Oxidation of adamantane with 1 atm molecular oxygen by vanadium-substituted polyoxometalates

Article abstract of DOI:10.1016/j.jcat.2005.04.012

The oxidation of adamantane with molecular oxygen as a sole oxidant was efficiently promoted by catalyst precursors of vanadium-substituted Keggin-type phosphomolybdates such as H₄PVMo₁₁O₄₀, H ₅PV₂Mo₁₀O₄₀, and H ₆PV₃Mo₉O₄₀ in butyronitrile. The major product was a tertiary CH bond oxygenated product of 1-adamantanol, and secondary CH bond oxygenated products were also formed. The total yield of oxygenated products for the oxidation of adamantane in the presence of H ₅PV₂Mo₁₀O₄₀ reached 84%. NMR and IR data show that the vanadium-substituted phosphomolybdates, such as H ₄PVMo₁₁O₄₀ and H₅PV ₂Mo₁₀O₄₀, decompose to form the monomeric vanadium species (VVO2+ (main) and V^IVO²⁺) and PMo12O403- Keggin anion. The reaction mechanism involving a radical species was proposed from ESR and kinetic data. The catalysts initially abstract the hydrogen of adamantane to form the adamantyl radical and reduced catalysts. This step would be promoted mainly by the vanadium species, such as VVO2+, and the phosphomolybdates, PMo12O40n-, enhance the activity. The adamantyl radical formed promotes the successive formation of the key intermediates, such as adamantyl radical and hydroperoxide species.

Full text of DOI:10.1016/j.jcat.2005.04.012

Journal of Catalysis 233 (2005) 81–89
www.elsevier.com/locate/jcat
Oxidation of adamantane with 1 atm molecular oxygen
by vanadium-substituted polyoxometalates
Satoshi Shinachi, Mitsunori Matsushita, Kazuya Yamaguchi, Noritaka Mizuno ∗
Department of Applied Chemistry, School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
Received 24 February 2005; revised 5 April 2005; accepted 11 April 2005
Available online 23 May 2005
Abstract
The oxidation of adamantane with molecular oxygen as a sole oxidant was efficiently promoted by catalyst precursors of vanadium-
substituted Keggin-type phosphomolybdates such as H PVMo , H PV Mo , and H PV Mo O in butyronitrile. The major
O
11 40
O
10 40
4
5
2
6
3
9 40
product was a tertiary C–H bond oxygenated product of 1-adamantanol, and secondary C–H bond oxygenated products were also formed.
The total yield of oxygenated products for the oxidation of adamantane in the presence of H PV Mo reached 84%. NMR and IR data
O
10 40
5
2
show that the vanadium-substituted phosphomolybdates, such as H PVMo
O
and H PV Mo O , decompose to form the monomeric
4
11 40 5 2 10 40
V
+
IV 2+
3−
O
12 40
vanadium species (V O (main) and V
O
) and PMo
Keggin anion. The reaction mechanism involving a radical species was
2
proposed from ESR and kinetic data. The catalysts initially abstract the hydrogen of adamantane to form the adamantyl radical and reduced
V
+
n−
O ,
12 40
catalysts. This step would be promoted mainly by the vanadium species, such as V O , and the phosphomolybdates, PMo
2
enhance the activity. The adamantyl radical formed promotes the successive formation of the key intermediates, such as adamantyl radical
and hydroperoxide species.
2005 Elsevier Inc. All rights reserved.
Keywords: Adamantane; Molecular oxygen; Oxidation; Polyoxometalate; Vanadium
1. Introduction
cular bromine, followed the hydrolysis of the correspond-
ing brominated ones. Secondary mono-oxygenated adaman-
tanes such as 2-adamantanol are synthesized by the re-
arrangement of tertiary hydroxylated adamantanes with the
use of concentrated H2SO4 [8–10]. From the standpoint of
green chemistry, these conventional methods are undesir-
able because they need multistep reactions and hazardous
reagents to produce large quantities of by-products [8–10].
Adamantane has four relatively weak tertiary C–H bonds,
and the autooxidation may take place via a radical-chain
mechanism at elevated temperatures. However, even in the
case of adamantane, it is known that such autooxidation
cannot be efficiently catalyzed by simple transition-metal
salts such as Co(acac)2, Mn(OAc)2, and Mn(acac)3 under
mild reaction conditions (below 373 K) [11,12]. In these
contexts, there is a great demand for the development of
one-step catalytic adamantane oxidation systems with an en-
vironmentally friendly oxidant of molecular oxygen without
The development of efficient methods for the oxidation
of unactivated C–H bonds of alkanes with molecular oxy-
gen is of great importance from industrial and synthesis
standpoints [1–3]. Recently, much attention has been paid
to the oxidation of adamantane because the substituted-
adamantane derivatives, especially mono- or di-substituted
ones, can be used as important precursors for photore-
sists and medicines [4–6]. Although many polysubstituted
adamantanes have been easily synthesized, the synthesis of
mono- or di-oxygenated ones is very difficult [7]. Typi-
cally, tertiary mono- and di-oxygenated adamantanes, such
as 1-adamantanol and 1,3-adamantanediol, are synthesized
by the bromination of adamantane with the use of mole-
*
Corresponding author. Fax: +81 3 5841 7220.
E-mail address: tmizuno@mail.ecc.u-tokyo.ac.jp (N. Mizuno).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2005.04.012
2005 Elsevier Inc. All rights reserved.

82
S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
hazardous reagents which can reduce environmentally unde-
sirable wastes.
were measured at 70.75 MHz in butyronitrile. VOCl3 was
used as an external standard. IR spectra were measured on a
Jasco FT/IR-460 Plus with KBr disks. ESR measurements
were made at 100 K (X-band) with a JEOL JES-RE-1X
spectrometer. The microwave power, modulation width, and
time constant were 1.0 mW, 1.0 mT, and 0.1 s, respectively.
