oxone (1 equiv) afforded the cis-diol product (8.4 g) in 84% yield with 85% substrate conversion. After
catalysis, the L-N4Me2 ligand released due to demetalation can be reused to react with newly added
Fe(ClO4)2 ·4H2O to generate an iron catalyst in situ, which could be used to restart the catalytic alkene
cis-dihydroxylation. Mechanistic studies by ESI-MS, isotope labeling studies, and DFT calculations on the
1-catalyzed cis-dihydroxylation of dimethyl fumarate with oxone reveal possible involvement of cis-
HO-FeVdO and/or cis-OdFeVdO species in the reaction; the cis-dihydroxylation reactions involving cis-
HO-FeVdO and cis-OdFeVdO species both proceed by a concerted but highly asynchronous mechanism,
with that involving cis-HO-FeVdO being more favorable due to a smaller activation barrier.
Introduction
osmium compounds are expensive, there is an increasing interest
in developing cis-dihydroxylation catalysts of other transition
Metal-catalyzed cis-dihydroxylation of alkenes is an important
method for C-O bond formation reactions in organic synthesis,
as elegantly demonstrated by tremendous studies on such cis-
dihydroxylation reactions catalyzed by OsO4,1 particularly the
Sharpless asymmetric dihydroxylation protocol, which has wide
metals such as ruthenium,3-5 manganese,6,7 and iron,8-12 of
which only ruthenium2b,5 and manganese7a catalysts have been
(3) (a) Shing, T. K. M.; Tai, V. W.-F.; Tam, E. K. W. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 2312. (b) Shing, T. K. M.; Tam, E. K. W.;
Tai, V. W.-F.; Chung, I. H. F.; Jiang, Q. Chem.sEur. J. 1996, 2, 50.
(c) Shing, T. K. M.; Tam, E. K. W. Tetrahedron Lett. 1999, 40, 2179.
(4) (a) Plietker, B.; Niggemann, M. Org. Lett. 2003, 5, 3353. (b) Plietker,
B.; Niggemann, M.; Pollrich, A. Org. Biomol. Chem. 2004, 2, 1116.
(c) Plietker, B.; Niggemann, M. Org. Biomol. Chem. 2004, 2, 2403.
(d) Plietker, B.; Niggemann, M. J. Org. Chem. 2005, 70, 2402. (e)
Plietker, B. Synthesis 2005, 2453.
applications1b-h and can be performed on large scales.2
A
number of OsO4-catalyzed large-scale cis-dihydroxylation reac-
tions using N-methylmorpholine N-oxide (NMO) as terminal
oxidant have also been known.2 Since OsO4 is highly toxic and
(1) (a) Schro¨der, M. Chem. ReV. 1980, 80, 187. (b) Kolb, H. C.; Van
Nieuwenhze, M. S.; Sharpless, K. B. Chem. ReV. 1994, 94, 2483. (c)
Gao, Y. In Encyclopedia of Reagents for Organic Synthesis; Paquette,
L. A., Ed.; Wiley: New York, 1995; Vol. 6, p 3801. (d) Johnson,
R. A.; Sharpless, K. B. In Catalytic Asymmetric Synthesis, 2nd ed.;
Ojima, I., Ed.; Wiley-VCH: New York, 2000; p 357. (e) Bolm, C.;
Hildebrand, J. P.; Mun˜iz, K. In Catalytic Asymmetric Synthesis, 2nd
ed.; Ojima, I., Ed.; Wiley-VCH: New York, 2000; p 399. (f) Kolb,
H. C.; Sharpless, K. B. In Transition Metals for Organic Synthesis;
Beller, M., Bolm, C., Eds.; Wiley-VCH: Weinheim, Germany, 2004;
Vol. 2, p 275. (g) Zaitsev, A. B.; Adolfsson, H. Synthesis 2006, 1725.
(h) Salvador, J. A. R.; Silvestre, S. M.; Moreira, V. M. Curr. Org.
Chem. 2008, 12, 492.
(5) (a) Ho, C.-M.; Yu, W.-Y.; Che, C.-M. Angew. Chem., Int. Ed. 2004,
43, 3303. (b) Yip, W.-P.; Ho, C.-M.; Zhu, N.; Lau, T.-C.; Che, C.-M.
Chem.-Asian J. 2008, 3, 70.
(6) De Vos, D. E.; de Wildeman, S.; Sels, B. F.; Grobet, P. J.; Jacobs,
P. A. Angew. Chem., Int. Ed. 1999, 38, 980.
(7) (a) Brinksma, J.; Schmieder, L.; van Vliet, G.; Boaron, R.; Hage, R.;
De Vos, D. E.; Alsters, P. L.; Feringa, B. L. Tetrahedron Lett. 2002,
43, 2619. (b) de Boer, J. W.; Brinksma, J.; Browne, W. R.; Meetsma,
A.; Alsters, P. L.; Hage, R.; Feringa, B. L. J. Am. Chem. Soc. 2005,
127, 7990.
(8) (a) Chen, K.; Que, L., Jr. Angew. Chem., Int. Ed. 1999, 38, 2227. (b)
Costas, M.; Tipton, A. K.; Chen, K.; Jo, D.-H.; Que, L., Jr. J. Am.
Chem. Soc. 2001, 123, 6722. (c) Ryu, J. Y.; Kim, J.; Costas, M.; Chen,
K.; Nam, W.; Que, L., Jr. Chem. Commun. 2002, 1288. (d) Fujita,
M.; Costas, M.; Que, L., Jr. J. Am. Chem. Soc. 2003, 125, 9912. (e)
Oldenburg, P. D.; Shteinman, A. A.; Que, L., Jr. J. Am. Chem. Soc.
2005, 127, 15672. (f) Suzuki, K.; Oldenburg, P. D.; Que, L., Jr. Angew.
Chem., Int. Ed. 2008, 47, 1887.
(2) (a) Catalysts for Fine Chemical Synthesis; Roberts, S. M., Poignant,
G., Eds.; John Wiley & Sons, Ltd.: New York, 2002; Vol. 1
(Hydrolysis, Oxidation and Reduction), p 105. (b) Caron, S.; Dugger,
R. W.; Ruggeri, S. G.; Ragan, J. A.; Ripin, D. H. B. Chem. ReV. 2006,
106, 2943.
9
10.1021/ja100967g  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 13229–13239 13229

A R T I C L E S
Chow et al.
Chart 1
applied to cis-dihydroxylation of alkenes on >1 g scale. In view
of the inexpensiveness and biocompatibility of iron, the
development of practical iron catalysts for alkene cis-dihy-
droxylation would be of high interest.
cis-Dihydroxylation of alkenes catalyzed by iron complexes
was pioneered by Que and co-workers.8 Over the past decade,
a series of iron-catalyzed alkene cis-dihydroxylation reactions
using H2O2 as terminal oxidant,8-12 along with in-depth
mechanistic studies,13-16 have been reported, predominantly by
Que and co-workers, mainly to develop functional models for
nonheme iron enzymes Rieske dioxygenases.17,18 These iron-
catalyzed reactions display the following features: (i) excess
alkene substrates were used, (ii) significant amounts of epoxides
were formed as byproducts, (iii) the reactions were performed
on small scales, and (iv) no recycling of catalysts was studied.
Exceptions to feature (ii) are the high cis-diol selectivities,
obtained using excess alkene substrates (with H2O2 as terminal
oxidant), in the cis-dihydroxylation of electron-deficient alkenes
reported by Que and co-workers8d,f and in the cis-dihydroxy-
lation of cyclooctene reported by Costas and co-workers.11
Notably, as an exception to feature (i), the cis-dihydroxylation
reactions of electron-rich aliphatic alkenes (such as oct-1-ene
and hept-2-ene) and cyclohexene reported by Que and co-
workers8c were performed using limiting amounts of alkene
substrates, with cis-diol/epoxide ratios of (1.5-4.3):1; this is
hitherto the sole report on iron-catalyzed cis-dihydroxylation
in which alkene substrates were used in limiting amounts. It
remains a formidable challenge to realize iron-catalyzed cis-
dihydroxylation of alkenes that not only uses limiting amounts
of alkene substrates but also has high cis-diol selectivity and is
applicable to large-scale synthesis.
complexes.5,19 We have demonstrated that cis-dioxorutheni-
um(VI) complexes react with alkenes to afford cis-diols through
a [3 + 2] cycloaddition mechanism,19 which suggests that other
reactive cis-dioxometal complexes, including those of naturally
abundant metals such as iron, could also undergo cis-dihy-
droxylation of alkenes with a similar mechanism. We therefore
endeavored to explore the possibility of extending the oxidation
chemistry of cis-dioxometal complexes from ruthenium to iron.
