Oxidative Cleavage of Vicinal Diols with the Combination of Platinum and Vanadium Catalysts and Molecular Oxygen

Article abstract of DOI:10.1002/cctc.201600153

The combination of Pt/C and V₂O₅ catalysts gave good performance for the oxidative cleavage of trans-1,2-cyclohexanediol into adipic acid via 2-hydroxycyclohexanone. The yield of adipic acid reached 90 % in the one-pot oxidative cleavage of trans-1,2-cyclohexanediol. The yield was higher than that obtained in the oxidation of 2-hydroxycyclohexanone with V catalyst, and the higher yield was due to the low 2-hydroxycyclohexanone concentration during the one-pot oxidation of trans-1,2-cyclohexanediol. Cyclic vicinal diols having a six-membered ring and linear vicinal diols having two secondary OH groups were converted into dicarboxylic acids and two carboxylic acids, respectively. The activity of the Pt catalyst decreased during the reaction, and the activity was partially restored by treating the used catalyst with H₂ at 573K. Even without regeneration, the turnover number based on total Pt could reach ≈1000 in a long reaction with an increased amount of trans-1,2-cyclohexanediol.

Full text of DOI:10.1002/cctc.201600153

DOI: 10.1002/cctc.201600153
Full Papers
Oxidative Cleavage of Vicinal Diols with the Combination
of Platinum and Vanadium Catalysts and Molecular
Oxygen
Naoyuki Obara, Shota Hirasawa, Masazumi Tamura, Yoshinao Nakagawa,* and
[a]
Keiichi Tomishige*
The combination of Pt/C and V O catalysts gave good per-
ring and linear vicinal diols having two secondary OH groups
were converted into dicarboxylic acids and two carboxylic
acids, respectively. The activity of the Pt catalyst decreased
during the reaction, and the activity was partially restored by
2
5
formance for the oxidative cleavage of trans-1,2-cyclohexane-
diol into adipic acid via 2-hydroxycyclohexanone. The yield of
adipic acid reached 90% in the one-pot oxidative cleavage of
trans-1,2-cyclohexanediol. The yield was higher than that ob-
tained in the oxidation of 2-hydroxycyclohexanone with V cat-
alyst, and the higher yield was due to the low 2-hydroxycyclo-
hexanone concentration during the one-pot oxidation of trans-
treating the used catalyst with H at 573 K. Even without re-
2
generation, the turnover number based on total Pt could
reach ꢁ1000 in a long reaction with an increased amount of
trans-1,2-cyclohexanediol.
1
,2-cyclohexanediol. Cyclic vicinal diols having a six-membered
Introduction
[
3]
Oxidative cleavage of vicinal diols selectively dissociates one
CÀC bond in the substrate. In particular, oxidative cleavage of
cyclic vicinal diols to the corresponding dicarboxylic acids pro-
ceeds without decrease in the length of the carbon chain
atom-economical chemistry. A typical substrate is trans-1,2-
cyclohexanediol (trans-1,2-CHD), and the target product is
adipic acid, which is industrially one of the most important di-
[
4]
carboxylic acids. Systems using molecular oxygen as an oxi-
dant for oxidation of 1,2-CHD to adipic acid are limited (see
(
Scheme 1).
[
5–10]
the Supporting Information, Table S1).
ruthenium pyrochlore oxides, A2+xRu
Ru catalysts (e.g.,
(A=Pb or Bi;
O
2Àx 7Ày
[
5,6]
0
<x<1; 0<yꢀ0.5)) were reported to be effective;
howev-
er, bases such as NaOH and KOH were necessary for high activ-
ity and the product was adipate instead of adipic acid. In addi-
tion, the turnover number (TON) of adipate formation was not
very high (ꢀ44; based on total Ru). To the best of our knowl-
edge, high yields of free adipic acid in the oxidation of 1,2-
CHD with O2 have not been reported. On the other hand,
good (ꢀ90%) yields of adipate or adipic acid have been re-
ported for oxidation of 2-hydroxycyclohexanone (2-HCO) with
Scheme 1. Oxidative cleavage of cyclic vicinal diols to the corresponding di-
carboxylic acid.
Oxidative cleavage of the CÀC bonds of diols is also one of
[
1]
the methods to convert large compounds to smaller ones.
This reaction can be applied to biomass conversion, because
biomass generally has many oxygen atoms, especially as hy-
[
11]
P/Mo/V polyoxometalate catalysts and molecular oxygen. It
was hypothesized that the reaction begins with the formation
of a catalytically active oxovanadium(V) species, and the oxida-
[
2]
droxy groups. This reaction is traditionally performed with
stoichiometric oxidants, such as lead tetraacetate, periodic acid
[
1]
V
IV
[11–13]
and its salts. However, these oxidants are expensive and their
reactions result in the formation of toxic inorganic waste com-
pounds. The use of a catalyst and a clean oxidant such as mo-
lecular oxygen is desired from the viewpoint of green and
tion reaction is catalyzed by the V /V redox system.
Some
vanadium compounds are also known as effective catalysts for
[
14–20]
oxidative cleavage of lignin model compounds.