A reactor directly connected to an ESR tube was used to
avoid exposing the sample to air. After a certain catalytic re-
action period, the reaction solution was transferred into the
ESR tube, and the tube was sealed by firing. The simulation
was carried out with the assumption of axial symmetry for
vanadium, according to the literature [50].
Although there have been several reports on the liquid-
phase oxidation of adamantane with molecular oxygen,
these systems need reductants such as hydroquinone [13],
metals of zinc [14–16] and iron [17], and aldehydes [18–21].
Ishii and co-workers have developed an efficient alkane oxi-
dation system with a combined catalyst of N-hydroxyphtha-
limide (NHPI) and Co(acac)2 [11,12], and the NHPI almost
decomposed to phthalic acid and the other compounds (used
by our group). Therefore, a large amount of NHPI (at least
10 mol%) was required to achieve the high yields of the cor-
responding oxygenated products, and the catalyst could not
be recycled. Thus, there are only limited examples of the
oxidation of adamantane with molecular oxygen under mild
reaction conditions with only a small amount of transition-
metal catalysts [22–25].
Solvents were analytical grade (Tokyo Kasei) and were
purified before use [51]. Adamantane and 1,3-dimethylada-
mantane were commercially obtained from Tokyo Kasei
(reagent grade) and purified before use [51].
Polyoxometalates catalyze the transformation of vari-
ous kinds of functional groups because the redox and acid
properties can be controlled at the atomic or molecular
level through a change of the constituent elements [26–31].
Therefore, the catalytic function of polyoxometalates has
attracted much attention. Various kinds of polyoxometalate-
catalyzed liquid-phase oxidation reactions have been re-
ported [32–44], and Keggin-type phosphovanadomolybda-
tes such as H4PVMo11O40, H5PV2Mo10O40, and H6PV3-
Mo9O40 have been reported to be catalytically active for
the reactions [45–49]. However, to our knowledge, there
has been no application of the vanadium-substituted poly-
oxometalates to the oxidation of adamantane with mole-
cular oxygen as a sole oxidant. In this paper, we report
on the oxidation of adamantane with 1 atm of molec-
ular oxygen in butyronitrile without any additives pro-
moted by vanadium-substituted polyoxometalates. Notably,
the vanadium-substituted polyoxometalates showed higher
catalytic activity than did transition-metal salts such as
Co(OAc)2, Mn(OAc)2, and VO(acac)2, which are widely
used as catalysts for free radical autooxidations [1–3,11,12].
We also investigated the catalyst state and possible reaction
path.
2.2. Catalysts
Phosphometalates (except for H3PMo12O40, H4PV
Mo11O40, [VIVO(H2O)5]H[PMo12O40], and (NH4)5[PVIV
Mo11O40]) were supplied by Nippon Inorganic Colour
and Chemical Co., Ltd. and used after recrystallization
from water. H3PMo12O40 (Kanto) was used as received.
H4PVMo11O40 [52], [VIVO(H2O)5]H[PMo12O40] [53], and
(NH4)5[PVIVMo11O40] [54] were synthesized with proce-
dures described in the literature.
2.2.1. H4PVMo11O40
Disodium hydrogen phosphate (12.5 mmol) was dis-
solved in 25 ml of water and mixed with sodium meta-
vanadate (12.5 mmol) that had been dissolved in 25 ml of
boiling water. The mixture was cooled and acidified with
1.3 ml of concentrated sulfuric acid. To this mixture was
added an aqueous solution (50 ml) of sodium molybdate
dihydrate (138 mmol). Finally, 21 ml of concentrated sul-
furic acid was slowly added with vigorous stirring, which
was followed by a color change from dark red to light red.
The heteropoly acid was then extracted with 100 ml of di-
ethyl ether. In this extraction, the heteropoly etherate was
present in the middle layer. After the separation, a stream of
air was passed through the heteropoly etherate layer to free it
of ether. The orange solid obtained was dissolved in 13 ml of
water and concentrated in a vacuum desiccator until the crys-
tals appeared. The orange crystals were filtered, washed with
water, and dried under vacuum (23% yield). Anal. calcd. for
H4PVMo11O40 · 17H2O: P, 1.48; V, 2.44; Mo, 50.56. Found:
P, 1.61; V, 2.89; Mo, 49.82. IR (KBr) (cm−1): 1078 (sh,
νas(P–Oa–V)), 1061 (s, νas(P–Oa)), 960 (s, νas(Mo–Od)),
866 (s, νas(Mo–Ob–M)) (M = Mo, V), 786 (s, νas(Mo–
Oc–M)) (Oa, inner oxygen; Ob, corner-shared oxygen; Oc,
edge-shared oxygen; Od, terminal oxygen). 31P NMR (buty-
ronitrile, 109.24 MHz, 0.67 mM): −2.06 ppm (the purity by
31P NMR ꢀ 80%). UV–vis (acetonitrile, 109.24 MHz, 0.02
mM): λmax 308 nm (ε = 2.22 × 105 cm−1 M−1).
2. Experimental
2.1. General
GC analyses were performed on a Shimadzu GC-17A
with a flame ionization detector equipped with a DB-WAX
capillary column (internal diameter 0.25 mm, length 30 m).