Among our hypotheses is the potential utility of alkene cis-
dihydroxylation by cis-dioxoiron(V) species. Previously we have
isolated and structurally characterized a cis-dioxoruthenium(V)
complex bearing a tetradentate tertiary amine ligand;20 this,
along with earlier reports on bent [OdFeVdO]+ (O-Fe-O
angle: ∼85°) generated in gas phase and characterized by mass
spectrometry and theoretical calculations,21 lends credence to
the possible existence of cis-dioxoiron(V) species in solution.
A strategy is to use macrocyclic tertiary amine ligands to support
cis-dioxoiron units, similar to the use of such auxiliary ligands
in the reactive cis-dioxoruthenium complex [(Me3tacn)(CF3-
CO2)RuVIO2]ClO4 (Me3tacn ) N,N′,N′′-trimethyl-1,4,7-triaza-
cyclononane)19 and in the catalyst [RuIII(Me3tacn)Cl3]; the latter
can catalyze cis-dihydroxylation of alkenes with H2O2 on 100 g
scale.5b However, we found that changing the ruthenium of
[RuIII(Me3tacn)Cl3] to iron rendered the complex ineffective in
catalyzing alkene cis-dihydroxylation with H2O2.
During our quest for practical iron catalysts for cis-dihy-
droxylation of alkenes, we came across a highly efficient cis-
dihydroxylation of electron-deficient alkenes by oxone (potas-
sium peroxymonosulfate, 2KHSO5 ·KHSO4 ·K2SO4)22 catalyzed
by an iron(III) complex, [FeIII(L-N4Me2)Cl2]+ (1, L-N4Me2 )
N,N′-dimethyl-2,11-diaza[3.3](2,6)pyridinophane, Chart 1), which
bears a macrocyclic tetraaza ligand L-N4Me223 containing both
tertiary amine and pyridyl units. This iron complex, together
with its catecholate analogues containing the same L-N4Me2
ligand, was first reported by Kru¨ger and co-workers.24 The use,
in this work, of oxonesan inexpensive, commercially available,
and environmentally benign oxidantswas inspired by previous
Our interest in iron-catalyzed cis-dihydroxylation of alkenes
was initiated on the basis of our previous works on this type of
organic transformations mediated/catalyzed by ruthenium
(9) Klopstra, M.; Roelfes, G.; Hage, R.; Kellogg, R. M.; Feringa, B. L.
Eur. J. Inorg. Chem. 2004, 846.
(10) Bruijnincx, P. C. A.; Buurmans, I. L. C.; Gosiewska, S.; Moelands,
M. A. H.; Lutz, M.; Spek, A. L.; van Koten, G.; Klein Gebbink,
R. J. M. Chem.sEur. J. 2008, 14, 1228.
(11) Company, A.; Go´mez, L.; Fontrodona, X.; Ribas, X.; Costas, M.
Chem.sEur. J. 2008, 14, 5727.
(12) Barry, S. M.; Rutledge, P. J. Synlett 2008, 14, 2172.
(13) (a) Chen, K.; Costas, M.; Kim, J.; Tipton, A. K.; Que, L., Jr. J. Am.
Chem. Soc. 2002, 124, 3026. (b) Mairata i Payeras, A.; Ho, R. Y. N.;
Fujita, M.; Que, L., Jr. Chem.sEur. J. 2004, 10, 4944. (c) Mas-
Balleste´, R.; Fujita, M.; Hemmila, C.; Que, L., Jr. J. Mol. Catal. A:
Chem. 2006, 251, 49. (d) Mas-Balleste´, R.; Costas, M.; van den Berg,
T.; Que, L., Jr. Chem.sEur. J. 2006, 12, 7489. (e) Mas-Balleste´, R.;
Que, L., Jr. J. Am. Chem. Soc. 2007, 129, 15964.
(14) Bukowski, M. R.; Comba, P.; Lienke, A.; Limberg, C.; Lopez de
Laorden, C.; Mas-Balleste´, R.; Merz, M.; Que, L., Jr. Angew. Chem.,
Int. Ed. 2006, 45, 3446.
(15) Bautz, J.; Comba, P.; Lopez de Laorden, C.; Menzel, M.; Rajaraman,
G. Angew. Chem., Int. Ed. 2007, 46, 8067.
(16) (a) Bassan, A.; Blomberg, M. R. A.; Siegbahn, P. E. M.; Que, L., Jr.
J. Am. Chem. Soc. 2002, 124, 11056. (b) Quin˜onero, D.; Musaev,
D. G.; Morokuma, K. Inorg. Chem. 2003, 42, 8449. (c) Bassan, A.;
Blomberg, M. R. A.; Siegbahn, P. E. M. J. Biol. Inorg. Chem. 2004,
9, 439. (d) Bassan, A.; Blomberg, M. R. A.; Siegbahn, P. E. M.; Que,
L., Jr. Angew. Chem., Int. Ed. 2005, 44, 2939. (e) Quin˜onero, D.;
Morokuma, K.; Musaev, D. G.; Mas-Balleste´, R.; Que, L., Jr. J. Am.
Chem. Soc. 2005, 127, 6548. (f) Comba, P.; Rajaraman, G. Inorg.
Chem. 2008, 47, 78.
(19) Yip, W.-P.; Yu, W.-Y.; Zhu, N.; Che, C.-M. J. Am. Chem. Soc. 2005,
127, 14239.
(20) Li, C.-K.; Che, C.-M.; Tong, W.-F.; Tang, W.-T.; Wong, K.-Y.; Lai,
T.-F. J. Chem. Soc., Dalton Trans. 1992, 2109.
(21) Schro¨der, D.; Fiedler, A.; Schwarz, J.; Schwarz, H. Inorg. Chem. 1994,
33, 5094.
(22) The Chemistry of Peroxides; Rappoport, Z., Ed.; John Wiley & Sons,
Ltd.: Chichester, U.K., 2006; Vol. 2.
(17) For reviews, see: (a) Rohde, J.-U.; Bukowski, M. R.; Que, L., Jr. Curr.
Opin. Chem. Biol. 2003, 7, 674. (b) Oldenburg, P. D.; Que, L., Jr.
Catal. Today 2006, 117, 15. (c) Shan, X.; Que, L., Jr. J. Inorg.
Biochem. 2006, 100, 421. (d) Que, L., Jr.; Tolman, W. B. Nature 2008,
455, 333. (e) Bruijnincx, P. C. A.; van Koten, G.; Klein Gebbink,
R. J. M. Chem. Soc. ReV. 2008, 37, 2716.
(23) Bottino, F.; Grazia, M. D.; Finocchiaro, P.; Fronczek, F. R.; Mamo,
A.; Pappalardo, S. J. Org. Chem. 1988, 53, 3521.
(24) (a) Koch, W. O.; Kru¨ger, H.-J. Angew. Chem., Int. Ed. Engl. 1995,
34, 2671. (b) Koch, W. O.; Schu¨nemann, V.; Gerdan, M.; Trautwein,
A. X.; Kru¨ger, H.-J. Chem.sEur. J. 1998, 4, 686. (c) Koch, W. O.;
Schu¨nemann, V.; Gerdan, M.; Trautwein, A. X.; Kru¨ger, H.-J.
Chem.sEur. J. 1998, 4, 1255.
(18) Feng, Y.; Ke, C.-y.; Xue, G.; Que, L., Jr. Chem. Commun. 2009, 50.
9
13230 J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010

cis-Dihydroxylation of Alkenes
A R T I C L E S
works of Yang,25 Shi,26 and co-workers on ketone-catalyzed
oxidation using oxone as a terminal oxidant.