In addi-
tion, it is well known that noble metal catalysts are active in al-
[
21–25]
cohol oxidation with molecular oxygen,
and 2-HCO is the
oxidation product of 1,2-CHD at one of the two OH groups.
Various noble metal catalysts including conventional carbon-
supported ones have been tested for alcohol oxidation. For ex-
[
a] N. Obara, S. Hirasawa, Dr. M. Tamura, Dr. Y. Nakagawa,
Prof. Dr. K. Tomishige
Department of Applied Chemistry, School of Engineering
Tohoku University
ample, propylene glycol oxidation to hydroxyacetone proceeds
6
-6-07 Aoba, Aramaki, Aoba-ku, Sendai, 980-8579 (Japan)
[26,27]
over conventional Pt/C and Pd/C catalysts.
Although typi-
E-mail: yoshinao@erec.che.tohoku.ac.jp
tomi@erec.che.tohoku.ac.jp
[
26,27]
cal systems including these
such as NaOH or basic supports,
also use homogeneous bases
[
5,6,28–31]
there are some reports
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cctc.201600153.
showing that Pt catalysts have higher activity in the oxidative
ChemCatChem 2016, 8, 1732 – 1738
1732
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
or oxidant-free dehydrogenation of alcohol than other noble
metal catalysts and V O gave much lower conversions of 1,2-
2
5
[32,33]
metal catalysts in base-free conditions.
In this report, we
CHD than in the case of Pt/C (entries 9–11). The low activity of
Pd/C and Ru/C catalysts was in contrast to the reported high
performance of Pd- and Ru-based catalysts in basic condi-
conducted the one-pot oxidation of vicinal diols including 1,2-
CHD to the corresponding carboxylic acids with the combina-
tion of Pt and V catalysts. Very high yields (83–92%) of acids
and high TON values were obtained in the oxidation of 1,2-
CHD and linear diols.
[
5,6,28–31]
tions.
The neutral or acidic conditions in this system can
be related to the higher performance of Pt/C than other noble
[
32,33]
metal catalysts.
We selected the combination of Pt/C+
V O because of the simplicity and high performance.
2
5
When 2-HCO was used as the substrate, adipic acid was pro-
Results and Discussion
duced with the V O catalyst (entry 12), similarly to previous
2
5
[
11–13]
First of all, various combinations of catalysts were tested for
studies.
Over Pt/C alone, both the conversion of 2-HCO
the oxidation of aqueous trans-1,2-CHD in a 190 mL stainless-
and selectivity to adipic acid were very low (entry 13). These
data suggest that the conversion of trans-1,2-CHD to adipic
acid proceeded by Pt-catalyzed oxidative dehydrogenation of
trans-1,2-CHD to 2-HCO and V-catalyzed oxidative cleavage of
2-HCO to adipic acid.
steel autoclave with 0.3 MPa O at 353 K (Table 1). The metal
2
dispersion of catalysts is shown in the Supporting Information,
Table S2. The HPLC charts of the standard materials (Fig-
ure S1a) and typical reaction mixture (Figure S1b) are also
shown in the Supporting Information. The combination of Pt/
Although the conversion of 2-HCO with the V O catalyst
2
5
C+V O5 showed high selectivity for adipic acid (Table 1,
(entry 12) was ꢁ2 times higher than that of trans-1,2-CHD over
Pt/C (entry 1), the rate ratio cannot explain the result that 2-
HCO was hardly observed in the trans-1,2-CHD oxidation with
Pt/C+V O catalysts (<1% selectivity; entry 1). In addition, it
2
entry 1). We found that trans-1,2-CHD was converted to 2-HCO
over Pt/C alone, whereas trans-1,2-CHD was not converted
over V O alone (entries 2 and 3). Although some systems with
2
5
2
5
V catalysts have been reported to be active in alcohol oxida-
should be noted that the adipic acid selectivity of 2-HCO oxi-
dation (entry 12) was lower than that of trans-1,2-CHD oxida-
tion with Pt/C+V O (entry 1). More than twenty kinds of by-
[
34–40]
[34–37]
tion,
water is generally unfavorable as the solvent.
All
the Pt/C+vanadium compound systems showed high selectiv-
ity (81–92%) for adipic acid (entries 1 and 4–7), although the
activity of Pt, which was reflected in the trans-1,2-CHD conver-
sion, was more or less decreased by addition of vanadium
2
5
products were formed in the 2-HCO oxidation (entry 12). Ac-
cording to the GC-MS spectra, most byproducts were oxygen-
ates with C4 or C5 carbon chains, including esters. Then, we
tested the effect of co-existing activated carbon in the oxida-
tion of 2-HCO with a V catalyst. Addition of activated carbon
to the V O solution increased the activity in the oxidation of
compounds. The Pt/C+NaVO system (entry 7) showed much
3
lower activity than Pt/C+V O ; however, when the pH value of
2
5
the reaction mixture was adjusted by addition of H SO4 to
2
2
5
a similar value to that of the Pt/C+V O system (pH 6–7), the
2-HCO (entry 14). The high activity of V O +C agrees with the
2 5
2
5
system gave a similar performance to that of the Pt/C+V O5
absence of 2-HCO in the trans-1,2-CHD oxidation in the Pt/C+
V O system; that is, Pt/C-catalyzed oxidation of trans-1,2-CHD
2
system (entries 1 and 8). The combinations of other noble
2
5
Table 1. Oxidation of trans-1,2-CHD, 2-HCO, or adipic acid over various catalysts with molecular oxygen.