Mass spectra were recorded on a Shimadzu GCMS-QP2010
at an ionization voltage of 70 eV equipped with a DB-
WAX capillary column (internal diameter 0.25 mm, length
30 m). NMR spectra were recorded on a JEOL JNM-EX-
270. 31P NMR spectra of polyoxometalates were measured
at 109.24 MHz in butyronitrile. H3PO4 (85%) was used as
an external standard. 51V NMR spectra of polyoxometalates

S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
83
2.2.2. [VIVO(H2O)5]H[PMo12O40]
777 (s, νas(Mo–Oc–M)). UV–vis (acetonitrile, 0.02 mM):
λmax 672 nm (ε = 8.55 × 103 cm−1 M−1), 308 nm (ε =
2.22 × 105 cm−1 M−1).
H3PMo12O40 (4.85 mmol) was dissolved in 10 ml of wa-
ter. Barium hydroxide octahydrate (4.85 mmol) was added
to the solution in small portions to keep the pH constant
(pH 0.75). Vanadyl sulfate pentahydrate (4.85 mmol) was
then quickly added, and the solution was kept at room
temperature for 0.5 h. The barium sulfate precipitate was
filtered off, and then the resultant solution was kept at
277 K to yield green crystals (27% yield). Anal. calcd for
[VO(H2O)5]H[PMo12O40] · 8H2O: P, 1.46; V, 2.40; Mo,
54.18. Found: P, 1.71; V, 2.36; Mo, 54.15. IR (KBr) (cm−1):
1065 (s, νas(P–Oa)), 962 (s, νas(Mo–Od)), 870 (s, νas(Mo–
Ob–M)) (M = Mo, V), 790 (s, νas(Mo–Oc–M)). 31P NMR
(butyronitrile, 109.24 MHz, 0.67 mM): −2.50 ppm. UV–
vis (acetonitrile, 0.02 mM): λmax 308 nm (ε = 2.19 ×
105 cm−1 M−1).
2.3. Procedure for oxidation of adamantane and
1,3-dimethyladamantane
Oxidation of adamantane (or 1,3-dimethyladamantane)
was carried out in a glass vial containing a magnetic stir bar.
A typical procedure for the adamantane oxidation was as fol-
lows. Into a glass vial were successively placed adamantane
(1 mmol), catalyst (2 µmol), and butyronitrile (3 ml). Then
1 atm of molecular oxygen was introduced into the system.
The reaction mixture was heated at 356 K. The yields were
determined by GC analyses with naphthalene as an internal
standard. All of the products were confirmed by GC analysis
in combination with mass spectroscopy.
2.2.3. (NH4)5[PVIVMo11O40]
Ammonium heptamolybdate (7.8 mmol) was dissolved in
51 ml of hot water. The solution was cooled to room tem-
perature, followed by acidification with 1.5 ml of glacial
acetic acid. We prepared the second solution by dissolv-
ing vanadium oxide sulfate pentahydrate (5 mmol) in 15 ml
of water containing 0.41 ml of phosphoric acid (75% by
volume). The two solutions were cooled to 277 K, and
then a solution of vanadyl sulfate was added to that of
molybdate. After 0.5 h, the dark blue solution was fil-
tered, and ammonium nitrate (56.3 mmol) was added to
the filtrate, which was followed by the formation of a dark
blue crystalline solid. The solid was filtered and washed
with cold water to eliminate a trace of ammonium nitrate
(28% yield). Anal. calcd. for (NH4)5[PVIVMo11O40] · H2O:
H, 1.87; N, 3.45; P, 1.53; V, 2.51; Mo, 52.00. Found:
H, 2.02; N, 3.54; P, 1.71; V, 2.73; Mo, 49.24. IR (KBr)
(cm−1): 1066 (sh, νas(P–Oa–V)), 1051 (s, νas(P–Oa)), 945
(s, νas(Mo–Od)), 865 (s, νas(Mo–Ob–M)) (M = Mo, V),
2.4. Kinetic study
Oxidation of adamantane was performed with the same
procedure as described above. The reaction conditions are
given in the figure captions. Reaction rates (R0) for the
kinetic analyses were determined from the slope of reac-
tion profiles (conversion vs. time plots) at low conversion
(< 10%) of the substrate after the induction period.
3. Results and discussion
3.1. Oxidation of adamantane
First, the oxidation of adamantane was carried out with
H5PV2Mo10O40 at 356 K under 1 atm of molecular oxy-
gen in various solvents. The results are summarized in
Table 1. The oxidation did not proceed in the absence of
Table 1
a
Oxidation of adamantane with molecular oxygen in various solvents
H PV Mo
O
5
2
10 40
−−−−−−−−−−−−→
+
+
+
+
O
2
1
2
3
4
5
Entry
1
Solvent
Yield
Selectivity (%)
(%)
1
2
3
4
5
Butyronitrile
Butyronitrile
Diethylketone
Cyclopentanone
Dimethylformamide
Acetic acid
Acetonitrile
Toluene
1,2-Dichloroethane
46
84
30
22
12
54
43
65
68
80
65
–
17
24
11
14
–
–
–
–
–
14
11
17
12
19
15
–
16
20
17
16
11
20
–
19
12
–
–
–
–
–
–
–
b
2
3
4
5
6
7
8
9
11
< 1
< 1
< 1
–
–
–
–
–
–
a
Reaction conditions: H PV Mo
O
10 40
(2 µmol), adamantane (1 mmol), solvent (3 ml), 356 K, 96 h under 1 atm of molecular oxygen. Yields and selectiv-
5
2
ities were determined by gas chromatographic analysis using naphthalene as an internal standard.
b
288 h.

84
S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
ondary C–H bond oxygenated products were also formed.