Table 1. Oxidation of Cyclooctene (2a) with Oxone Using Various
Iron Catalystsa
Herein is reported the 1-catalyzed cis-dihydroxylation of a
variety of alkenes, particularly electron-deficient ones, by oxone
with high cis-diol selectivity. These highly selective reactions
are striking in several aspects: (i) The alkene substrates were
used in limiting amounts. (ii) cis-Diol products could be
prepared on ∼10 g scale. (iii) Replacing oxone with H2O2
rendered the cis-dihydroxylation ineffective, in contrast to the
general use of H2O2 in previously reported iron-8-12 and
ruthenium-catalyzed5b cis-dihydroxylation of alkenes. (iv) Oxone
is an ineffective oxidant in the alkene cis-dihydroxylation using
ruthenium catalysts,5 and, to the best of our knowledge, no
efficient cis-dihydroxylation of alkenes with oxone catalyzed
by a metal complex has been reported in literature. Also reported
here are mechanistic studies on the 1-catalyzed alkene cis-
dihydroxylation reactions by means of electrospray ionization
mass spectrometry (ESI-MS) and DFT calculations.
yield (%) based on conv
conv
(%)b
mass
I/II balance (%)
entry
catalyst
Ib
IIc
IIIb
1
2
3
4
5
6
[FeIII(Me3tacn)Cl3]
[FeII(qpy)Cl2]
55
42
0
3
d
d
27
58
d
0.1:1
30
58
d
d
d
[FeII(6-Me2-bpmcn)Cl2]
[FeIII(L-N4H2)Cl2]+
38
30
3
33
73
61
77
44
d
[FeIII(L-N4Me2)Cl2]+ (1) 99
53
45
60
d
8
4
5
34
12
6.6:1
[FeIII(L-N4Me2)Br2]+
96
12 11.3:1
12
9
e
f
[FeIII(L-N4Me2)Cl2]+ (1) 96
[FeIII(L-N4Me2)Cl2]+ (1) 37
12:1
7
g
8
a Reaction conditions: cyclooctene (0.5 mmol), catalyst (3.5 mol %)
in MeCN (6 mL), oxone (2 equiv) and NaHCO3 (6 equiv) in H2O (6
mL) added in 2 portions, RT, 5 min. b Determined by 1H NMR.
c Determined by GC. d Not detected in the reaction mixture. e Catalyst
loading: 0.7 mol %. f Isolated yield. g H2O2 (35% aq, 6 equiv), instead
of oxone, was used, added in 3 portions within 90 min.
Results and Discussion
Catalyst Screening. Iron complexes of macrocyclic tetraaza
ligands L-N4R2 (Chart 1), including [FeIII(L-N4Me2)(cat)]+ (cat
) catecholate ligands) reported by Kru¨ger and co-workers24 and
[FeIII(L-N4H2)Cl2]+ reported by Girerd, Banse, and co-workers,27
were previously studied as model systems for catechol dioxy-
genases (which catalyze oxidative cleavage of C-C bonds of
catechols with molecular oxygen).24,27 In this work, we screened
complexes [FeIII(L-N4Me2)Cl2]+ (1) and [FeIII(L-N4H2)Cl2]+,
together with [FeIII(Me3tacn)Cl3], [FeII(qpy)Cl2] (qpy ) 2,2′:
6′,2′′:6′′,2′′′:6′′′,2′′′′-quinquepyridine), and [FeII(6-Me2-bpmcn)-
Cl2] (6-Me2-bpmcn ) N,N′-dimethyl-N,N′-bis(6-methyl-2-
pyridylmethyl)-trans-1,2-diaminocyclohexane), for their catalytic
properties toward alkene cis-dihydroxylation. Both FeCl4- and
groups susceptible to oxidation by oxone.28 The FeCl4 and
ClO4 salts of 1 displayed similar catalytic activities toward
-
-
this reaction. Control experiments using [Et4N][FeCl4] (3.5 mol
%) as catalyst resulted in only ∼16% substrate conversion, and
no cis-diol I was detected in the reaction mixture; the corre-
sponding epoxide II was formed in ∼6% yield based on the
-
starting substrate. As the preparation of the FeCl4 salt of 1 is
more convenient, this salt of 1 was used in subsequent catalytic
studies.
The 1-catalyzed cis-dihydroxylation of cyclooctene with
oxone at room temperature resulted in a 99% substrate conver-
sion within 5 min, and the cis-diol I was formed as the major
product in 53% yield (based on the consumed substrate) whereas
epoxide II in 8% yield, giving a I/II ratio of 6.6:1 (entry 5,
Table 1). A higher I/II ratio of 11.3:1 was obtained using
[FeIII(L-N4Me2)Br2]+ (FeBr4- salt) as catalyst, but this catalyst
gave cis-diol I in a lower yield of 45% (entry 6, Table 1). The
mass balance obtained using catalyst 1 is 73% (entry 5, Table
1), substantially higher than those (30-61%) obtained using
the other iron catalysts (entries 1, 2, 4, 6; Table 1). No cis-diol
was detected in the reaction mixture when the reaction was
performed in the absence of 1 under the same conditions.
Decreasing the loading of 1 from 3.5 mol % to 0.7 mol %
resulted in a slight increase of mass balance, with the cis-diol
selectivity (I/II ratio) increasing from 6.6:1 to 12:1 (cf entries
5, 7; Table 1). By changing the terminal oxidant to H2O2, the
reaction (loading of 1: 3.5 mol %) afforded epoxide in 34%
yield (entry 8, Table 1); no cis-diol I was detected in the reaction
mixture. Regardless whether oxone or H2O2 was used as
terminal oxidant, R-hydroxy ketone III was formed as a minor
product (12% and 9% yields, respectively) in the 1-catalyzed
oxidation of cyclooctene (entries 5, 8; Table 1).
-
ClO4 salts of 1 were used; the crystal structures of 1·FeCl4
and 1·ClO4 ·MeCN have been determined by X-ray crystal-
lography (see Table S1 and Figures S1 and S2 in the Supporting
Information),whichhaveFe-Npy distancesof2.116(2)-2.1288(16)
Å and Fe-Namine distances of 2.219(2)-2.237(2) Å similar to
those of 1 previously reported by Kru¨ger and co-workers
(Fe-Npy 2.130 Å, Fe-Namine 2.232 Å).24b
At the outset, we examined the reaction of cyclooctene (2a)
with oxone (2 equiv) at room temperature using [FeIII(Me3tacn)-
Cl3] (3.5 mol %) as catalyst, in view of the catalytic activity of
[RuIII(Me3tacn)Cl3] for alkene cis-dihydroxylation.5b After 5 min,
the [FeIII(Me3tacn)Cl3]-catalyzed reaction gave a 55% substrate
conversion but afforded cis-diol product I in only 3% yield,
and the major product is epoxide II, which was formed in 27%
yield based on consumed substrate (entry 1, Table 1). When
the catalyst was changed to [FeII(qpy)Cl2], [FeII(6-Me2-bpmc-
n)Cl2], or [FeIII(L-N4H2)Cl2]+, no cis-diol I was detected in the
reaction mixtures (entries 2-4, Table 1).
Interestingly, replacing the NH group (in the L-N4H2 ligand)
of [FeIII(L-N4H2)Cl2]+ with NMe group to form [FeIII(L-
N4Me2)Cl2]+ (1) endowed the catalyst with a good activity
toward cis-dihydroxylation of cyclooctene with oxone, probably
due to enhanced stability of 1 toward oxidative deterioration,
as L-N4Me2, unlike L-N4H2, does not bear secondary amine NH
We next studied the reaction of methyl cinnamate (3a) with
oxone (2 equiv) at room temperature in the presence of catalyst
1 at different loadings (Table 2). Remarkably, using 3.5 mol %
of 1, the reaction afforded cis-diol product I in 81% yield with
98% substrate conversion within 5 min, together with trace
amounts of epoxide II, R-hydroxy ketone III, and benzaldehyde
(25) Yang, D. Acc. Chem. Res. 2004, 37, 497.
(26) (a) Shi, Y. Acc. Chem. Res. 2004, 37, 488. (b) Wong, O. A.; Shi, Y.
Chem. ReV. 2008, 108, 3958.
(27) Raffard, N.; Carina, R.; Simaan, A. J.; Sainton, J.; Rivie`re, E.;
Tchertanov, L.; Bourcier, S.; Bouchoux, G.; Delroisse, M.; Banse,
F.; Girerd, J.-J. Eur. J. Inorg. Chem. 2001, 2249.
(28) Gella, C.; Ferrer, E.; Alibe´s, R.; Busque´, F.; de March, P.; Figueredo,
M.; Font, J. J. Org. Chem. 2009, 74, 6365.
9
J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010 13231

A R T I C L E S
Chow et al.
Table 2. cis-Dihydroxylation of Methyl Cinnamate (3a) with Oxone Catalyzed by 1 at Different Catalyst Loadingsa
yield (%) based on conv
entry
catalyst loading (%)
conv (%)
Ib
IIc
IIIb
IVc
I/II
mass balance (%)
1
2
3
4
5
3.5
1.4
0.7
0.35
0
98
94
49
52
0
81
84
86
85
d
4
5
6
10
d
4
d
d
d
d
2
3
1
4
d
20.3:1
16.8:1
14.3:1
8.5:1
91
92
93
99
a Reaction conditions: substrate (0.5 mmol), 1 (0-3.5 mol %) in MeCN (6 mL), oxone (2 equiv) and NaHCO3 (6 equiv) in H2O (6 mL) added in 2
b
portions, RT, 5 min. Determined by H NMR. c Determined by GC. d Not detected in the reaction mixture.