Entry
Substrate
Catalyst
Conversion
%]
Selectivity [%]
[
Adipic acid
2-HCO
Glutaric acid
Succinic acid
CO
x
Others
1
2
3
4
5
6
7
8
9
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
trans-1,2-CHD
2-HCO
Pt/C+V
Pt/C
2
O
5
40
92
6
–
<1
84
–
<1
<1
4
2
1
2
<1
–
–
–
–
–
–
–
–
2
3
–
3
5
2
3
2
1
<1
–
3
17
3
2
1
–
–
<1
2
–
<1
2
<1
1
<1
<1
<1
–
2
4
1
2
–
4
2
3
4
2
1
<1
–
6
14
6
5
5
–
–
4
3
–
5
8
6
3
2
6
53
<1
22
25
8
27
38
10
1
<1
82
17
98
>99
>99
<1
2
V O
5
Pt/C+VCl
Pt/C+VOSO
Pt/C+NH VO
Pt/C+NaVO
Pt/C+NaVO
Pd/C+V
Ru/C+V
Rh/C+V
3
88
82
86
86
92
90
ꢁ70
–
63
2
62
82
84
–
4
·6H
2
O
4
3
3
[
a]
3
+H
2
SO
4
2
2
2
O
O
O
5
5
5
[
b]
1
0
ꢁ30
–
25
63
27
8
1
1
1
1
1
1
1
1
1
2
3
4
5
6
7
8
2
V O
5
2-HCO
Pt/C
[
c]
2-HCO
V
V
V
2
O
2
O
2
O
5
5
5
+C
+C
+C
2
2
2
–
[
d]
e]
[c]
[c]
2-HCO
2-HCO
2-HCO
[
7
–
–
none
Pt/C+V
adipic acid
2
O
5
<1
–
–
Reaction conditions: Substrate, 4.3 mmol; 5 wt% metal/C, 26 mmol or 0 mmol metal; vanadium compound, 110 mmol or 0 mmol V; water, 10 g; O
53 K; 4 h. CHD=Cyclohexanediol. HCO=Hydroxycyclohexanone. [a] H SO
ones as the low conversion precludes the obtained data from being comparable with other entries. [c]Activated carbon (Shirasagi FAC-10), 0.1 g.
d] 2-HCO, 2.2 mmol. [e] 2-HCO, 1.1 mmol.
2
, 0.3 MPa;
3
2
4
, 5 mmol; pH before the reaction, 6.8. [b] Selectivities in the entry are nominal
[
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1733
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
to 2-HCO is much slower than the V O +C catalyzed oxidation
2
5
of 2-HCO to adipic acid. However, the selectivity to adipic acid
in entry 14 was still lower than the trans-1,2-CHD oxidation
with Pt/C+V O . Then, we focused on the concentration of 2-
2
5
HCO as it was kept extremely low in the one-pot reaction. In
fact, the selectivity to adipic acid increased with decreasing 2-
HCO concentration in the V O +C system (entries 14–16). We
2
5
also conducted the reaction of 2-HCO over V O +C at short
2
5
reaction times (see the Supporting Information, Table S3). The
reaction can proceed without significant decrease in the rate
under the conditions of lower 2-HCO concentration. The rate
ratio of the Pt-catalyzed oxidation of trans-1,2-CHD (see the
Supporting Information, Table S3, entry 5) to (V O +C)-cata-
2
5
lyzed oxidation of 2-HCO was estimated to be around 4. There-
fore, the key to high adipic acid selectivity is the low concen-
tration of the 2-HCO intermediate during the one-pot conver-
sion of trans-1,2-CHD to adipic acid because of the high reac-
tion rate of 2-HCO oxidation.
Figure 1. Time course of the trans-1,2-cyclohexanediol oxidation over Pt/
C+V
none (~), glutaric acid+succinic acid (
tion conditions: Substrate, 0.5 g (4.3 mmol); 5 wt% Pt/C, 0.1 g (26 mmol Pt);
, 0.01 g (110 mmol V); water, 10 g; O , 0.3 MPa; 353 K. Numerical data
2
O
5
. Conversion (*), selectivity to adipic acid (^), 2-hydroxycyclohexa-
&
), CO+CO
(*), and others (^). Reac-
2
The time course of the trans-1,2-CHD oxidation over Pt/C+
V O is shown in Figure 1. The numerical data are shown in the
V
2
O
5
2
2
5
are shown in Table S4 in the Supporting Information.