After ca. 36 h, the successive oxidation of 1-adamantanol
to 1,3-adamantanediol proceeded. In the present oxidation,
N-(1-adamantyl)butylamide was also produced by the Ritter
reaction of 1-adamantanol with butyronitrile. When the re-
action time was prolonged to 288 h, the total yield of the
products reached 84%. From the standpoint of the prod-
uct yields, this value is higher than those reported so far
for the oxidation of adamantane with molecular oxygen
as a sole oxidant under mild conditions (below 373 K):
Ru-substituted sandwich-type polyoxometalate[(n-C8H17)3
CH3N]11 [WZnRu2(OH)(H2O)(ZnW9O34)2] (57%) [23],
Ru-substituted tungstosilicate [(n-C4H9)4N]4H[SiW11Ru
(H2O)O39] (42%) [22], Ni and Fe-substituted tungstophos-
phate [(n-C4H9)N]4H6[PW9O37{Fe2Ni(OAc) 3}] (29%)
[24], and K[Ru(saloph)Cl2] (13%) [25]. After the oxidation
was finished, adamantane was re-added to the reaction so-
lution, and the solution was again heated at 356 K in the
presence of molecular oxygen. The oxidation again pro-
ceeded without an induction period, and with almost the
same rate and product distribution as those observed for the
first run, showing that H5PV2Mo10O40 is intrinsically recy-
clable.
Fig. 1. Reaction profiles of the oxidation of adamantane with mole-
cular oxygen catalyzed by H PV Mo O
. Total yield (+), yields
5
2
10 40
of 1-adamantanol ("), 1,3-adamantanediol (Q), 2-adamantanol (×),
2-adamantanone (F), and N-(1-adamantyl)butylamide (2). Reaction con-
ditions: H PV Mo
O
10 40
(2 µmol, 0.67 mM), adamantane (1 mmol), buty-
5
2
ronitrile (3 ml), 356 K under 1 atm of molecular oxygen.
Table 2 lists the results for the oxidation of adaman-
tane with molecular oxygen in the presence of various
catalysts. Among the polyoxometalates tested, vanadium-
substituted polyoxometalates were the most effective cata-
H5PV2Mo10O40 in any solvents. Among the solvents tested,
butyronitrile gave the highest yields of the corresponding
oxygenated products. 1,2-Dichloroethane, toluene, and ace-
tonitrile, in which adamantane and/or catalyst were not
soluble, were poor solvents. Therefore, the oxidation was
carried out thereafter in butyronitrile. Fig. 1 shows the re-
action profiles for butyronitrile. The oxidation proceeded
with an induction period (10–11 h). The major product
was tertiary C–H bond oxygenated 1-adamantanol; sec-
lysts, and the yield decreased in the order H5PV2Mo10O40
≈
H6PV3Mo9O40 ꢀ H4PVMo11O40 > H4PVW11O40 ꢀ H3-
PMo12O40 > H3PW12O40. The vanadium-substituted poly-
oxometalates exhibited higher catalytic activity than those
of H3PMo12O40 and H3PW12O40, showing that vanadium
is an indispensable component for attaining high yields of
Table 2
a
Oxidation of adamantane with molecular oxygen in the presence of various catalysts
Catalyst
−−−−−−−−−−−−→
+
+
+
+
O
2
1
2
3
4
5
3
2
b
Entry
Catalyst
Yield
Selectivity (%)
C –H/C –H
(%)
1
2
3
4
5
1
2
3
4
5
6
8
9
10
11
H PV Mo O
10 40
46
46
39
26
7
< 1
29
17
< 1
< 1
54
54
54
62
50
–
78
80
–
17
17
16
15
8
–
–
2
–
14
14
15
16
8
–
–
7
–
16
16
15
15
12
–
22
11
–
19
19
20
12
22
–
–
–
–
–
13.1
13.1
12.9
12.0
13.2
–
10.6
14.0
–
5
2
H PV Mo O
6
3
9 40
H PVMo
O
11 40
4
H PVW
4
O
11 40
H PMo
O
12 40
3
H PW
3
O
12 40
VO(acac)
2
Co(OAc)
2
Mn(OAc)
None
2
–
–
–
–
–
a
Reaction conditions: catalyst (2 µmol), adamantane (1 mmol), butyronitrile (3 ml), 356 K, 96 h under 1 atm of molecular oxygen. Yields and selectivities
were determined by gas chromatographic analysis using naphthalene as an internal standard. Carbon balance for each reaction was more than 93%.
b
The selectivity parameter defined by the relative reactivity of tertiary C–H bonds to secondary C–H bonds (= {(1 + 2 × 2 + 5)/4}/{(3 + 4)/12}).

S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
85
oxygenated adamantanes.1 It was noted that the vanadium-
substituted polyoxometalates showed higher catalytic activ-
ity than those of transition-metal salts such as Co(OAc)2,
Mn(OAc)2, and VO(acac)2, which are widely used as cata-
lysts for free radical autooxidations.
3.2. State of the catalyst
We investigated the state of the catalyst with ESR
spectroscopy in more detail. In the measurements, 1,3-
dimethyladamantane was used as a substrate because the
solubility of adamantane in butyronitrile was low (below
298 K) and the added adamantane was partly insoluble.
When H5PV2Mo10O40 was used in the following ESR mea-
surements, the spectrum was very complicated, as shown
in Fig. 2a, and could not be well reproduced by the simula-
tion with parameters of one species because H5PV2Mo10O40
was an inseparable mixture of five isomers, as has been
reported [55–57]. Therefore, the following ESR measure-
ments were carried out with H4PVMo11O40. The butyroni-
trile solution of H4PVMo11O40 (0.67 mM) without 1,3-
dimethyladamantane was completely ESR silent, even after
treatment at 356 K under argon. Five hundred equivalents
of 1,3-dimethyladamantane with respect to H4PVMo11O40
were added to the solution, which was kept at 356 K
under 1 atm of argon. After 6 h, the color of the solu-
tion had changed from yellow to yellow-green, suggest-
ing the reduction of H4PVMo11O40. When adamantane
was used as a substrate, the same color change was ob-
served. The yellow-green color was not the same as that
of (NH4)5[PVIVMo11O40] (blue), but the same as that of
[VIVO(H2O)5]H[PMo12O40] in butyronitrile. Then the ESR
spectrum was measured at 100 K (Fig. 2b). The hyper-
fine structure of vanadium was observed, and the ESR
spectrum was well reproduced by simulation with the fol-
Fig. 2. ESR spectra of vanadium-containing polyoxometalates at 100 K.