1
IV, with a high cis-diol selectivity (I/II) of 20.3:1 and mass
balance of 91% (entry 1, Table 2). The cis-diol selectivity and
substrate conversion decreased, although the mass balance
increased, with decreasing loading of catalyst 1 (entries 1-4,
Table 2). At catalyst loadings of 0.35-1.4 mol %, no R-hydroxy
ketone III was detected in the reaction mixture (entries 2-4,
Table 2).
Substrate Scope. The 1-catalyzed cis-dihydroxylation of
alkenes with oxone (2 equiv) at room temperature could be
applicable to a variety of alkene substrates, from electron-
rich ones including cyclic alkene (2a), aryl alkenes (2b-e),
and linear alkenes (2f-h), to electron-deficient ones including
R,ꢀ-unsaturated esters (3a,f-j) and R,ꢀ-unsaturated ketones
(3b-e) (Table 3).
R-hydroxy ketones III formed in the cis-dihydroxylation of 2a-e
and 3a (Table 3) could be due to overoxidation and/or a direct
ketohydroxylation of these alkenes. We treated cis-diol I (obtained
from cis-dihydroxylation of cyclooctene) with oxone (2 equiv) in
the presence of 1 (0.7 mol %) at room temperature for 5 min; this
reaction afforded the corresponding III in 23% yield with 68%
substrate conversion (Scheme S1 in Supporting Information). We
also examined the mass flow during the cis-dihydroxylation of
cyclooctene with oxone (2 equiv) catalyzed by 1 (3.5 mol %), and
found that the mass balance, and the yield of cis-diol I as well,
markedly decreased with increasing reaction time (Figure S3 in
the Supporting Information). Provided that such loss in mass
balance is mainly due to overoxidation, the corresponding over-
oxidation should largely give other unidentified products, as the
yield of III was little changed with reaction time.
For the electron-rich alkenes, the highest yields of cis-diols
I and highest substrate conversion and mass balances were
obtained for terminal aryl alkenes 2b,d, which were converted
to cis-diols I in 65-67% yields with >99% substrate conversion
within 5 min, giving the I/II ratios of up to 16.8:1 and mass
balances of 95-100% (entries 2, 4; Table 3). In the cis-
dihydroxylation of 2a-e, R-hydroxy ketones III were formed
in 8-23% yields (entries 1-5, Table 3), and, for 2b-e,
aldehydes IV were also formed (in 5-9% yields). Notably,
byproducts II-IV were not detected in the reaction mixtures
for the oxidation of linear terminal alkenes 2f,g (entries 6, 7;
Table 3); these two alkenes were converted to cis-diols I in
64% and 53% yields, respectively, with moderate substrate
conversions (52-71%) and mass balances (53-64%).
Electron-deficient alkenes 3a-j, except 3h, were generally
converted into cis-diols I with good-to-excellent values of cis-
diol yields (60-99%), substrate conversions (84 to >99%), and
mass balances (60-99%) (entries 9-15, 17, 18; Table 3) and
high I/II ratios ((8.5-20.3):1, entries 9-13 in Table 3). No
R-hydroxy ketones III were detected in the reaction mixtures,
except for 3a; epoxides II and aldehydes IV were formed in
only up to 8% yield or in about half the cases not detectable.
Strikingly, the 1-catalyzed oxidation of dimethyl fumarate (3i)
with oxone at room temperature for 5 min produced no
detectable amounts of byproducts II-IV and afforded cis-diol
I in 99% yield with >99% substrate conversion and 99% mass
balance (entry 17, Table 3).
While 1 can catalyze alkene cis-dihydroxylation with oxone
using limiting amounts of alkene substrates, higher product
turnover numbers (TON) could be obtained by using excess
amounts of alkenes, partly due to minimization of overoxidation.
For the cis-dihydroxylation of cyclooctene (2a) catalyzed by 1
(0.7 mol %), the TON of cis-diol product I increased from 82
to 158 upon changing substrate/oxidant ratio from 1:2 to 5:1
(entries 1, 2; Table S2 in the Supporting Information); R-hy-
droxy ketone III was still formed, but the ratio of III/I decreased
from 1:5.1 to 1:8.8. An increase in product TON, upon changing
substrate/oxidant ratio from 1:2 to 5:1, was also observed for
the cis-dihydroxylation of 2e,h and 3g (entries 3-8, Table S2
in the Supporting Information).
Large-Scale Reactions. We explored the feasibility of scaling
up the 1-catalyzed cis-dihydroxylation to ∼10 g scale. A one-
pot reaction of 9.7 g of methyl cinnamate (3a) with oxone (1
equiv) in the presence of 1 at room temperature for 10 min
gave 8.4 g of cis-diol I (84% yield, 85% substrate conversion,
Scheme 1). This large-scale reaction was performed by adding
each of the substrate and catalyst in two equal portions with
each portion reacting for 5 min (see Experimental Section);
interestingly, in adding the second portion of catalyst,
Fe(ClO4)2 ·4H2O was used, without the need to use complex 1.
A control experiment revealed that Fe(ClO4)2 ·4H2O alone is
not an active catalyst for the alkene cis-dihydroxylation. We
speculate that at the end of the 1-catalyzed oxidation of 3a with
oxone, the ligand L-N4Me2 was released due to demetalation,
and the cis-dihydroxylation of 3a could also be catalyzed by
an iron complex generated in situ from reaction of L-N4Me2
with Fe(ClO4)2 · 4H2O. If such is the case, multiple reuse of
L-N4Me2 for complexation with new batch of Fe(ClO4)2 ·4H2O
in the catalysis could be feasible.
Overoxidation Issue. Owing to the use of limiting amounts of
substrates, overoxidation could occur in the presence of excess
oxidant; however, this was not found to be a serious problem in
the 1-catalyzed oxidation of alkenes with oxone (2 equiv) at room
temperature, at least for the alkene substrates giving good-to-
excellent mass balances as depicted in Table 3. The byproducts
9
13232 J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010

cis-Dihydroxylation of Alkenes
A R T I C L E S
Table 3. cis-Dihydroxylation of Alkenes (2a-h and 3a-j) with Oxone Catalyzed by 1a
yield (%) based on conv
entry
substrate
conv (%)b
Ic
IIb
IIId
IVb,d
I/II
mass balance (%)
1
2
3
4
5
6
7
8
2a
2b
2c
2d
2e
2f
2g
2h
3a
3b
3c
3d
3e
3f
96
60
67
46
65
44
64
53
5
4
e
9
19
e
12
17c
23c
18c
8
e
e
e
4
e
e
e
e
e
e
e
e
e
7
9
8
5
e
e
e
2
8
2
2
2
2
e
e
e
e
12:1
16.8:1
77
95
78
100
76
64
53
18
91
79
80
80
78
63
60
56
99
77
>99
>99
>99
89
71
52
97
94
92
90
90
>99
84
94
34
>99
94
7.2:1
2.3:1
e
16 (36f)
81
2
4
6
5
6
8
e
e
e
e
e
8:1
20.3:1
10.8:1
14.6:1
12:1
9
10
11
12
13
14
15
16
17
18
65
73
72
68 (75f)
61
8.5:1
3g
3h
3i
60 (67f)
56 (65f)
99
3j
77
e
a Reaction conditions: substrate (0.5 mmol), 1 (0.7 mol % for entries 1, 5, 8, 15; 3.5 mol % for the other entries) in MeCN (6 mL), oxone (2 equiv)
d
and NaHCO3 (6 equiv) in H2O (6 mL) added in 2 portions, RT, 5 min. b Determined by GC. c Isolated yield. Determined by H NMR. e Not detected
in the reaction mixture. Yield determined by H NMR.
1
f
1
Scheme 1. Large-Scale cis-Dihydroxylation of Methyl Cinnamate (3a) with Oxone (2 equiv) Catalyzed by 1a
a Benzaldehyde (IV) was formed in <1% yield.
Reuse of L-N4Me2 Ligand. Before focusing on the reuse of
L-N4Me2, we examined the reactions of cyclooctene (2a) and
methyl cinnamate (3a) with oxone (2 equiv) at room temperature
using a 1:1 mixture of Fe(ClO4)2 ·4H2O and L-N4Me2 as catalyst
at different catalyst loadings (0.35-3.5 or 0.7-3.5 mol %).