Supporting Information, Table S4. The reaction proceeded to
some extent during the initial heating to the reaction tempera-
ture. The conversion reached higher than 99%, and 90% yield
of adipic acid was obtained at 48 h. The selectivity to adipic
acid remained at about 90% after 4 h. 2-HCO was detected at
short reaction times, and it almost disappeared after 4 h. This
behavior agreed with the idea that 2-HCO is the intermediate
of adipic acid formation. The selectivities to glutaric acid and
succinic acid as well as adipic acid hardly changed, indicating
that the acids are stable under the reaction conditions. The
low reactivity of adipic acid under the reaction conditions was
also confirmed (Table 1, entry 18).
dary OH groups, were converted to acetic acid, propionic acid,
and benzoic acid, respectively, in good yields (ꢂ84%; en-
tries 4–6). Propylene glycol, which has one primary OH group
and one secondary OH group, was converted to acetic acid
and CO (entry 7). The carbon-based yield of acetic acid (55%)
2
corresponds to 83% of the theoretical value [(carbon number
of acetic acid)/(carbon number of propylene glycol)=66.7%].
Sugar alcohols (ethylene glycol and glycerol) were almost com-
pletely converted to CO (entries 8 and 9). For the substrates
2
with primary OH group(s), formic acid was probably formed in
the initial stage, and the formic acid would be rapidly oxidized
The Pt/C+V O system was applied to various related sub-
2
5
[
41]
strates (Table 2). The data for shorter reaction times are shown
to CO2 over the Pt catalyst.
We confirmed that CO2 was
in the Supporting Information, Table S5. Both cis- and trans-
formed from formic acid under the reaction conditions
(entry 10). Five-membered-ring diols are unsuitable substrates
for this system. 1,2-Cyclopentanediols were converted to gluta-
ric acid in lower yields (ꢀ48%; entries 11 and 12). Large
1
9
,2-CHD were converted to adipic acid in good yields (83–
0%; entries 1–3). Among the linear diols, 2,3-butanediol, 3,4-
hexanediol, and meso-hydrobenzoin, which have two secon-
2 5
Table 2. Oxidation of various substrates over Pt/C+V O .
Entry
Substrate
t [h]
Conversion [%]
Product (carbon-based selectivity [%])
1
2
3
trans-1,2-CHD
cis-1,2-CHD
1,2-CHD (cis/trans-mixture,
cis/trans=4:1)
48
48
48
>99
99
>99
adipic acid (90), glutaric acid (3), succinic acid (2), CO
adipic acid (83), glutaric acid (4), succinic acid (4), CO
adipic acid (87), glutaric acid (3), succinic acid (3), CO
2
2
2
(4), others (1)
(7), others (1)
(5), others (1)
4
2,3-butanediol
3,4-hexanediol
meso-hydrobenzoin
propylene glycol
ethylene glycol
glycerol
formic acid
trans-1,2-cyclopentanediol
cis-1,2-cyclopentanediol
1,4-anhydroerythritol
24
72
48
4
4
4
>99
98
acetic acid (92), CO
propionic acid (86), acetic acid (4), CO
benzoic acid (92), benzaldehyde (3), CO
acetic acid (55), CO (42), others (2)
2
(6), others (2)
[
a]
5
6
7
8
9
1
2
(4), others (5)
>99
>99
>99
>99
>99
>99
>99
98
2
(4), others (2)
2
CO
CO
CO
2
2
2
(98), others (0.8)
(98), others (1)
(99), others (1)
0
4
1
1
48
48
8
glutaric acid (35), succinic acid (28), CO
glutaric acid (48), succinic acid (15), CO
2
(28), others (7)
1
1
2
3
2
(18), others (18)
(28), others (3)
diglycolic acid (69), glycolic acid (0.7), CO
2
Reaction conditions: Substrate, 4.3 mmol; 5 wt% Pt/C, 0.1 g (26 mmol Pt); V
2
O
5
, 0.01 g (110 mmol V); water, 10 g; O , 0.3 MPa; 353 K. [a] 333 K. CHD=Cyclo-
2
hexanediol.
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1734
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
amounts of smaller dicarboxylic acid (succinic acid) and CO2
were formed than in the cases of 1,2-CHDs. The selectivity to
glutaric acid was even lower at short reaction times (see the
Supporting Information, Table S5, entries 8 and 9). When 1,2-
cyclopentanediols were oxidized over the Pt/C catalyst alone,
more than ten kinds of byproducts with a combined selectivity
of ꢂ50% were formed in addition to the corresponding a-
ketol. The lower glutaric acid yield in Table 2, entries 11 and 12
was due to this low a-ketol selectivity in the Pt-catalyzed reac-
tion. 1,4-Anhydroerythritol (cis-3,4-tetrahydrofurandiol) was
vanadium species were not observed in the XRD pattern of Pt/
C+V O after the catalytic use without reduction and replen-
2
5
ishment with fresh V O (Figure 3d), suggesting that the vana-
2
5
dium species in the recovered solid catalyst were highly dis-
persed or amorphous. Poisoning with CO or organic molecules
can cause deactivation of the Pt catalyst. We tried to remove
the adsorbed species by treating the used catalysts with a flow
of hot H , as calcination treatments cannot be applied to this
2
carbon-supported catalyst. Davis et al. reported the regenera-
tion of Pt catalysts by reduction with H at 473 K after use in
2
[
42]
converted to diglycolic acid and CO in 69% and 28% yields,
the oxidation of alcohols. We reused the Pt/C+V O catalyst
2 5
2
respectively (Table 2, entry 13). Unlike cyclopentanediols, at
low conversion the selectivity to diglycolic acid was higher
with H treatment at 473 K before each reuse (Figure 2C). The
2
activity of the catalyst was slightly recovered by H treatment.