(a) H PV Mo
O
10 40
(0.67 mM) was treated with 1,3-dimethyladaman-
5
2
tane (500 eq) in butyronitrile under 1 atm of argon at 356 K for 6 h.
The signal intensity corresponds to the extent of reduction of 0.5 elec-
trons per polyanion. (b) H PVMo
O
11 40
(0.67 mM) in butyronitrile was
4
treated with 1,3-dimethyladamantane (500 eq) under 1 atm of argon at 356
K for 6 h. The signal intensity corresponds to the extent of reduction of
IV
0.3 electrons per polyanion. (c) [V O(H O) ]H[PMo
O ] in butyroni-
2
5
12 40
IV
trile. (d) (NH ) [PV Mo O ] in acetonitrile. (e) The above solution (b)
11 40
4 5
was treated with 1 atm of molecular oxygen at 356 K for 10 min. The sig-
nal intensity corresponds to the extent of reduction of 0.03 electrons per
polyanion. The dotted line in (b)–(d) were obtained by the simulation (see
text). Six asterisks in (b) were assigned to the signals of the Mn marker.
lowing parameters: g⊥ = 1.984, g = 1.945, A⊥ = 6.4 mT,
ꢁ
A = 17.1 mT (dotted line in Fig. 2b). These parameters
ꢁ
are typical of the parallel and perpendicular features of
V4+ in an axial ligand field [54,58]. The ESR spectra of
[VIVO(H2O)5]H[PMo12O40] and (NH4)5[PVIVMo11O40]
at 100 K are shown in Figs. 2c and d, respectively. The
spectra were well reproduced with the following parame-
ters: [VIVO(H2O)5]H[PMo12O40] (g = 1.982, g = 1.938,
found that the parameters of the hyperfine structure of
H4PVMo11O40 reduced with 1,3-dimethyladamantane were
not similar to those of (NH4)5[PVIVMo11O40], but to
those of [VIVO(H2O)5]H[PMo12O40]. The ESR spectrum
of H5PV2Mo10O40 reduced with 1,3-dimethyladamantane
(Fig. 2a) was similar to that of [VIVO(H2O)5]H[PMo12O40].
31P and 51V NMR measurements were carried out with
the same sample (H4PVMo11O40 + 1,3-dimethyladaman-
tane + butyronitrile) as that used for the ESR measurements.
The fresh H4PVMo11O40 in butyronitrile showed a 31P
NMR signal at −2.06 ppm (Fig. 3a). The 31P NMR spec-
trum of H4PVMo11O40 obtained after it was used for the
oxidation of 1,3-dimethyladamantane showed an intense sig-
nal at −2.59 ppm (Fig. 3b). The signal at −2.59 ppm can
probably be assigned to PMo12O403−, since the 31P NMR
⊥
ꢁ
A
⊥
= 6.7 mT, A = 17.7 mT, dotted line in Fig. 2c);
ꢁ
(NH4)5[PVIVMo11O40] (g = 1.977, g = 1.939, A
=
⊥
ꢁ
⊥
5.3 mT, A = 14.8 mT, dotted line in Fig. 2d).2 It was
ꢁ
1
The oxidation of adamantane was carried out with VO(acac) under
2
the conditions in Table 2. The yields of 1-adamantanol, 1,3-adamantanediol,
2-adamantanol, and 2-adamantanone were 23, < 1, < 1, and 6%, respec-
tively, and the total yield was lower than those with vanadium-substituted
phosphomolybdates. The butyronitrile solution of VO(acac) was com-
2
51
51
pletely V NMR silent. The V NMR spectrum of VO(acac) after the
2
oxidation of 1,3-dimethyladamantane under the conditions in Fig. 2 showed
V
+
a signal at −360 ppm which may be assigned to V O species in the pres-
2
IV
ence of the acetyl acetonato ligand.
was confirmed for [V O(H O) ]H[PMo
O ] that the parameters of the
12 40
2
5
2
IV
The ESR spectrum of (NH ) [PV Mo
O ] was measured in ace-
11 40
hyperfine structure of vanadium in acetonitrile was the same as those in bu-
tyronitrile.
4 5
IV
tonitrile because (NH ) [PV Mo
O ] was insoluble in butyronitrile. It
4 5
11 40

86
S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
51
V
+
and V O species in buty-
2
31
Fig. 4. V NMR spectra of H PVMo
O
11 40
Fig. 3.
P NMR spectra of H PVMo
O
11 40
in butyronitrile before
4
4
ronitrile. (a) As-synthesized H PVMo
idation of 1,3-dimethyladamantane. (c) V O
O
(0.67 mM). (b) After the ox-
and after the oxidation of 1,3-dimethyladamantane. (a) As-synthesized
H PVMo (0.67 mM). (b) After the oxidation of 1,3-dimethy-
4
11 40
V
+
species (0.67 mM). Reac-
O
11 40
2
4
tion conditions: H PVMo
O
11 40
(2 µmol, 0.67 mM), 1,3-dimethyladaman-
ladamantane. Reaction conditions: H PVMo
O
11 40
(2 µmol, 0.67 mM),
4
4
tane (1 mmol), butyronitrile (3 ml), 356 K, 24 h, under 1 atm of molecular
oxygen.