These reactions indeed mainly converted both 2a and 3a to the
corresponding cis-diol I products, with 53-96% substrate
conversions, 54-63% cis-diol yields, (4.2-9.2):1 I/II ratios,
and 71-82% mass balances for 2a, and 56-96% substrate
conversions, 78-86% cis-diol yields, (21.5-28.7):1 I/II ratios,
and 82-94% mass balances for 3a (Table S3 in the Supporting
Information), demonstrating a similar catalytic activity of
“Fe(ClO4)2 ·4H2O + L-N4Me2” to that of 1.
Given the highest cis-diol selectivity (I/II ratio: 28.7:1),
coupled with excellent substrate conversion and mass balance,
in the cis-dihydroxylation of 3a with oxone (2 equiv) at a
catalyst loading of 1.4 mol % (entry 5, Table S3 in the
Supporting Information), we studied the reuse of L-N4Me2 for
the same reaction at this catalyst loading. The first run was
carried out using catalyst 1; in subsequent runs, the catalyst
was generated in situ by direct reaction of added
Fe(ClO4)2 ·4H2O with the L-N4Me2 released during previous run
(no isolation of the released L-N4Me2 was required). Upon
reusing L-N4Me2 five times, the reaction still gave good overall
substrate conversion of 85% and cis-diol yield of 89% (entry
3, Table 4). This demonstrates that the macrocyclic tetraaza
9
J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010 13233

A R T I C L E S
Chow et al.
Table 4. Reuse of L-N4Me2 in cis-Dihydroxylation of Methyl Cinnamate (3a) with Oxone Catalyzed by 1 or “Fe(ClO4)2 ·4H2O + L-N4Me2”a
accumulated yield (%) based on conv
entry
times reused
accumulated conv (%)b
Ib
IIc
IIIb
IVc
I/II
mass balance (%)
1
2
3
1
3
5
98
95
85
84
84
89
5
9
d
d
3
3
3
1
3
16.8:1
9.3:1
92
100
98
a Reaction conditions for the 1st run: methyl cinnamate (0.5 mmol), 1 (1.4 mol %) in MeCN (6 mL), oxone (2 equiv) and NaHCO3 (6 equiv) in H2O
(6 mL) added in 2 portions, RT, 5 min; reaction conditions for the other runs: methyl cinnamate (0.5 mmol) and Fe(ClO4)2 ·4H2O (1.4 mol %) in MeCN
b
(6 mL), oxone (2 equiv) and NaHCO3 (6 equiv) in H2O (6 mL) added in 2 portions, RT, 5 min. Determined by H NMR. c Determined by GC. d Not
1
detected in the reaction mixture.
Scheme 2. Possible Pathways of 1-Catalyzed cis-Dihydroxylation of Alkenes with Oxone
ligand L-N4Me2 is robust toward the cis-dihydroxylation of
alkenes with oxone.
ions, oxone is proposed to act as a single oxygen atom donor
by undergoing heterolytic O-O bond cleavage,29 resulting in
two-electron oxidation of metal ions, examples of which include
Mechanistic Aspects. Depending on the ligands in the iron
catalysts, iron-catalyzed cis-dihydroxylation of alkenes with
H2O2 is documented to involve several types of active inter-
mediates, including cis-HO-FeVdO and FeIII-(η2-OOH) inter-
mediates in water-assisted (wa) and nonwater-assisted (nwa)
mechanisms, respectively, established by Que and co-workers,17b
together with a cis-FeIV(OH)2 intermediate recently proposed
by Comba and co-workers.15 Similar species could also be
considered as candidates for the active intermediate in the
1-catalyzed alkene cis-dihydroxylation with oxone. Another
candidate is the iron analogue of the cis-dioxoruthenium
complexes that are reactive toward cis-dihydroxylation of
alkenes as described in our previous work.19
the generation of HO-MVdO (M ) Fe29c and Mn29b,d
)
intermediates, previously proposed by Meunier,29b,c Groves,29d
and co-workers, from reaction of iron(III) or manganese(III)
porphyrins with oxone in aqueous solutions at pH 7-8, with
the second oxygen atom coming from H2O.29b,c Similar genera-
tion of [FeV(L-N4Me2)(O)(OH)]2+ (cis-HO-FeVdO species A,
Scheme 2) from treatment of the iron(III) complex 1 with oxone
in MeCN/H2O (1:1 v/v) under the catalytic conditions (pH 7)
was proposed in this work (reaction 1 in Scheme 2) as one of
the possible mechanisms. Another mechanism we proposed here
is the reaction of 1 with more than one equivalent of oxone to
generate [FeV(L-N4Me2)O2]+ (cis-OdFeVdO species B, Scheme
2) with its oxygen atoms both coming from oxone (reaction 2
in Scheme 2); however, this mechanism could be complex,
which might involve homolytic O-O bond cleavage. The cis-
HO-FeVdO species A could be in equilibrium with the cis-
OdFeVdO species B through deprotonation and protonation,
respectively, analogous to a HO-MnVdO porphyrin species
in equilibrium with an OdMnVdO porphyrin species in solution
as reported by Groves and co-workers.31 Intermediates A and
B could also be generated in the catalytic reactions using
“Fe(ClO4)2 ·4H2O + L-N4Me2” via initial one-electron oxidation
In literature, oxone is known to show various oxidation
behaviors and can undergo heterolytic and homolytic O-O bond
cleavage to effect two-29 and one-electron30 oxidation of metal
ions, respectively. In most of its reactions with trivalent metal
(29) For examples, see: (a) Labat, G.; Meunier, B. J. Org. Chem. 1989,
54, 5008. (b) Pitie´, M.; Bernadou, J.; Meunier, B. J. Am. Chem. Soc.
1995, 117, 2935. (c) Sorokin, A.; Meunier, B. Eur. J. Inorg. Chem.
1998, 1269. (d) Jin, N.; Groves, J. T. J. Am. Chem. Soc. 1999, 121,
2923.
(30) Lente, G.; Kalma´r, J.; Baranyai, Z.; Kun, A.; Ke´k, I.; Bajusz, D.;
Taka´cs, M.; Veres, L.; Fa´bia´n, I. Inorg. Chem. 2009, 48, 1763.
9
13234 J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010

cis-Dihydroxylation of Alkenes
A R T I C L E S
Figure 1. ESI mass spectrum of a reaction mixture of 1, oxone (16 equiv), and dimethyl fumarate (3i, 10 equiv) in MeCN/H2O (5:1 v/v) at room temperature.
of iron(II) to iron(III) by oxone (note a recent report on the
oxidation of iron(II) to iron(III) with oxone30), similar to the
preoxidation of iron(II) to iron(III) involved in the iron(II)-
catalyzed cis-dihydroxylation with H2O2 reported by Que and
co-workers.13a However, the formation of cis-FeIV(OH)2 species
from “Fe(ClO4)2 ·4H2O + L-N4Me2” and oxone could not be
excluded but might be less favored, as the DFT calculations by
Comba and co-workers revealed that cis-FeIV(OH)2 species is
generated from an FeII(H2O2) species by O-O bond homolysis
of the coordinated H2O2.15
Two possible pathways for the cis-dihydroxylation with oxone
catalyzed by 1 are depicted in Scheme 2, one of which features
cis-HO-FeVdO intermediate A whereas the other involves cis-
OdFeVdO intermediate B; both A and B might react with
alkene to afford cis-diol I, via intermediates C and D,
respectively.
substrate (unless otherwise indicated), as this alkene gave the
best results in the 1-catalyzed cis-dihydroxylation with oxone
(see Table 3).
i. ESI-MS. Upon treatment of 1 with oxone (16 equiv) and
dimethyl fumarate (10 equiv) in MeCN/H2O (5:1 v/v) at room
temperature for ∼30 s, we examined the reaction mixture by
ESI-MS. The spectrum (Figure 1) shows prominent cluster peaks
at m/z 536.11 attributable to {C·Cl-} (Calcd m/z 536.11) and
at m/z 500.14 attributable to D or [{C·Cl-} - HCl] (Calcd m/z
500.14), as well as an intense cluster peak at m/z 394.03 assigned
to 1. The observed isotope patterns for the three cluster peaks
well match their calculated ones (see Figures S4-S6 in the
Supporting Information). Collision-induced dissociation of the
ion with m/z 536.11 resulted in the formation of the ion with
m/z 500.14 (Figure S7 in the Supporting Information), suggest-
ing a fragmentation of {C·Cl-} to D, apparently by loss of
HCl. For the ion with m/z 500.14, collision-induced dissociation
generated an ion with m/z 412.12, possibly due to C-C bond
cleavage of the coordinated diolate ligand in D (Figures S8 and
S9 in the Supporting Information). ESI-MS analysis of the
reaction mixture of 1 with oxone (4 equiv) and cyclooctene (2a,
20 equiv) was also performed, revealing a cluster peak at m/z
466.21 attributable to D or [C - H+] (Calcd m/z 466.20) (Figure
S10 in the Supporting Information). This species showed a
shorter lifetime than that of the corresponding species in the
To gain further insight into the mechanism of the 1-catalyzed
alkene cis-dihydroxylation with oxone, we have made attempts
to detect the reaction intermediates by ESI-MS and performed
isotope labeling studies. DFT calculations on the reaction
pathways depicted in Scheme 2 have also been conducted. In
all these studies, dimethyl fumarate (3i) was selected as the
(31) (a) Lahaye, D.; Groves, J. T. J. Inorg. Biochem. 2007, 101, 1786. (b)
Jin, N.; Ibrahim, M.; Spiro, T. G.; Groves, J. T. J. Am. Chem. Soc.