2
(
89% at 69% conversion; see the Supporting Information,
The XRD analysis (Figure 3e) showed that the linewidth of Pt
peaks was almost unchanged (4.8 nm to 5.4 nm). The adsorbed
poisonous species was probably not removed. To complete
Table S5, entry 10).
The reuse experiment was conducted to see the stability of
the catalyst (Figure 2). The used Pt/C with or without V O was
the regeneration, we increased the H treatment temperature
2
5
2
collected by filtration, dried in the air, and reused for another
catalytic run. When Pt/C+V O catalyst was reused without re-
2
5
generation, the catalyst showed good selectivity to adipic acid
Figure 2A), although the selectivity to 2-HCO was a little in-
(
creased. The amount of eluted metal into the reaction solution
was analyzed by inductively coupled plasma optical emission
spectrometry (ICP-OES). ICP analysis of the filtrate showed that
about half of the vanadium still remained in the solid catalyst
in each run (the percentage of dissolved V: 45%!19%!9%,
in the order of fresh, reuse 1, and reuse 2). It is indicated that
the remaining vanadium species had enough catalytic activity
to oxidize most of the 2-HCO. The leaching of dissolved Pt was
very small (the percentage of dissolved Pt: 0.3%!0.1%!
Figure 3. XRD patterns of Pt/C, Pt/C+V
action, (b) Pt/C+V before the reaction (mixed with a Pt/V=26:110 molar
ratio) (c) Pt/C after the catalytic use three times (Figure 2B), (d) Pt/C+V O5
2 5 2 5
O , and V O . (a) Pt/C before the re-
<
0.1%, in the order of fresh, reuse 1, and reuse 2), however,
2
O
5
the conversion of trans-1,2-CHD, which reflects the activity of
Pt, was decreased in each reuse. Over Pt/C alone, the conver-
sion was also decreased in each reuse (Figure 2B), which is
2
after catalytic use three times (not reduced; Figure 2A), (e) Pt/C+V
catalytic use three times (reduced at 473 K before the catalytic use; Fig-
ure 2C), (f) Pt/C+V after catalytic use three times (reduced at 573 K
before the catalytic use; Figure 2D), (g) V O before the reaction. d was
2 5
O after
2
O
5
similar to the case of Pt/C+V O , suggesting that Pt/C was de-
2
5
2
5
XRD
activated. The XRD analyses (Figure 3) indicated that the Pt
particle aggregation was not a cause of deactivation. Peaks for
calculated by Scherrer’s equation and the width of the Pt (111) peak at
9.28.
3
Figure 2. Reuse experiments of Pt/C+V
C) Pt/C+V with H treatment at 473 K before each reuse. Reduction conditions: H
C+V with H treatment at 573 K before each reuse. Reduction conditions: H flow, 573 K, 1 h. Fresh (4.5 mg) V
*), adipic acid (
), 2-hydroxycyclohexanone ( ), glutaric acid+succinic acid ( ), others ( ). Reaction conditions: trans-1,2-Cyclohexanediol, 0.5 g
4.3 mmol); 5 wt% Pt/C, 0.1 g (26 mmol Pt); V , 0.01 g (110 mmol V); water, 10 g; O , 0.3 MPa; 353 K; 4 h.
2
O
5
or Pt/C. (A) Pt/C+V
2
O
5
without regeneration and replenishment with fresh V
flow, 473 K, 2 h. Fresh V (4.5 mg) was added for each reuse. (D) Pt/
was added for each reuse. Conversion
2 5
O . (B) Pt/C without regeneration.
(
2
O
5
2
2
2 5
O
2
O
5
2
2
2 5
O
(
(
&
&
2
O
5
2
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1735
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
from 473 K to 573 K (Figure 2D). The activity of the used cata-
lyst was much improved by treating at 573 K. However, the Pt
particle size was significantly increased (Figure 3 f, 4.8 nm to
the conversion was almost the same (data shown in the Sup-
porting Information, Figure S2). Considering the lower cost of
V than Pt, a higher V/Pt ratio such as 10 mg V O with 0.1 g Pt/
2
5
6
.7 nm). The increase in Pt particle size can explain the remain-
C is a better choice than 2 mg V O from the view of organic
2 5
ing activity loss. Fine optimization of the reductive regenera-
tion conditions may further improve the catalyst life.
synthesis. We carried out an activity test of the catalyst after
the run for 168 h in Figure 4. The result showed that the cata-
lyst after the run was almost totally deactivated (see the Sup-
porting Information, Table S7, entry 2), indicating that the recy-
cling procedure was not the cause of deactivation. The deacti-
vation was significantly suppressed by treatment with flowing
We tested a long reaction with a larger substrate amount to
evaluate the limit of the catalyst life without regeneration. In
this experiment, higher O pressure (0.3!2 MPa) and a higher
2
reaction temperature (353!363 K) were applied to increase
the reaction rate. The vanadium amount was decreased to try
to increase the V-based turnover number. The results are
shown in Figure 4 and the numerical data are shown in the
H at 573 K (see the Supporting Information, Table S7, entry 3),
2
as in the case of the reuse experiment of Pt/C+V O under
2
5
standard reaction conditions (Figure 2). These data agreed
with the idea that the poisoned Pt surface was regenerated by
the reductive heat treatment.