1,3-dimethyladamantane (1 mmol), butyronitrile (3 ml), 356 K, 24 h, un-
der 1 atm of molecular oxygen.
spectra of H3PMo12O40 and [VIVO(H2O)5]H[PMo12O40]
in butyronitrile showed signals at −2.54 and −2.50 ppm
(see Fig. S1 in the supplementary material), respectively,
and the positions were close to −2.59 ppm. The 31P NMR
spectrum of H4PVMo11O40 obtained after it was used for
adamantane oxidation was measured by the removal of in-
soluble adamantane with decantation after the reaction; it
also showed an intense signal at −2.63 ppm (see Fig. S2
in the supplementary material). The 51V NMR spectrum
of fresh H4PVMo11O40 in butyronitrile showed a sig-
nal at −560 ppm (Fig. 4a). The 51V NMR spectrum of
H4PVMo11O40 obtained after it was used for the oxidation
of 1,3-dimethyladamantane showed a signal at −542 ppm
V)), 1061 (s, νas(P–Oa)), 960 (s, νas(Mo–Od)), 866 (s,
νas(Mo–Ob–M)) (M = Mo, V), and 786 cm−1 (s, νas(Mo–
Oc–M)). After the adamantane oxidation, the band position
and intensities of the recovered catalyst were almost un-
changed, except that the shoulder νas(P–Oa–V) band inten-
sity was much decreased.
These ESR, NMR, and IR data show the elimination of
the vanadium from the Keggin anion framework to form
+
3−
monomeric VVO2 species and PMo12O40 Keggin an-
ions during the oxidation, which are likely catalytically ac-
tive for the present adamantane oxidation.
After H4PVMo11O40 was reduced with 1,3-dimethylada-
mantane in butyronitrile, 1 atm of molecular oxygen was in-
troduced to the solution, resulting in a quick change (within
10 min) of the color from yellow-green to yellow and the
near disappearance of ESR signals (Fig. 2e). These facts in-
dicate that the reduced catalyst is easily reoxidized by mole-
cular oxygen.
+
(Fig. 4b). The signal at −542 ppm was assigned to VVO2
species, since the chemical shift was almost the same as
+
that of VVO2 species in butyronitrile (−541 ppm, Fig. 4c)
prepared with methods described in the literature [59]. The
51V NMR spectrum of H4PVMo11O40 obtained after it was
used for adamantane oxidation was measured in the same
way as that for the 31P NMR spectrum measurement and
showed a signal at −541 ppm (see Fig. S3 in the supple-
mentary material).
3.3. Reaction mechanism
As mentioned, the oxidation of adamantane proceeded
with an induction period (10–11 h). When a free radical
inhibitor of 2,6-di-tert-butyl-p-cresol was added to the re-
The IR spectrum of fresh H4PVMo11O40 showed bands
characteristic of the Keggin structure: 1078 (sh, νas(P–Oa–

S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
87
Fig. 5. Dependence of reaction rate (R ) on the partial pressure of mole-
0
Fig. 6. Dependence of reaction rate (R ) on concentration of
0
cular oxygen (p ). Reaction conditions: H PV Mo
O
10 40
(2 µmol),
O
5
2
2
H PV Mo
O
. Reaction conditions: H PV Mo O
(∼ 3.33 mM),
5
2
10 40
5
2
10 40
adamantane (1 mmol), butyronitrile (3 ml), 356 K. p = 0.5–1.0 atm.
O
values were determined from the reaction profiles (co2nversion vs. time
adamantane (1 mmol), butyronitrile (3 ml), 356 K, 96 h under 1 atm of
R
0
molecular oxygen. R values were determined from the reaction profiles
0
curves) at low conversion (< 10%) of adamantane.
(conversion vs. time curves) at low conversion (< 10%) of adamantane.
action solution, the oxidation of adamantane was imme-
diately stopped. When a radical initiator (α, αꢂ-azobisiso-
butyronitrile (AIBN) or tert-butylhydroperoxide (TBHP))
was added to the reaction solution in the presence of
H5PV2Mo10O40, the induction period almost disappeared
(see Fig. S5 in the supplementary material). The reaction
rate obtained by the slope of the linear line after the induc-
tion period was independent of the pressure of the molecular
with molecular oxygen to form ROO· [Eq. (3)], which in
turn reacts either with adamantane [Eq. (4)] or with a sec-
ond ROO· species [Eq. (5)].4 The observed induction period
shows that the reaction requires the accumulation of reactive
species. A decrease in the induction period with the addi-
tion of AIBN or TBHP suggests that R· and hydroperoxide
species (ROOH) are the reactive species in the present oxi-
dation. Thus, the induction period would be needed to build
up sufficient chain carriers such as R· and ROO· by the re-
action among catalyst, adamantane, and molecular oxygen
according to Eqs. (1)–(5). The initial formation of R· is a
key step in the present oxidation and may take place mainly
on the vanadium
oxygen, as shown in Fig. 5 (pO ꢀ 0.5 atm). The reac-
2
tion rate increased with the increase in concentration of
H5PV2Mo10O40, reached the maximum at 0.67 mM, and
then decreased (Fig. 6) [60]. As shown in Table 2, the C3–
H/C2–H values were in the range of 10.6–14.0, which is
characteristic of the radical-mediated oxidations [61]. The
catalystox + RH → catalystred + H+ + R·,
(1)
(2)
(3)
(4)
(5)
independence of the rate on pO , the bell-shaped depen-
2
dence of the rate on the concentration of catalyst, and the
C3–H/C2–H values strongly suggest that the present ox-
idation proceeds via a free radical autooxidation mecha-
nism. First, the oxidation is initiated via the one-electron
transfer from adamantane to the catalyst to give the re-
duced form and adamantyl radical species (R·) according
1
1
catalystred + H+ + O2 → catalystox + H2O,
4
2
R·+ O2 → ROO·,
ROO·+ RH → R·+ ROOH,
ROO·+ ROO·→ ROOOOR,
to Eq. (1) (catalystox = V5+ + PMo12O403−, catalystred
=
+
species such as VVO2
,
and phosphomolybdate,
V
4+ + PMo12O403− or V5+ + PMo12O404−).3 The reduced
PMo12O40n−, enhances the activity. Another important role
of the catalyst is likely the decomposition of ROOH species
into free radical species.