2007, 129, 12416.
9
J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010 13235

A R T I C L E S
Chow et al.
Figure 2. Upper: simulated isotope pattern for species B. Lower: observed isotope pattern in high-resolution ESI mass spectrum for the cluster peak at m/z
356.09 observed by ESI-MS analysis of a reaction mixture of [Fe(L-N4Me2)Br2]+ with oxone (8 equiv) in MeCN/H2O (5:1 v/v) at room temperature.
reaction of 3i, and was partially converted to 1 upon collision-
induced dissociation (Figure S11 in the Supporting Information).
Efforts were also made to detect, by ESI-MS, oxoiron(V)
species A and/or B (Scheme 2) using a reaction mixture of 1
with oxone in the absence of alkene substrate. As B has a
calculated m/z value of 356.09, which is too close to the m/z
359.07 of the major fragment generated by collision-induced
dissociation of 1 (Figure S12 in the Supporting Information),
we changed the iron complex from 1 to [FeIII(L-N4Me2)Br2]+
to avoid any overlap of the cluster peak of B with that of the
major fragment of 1. Importantly, analysis of a reaction mixture
of [Fe(L-N4Me2)Br2]+ with oxone (8 equiv) in MeCN/H2O (5:1
v/v) at room temperature by ESI-MS revealed a weak cluster
peak at m/z 356.09 with an isotope pattern largely matching
that simulated for B having a [Fe(L-N4Me2)O2]+ formulation
(Figure 2). We assign this cluster peak to intermediate B, as no
cluster peak ascribable to A (Calcd m/z 178.55) or {A·Br-}
(Calcd m/z 436.02) was found. However, the possible formation
of B from A upon loss of H+ under mass spectrometric
conditions could not be excluded.
When H2O2, instead of oxone, was employed to oxidize 2a
in the presence of catalyst 1 under similar conditions, ESI-MS
analysis of the reaction mixture did not reveal cluster peaks
ascribable to intermediate C or D. Such is also the case if
[FeIII(L-N4H2)Cl2]+, rather than 1, was used as catalyst, ir-
respective of whether oxone or H2O2 was used as oxidant. In
the ESI mass spectrum of the reaction mixture of [Fe(L-
N4Me2)Br2]+ with H2O2, there was no cluster peak ascribable
to intermediate A or B. These may rationalize the ineffective
cis-dihydroxylation of 2a involving H2O2 or [FeIII(L-N4H2)Cl2]+
performed in this work (Table 1).32 The higher activity of oxone
relative to H2O2 in the 1-catalyzed cis-dihydroxylation is
reminiscent of a similar phenomenon previously reported by
Meunier and co-workers29a on the oxidation of lignin model
compounds with oxone or H2O2 catalyzed by iron and manga-
nese porphyrins, which is considered to stem from highly
favored heterolytic O-O bond cleavage of oxone, compared
with H2O2, for generation of high-valent metal-oxo intermedi-
ates, owing to more unsymmetrical O-O bond in oxone,
together with SO42- being a better leaving group than OH-.29a
ii. Isotope Labeling Studies. We analyzed the reaction mixture
of 1 with oxone (16 equiv) and dimethyl fumarate (10 equiv)
in MeCN/H218O (5:1 v/v) at room temperature in the presence
of H218O (∼485 equiv). ESI-MS measurements suggested ∼16%
of 18O incorporation into D to form singly labeled species D-18O
and ∼10% of 18O incorporation into {C·Cl-} to form singly
labeled species {C·Cl-}-18O (Chart 2; no doubly 18O labeled
species were detected); these values of 18O incorporation were
∼41% and ∼38%, respectively, for the reaction of 1 with 2
equiv of oxone under the same conditions. When mixtures of
H218O and H216O containing different amounts of H216O were
used, the 18O incorporation decreased with increasing content
of H216O. Therefore, the 1-catalyzed cis-dihydroxylation of
(32) A facile oxidation (by oxone) of the secondary amine NH groups in
the L-N4H2 ligand of [FeIII(L-N4H2)Cl2]+, as mentioned in an earlier
section, would significantly change the binding behavior of this
macrocyclic ligand and hampers formation of the intermediates
corresponding to A-D in Scheme 2. Indeed, in contrast to the easy
recovery of L-N4Me2 at the end of 1-catalyzed oxidation with oxone,
neither L-N4H2 nor its iron complex was recovered after the corre-
sponding catalysis using [FeIII(L-N4H2)Cl2]+. For the 1-catalyzed
oxidation using H2O2, the two coordinated Cl- ligand in the starting
catalyst might have a deleterious effect (see ref 17b) toward the cis-
dihydroxylation (note that oxone can oxidize Cl-, see: Dieter, R. K.;
Nice, L. E.; Velu, S. E. Tetrahedron Lett. 1996, 37, 2377. Therefore,
“Fe(ClO4)2 ·4H2O + L-N4Me2” (3.5 mol% each) were employed to
catalyze the reaction of 2a with H2O2 (4 equiv) in MeCN/H2O (5:1
v/v) at room temperature for 1 h, which gave a 73% substrate
conversion and a 74% mass balance (both substantially higher than
those obtained for catalyst 1 (entry 8, Table 1)) but still afforded
epoxide II as the major product (49% yield based on consumed
substrate), with cis-diol I and R-hydroxy ketone III formed in 9%
and 16% yields, respectively.
9
13236 J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010

cis-Dihydroxylation of Alkenes
A R T I C L E S
Chart 2
dimethyl fumarate (3i) with oxone is considerably assisted by
water, providing evidence for the involvement of reaction 1
depicted in Scheme 2.
previous experimental investigations.19 DFT calculations on the
mechanisms of iron-catalyzed analogues are sparse,16d,15 reveal-
ing cis-dihydroxylation of alkenes (such as 2-butene) by cis-
HO-FeVdO species via an oxoiron-containing carbon radical
(which is converted to the cycloadduct without any significant
energy barrier) as reported by Que and co-workers,16d and the
concerted cis-dihydroxylation of ethene by cis-FeIV(OH)2 species
as reported by Comba and co-workers.15
In previously reported isotope labeling studies on iron-
catalyzed cis-dihydroxylation with H2O2 involving water-assisted
mechanism, the extent of 18O incorporation from H218O into the cis-
diol products (singly labeled) is typically 60-97%;8d,9,11,13a,c,d lower
values of 33% and 29% were observed by Que,8e Feringa,9 and
co-workers, respectively, in some cases, which can be rational-
ized by water incorporation via nonwater-assisted pathway8e or
by the likely existence of several competing pathways.9 The
up to ∼41% 18O incorporation observed for the 1-catalyzed cis-
dihydroxylation with oxone is indicative of the involvement of
both reactions 1 and 2 (Scheme 2) in this catalytic process.
In addition, we examined a reaction of 1 (0.1 mmol) with
oxone (1 equiv) and 3a (1 equiv) in MeCN/H2O (2 mL each)
at room temperature for 5 min, which gave the cis-diol I in
55% yield (determined by 1H NMR; substrate conversion:
100%), considerably lower than the 81% yield obtained under
the catalytic conditions. This may imply that more than one
equivalent of oxone is needed to convert 1 to the reactive
oxoiron(V) species in these oxidation processes and is supportive
of both reactions 1 and 2 depicted in Scheme 2.