Conclusions
Pt/C combined with V O can selectively convert vicinal diols
2
5
into the corresponding carboxylic acids with molecular
oxygen. The yield of adipic acid from trans-1,2-cyclohexanediol
reached 90% over Pt/C+V O . This value was higher than the
2
5
yield obtained in the a-ketol (2-hydroxycyclohexanone) oxida-
tion conducted separately. The turnover number for adipic
acid formation based on total Pt can reach ꢁ1000. The Pt cata-
lyst was gradually deactivated, and the activity was partially re-
stored by treating the used Pt catalyst with H at 573 K. The V-
2
catalyzed a-ketol oxidation gave higher selectivity to carboxyl-
ic acids when the concentration of a-ketol was lower. The
combination with Pt-catalyzed diol oxidation can keep the
concentration of a-ketol low enough to obtain good selectivity
to carboxylic acids. Linear vicinal diols having two secondary
OH groups were converted into two carboxylic acids in good
Figure 4. Time course of the trans-1,2-cyclohexanediol (trans-1,2-CHD) oxida-
tion over Pt/C+V
2
O
5
in the case of a large amount of substrate. Conversion
(
*), selectivity to adipic acid (^), 2-hydroxycyclohexanone (~
), glutaric
acid+succinic acid (
trans-1,2-CHD, 6 g (52 mmol); 5 wt% Pt/C, 0.1 g (26 mmol Pt); V
22 mmol V); water, 40 g; O , 2 MPa; 363 K.
&
), CO+CO
2
(*), and others (^). Reaction conditions:
2
O
5
, 2 mg
(
2
yields. For substrates with a primary group, CO was formed in-
2
stead of formic acid from the ÀCH OH unit.
2
Supporting Information, Table S6. The selectivity to adipic acid
was low at short reaction times, however, it increased with de-
creasing selectivity to 2-HCO during the reaction’s progress.
The high selectivity to adipic acid (ꢁ80%) was maintained
after 24 h. Although the conversion was not completed at
Experimental Section
Catalyst
Carbon-supported noble metal catalysts (5 wt% loading, metal dis-
persion is shown in the Supporting Information, Table S2) and V O
were purchased from Wako Pure Chemical Industries, Ltd. The va-
2
5
1
68 h, the turnover number (TON) in adipic acid formation
reached up to 920 (based on total Pt) and 1100 (based on total
V). These values were approximately 20 times as large as the
highest one in the literature for 1,2-CHD oxidation to adipic
acid (see the Supporting Information, Table S1). Therefore,
even without regeneration the catalyst can give much higher
TON values for 1,2-CHD oxidation to adipic acid than literature
systems. However, the selectivity to adipic acid was slightly
lower than the case of the standard reaction conditions
nadium compounds used were commercial V O , VCl , VOSO ·nH O
2
5
3
4
2
(n=6 as determined by thermogravimetric and differential thermal
analysis (TG-DTA)), NaVO , and NH VO .
3
4
3
Activity tests
The oxidation of vicinal diols was performed in a 190 mL stainless-
steel autoclave with an inserted glass vessel. The noble metal cata-
lyst, vanadium compound, substrate, and water (solvent) were put
into the autoclave together with a spinner. The pH was measured
with a pH meter when necessary. After sealing, the reactor was
filled with 0.3 MPa oxygen. The autoclave was then heated to
(
Figure 1 and Table 2). One of the reasons could be the higher
-HCO concentration caused by the small amount of solvent
and V O catalyst (Figure 4). We checked the dependence on
2
2
5
the amount of V O catalyst in the oxidation of trans-1,2-CHD
2
5
3
53 K, and the temperature was monitored by using a thermocou-
under the same conditions as those used for Figure 4. In fact,
ple inserted in the autoclave. The heating took about 20 min.
During the experiment, the stirring rate was fixed at 500 rpm (mag-
netic stirring). After an appropriate reaction time, the reactor was
increasing the V O amount (2 mg!10 mg) increased the se-
2
5
lectivity to adipic acid from 81% (at 24 h) to 85%, although
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1736
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
cooled down and the gases were collected in a gas bag. The auto-
clave contents were diluted with 2-propanol and transferred to
a vial, while the catalyst was separated by filtration. The standard
conditions for the reaction were as follows: 4.3 mmol trans-1,2-cy-
clohexanediol (substrate), 0.1 g carbon-supported Pt catalyst
Acknowledgments
Part of this work is supported by the JSPS KAKENHI “Grant-in-Aid
for Scientific Research (C)” (10436545).