species then reacts with molecular oxygen to regenerate the
oxidized form according to Eq. (2). The radical R· reacts
3
When H PVMo
O
11 40
(1.67 mM) in acetic acid was reduced by 200
4
4
equivalents of 1,3-dimethyladamantane at 356 K for 4 h under 1 atm of
argon, the ESR spectrum of the solution at 4 K showed a signal at g =
2.005 in addition to the hyperfine structure of vanadium (see Fig. S4 in the
supplementary material). The signal may be assigned to an alkyl radical
species.
The reaction of two ROO· species [Eq. (5)] produces very unstable di-
alkyltetraoxide, ROOOOR, which quickly decomposes into molecular oxy-
gen and two RO· species. Then the RO· species can react with adamantane
to form the corresponding alcohol and R· species. In a side recombination
reaction, dialkylperoxide ROOR, can be also formed.

88
S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
4. Conclusion
[16] P. Battioni, J.F. Bartoli, P. Leduc, M. Fontecave, D. Mansuy, J. Chem.
Soc., Chem. Commun. (1987) 791.
[17] D.H.R. Barton, R.S. Hay-Motherwell, W.B. Motherwell, Tetrahedron
Lett. 24 (1983) 1979.
[18] N. Komiya, T. Naota, Y. Oda, S.-I. Murahashi, J. Mol. Catal. A: Chem.
117 (1997) 21.
[19] S.-I. Murahashi, T. Naota, N. Komiya, Tetrahedron Lett. (1995) 8059.
[20] R. Giannandrea, P. Mastrorilli, C.F. Nobile, G.P. Suranna, J. Mol.
Catal. 94 (1994) 27.
[21] S.-I. Murahashi, Y. Oda, T. Naota, J. Am. Chem. Soc. 114 (1992) 7913.
[22] K. Yamaguchi, N. Mizuno, New. J. Chem. 26 (2002) 972.
[23] R. Neumann, M. Dahan, J. Am. Chem. Soc. 120 (1998) 11969.
[24] N. Mizuno, M. Tateishi, T. Hirose, M. Iwamoto, Chem. Lett. (1993)
2137.
We could develop an adamantane oxidation system with
vanadium-substituted polyoxometalates and 1 atm of mole-
cular oxygen without any additives such as reductants and
radical initiators. Spectroscopic data show that elimination
of the vanadium from the framework of phosphomolybdate
Keggin anions occurs to form free vanadium species and
3−
PMo12O40 Keggin anions during the reaction. The reac-
tion would be promoted mainly by the vanadium species,
n−
and PMo12O40 enhances the activity. The catalyst gener-
ates adamantyl radicals by the abstraction of one electron
from adamantane.
[25] M.M.T. Khan, D. Chatterjee, S. Kumar, A.P. Rao, N.H. Khan, J. Mol.
Catal. 75 (1992) L49.
[26] I.V. Kozhevnikov, Catalysis by Polyoxometalates, Wiley, Chichester,
England, 2000.
Acknowledgments
[27] R. Neumann, Prog. Inorg. Chem. 47 (1998) 317.
[28] N. Mizuno, M. Misono, Chem. Rev. 98 (1998) 199.
[29] T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 41 (1996) 113.
[30] C.L. Hill, C.M. Prosser-McCartha, Coord. Chem. Rev. 143 (1995) 407.
[31] M.T. Pope, A. Müller, Angew. Chem. Int. Ed. Engl. 30 (1991) 34.
[32] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi, N.
Mizuno, Science 300 (2003) 964.
[33] W. Adam, P.L. Alsters, R. Neumann, C.R. Saha-Möller, D. Sloboda-
Rozner, R. Zhang, J. Org. Chem. 68 (2003) 1721.
[34] D. Sloboda-Rozner, P.L. Alsters, R. Neumann, J. Am. Chem. Soc. 125
(2003) 5280.
This work was supported in part by the Core Research
for Evolutional Science and Technology (CREST) program
of the Japan Science and Technology Agency (JST) and
a Grant-in-Aid for Scientific Research from the Ministry
of Education, Culture, Sports, Science, and Technology of
Japan.
Supplementary material
[35] N.M. Okun, T.M. Anderson, C.L. Hill, J. Am. Chem. Soc. 125 (2003)
3194.
The online version of this article contains additional sup-
plementary material.
[36] H. Weiner, A. Trovarelli, R.G. Finke, J. Mol. Catal. A 191 (2003) 253.
[37] A.M. Khenkin, R. Neumann, Adv. Synth. Catal. 344 (2002) 1017.
[38] S. Ellis, I.V. Kozhevnikov, J. Mol. Catal. A 187 (2002) 227.
[39] R. Ben-Daniel, L. Weiner, R. Neumann, J. Am. Chem. Soc. 124 (2002)
8788.
Please visit DOI:10.1016/j.jcat.2005.04.012.
[40] H. Tsuji, Y. Koyasu, J. Am. Chem. Soc. 124 (2002) 5608.
[41] J.T. Rhule, W.A. Neiwert, K.I. Hardcastle, B.T. Do, C.L. Hill, J. Am.
Chem. Soc. 123 (2001) 12101.
References
[1] R.A. Sheldon, J.K. Kochi, Metal Catalyzed Oxidation of Organic
Compounds, Academic Press, New York, 1981.
[2] C.L. Hill, in: A.L. Baumstark (Ed.), Advances in Oxygenated
Processes, vol. 3, JAI Press, London, 1998, p. 1.
[3] M. Hudlucky, Oxidations in Organic Chemistry, ACS Monograph Se-
ries, American Chemical Society, Washington, DC, 1990.
[4] X. Wan, M. Duncan, P. Nass, J.W. Harmon, Anticancer Res. 21 (2001)
2657.
[42] I.A. Weinstock, E.M.G. Barbuzzl, M.W. Wemple, J.J. Cowan, R.S.
Reiner, D.M. Sonnen, R.A. Heintz, J.S. Bond, C.L. Hill, Nature 414
(2001) 191.
[43] Y. Nishiyama, Y. Nakagawa, N. Mizuno, Angew. Chem. Int. Ed. 40
(2001) 3639.
[44] T. Hayashi, A. Kishida, N. Mizuno, Chem. Commun. (2000) 381.
[45] T. Yokota, M. Tani, S. Sakaguchi, Y. Ishii, J. Am. Chem. Soc. 125
(2003) 1476.
[5] N. Shida, T. Ushiroguchi, K. Asakawa, T. Okino, S. Saito, Y. Funaki,
A. Takaragi, K. Tsutsumi, K. Inoue, T. Nakao, J. Photopolym. Sci.
Technol. 13 (2000) 601.
[6] N. Matsuzawa, S. Takechi, T. Ohfuji, K. Kuhara, S. Mori, M. Endo,
K. Kamon, T. Morisawa, A. Yamaguchi, M. Sasago, Jpn. J. Appl.
Phys. 37 (1998) 5781.
[46] M. Vennat, P. Herson, J.-M. Brégeault, G.B. Shul’pin, Eur. J. Inorg.
Chem. (2003) 908.
[47] R. Ben-Daniel, R. Neumann, Angew. Chem. Int. Ed. 42 (2003) 92.
[48] T. Yokota, S. Sakaguchi, Y. Ishii, Adv. Syhth. Catal. 344 (2002) 849.
[49] A.M. Khenkin, L. Weiner, Y. Wang, R. Neumann, J. Am. Chem. Soc.
123 (2001) 8531.
[7] H. Stetter, Angew. Chem. 66 (1954) 217.
[50] C.P. Stewart, A.L. Porte, J. Chem. Soc., Dalton Trans. (1972) 1661.
[51] D.D. Perrin, in: W.L.F. Armarego (Ed.), Purification of Laboratory
Chemicals, third ed., Pergamon Press, Oxford, UK, 1988.
[52] G.A. Tsigdinos, C.J. Hallada, Inorg. Chem. 7 (1968) 437.
[53] R. Bayer, C. Marchal, F.X. Liu, A. Tézé, G. Hervé, J. Mol. Catal. A:
Chem. 110 (1996) 65.
[8] H. Stetter, M. Schwarz, A. Hirschhorn, Chem. Ber. 92 (1959) 1629.
[9] H. Stetter, E. Rauscher, Chem. Ber. 93 (1960) 1161.
[10] G.W. Smith, H.D. Williams, J. Org. Chem. 7 (1961) 2207.
[11] Y. Ishii, J. Mol. Catal. A 117 (1997) 123.
[12] Y. Ishii, T. Iwasawa, S. Sakaguchi, K. Nakayama, Y. Nishiyama,
J. Org. Chem. 61 (1996) 4520.
[54] C. Marchal-Roch, N. Laronze, N. Guillou, A. Tézé, G. Hervé, Appl.
Catal. A 199 (2000) 33.
[55] R. Neumann, M. Levin, J. Am. Chem. Soc. 114 (1992) 7278.
[56] L. Pettersson, I. Andersson, A. Selling, J.H. Grate, Inorg. Chem. 33
(1994) 982.
[13] T. Funabiki, H. Ishida, S. Yoshida, Chem. Lett. (1991) 1819.
[14] I. Yamanaka, K. Nakazaki, T. Akimoto, K. Otsuka, J. Chem. Soc.,
Perkin Trans. 2 (1996) 2511.
[15] N. Kitajima, M. Ito, H. Fukui, Y. Moro-oka, J. Chem. Soc., Chem.
Commun. (1991) 102.

S. Shinachi et al. / Journal of Catalysis 233 (2005) 81–89
89
[57] N. Mizuno, H. Yahiro, J. Phys. Chem. B 102 (1998) 437.
[58] J.J. Altenau, M.T. Pope, R.A. Prados, H. So, Inorg. Chem. 14 (1975)
417.
catalyst might act as an inhibitor at higher catalyst concentration. See
(a) A.M. Khenkin, A. Rosenberger, R. Neumann, J. Catal. 182 (1999)
82;
[59] C. Slebodnick, V.L. Pecoraro, Inorg. Chim. Acta 283 (1998) 37.
[60] A similar decrease in the rate with the increase in the catalyst
concentration was reported for the alkane oxidation system with
(b) O.A. Kholdeeva, V.A. Grigoreiv, G.M. Maksimov, M.A. Fedotov,
A.V. Golovin, K.I. Zamaraev, J. Mol. Catal. 114 (1996) 123.
[61] M. Bonchio, G. Scorrano, P. Toniolo, A. Proust, V. Artero, V. Conte,
Adv. Synth. Catal. 344 (2002) 841.
O /aldehydes involving radical species, and it is suggested that the
2