In this work, the mechanisms of cis-dihydroxylation of
dimethyl fumarate (3i) by oxoiron(V) species A and B (Scheme
2) were investigated with hybrid density functional theory at
the B3LYP/6-31G(d):lanl2dz for Fe level. Before exploring the
potential energy surface of the reactions, we conducted computa-
tional studies on the relative energy of the separated reactants “3i
+ A” and “3i + B” and the respective products C and D (Scheme
2) with different spin multiplicities. The calculated results reveal
that the ground states of A and B feature quartet FeV centers; this
spin state was considered in subsequent calculations. The computed
structure and conformation of the L-N4Me2 ligand in A and B are
consistent with those of L-N4Me2 in the crystal structures of
1·FeCl4 and 1·ClO4 ·MeCN, and the computed FeV)O distances
(1.613-1.638 Å) in A and B (Figure S13 in the Supporting
Information) are similar to those computed for the bent
[OdFeVdO]+ (1.63-1.70 Å),21 the cis-HO-FeVdO species
iii. DFT Calculations. There have been extensive theoretical
studies, such as DFT calculations, on the mechanism of alkene
cis-dihydroxylation catalyzed by OsO4,33 which generally favor
a concerted [3 + 2] cycloaddition mechanism. A concerted [3
+ 2] cycloaddition mechanism is also favored for the reaction
of RuO4 with ethene on the basis of DFT calculations by
Frenking and co-workers,34 and for the alkene cis-dihydroxy-
lation by cis-[RuVI(Me3tacn)O2(OCOCF3)]+ according to our
[FeV(TPA)(O)(OH)]2+ (1.66 Å, TPA
)
tris(2-pyridyl-
methyl)amine)16a and the oxoiron(V) complex [FeV(TAML)(O)]-
(1.60 Å, H4TAML ) a macrocyclic tetraamide).35
The reaction of 3i with A was found to involve initial
formation of a σ-complex CP1, which is then converted to a
radical intermediate CP2 via transition state TS,36 followed by
conversion of CP2 to cycloadduct C (Figure 3, upper). Similarly,
the reaction of 3i with B (Figure 3, lower) starts with a
σ-complex CP1′, with subsequent formation of a radical
intermediate CP2′ via transition state TS′, followed by formation
of cycloadduct D. The calculated structures, along with key bond
(33) (a) Pidun, U.; Boehme, C.; Frenking, G. Angew. Chem., Int. Ed. Engl.
1996, 35, 2817. (b) Dapprich, S.; Ujaque, G.; Maseras, F.; Lledo´s,
A.; Musaev, D. G.; Morokuma, K. J. Am. Chem. Soc. 1996, 118,
11660. (c) Torrent, M.; Deng, L.; Duran, M.; Sola, M.; Ziegler, T.
Organometallics 1997, 16, 13. (d) DelMonte, A. J.; Haller, J.; Houk,
K. N.; Sharpless, K. B.; Singleton, D. A.; Strassner, T.; Thomas, A. A.
J. Am. Chem. Soc. 1997, 119, 9907. (e) Ujaque, G.; Maseras, F.;
Lledo´s, A. J. Am. Chem. Soc. 1999, 121, 1317. (f) Norrby, P.-O.;
Rasmussen, T.; Haller, J.; Strassner, T.; Houk, K. N. J. Am. Chem.
Soc. 1999, 121, 10186. (g) Fristrup, P.; Tanner, D.; Norrby, P.-O.
Chirality 2003, 15, 360. (h) Ujaque, G.; Maseras, F.; Lledo´s, A. Eur.
J. Org. Chem. 2003, 833.
(35) Tiago de Oliveira, F.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal,
S.; Que, L., Jr.; Bominaar, E. L.; Mu¨nck, E.; Collins, T. J. Science
2007, 315, 835.
(36) In the conversion of CP1 to CP2 via TS, the alkene group of 3i is
attacked mainly by the oxo group of intermediate A. An alternative
pathway is the initial attack of the alkene group mainly by the hydroxo
group of A, but this pathway was found to have a higher barrier (for
more detail, see Figure S15 in the Supporting Information).
(34) Frunzke, J.; Loschen, C.; Frenking, G. J. Am. Chem. Soc. 2004, 126,
3642.
9
J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010 13237

A R T I C L E S
Chow et al.
Figure 4. Calculated structures of transition states TS and TS′ and
cycloadducts C and D with omission of hydrogen atoms except those bonded
to C1, C2, and O2 atoms. The key bond distances (Å) are indicated.
Figure 3. Relative free energy profiles in gas phase and selected parameters
of stationary points for the reactions of dimethyl fumarate (3i) with
oxoiron(V) species A (upper) and B (lower). The free energies of activation
(∆Gsolv in kcal mol-1) in MeCN solvent at CPCM(bondi)-B3LYP/6-31G(d):
lanl2dz for Fe level of theory are in the parentheses.
on the contrary, CP1 has a ∆G value lower than that of “3i +
A” by 8.1 kcal mol-1. (iii) The structure of CP1′ has identical
C(1)-O(1) and C(2)-O(2) distances (2.865 Å), whereas
different C(1)-O(1) and C(2)-O(2) distances (2.896 and 3.853
Å, respectively) are found in CP1. (iv) On going from CP1′ to
TS′, the C(2)-O(2) distance increases from 2.865 to 3.172 Å,
unlike a decrease of C(2)-O(2) distance from 3.853 to 3.345
Å on going from CP1 to TS. (v) The transition state TS′ has a
higher free energy of activation (16.6 kcal mol-1) than that (11.8
kcal mol-1) for the transition state TS (Figure 3).
As MeCN was used as a solvent in the 1-catalyzed cis-
dihydroxylation of 3i, the solvation effect on the reaction
energetics for the oxidation of 3i by B was examined by SCRF
calculations (CPCM) using the gas-phase geometry with MeCN
as the solvent. The results of bulk effect show that the barrier
of the first oxidation step decreases from 16.6 (∆Ggas) to 13.1
(∆Gsolv) kcal mol-1 (Figure 3, the data in parentheses). From
the value (13.1 kcal mol-1) of the free energy of activation in
the rate-determining conversion of the σ-complex CP1′ to the
radical intermediate CP2′, one can roughly estimate that the
half lifetime of the reactants is on the scale of 10-3 s according
to the transition state theory and by supposing the reaction is
of first-order. Therefore, the present computational results
suggest that a cis-dioxoiron(V) species could be a reaction
intermediate for direct cis-dihydroxylation of alkenes.
distances, are depicted in Figure 4 for TS, C, TS′, and D, and
in Figure S13 (see the Supporting Information) for CP1, CP2,
CP1′, and CP2′.37
Both the mechanisms involving A and B have an asymmetric
transition state (TS or TS′), in which the C(1)-O(1) distance
is significantly shorter than that in the σ-complex (1.931 vs 2.896
Å for TS vs CP1, 1.974 vs 2.865 Å for TS′ vs CP1′). Also, the
iron center changes its oxidation state from FeV to FeIV upon
conversion of the σ-complex (CP1 or CP1′) to the radical
intermediate (CP2 or CP2′), and from FeIV to FeIII in subsequent
conversion of the radical intermediate to the cycloadduct (C or
D). All these conversions are exothermic processes.
However, compared with the reaction of 3i with A, the
reaction of the same alkene with B shows several differences:
(i) In terms of relative energy (∆E), the σ-complex (CP1′) is
more stable than the separate reactants (3i + B) by 7.2 kcal
mol-1, a value smaller than that of 19.0 kcal mol-1 (CP1 vs
“3i + A”) in the reaction with A. (ii) The relative free energy
(∆G) of CP1′ is 3.1 kcal mol-1 higher than that of “3i + B”;
(37) Concerning the conversions of CP2 f C and CP2′ f D, one transition
state (TS2) was located on quartet state in the former case (no transition
state was successfully located for the latter conversion at the B3LYP/
6-31G (d):lanl2dz level of theory). However, the transition state TS2
was found to be heavily contaminated by other spin states (〈S2〉 )
4.63), rendering it hard to obtain an accurate energy of TS2 by the
theoretical method (B3LYP) employed in our study (the calculated
free energy of TS2 is lower by 4.2 kcal/mol than that of CP2 due to
the underestimation of TS2 energy resulting from spin contamination).
For this reason, TS2 is not shown in Figure 3. The calculated structure
of this transition state is depicted in Figure S14 in the Supporting
Information.