(
26 mmol Pt), 0.01 g V O (110 mmol V), 10 g water (solvent),
2 5
0
4
.3 MPa initial oxygen pressure, 353 K reaction temperature, and
h reaction time. Under these conditions, the solubility limit of
Keywords: molecular oxygen · oxidative cleavage · platinum ·
vanadium · vicinal diols
adipic acid in water at 293 K corresponds to 30% yield. For the
runs with higher adipic acid yield, crystalline adipic acid was
formed, and dilution is essential for correct analysis. Details of the
reaction conditions are described for each result. Analyses were
conducted by means of FID-GC, GC-MS, and HPLC. The FID-GC
[1] T. K. M. Shing in Comprehensive Organic Synthesis Vol. 7 (Eds.: B. M.
Trost, I. Fleming), Pergamon Press, Oxford, 1993, pp. 703.
[
[
2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502.
3] P. T. Anastas, L. B. Bartlett, M. M. Kirchhoff, T. C. Williamson, Catal. Today
(
Shimadzu GC-2014) for analysis of liquids and GC-MS (Shimadzu
GCMS-QP2010 Ultra) were equipped with a CP-Sil 5 CB or HP-FFAP
capillary column. They were mainly used when the linear vicinal
diols were used as substrates. A Porapak N packed column and
a methanator were used for the separation of products in the gas
phase such as CO , CO, and CH . The HPLC (Shimadzu Prominence)
was equipped with a UV/Vis detector (wavelength 210 nm), refrac-
tive index detector, and an Aminex HPX-87H column (Bio-Rad). Di-
luted H SO (0.01m) was employed as eluent. The HPLC charts of
2
000, 55, 11–22.
4] S. Van de Vyver, Y. Romµn-Leshkov, Catal. Sci. Technol. 2013, 3, 1465–
479.
[
1
[5] a) T. R. Felthouse, P. B. Fraundorf, R. M. Friedman, C. L. Schosser, J. Catal.
1991, 127, 393–420; b) T. R. Felthouse, J. Am. Chem. Soc. 1987, 109,
7566–7568.
2
4
[
6] S. J. H. F. Arts, F. van Rantwijk, R. A. Sheldon, J. Carbohydr. Chem. 1996,
5, 317–329.
1
2
4
[7] J.-M. BrØgeault, B. ElAli, J. Mercier, J. Martin, C. Martin, C. R. Acad. Sci.
Ser. II 1989, 309, 459–462.
the standard materials (Figure S1a) and a typical reaction mixture
Figure S1b) are shown in the Supporting Information. HPLC analy-
(
[
[
8] E. Takezawa, S. Sakaguchi, Y. Ishii, Org. Lett. 1999, 1, 713–715.
9] T. Iwahama, Y. Yoshino, T. Keitoku, S. Sakaguchi, Y. Ishii, J. Org. Chem.
sis was used when cyclic vicinal diols were used as substrates. The
conversion and the selectivity were calculated based on carbon
and defined as follows: Conversion (%)={(mol of consumed sub-
strate)/(mol of substrate)}”100. Selectivity (%)=[{(mol of prod-
uct)”(number of carbon atoms in a product molecule)}/{(mol of
consumed substrate)”(number of carbon atoms in a substrate
molecule)}]”100. Turnover number (TON) was calculated by the
following equation for each metal (Pt or V): TON=(mol of adipic
acid)/(mol of metal). The carbon balance was also confirmed for
each result, and the difference in the carbon balance was always in
the range of 100Æ10% unless noted. The amount of metal eluted
into the reaction solution was analyzed by ICP-OES (Thermo Fisher
Scientific Inc., iCAP6500). Cold water was used as the washing sol-
vent instead of 2-propanol for the reuse experiments and determi-
nation of eluted metal. In the reuse experiments, the adipic acid
yield was not very high in comparison with the solubility limit of
adipic acid, and washing with water can totally dissolve the
formed adipic acid. Metal dissolution was calculated as follows:
Metal dissolution (%)=(mol of metal in the recovered reaction so-
lution)/(mol of metal initially put into the autoclave).
2000, 65, 6502–6507.
[10] E. Rozhko, K. Raabova, F. Macchia, A. Malmusi, P. Righi, P. Accorinti, S.
Alini, P. Babini, G. Cerrato, M. Manzoli, F. Cavani, ChemCatChem 2013, 5,
1998–2008.
[11] M. Vennat, P. Herson, J.-M. BrØgeault, G. B. Shul’pin, Eur. J. Inorg. Chem.
2003, 908–917.
[
[
12] J.-M. BrØgeault, Dalton Trans. 2003, 3289–3302.
13] J.-M. BrØgeault, M. Vennat, L. Salles, J.-Y. Piquemal, Y. Mahha, E. Briot,
P. C. Bakala, A. Atlamsani, R. Thouvenot, J. Mol. Catal. A 2006, 250, 177–
189.
[
14] I. A. Weinstock, E. M. G. Barbuzzi, M. W. Wemple, J. J. Cowan, R. S.
Reiner, D. M. Sonnen, R. A. Heintz, J. S. Bond, C. L. Hill, Nature 2001, 414,
191–195.
