COMMUNICATION
Table 1. Oxidation of cis-4-methylcyclohexyl pivalate by different cata-
dation of these three substrates and account for 4–13%
yield.
lysts.[a]
Oxidation of the methylene site in simple cycloalkanes
(Table 2, entries 5 and 6) was also accomplished, affording
the corresponding ketone products in 54–62% yields, pre-
sumably by means of a 2-step oxidation involving initial for-
mation of the corresponding alcohol, which is also obtained
in minor amounts. In agreement with this proposal, submit-
ting cyclohexanol to the standard experimental conditions
yielded cyclohexanone quantitatively. A more challenging
linear alkane such as n-hexane is also oxidized to a 1:1 mix-
ture of corresponding C2 and C3 ketones in a remarkable
57% combined yield (Table 2, entry 10). On the other hand,
methylene oxidation is sensitive to the electronic properties
Entry
Cat.
Cat. [mol%]/
Oxidant [equiv]/
AcOH [equiv]
Oxidant
A [%][b]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
3Me,Me
3Me,Me
3Me,Me
1
1:1.2:0.5
(1:1.2:0.5)x3
3:3.6:1.5
3:3.6:1.5
3:3.6:1.5
H2O2
H2O2
H2O2
H2O2
31
51
54[d]
53
64
44
6
23
39
51
23
54
52
19
9
2
H2O2
3Me,Me
3Me,Me
3Me,Me
3Me,Me
3Me,Me
3Me,Me
3Me,Me
3Me,Me
1
3:3.6:1.5
3:3.6:1.5
CH3CO3H
Oxone
tBuOOH
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
À
of the C H bonds. This is shown in the oxidation of methyl
3:3.6:1.5
3:3.6:1.5[c]
5:3.6:1.5
hexanoate (Table 2, entry 11), which afforded preferential
oxidation at the most remote, electron-rich methylene site.
Modest yet significant steric effects are illustrated in the oxi-
dation of methylenic sites of 1,1-dimethylcyclohexane
(Table 2, entry 12). Ketone resulting from oxidation at the
sterically more demanding C2 site is obtained in 19% yield,
while C3 and C4 ketones are obtained in 26 and 13% yields,
respectively. These numbers, normalized for the number of
1:3.6:1.5
(2:2.4:9.0)[c] (1:1.2:0.5)[e]
3:3.6:0
3:3.6:0
3:3.6:0
2
[a] Equivalents refer to substrate. The reaction was performed by slow
syringe-pump addition, over 30 min, of a solution of H2O2 in acetonitrile
of catalyst and substrate at 08C in acetonitrile. [b] GC yield. [c] Catalyst
and H2O2 were simultaneously added through two parallel syringe pumps
over 30 min, into a solution of substrate at 08C in acetonitrile. [d] Meth-
ylenic site oxidation products were also detected with 9% yield. [e] A
third addition of catalyst, H2O2, and acetic acid was made.
À
C H bonds, indicate that C3 and C4 are equally reactive,
and slightly preferred over C2.
À
Oxidation of C H bonds can also lead to products arising
from a competition between hydroxylation and desaturation
1
reactions. H NMR analysis of a crude reaction mixture ob-
tained after oxidation of 4-methyl valeric acid (Table 2,
entry 13) under standard experimental conditions showed
the presence of C4 alcohol (32% yield), and 4,4’-dimethyl-g-
butyrolactone (9%). These products arise from an initial hy-
droxylation at C4. Following workup, 4,4’-dimethyl-g-butyr-
olactone was isolated in 42% yield. Besides hydroxylation,
products derived from a desaturation reaction are also ob-
served and include 4-methylpent-4-enoic acid (2%), the cor-
responding C4–C5 epoxide (2%), and 4-hydroxymethyl-4-
methyl-g-butyrolactone (11%). These products can be un-
derstood by assuming an initial desaturation, epoxidation of
the resulting olefin, and then intermolecular lactonization
through epoxide ring opening to yield 4-hydroxymethyl-4-
methyl-g-butyrolactone. Interestingly, the desaturation/hy-
droxylation product ratio depends on the nature of the iron
catalyst; with 3Me,Me it is 1:3.5, but 1 yields a substantially
larger discrimination towards hydroxylation (1:6.4).[13] The
origin of these differences is not understood at present.
Comparison between the regioselectivity in the oxidation
of cis- and trans-4-methylcyclohexyl-1-pivalate (Table 2, en-
tries 7 and 8) provides an illustration of the role of strain re-
Of these, only peracetic acid provides comparable product
yields and selectivity with regard to H2O2. We therefore re-
stricted our study to H2O2 as the more attractive oxidant.
Sequential addition of the catalyst, or higher catalyst load-
ings are not necessary (Table 1, compare entries 2 and 10),
and therefore the experimental procedure is largely simpli-
fied with respect to 1[8] and 2.[9] A further fundamental dif-
ference between the reactivity of these last two catalysts and
3Me,Me is that acetic acid appears not to be strictly required
for efficient catalysis (Table 1, compare entries 13–15).
Acetic acid has been shown to be a key additive to enhance
product yields in Fe-catalyzed oxidations,[11] by promoting
III
[12]
À
fast O O heterolysis of Fe -OOH intermediates. The ex-
ceptional behavior exhibited by 3Me,Me suggests that O O
À
lysis is remarkably facile for this catalyst.
The oxidation of a series of substrates using the above-de-
fined experimental conditions are shown in Table 2. Com-
bined product yields (with respect to the initial substrate)
range from 28 to 63%, adding 3Me,Me to the exclusive family
of iron catalysts that may find application in chemical syn-
thesis. Oxidation of 2,6-dimethyloctane occurs preferentially
at the two tertiary sites with a modest combined yield of
24%. An additional 4% is ketone byproduct (Table 2,
entry 1). On the other hand, when halide or ester groups are
À
lease effects in conditioning C H site selectivity. The terti-
ary C4 alcohol and C3 ketone were obtained in 39 and 22%
respective yields. As recently shown by Baran and Eschen-
moser,[14] breakage of tertiary C H sites in equatorial posi-
À
À
present, oxidation occurs preferentially at the remote C H
tions of cyclohexane skeletons is facilitated by strain release.
À
bond in 29–35% yields (Table 2, entries 2–4). This indicates
that the oxidant has an electrophilic nature. Products result-
ing from methylene oxidation were also observed in the oxi-
In the absence of this factor, tertiary C H bonds in axial po-
sitions are stronger, and their oxidation can compete with
[14]
À
that of secondary C H sites.
Chem. Eur. J. 2013, 19, 1908 – 1913
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1909