On the basis of above ESI-MS, isotope labeling studies, and
DFT calculations, we propose that the 1-catalyzed cis-dihy-
droxylation of alkenes with oxone could involve cis-
HO-FeVdO and/or cis-OdFeVdO intermediates, as the cis-
dioxoiron(V) species results in a reasonable barrier (16.6 kcal
mol-1), although this larger barrier than that for cis-HO-FeVdO
(11.8 kcal mol-1) would render the cis-dihydroxylation by cis-
9
13238 J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010

cis-Dihydroxylation of Alkenes
A R T I C L E S
OdFeVdO a less favorable pathway. We note that a re-
lated structurally characterized cis-OdRuVdO complex, cis-
[RuVLO2]+ (L ) N,N,N′,N′,3,6-hexamethyl-3,6-diazaoctane-1,
8-diamine), is stable toward protonation at pH g 3.0; thus it is
not unreasonable to conceive that the cis-OdFeVdO species B
is the major oxoiron(V) species present in the solution under
the catalytic conditions (pH 7).38 The reaction of cis-
HO-FeVdO and cis-OdFeVdO intermediates with alkene to
give cycloadducts C and D via CP2 and CP2′, respectively,
has no significant activation barrier in the conversion of CP2/
CP2′ to C/D, and therefore can be described as a concerted but
highly asynchronous process, in which the formation of the
second C-O bond is coupled to the formation of the first one.
This concerted mechanism revealed by DFT calculations is
consistent with the absence of appreciable trans-diol product
in the catalytic reactions.
to a solution of alkene (0.5 mmol) and 1 (0.7-3.5 mol %) in
acetonitrile (6 mL) at room temperature within 5 min. Upon
quenching by aqueous saturated Na2SO3 solution, the mixture was
extracted with ethyl acetate (10 mL) containing 1,4-dichlorobenzene
(GC internal standard, 0.1 mmol). An aliquot of the extract was
analyzed by GC. Afterward, the mixture was further extracted with
ethyl acetate (3 × 50 mL). The organic products (diol and
R-hydroxy ketone) were identified and quantified by 1H NMR
analysis, or isolated after purification by column chromatography
on silica gel.
Large-Scale cis-Dihydroxylation of Methyl Cinnamate (3a)
with Oxone Catalyzed by 1. A solution of oxone (30 mmol) and
NaHCO3 (90 mmol) in water (300 mL) was added in two portions
to a solution of methyl cinnamate (30 mmol) and 1 (1.4 mol %) in
acetonitrile (300 mL) at room temperature within 5 min. To the
reaction mixture was added a solution of methyl cinnamate (30
mmol) and Fe(ClO4)2 ·4H2O (1.4 mol %) in acetonitrile (300 mL),
followed by addition of a solution of oxone (30 mmol) and NaHCO3
(90 mmol) in water (300 mL) in the same manner as described
above. After quenching by aqueous saturated Na2SO3 solution, the
reaction mixture was extracted with ethyl acetate (4 × 500 mL),
Conclusion
By using iron complexes of a macrocyclic tetraaza ligand,
we have developed an efficient and selective metal-catalyzed
cis-dihydroxylation of alkenes with oxone as terminal oxidant.
This iron-catalyzed reaction proceeds rapidly at room temper-
ature, uses limiting amounts of alkene substrates, and can be
performed on ∼10 g scale, providing a convenient protocol for
practical cis-dihydroxylation of electron-deficient alkenes (such
as R,ꢀ-unsaturated esters). High-valent oxoiron(V) species cis-
HO-FeVdO and/or cis-OdFeVdO might be involved in the
alkene cis-dihydroxylation with oxone catalyzed by [FeIII(L-
N4Me2)Cl2]+ (1). According to DFT calculations, the cis-dihy-
droxylation reactions by cis-HO-FeVdO and cis-OdFeVdO both
occur by a concerted but highly asynchronous pathway, with the
cis-dioxoiron(V) species resulting in a larger activation barrier.
Taking also into account the [3 + 2] cycloaddition of cis-
dioxoruthenium complexes with alkenes, it is apparent that reactive
cis-dioxometal species other than that of osmium could be
harnessed for alkene cis-dihydroxylation.
1
and the extract was subjected to GC and H NMR analysis for
product identification and quantification. The cis-diol product was
isolated after purification by column chromatography on silica gel.
Reuse of L-N4Me2 Ligand in Iron-Catalyzed cis-Dihydroxy-
lation of Methyl Cinnamate (3a) with Oxone. A solution of oxone
(1.0 mmol) and NaHCO3 (3.0 mmol) in water (6 mL) was added
in two portions to a solution of methyl cinnamate (0.5 mmol) and
1 (1.4 mol %) in acetonitrile (6 mL) at room temperature within 5
min. Subsequent runs were performed by adding a solution of
methyl cinnamate (0.5 mmol) and Fe(ClO4)2 ·4H2O (1.4 mol %) in
acetonitrile (6 mL) to the reaction mixture followed by addition of
a solution of oxone (1.0 mmol) and NaHCO3 (3.0 mmol) in water
(6 mL) in the same manner as described above. The reaction was
quenched by saturated aqueous Na2SO3 solution after all desired
runs were completed. The reaction mixture was extracted with ethyl
acetate (3 × 100 mL), and the extract was subjected to GC and 1H
NMR analysis for product identification and quantification.
Computational Details. B3LYP/6-31G(d):lanl2dz for Fe density
functional theory method was employed to optimize local minima
on the potential energy hypersurfaces corresponding to the reactant
complex, the reaction intermediates, the final complex, and other
fragment species. Vibrational analyses were systematically carried
out to ensure that the optimized geometries correspond either to a
local minimum that has no imaginary frequency mode or to a saddle
point that has only one imaginary frequency mode. Cartesian
coordinates for stationary structures are included in the Supporting
Information (Table S4). The bulk effect on the dihydroxylation of
alkene by cis-dioxoiron(V) species B was investigated by the SCRF
calculation using CPCM-bondi model at the B3LYP/6-31G(d):
lanl2dz for Fe level of theory.
Experimental Section
General. Oxone and NaHCO3 (purchased from Sigma-Aldrich)
were used as received. Alkene substrates 2a-h and 3a-j were
purified by column chromatography on silica gel or Al2O3.
27
Complexes [FeIII(Me3tacn)Cl3]39 and [FeIII(L-N4H2)Cl2]+ were
prepared as described in the literature. [FeII(qpy)Cl2] and [FeII(6-
Me2-bpmcn)Cl2] were prepared according to the same procedures
40
as those reported for [FeII(qpy)(MeCN)2](ClO4)2 and [FeII(bpm-
cn)Cl2],41 respectively, except that FeCl2 (instead of Fe(ClO4)2) was
used in the former case and 6-Me2-bpmcn (instead of bpmcn) was
used in the latter case. Catalysts [FeIII(L-N4Me2)Cl2]+ (1) and
[FeIII(L-N4Me2)Br2]+ were prepared from reactions of L-N4Me2 with
FeCl3 ·6H2O and FeBr3, respectively (see the Supporting Informa-
tion for detailed procedures).
Acknowledgment. This work was supported by The University
of Hong Kong (University Development Fund), the University Grants
Council of HKSAR (the Area of Excellence Scheme: AoE 10/01P),
and the Hong Kong Research Grants Council (HKU 1/CRF/08, HKU
7007/08P). We are grateful to Dr. Nianyong Zhu for assistance in
solving the X-ray crystal structures of 1·FeCl4 and 1·ClO4 ·MeCN.
General Procedure for cis-Dihydroxylation of Alkenes with
Oxone Catalyzed by 1. A solution of oxone (1.0 mmol) and
NaHCO3 (3.0 mmol) in water (6 mL) was added in two portions
(38) We examined the pH dependence of the 1-catalyzed cis-dihydroxy-
lation of 2e with oxone (by adjusting the amount of NaHCO3 used in
the reaction). Upon increasing pH from 7 to 8-9, a decrease in the
efficiency of cis-dihydroxylation was found (substrate conversion: from
89% to 60%, yield of cis-diol I: from 44% to 25%, yield of epoxide:
from 19% to 22%). We noted that 1 became significantly less stable
at pH 8-9.
Supporting Information Available: Physical measurements,
procedures for catalyst preparation and product characterization,
Tables S1-S4, Scheme S1, Figures S1-S15, and CIF files for
the crystal structures of 1·FeCl4 and 1·ClO4 ·MeCN. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(39) Chaudhuri, P.; Winter, M.; Wieghardt, K.; Gehring, S.; Haase, W.;
Nuber, B.; Weiss, J. Inorg. Chem. 1988, 27, 1564.
(40) Wong, E. L.-M.; Fang, G.-S.; Che, C.-M.; Zhu, N. Chem. Commun.
2005, 4578.
(41) Costas, M.; Que, L., Jr. Angew. Chem., Int. Ed. 2002, 41, 2179.
JA100967G
9
J. AM. CHEM. SOC. VOL. 132, NO. 38, 2010 13239