[15] S. K. Hanson, R. T. Baker, J. C. Gordon, B. L. Scott, D. L. Thorn, Inorg.
Chem. 2010, 49, 5611–5618.
[
[
[
16] S. Son, F. D. Toste, Angew. Chem. Int. Ed. 2010, 49, 3791–3794; Angew.
Chem. 2010, 122, 3879–3882.
17] B. Sedai, C. Díaz-Urrutia, R. T. Baker, R. Wu, L. A. Silks, S. K. Hanson, ACS
Catal. 2011, 1, 794–804.
18] S. K. Hanson, R. Wu, L. A. Silks, Angew. Chem. Int. Ed. 2012, 51, 3410–
3413; Angew. Chem. 2012, 124, 3466–3469.
[
[
19] R. Ma, Y. Xu, X. Zhang, ChemSusChem 2015, 8, 24–51.
20] J. Mottweiler, M. Puche, C. Räuber, T. Schmidt, P. Concepción, A. Corma,
C. Bolm, ChemSusChem 2015, 8, 2106–2113.
Catalyst characterization
[
[
21] T. Mallat, A. Baiker, Chem. Rev. 2004, 104, 3037–3058.
22] T. Mallat, A. Baiker, Catal. Today 1994, 19, 247–284.
X-ray diffraction (XRD) patterns were measured by using a Rigaku
MiniFlex600 diffractometer. CuKa (l=0.154 nm, 40 kV, 15 mA) radia-
tion was used as an X-ray source. The particle size on Pt/C was cal-
culated by the Scherrer’s equation. The amount of CO chemi-
sorption was measured in a high-vacuum system by using a volu-
metric method. Before adsorption measurements, the catalysts
[23] P. Gallezot, Catal. Today 1997, 37, 405–418.
[24] M. Besson, P. Gallezot, Catal. Today 2000, 57, 127–141.
[25] C. Parmeggiani, F. Cardona, Green Chem. 2012, 14, 547–564.
[
[
[43]
26] L. Prati, M. Rossi, J. Catal. 1998, 176, 552–560.
27] H. H. C. M. Pinxt, B. F. M. Kuster, G. B. Marin, Appl. Catal. A 2000, 191,
4
5–54.
28] G.-J. ten Brink, I. W. C. E. Arends, R. A. Sheldon, Science 2000, 287, 1636–
639.
were treated with H at 373 K for 1 h. Subsequently, the adsorption
2
[
[
was performed at room temperature. The gas pressure at adsorp-
tion equilibrium was about 1.2 kPa. The sample weight was about
1
29] K. Yamaguchi, N. Mizuno, Angew. Chem. Int. Ed. 2002, 41, 4538–4542;
Angew. Chem. 2002, 114, 4720–4724.
3
0
.1 g. The dead volume of the apparatus was about 60 cm . The
adsorption amount of CO is represented as the molar ratio to the
amount of noble metal.
[30] Y. Uozumi, R. Nakao, Angew. Chem. Int. Ed. 2003, 42, 194–197; Angew.
Chem. 2003, 115, 204–207.
[31] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc.
2004, 126, 10657–10666.
[
[
[
32] T. Wang, H. Shou, Y. Kou, H. Liu, Green Chem. 2009, 11, 562–568.
33] K. Kon, S. M. A. H. Siddiki, K. Shimizu, J. Catal. 2013, 304, 63–71.
34] S. Velusamy, T. Punniyamurthy, Org. Lett. 2004, 6, 217–219.
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1737
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
[
[
[
[
35] Z. Du, J. Ma, H. Ma, M. Wang, Y. Huang, J. Xu, Catal. Commun. 2010, 11,
32–735.
36] A. Dewan, T. Sarma, U. Bora, D. K. Kakati, Tetrahedron Lett. 2011, 52,
563–2565.
37] G. Sarmah, S. K. Bharadwaj, A. Dewan, A. Gogoi, U. Bora, Tetrahedron
Lett. 2014, 55, 5029–5032.
38] M. Kirihara, Y. Ochiai, S. Takizawa, H. Takahata, H. Nemoto, Chem.
Commun. 1999, 1387–1388.
39] Y. Maeda, N. Kakiuchi, S. Matsumura, T. Nishimura, T. Kawamura, S.
Uemura, J. Org. Chem. 2002, 67, 6718–6724.
[40] N. Jiang, A. J. Ragauskas, J. Org. Chem. 2007, 72, 7030–7033.
[41] J. M. A. Harmsen, L. Jelemensky, P. J. M. Van Andel-Scheffer, B. F. M.
Kuster, G. B. Marin, Appl. Catal. A 1997, 165, 499–509.
[42] M. S. Ide, D. D. Falcone, R. J. Davis, J. Catal. 2014, 311, 295–305.
[43] Z. Hu, H. Nakamura, K. Kunimori, H. Asano, T. Uchijima, J. Catal. 1988,
112, 478–488.
7
2
Received: February 8, 2016
Published online on April 21, 2016
ChemCatChem 2016, 8, 1732 – 1738
www.chemcatchem.org
1738
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim