Oxidative C?C Cleavage of Ketols over Vanadium–Carbon Catalysts

Article abstract of DOI:10.1002/cctc.201700566

The oxidation of 2-hydroxycyclohexanone and carbohydrates to adipic acid and formic acid, respectively, was performed with a combination of a vanadium catalyst, carbon as the cocatalyst, and O₂ in water under mild conditions (353 K, 0.1–0.3 MPa). The catalytic activity of aqueous V₂O₅ was increased significantly by the addition of activated carbon (C), whereas the addition of carbon black, graphene oxide, and carbon nanotubes has a much smaller effect. UV/Vis and inductively coupled plasma optical emission spectroscopy were used to show that most V^V species are adsorbed on C at least in a short reaction time. The V^IV species (VO²⁺) was not adsorbed on C. The order of activity of vanadium species was V^V on C>dissolved free V^V?dissolved free V^IV. To help the oxidation of V^IV, the further addition of phosphomolybdate (PMo₁₂O₄₀ ^3?; PMo) was also tested, and the activity was improved. The selectivity and yield of adipic acid from 2-hydroxycyclohexanone was also improved slightly by the addition of PMo. However, in the oxidation of glucose, the addition of PMo did not improve the final formic acid yield. The oxidation of glucose with the V+C system gave a 42 % yield of formic acid, which was comparable to the values reported with a more expensive PVMo polyoxometalate catalyst. A reaction mechanism is proposed in which the reversibly formed “biradical” state of the vanadium ketol complex reacts with O₂ and accompanying rearrangement dissociates the C?C bond.

Full text of DOI:10.1002/cctc.201700566

Accepted Article
Title: Oxidative C-C Cleavage of Ketols over Vanadium-Carbon
Catalysts
Authors: Yoshinao Nakagawa, Dai Sekine, Naoyuki Obara, Masazumi
Tamura, and Keiichi Tomishige
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201700566
Link to VoR: http://dx.doi.org/10.1002/cctc.201700566
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201700566
Oxidative C-C Cleavage of Ketols over Vanadium-Carbon Catalysts
[
a,b]
[a]
[a]
[a,b]
Yoshinao Nakagawa,*
Dai Sekine, Naoyuki Obara, Masazumi Tamura,
Keiichi
[
a,b]
Tomishige*
[a]
Dr. Y. Nakagawa, D. Sekine, N. Obara, Dr. M. Tamura and Prof. K. Tomishige: Department of
Applied Chemistry, School of Engineering, Tohoku University, 6-6-07, Aoba, Aramaki, Aoba-ku,
Sendai 980-8579, Japan
[b]
Dr. Y. Nakagawa, Dr. M. Tamura and Prof. K. Tomishige: Research Center for Rare Metal and
Green Innovation, Tohoku University, 468-1, Aoba, Aramaki, Aoba-ku, Sendai 980-0845, Japan
*
Corresponding author: Yoshinao Nakagawa and Keiichi Tomishige
School of Engineering, Tohoku University,
6
-6-07, Aoba, Aramaki, Aoba-ku, Sendai, 980-8579, Japan
E-mail: yoshinao@erec.che.tohoku.ac.jp, tomi@erec.che.tohoku.ac.jp
1
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Abstract
Oxidation of 2-hydroxycyclohexanone and carbohydrates to adipic acid and formic acid, respectively,
was carried out with combination of vanadium catalyst, carbon as co-catalyst and molecular oxygen
in water under mild conditions (353 K, 0.1-0.3 MPa). The catalytic activity of aqueous V O was
2
5
significantly increased by addition of activated carbon (C), while addition of carbon black, graphene
oxide and carbon nanotube has much smaller effect. The UV-vis and ICP analyses showed that most
2
+
vanadium(V) species is adsorbed on C at least at short reaction time. Vanadium(IV) species (VO )
was not adsorbed on C. The order of activity of vanadium species was vanadium(V) on C >
dissolved free vanadium(V) >> dissolved free vanadium(IV). In order to help oxidation of
3
40
-
vanadium(IV), further addition of phosphomolybdate (PMo O ; PMo) was also tested, and the
1
2
activity was actually improved. The selectivity and yield of adipic acid from
-hydroxycyclohexanone was also slightly improved by addition of PMo. However, in the oxidation
2
of glucose, the addition of PMo did not improve the final formic acid yield. Oxidation of glucose
with vanadium + C system gave 42% yield of formic acid, which was comparable to the values
reported with more expensive P-V-Mo polyoxometalate catalyst. A reaction mechanism is proposed
where reversibly formed “biradical” state of vanadium-ketol complex reacts with molecular oxygen,
and accompanying rearrangement dissociates the C-C bond.
Keywords
Vanadium; oxidation; carboxylic acid; activated carbon
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
1
. Introduction
Selective oxidative C-C cleavage of poly-functionalized compounds to aldehydes or carboxylic
[
1-3]
acids attracts growing interest, especially in the research area of biomass conversions.
Oxidative
C-C cleavage of vicinal diols or ketols (hydroxyketones) is typically performed with stoichiometric
[
1,4,5]
oxidants such as lead tetraacetate and periodates.
Because of the non-green nature of such
oxidants, catalytic systems using molecular oxygen as an oxidant have been searched. Typical
[6-10]
[11]
catalysts for oxidative C-C cleavage with molecular oxygen are ruthenium-,
manganese-,
[
12-14]
[10,13,15-22]
copper,
and vanadium-based ones.
Ruthenium and manganese catalysts can convert
aromatic vicinal diol to aldehydes in good yields, and they are less effective for aliphatic substrates,
especially in neutral or acidic conditions. Substrates for Cu-catalyzed C-C oxidation are also
aromatic compounds, typically those with benzoyl group. On the other hand, homogeneous
vanadium catalysts are active in oxidative C-C cleavage of aliphatic ketols such as
[
10,15,16]
[17-22]
2
-hydroxycyclohexanone
and sugars
to adipic acid and formic acid, respectively
[
23,24]
[25-28]
(Scheme 1). As related systems, oxidations of catechols
and lignin-model compounds
with
homogeneous vanadium catalysts have been also intensively investigated. The most investigated
vanadium-based catalyst in oxidative C-C cleavage of ketols is Keggin-type polyoxometalate,
[
10,15a,16-18,20,22,29,30]
H PV Mo O (denoted as PVMo).
PVMo polyoxometalates are redox-active:
3
+n
n
12-n 40
vanadium(V) in PVMo is easily reduced to tetravalent, and the vanadium(IV) in reduced PVMo is
[
31]
easily re-oxidized by molecular oxygen.
On the other hand, oxidation of simple dissolved
2
+
vanadium(IV) species such as VO to vanadium(V) is slow with molecular oxygen. The reactivity
of reduced PVMo with molecular oxygen enables PVMo to be good oxidation catalysts in various
[
31,32]
reactions.
However, aqueous solution of PVMo is a mixture of polyoxometalates with different
[31]
V/Mo ratio and monomeric metal-oxo species in equilibrium. Because of the variety of vanadium
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
species in PVMo solution, the amount of active vanadium species at one time should be much lower
than the total vanadium amount in the system, and there can be vanadium catalysts with more activity.
Very recently, we have developed a catalytic system with combination of Pt/C + V O solution for
2
5
[33]
oxidation of vicinal diols to two carboxylic acids.
Platinum works as the oxidation catalyst for
vicinal diols to ketols, and vanadium works as the oxidation catalyst for ketols to carboxylic acids.
During this research, we found that the catalytic activity of vanadium in the latter reaction is
enhanced by the presence of carbon, which has been added to the system as a support of Pt. However,
[
33]
in that paper we did not further investigate the effect of carbon because the ketol oxidation is not
the rate-determining step in vicinal diol oxidation to carboxylic acids. In this study, we focus on the
vanadium-catalyzed ketol oxidation, which itself is a valuable reaction, in the presence of carbon. We
found that pentavalent vanadium species adsorbed on the activated carbon have high activity. The
effect of further addition of polyoxometalate salt to vanadium-carbon system has been also
evaluated.
2
. Results and discussion
2
.1. Effect of carbon on vanadium-catalyzed oxidation of 2-hydroxycyclohexanone
Oxidation of 2-hydroxycyclohexanone to adipic acid (AA) was selected as a model reaction, since
[
10,15,16]
this substrate has been one of the most tested ones for ketol oxidation.
In addition, the product,
[
34]
adipic acid, has high importance in chemical industry.
oxidation with V O catalyst and various carbon additives are shown in Table 1. The amount of V O
5
The results of 2-hydroxycyclohexanone
2
5
2
(<0.01 g/ 10 mL) was below the solubility in water solvent (0.8 g/ 100 mL at 293 K, according to the
safety data sheet by the supplier); no V O powder was left in the reaction media of all runs. V O
2
5
2
5
solution has some catalytic ability (Table 1, entry 2). The addition of activated carbon (FAC-10 and
4
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Carboraffin) increases the catalytic activity (entries 3 and 4), and FAC-10 showed the best
promotional effect. On the other hand, carbon black (XC-72), graphene oxide (entry 6) and carbon
nanotube (entry 7) have almost no promotional effect. It is well known that activated carbon have
various types of functional groups such as carboxylic acids, phenols, ketones and so on. On the other
[
35]
hand, typical functional groups in graphene oxide are alcoholic OH and epoxide. Carbon black has
smaller amount of oxygen-containing functional groups and carbon nanotubes have very few
functional groups. Functional groups which are not major in graphene oxide such as phenolic OH
and aromatic carboxyl may be the ligands for active vanadium species. Without V O , the carbon
2
5
materials hardly enhanced the formation rate of adipic acid (entries 8-12), indicating that vanadium is
essential in the catalysis. Activated carbon FAC-10 was selected as the additive in the following
studies because of the highest promotional effect.
The selectivity to adipic acid with V O + carbon (FAC-10) catalysts (63%) was not higher than
2
5
V O alone catalyst system (80%) in Table 1. However, at higher conversion level the selectivity of
2
5
V O alone catalyst system was decreased and the selectivity order was reversed, as discussed later.
2
5
The main by-products in Table 1 were glutaric acid (C5 dicarboxylic acid; GA), CO and several
2
unidentified compounds detected by RID-HPLC. Succinic acid (C4 dicarboxylic acid; SA) was
sometimes produced; however the formation was negligible in the case of V O + FAC-10 catalyst
2
5
system.
Next, the effect of pH value was checked (Table 2). Carbon materials with many functional groups
have some acidity and make buffer solution when they are added to water. It is well known that the
[
36]
structure of vanadium(V) species much depends on the pH value. The purpose of this experiment
is to check whether the promotional effect by carbon addition is only due to the pH control or not.
The aqueous V O solution had a pH value of 3.7 (Table 2, entry 1), and the value was increased to
2
5
5
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
5
.0 when carbon (FAC-10) was added (entry 2). When NaOH was added to V O solution to control
2 5
similar pH value of V O + carbon system (entry 3), the change of activity was much smaller than
2
5
the case of carbon addition. Conversely, when H SO was added to the V O + C system to decrease
2
4
2
5
the pH values (entries 4 and 5), the reaction rate was almost unchanged, although the selectivity to
adipic acid was slightly decreased. These data show that the addition of carbon has more effect than
pH control.
The catalytic performance of V O + C was further compared with that of H PV Mo O , which
2
5
5
2
10 40
is a typical PVMo polyoxometalate (Table 2, entry 6). While the conversion rate was similar between
H PV Mo O and V O + C catalysts, the selectivity to adipic acid was very low (34%) with
5
2
10 40
2
5
H PV Mo O . Low selectivity (<35%) to adipic acid from aqueous 2-hydroxycyclohexanone with
5
2
10 40
[
10]
PVMo catalyst has been also reported in the literature. The V O + C catalyst system is superior to
2
5
PVMo catalyst for simple ketol in water solvent, while in methanol solvent PVMo catalyst has been
[
15a,16]
reported to give good selectivity to dimethyl adipate.
2
.2. The state of vanadium during the reaction
The state of vanadium species during the reaction was investigated. As a preliminary experiment,
hot filtration at 353 K and ICP-OES analysis of the filtrate was conducted for the mixture of V O +
2
5
C + water (2 mg + 20 mg + 10 g) system. It has been known that main species of acidic aqueous
6
-
+
[36]
V O solution are decavanadate (V O ) and VO , both of which have yellow~orange color.
2
5
10 28
2
The yellow color of dissolved vanadium(V) almost disappeared by carbon addition and filtration, and
only 3% of vanadium was dissolved in the filtrate. Then we determined the state of vanadium in the
reaction mixture for 2-hydroxycyclohexanone oxidation after hot filtration. In order to conduct the
hot filtration, the reaction was carried out in a test tube under 0.1 MPa O . The results were
2
6
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
summarized in Table 3. The reaction time was set as 10 and 60 min, and for all cases (V O ,
2
5
IV
V OSO and V O + C catalysts) the conversion at 60 min was higher than that at 10 min, indicating
4
2
5
that the catalyst was active at least at 10 min. In the case of V O alone system with low activity
2
5
(Table 3, entries 1 and 2), the amount of dissolved vanadium(V) species was gradually decreased,
and the amount of dissolved vanadium(IV) species increased. In addition, significant amount of
black solid was formed which probably contained the rest of vanadium species. However, significant
amount of dissolved vanadium(V) species was still present at 10 min (34% of initial amount),
indicating that dissolved vanadium(V) species as well as the black solid have low activity. In the case
of VOSO (entries 3 and 4), most vanadium(IV) species remained in the homogeneous phase during
4
the reaction. The activity of VOSO was even lower than that of V O . In the cases of V O + C
4
2
5
2
5
systems with higher activity (entries 5-7), vanadium was present as adsorbed species on carbon
support and homogeneous (dissolved) vanadium(IV) species. The combination of vanadium(IV)
species + C (entry 9) gave lower activity than V O + C, and the most vanadium species was
2
5
dissolved vanadium(IV) with very low activity. The activity of vanadium(IV) species + C (entry 8)
can be explained by the small amount of carbon-adsorbed vanadium(V) species produced by
oxidation of vanadium(IV), since dissolved vanadium(IV) was indeed decreased during the reaction
(entries 8 and 9). Based on these results, we concluded that adsorbed vanadium(V) species on
activated carbon have much higher activity in ketol oxidation than homogeneous (dissolved)
vanadium(V) and vanadium(IV) species.
Yao Fu et al. and Ye Wang et al. reported that VOSO has good activity in sugar oxidation to
4
[
18,19]
formic acid;
however, the oxygen pressure was much higher (2 MPa) than our case (0.1-0.3
MPa), and the vanadium(IV) species can be readily oxidized in the literature systems.
7
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
2
.3. Effect of addition of polyoxometalate
As described in the previous section, formation of vanadium(IV) species will decrease the
catalytic activity. To assist the oxidation of vanadium(IV) species to vanadium(V), we conducted the
reaction tests with addition of polyoxometalate (phosphomolybdate), although the system became
more complex due to the complexity and the reactions of polyoxomealate. As well as acid-type
phosphomolybdate (H PMo O ; HPMo), cesium phosphomolybdate (Cs H PMo O ; CsPMo)
3
12 40
2.5 0.5
12 40
was used because of the low solubility in water and large surface area of cesium
[
32,37]
phosphomolybdate.
The reaction results are summarized in Table 4. The addition of CsPMo or
HPMo slightly increases the activity of V O + C system (entries 3, 5 and 6), and the selectivity to
2
5
adipic acid was also increased. The promotional effect of CsPMo was larger than that of HPMo.
Without carbon, the addition of CsPMo or HPMo also increased the activity of V O ; however, the
2
5
order was opposite: V O + HPMo system has higher activity than V O + CsPMo (entries 7 and 8).
2
5
2
5
It is well known that dissolved vanadium(V) and phosphomolybdate quickly react each other to form
[
31]
phophomolybdovanadates (PVMo), and the activity increase can be explained by the formation of
PVMo with higher activity than V O (cf. Table 2). In V O + C + HPMo system, the formation of
2
5
2
5
PVMo in homogeneous phase may decrease the amount of carbon-adsorbed vanadium(V) species,
limiting the promotional effect of phosphomolybdate addition (Table 4, entries 3 and 6; conversion
+7%) in comparison with that of CsPMo addition (entries 3 and 5; conversion +11%) and that of
HPMo addition to carbon-free system (entries 2 and 8; conversion +14%). The homogeneous phase
after the reaction showed yellow color which is indicative of fully oxidized vanadium and/or
polyoxometalate species for all entries 5-8, and the UV-Vis spectrum has negligible band for
vanadium(IV) species (Figure S2). However, significant amount of vanadium species was still
present in the homogeneous phase after the reactions with V O + C + CsPMo or HPMo (entries 5
2
5
8
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
and 6). At higher conversion level, some products may coordinate the vanadium species and desorb
the vanadium(V) from carbon surface. In addition, vanadium-containing polyoxometalates produced
by the reaction of phosphomolybdate with vanadium species were likely to be present in the
homogeneous phase, especially in the case using HPMo (entries 6 and 8).
According to Table 4, V O + C system has good performance, and V O + C + CsPMo system has
2
5
2
5
slightly higher performance. The reaction tests at longer reaction time were carried out to complete
the reaction for both systems (Scheme 2). The selectivity was almost unchanged until the total
conversion of 2-hydroxycyclohexanone. The V O + C system gave 60% yield of adipic acid (at 96%
2
5
conversion) and the V O + C + CsPMo system gave 73% yield (at 97% conversion). On the other
2
5
hand, V O catalyst without any additive decreased its selectivity at higher conversion level (80% →
2
5
<
60%), and the adipic acid yield was lower than those of V O + C and V O + C + CsPMo systems.
2 5 2 5
2
.4. Application to other substrates
We applied the V O + C and V O + C + CsPMo systems to carbohydrate substrates. The main
2
5
2
5
products of carbohydrate oxidation by vanadium species are formic acid and CO . Table 5 shows the
2
results of dihydroxyacetone oxidation. Main products of dihydroxyacetone are glycolic acid, formic
acid and CO . Small amount of acetic acid was also detected. At short reaction time the products are
2
mainly glycolic acid and formic acid. At longer reaction time the yield of glycolic acid decreased and
those of formic acid and CO increased. As an ideal case one dihydroxyacetone molecule is
2
converted to two formic acid molecules and one CO (Scheme 3) although the oxidation of glycolic
2
acid was slow. The slow oxidation of glycolic acid was verified by the reaction tests with glycolic
acid as the substrate (Table S1, Supporting Information): the reaction rate of glycolic acid was about
1
order slower than that of dihydroxyacetone, and about 1:1 formation of formic acid and CO from
2
9
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
glycolic acid was confirmed. The order of conversion rate of dihydroxyacetone with different
catalyst systems was as follows: V O + C + CsPMo > V O + C > V O + CsPMo > V O , which
2
5
2
5
2
5
2
5
was similar to the case of 2-hydroxycyclohexanone substrate. However, at longer reaction time the
yields were similar between V O + C + CsPMo and V O + C.
2
5
2
5
Table 6 shows the results of oxidation of fructose and glucose. Fructose oxidation was much faster
than glucose oxidation (entries 1-8). The identified products of fructose oxidation were formic acid,
CO and glycolic acid, while glycolic acid can be the intermediate of formic acid and CO as
2
2
described above. The order of conversion rate of fructose with different catalyst systems was also
almost the same as that for 2-hydroxycyclohexanone and dihydroxyacetone (V O + C + CsPMo ≥
2
5
V O + C > V O + CsPMo > V O ) (entries 1-4). However, the oxidation of glucose showed a
2
5
2
5
2
5
different trend. The order of conversion rate was much higher when CsPMo was present in the
system (entries 5-8). An important point is that significant amount of fructose was formed in V O +
2
5
CsPMo and V O + C + CsPMo systems (entries 7 and 8). With only CsPMo, fructose as the sole
2
5
product was formed (entry 9). Considering the higher reactivity of fructose, fructose works as an
intermediate of glucose oxidation to formic acid, and CsPMo promote the isomerization of glucose to
fructose. Ye Wang et al. proposed that the main route of glucose oxidation with vanadium (IV)/(V)
catalyst is isomerization of glucose to fructose, retro-aldol reaction to two C3 units (glyceraldehyde
[19]
and dihydroxyacetone), and oxidation of the C3 units to formic acid and CO2. Our results agree
with this reaction route. Although the involvement of retro-aldol reaction was not confirmed in our
system, the formation of glycolic acid in both dihydroxyacetone and fructose oxidations (Tables 5
and 6) supports that dihydroxyacetone works as an intermediate of fructose oxidation. On the other
hand, under severer conditions to complete the reaction, the final yield of formic acid from glucose
was similar (~45%) among V O + C, V O + CsPMo, and V O + C + CsPMo catalyst systems
2
5
2
5
2
5
10
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
(entries 12-14). The yield values were comparable to those reported for glucose oxidation with
[
17a,18,20]
PVMo catalysts (~50%).
Nevertheless, the reaction rate of V O + C system was much higher
2
5
than that of H PV Mo O , a typical PVMo catalyst (entry 15) in our reaction conditions. The lower
5
2
10 40
activity of PVMo than V O + C was in contrast to the comparable activity for
2
5
2
-hydroxycyclohexanone oxidation between these catalysts (Table 2). The PVMo catalyst may have
low activity in glucose isomerization.
Therefore, for both dihydroxyacetone and glucose, the final yields are similar between V O + C
2
5
and V O + C + CsPMo systems, although CsPMo has some positive effect in substrate conversion.
2
5
V O + C system is attractive in view of catalyst cost.
2
5
2
.5. Reaction mechanism
Reaction of molecular oxygen is always a key step of oxidation with molecular oxygen under mild
conditions because of the biradical nature of molecular oxygen. We conducted the reaction under
different oxygen pressures to investigate the effect of oxygen pressure in the kinetics. The results are
shown in Table 7. Here, we added small amount of polyoxometalate to the system to oxidize
vanadium(IV) and then to exclude the effect of the formation amount of inactive vanadium(IV)
species under different oxygen pressure. Although the exact reaction order could not be calculated
because of the high conversion level even at short reaction time (0.5 h), the reaction rate was clearly
increased by increasing the oxygen pressure, and the effect can be regarded as first order. These
results indicate that the rate-determining step involves free or catalyst-bound molecular oxygen.
There is no widely-accepted mechanism of ketol oxidation with vanadium catalyst. Vennat et al.
proposed the mechanism that starts with abstraction of alpha-hydrogen of coordinated alcohol group
[
16]
(CH-OV) by vanadium(V) center.
The formed “biradical” consisting of vanadium(IV) and
11
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
．
． ．
R-C(-OV)-R’ quickly reacts with O=O to give alkylperoxovanadium(V) species without unpaired
electron. However, this mechanism does not agree with our data that oxygen pressure has clear
positive effect while hydrogen abstraction is a slow reaction (rate-determining step). There are other
proposed mechanisms for vanadium-catalyzed ketol oxidation which involve free vanadium(IV)
[
19,21,22]
species in the redox cycle.
vanadium(IV) has much lower activity. On the other hand, Finke et al. proposed a mechanism for
,5-di-tert-butylcatechol oxidation to dicarboxylic acids with homogeneous vanadium catalyst where
This type of mechanisms also do not agree with our data that free
3
one electron in the aromatic ring of coordinated substrate is transferred to vanadium(V) to give
semiquinone and vanadium(IV), and then molecular oxygen reacts with the semiquinone and
[
24]
vanadium(IV) to give peroxo-bridged quinone-vanadium(V) species.
The kinetics of the
3
3
,5-di-tert-butylcatechol oxidation follows first order for oxygen pressure. Inspired by the
,5-di-tert-butylcatechol oxidation mechanism, we propose a mechanism as shown in Scheme 4.
First, ketol is coordinated to vanadium(V) center with the -OH group as alkoxide (step (i)). Biradical
species is reversibly formed by transfer of one  electron in C=O to vanadium(V) (step (ii)).
Molecular oxygen reacts with the biradical state of vanadium-substrate complex (step (iii)), and the
rearrangement cleaves the C-C bond (step (iv)) to give vanadium-bound percarboxylate and aldehyde.
The steps (iii) and (iv) may proceed simultaneously, because the intermediate between steps (iii) and
(iv) has unstable positively-charged oxygen. The rate-determining step will be step (iii) (assumed
that this step (iii) is irreversible) or step (iv) (assumed that the step (iii) is reversible). The
vanadium-bound percarboxylate and aldehyde are converted to two carboxylic acids by
[
38]
vanadium-catalyzed Baeyer-Villiger-like reaction (step (v)).
As a side reaction, homolytic O-O
[
39]
dissociation of V(V)-O-O moiety produces free inactive vanadium(IV) species and free radicals,
which leads to various kinds of by-products by radical-chain autoxidation. The role of carbon support
12
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
may include a reservoir of electrons and it may stabilize the “biradical” state of vanadium-substrate
complex.
We also carried out DFT (density functional theory) calculation for the proposed scheme. The
activated carbon ligand was simulated by 1,8-naphthalenediolate. The transition states were not
calculated because the steps (i) and (v) involve solvent molecules and the step (iii) clearly has
multi-configurational nature that should not be treated by single-determinant theory. Nevertheless,
the geometry optimization for the step (iii) product leads to C-C dissociation and finally the step (iv)
product was obtained. The energy level of the step (iv) product is significantly lower (92 kJ/mol)
than the step (i) product. These calculations support that the steps (iii) and (iv) proceed
simultaneously. The detailed results of these calculations are described in Figure S3, Supporting
Information.
3
. Conclusions
Catalytic activity of aqueous V O for ketol oxidation with molecular oxygen is promoted by
2
5
addition of activated carbon. Addition of carbon materials with smaller amount of functional groups
such as carbon black and carbon nanotube is not effective on the catalytic activity of vanadium.
Although much amount of vanadium is present in homogenous phase during the reaction,
vanadium(V) species adsorbed on activated carbon has high activity. Vanadium(IV) species are much
less active. This point also suggests that the main reaction path of ketol oxidation does not include
free vanadium(IV) species. The vanadium(V) + carbon system can be applied to carbohydrate
oxidation to formic acid. The activity of vanadium(V) + carbon in glucose oxidation is much higher
than that of V O alone or acid-type P-V-Mo polyoxometalate (H PV Mo O ), while the activity of
2
5
5
2
10 40
vanadium(V) + carbon and H PV Mo O is similar in 2-hydroxycyclohexanone oxidation. The
5
2
10 40
13
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
final yield of formic acid (42%) from glucose is comparable to those reported for acid-type P-V-Mo
polyoxometalates. The addition of cesium phosphomolybdate (Cs H PMo O ) to vanadium(V) +
2
.5 0.5
12 40
carbon system further increases the activity. The activity increase may be due to the oxidation of
vanadium(IV) species assisted by polyoxometalate, although the chemistry of polyoxometalate
addition is very complex because of the reactivity, acidity and redox ability of polyoxometalate.
While the final yield of formic acid from glucose is almost unchanged by addition of cesium
phosphomolybdate, the yield of adipic acid from 2-hydroxycyclohexanone is slightly increased.
Based on the data that free vanadium(IV) is not involved in the main reaction cycle and high oxygen
pressure increases the activity, a reaction mechanism is proposed where reversibly formed “biradical”
state of vanadium-ketol complex reacts with molecular oxygen, and accompanying rearrangement
dissociates the C-C bond. The effective functional groups of carbon support have not been
determined, and the determination is one of the targets of future study such as by using model
ligands.
4
. Experimental section
Materials
Organic reagents and catalysts used in this study were commercially available and used as
received. The used carbon materials were: Osaka Gas Chemicals Shirasagi FAC-10 activated carbon
2
(
lignocellulosic, activated by steam, BET surface area 851 m /g), Osaka Gas Chemicals Carboraffin
2
activated carbon (lignocellulosic, activated by zinc chloride, 1215 m /g), Vulcan XC-72 carbon black
2
(
224 m /g), graphene oxide purchased from Aldrich, and carbon nanotube (multi-walled, 10-30 nm)
purchased from Wako. V O and H PMo O •23H O were purchased from Wako. VOSO •6H O
2
5
3
12 40
2
4
2
was purchased from Kanto Chemicals. H PV Mo O •25H O was kindly given by Nippon Inorganic
5
2
10 40
2
14
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Colour & Chemical Co., Ltd. Cs H PMo O (CsPMo) was prepared by slow addition of 0.08 M
2
.5 0.5
12 40
aqueous cesium carbonate to 0.06 M aqueous H PMo O in stoichiometric amount (molar ratio Cs :
3
12 40
[
40]
P = 2.5 : 1) at 323 K and subsequent evaporation and calcination at 573 K. The BET surface area
2
of prepared CsPMo was 15.7 m /g. XRD pattern of CsPMo confirmed the cubic structure of cesium
[
40]
phosphomolybdate. These measurements were conducted with Shimadzu Gemini 2360 (BET) and
Rigaku MiniFlex 600 (XRD) instruments. Dihydroxyacetone (Wako) and 2-hydroxycyclohexanone
(Aldrich) was received as the dimer solid. The molar amount of dihydroxyacetone and
2
-hydroxycyclohexanone was shown as that of the monomer unit.
Activity tests
The oxidation reaction was performed in a 190 mL stainless-steel autoclave with an inserted glass
vessel. The catalysts, substrate and water (10 mL) were put into the autoclave together with a spinner.
The amount of V O was below the solubility limit (0.8 g / 100 mL at 293 K) in all cases. The pH
2
5
was measured with a pH meter when necessary. After sealing, the reactor was filled with 0.3 MPa
oxygen. The autoclave was heated to 353 K (~20 min to reach, as monitored by thermocouple
inserted in the autoclave). The stirring rate was fixed at 500 rpm (magnetic stirring). After an
appropriate reaction time the reactor was cooled down by cold water bath, and the gases were
collected in a gas bag. The autoclave contents were diluted with 2-propanol and transferred to a vial,
while the catalyst was separated by filtration with membrane filter. Analysis of liquid phase was
conducted with HPLC (Shimadzu Prominence; Aminex HPX-87 H column (Bio-rad), 0.01 M
sulfuric acid as eluent, UV detector (210 nm) and RID). The gas phase was analyzed with FID-GC
(Shimadzu GC-2014, Porapak N packed column, FID combined with methanator). Conversion,
selectivity and yield were calculated as the following formula: Conversion (%) = {mol of consumed
15
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
substrate}/{mol of charged substrate} × 100; Selectivity (%) = {(mol of product)(number of carbon
atoms in the product)}/{(mol of consumed substrate)(number of carbon atoms in the substrate)} ×
1
00; Yield (%) = {(mol of product)(number of carbon atoms in the product)}/{(mol of charged
substrate)(number of carbon atoms in the substrate)} × 100. When the conversion level was very low
<10%), the conversion was calculated by the sum of product yields. The carbon balance was
(
1
00±10%.
In the case of oxidation under atmospheric pressure, a 15 mL test tube was used instead of an
autoclave. The removal of catalyst after the reaction was conducted without cooling (hot filtration).
Metal amount in the solution was analyzed by ICP-OES (Thermo Fisher Scientific Inc.,
iCAP6500). UV-Vis measurements were conducted with the Shimadzu 2450 UV-visible spectrometer,
using silica quartz square cell. Methanol (1 mL) and phthalate pH standard solution (pH 4.01; 1 mL)
was added to the sample (2 mL) in order to make the coordination mode of vanadium species the
same in different samples. The amount of dissolved vanadium(V) and (IV) species was determined
by the absorbance in 450-500 nm and 700-800 nm regions, respectively. V O and VOSO solutions
2
5
4
were used as standards. Solutions of V O and NH VO showed the same spectrum pattern in this
2
5
4
3
protocol. It was confirmed that mixed solution of NH VO and VOSO (33% and 67% of the original
4
3
4
concentration, respectively; similar concentrations observed in Table 3) gives the correct
concentration of each species. The raw spectra are shown in Figure S1 (Supporting Information). It
was confirmed that dissolved VOSO was not oxidized during the measurement processes.
4
3
[41,42]
DFT calculations were conducted with Materials Studio DMol software.
The used functional
was PW91 (default setting for GGA-level DFT in this program), and the basis sets were built-in
double numerical functions plus d-functions for non-hydrogen atoms (DND3.5; comparable to
[
43]
6
-31G(d)). Core electrons were represented by DSPP (DFT semi-core pseudopotentials) built in
16
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
the software. Orbitals for compounds with unpaired electrons (O ) were unrestricted. The initial state
2
was set as VO(OH)(C H O ) (C H O = 1,8-naphthalenediolato), and all the calculations were
1
0
6
2
10
6
2
carried out for gas-phase states.
Acknowledgments
Part of this work is supported by JSPS KAKENHI “Grant-in-Aid for Scientific Research (C)”
K15K065640.
References
[
1] T. K. M. Shing in Comprehensive Organic Synthesis Vol. 7 (Eds.: B. M. Trost, I. Fleming),
Pergamon Press, Oxford, 1994, pp. 703.
[
[
[
[
[
2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411-2502.
3] C. Li, X. Zhao, A. Wang, G. W. Huber, T. Zhang, Chem. Rev. 2015, 115, 11559-11624.
4] R. Floresca, M. Kurihara, D. S. Watt, A. Femir, J. Org. Chem. 1993, 58, 2196-2200.
5] I. Carrera, M. C. Brovetto, J. C. Ramos, G. A. Seoane, Tetrahedron Lett. 2009, 50, 5399-5402.
6] 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.
[
[
[
[
7] S. J. H. F. Arts, F. van Rantwijk, R. A. Sheldon, J. Carbohydr. Chem. 1996, 15, 317-329.
8] E. Takezawa, S. Sakuguchi, Y. Ishii, Org. Lett. 1999, 1, 713-715.
9] K. Yamaguchi, N. Mizuno, Chem. Eur. J. 2003, 9, 4353-4361.
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] V. Escande, C. H. Lam, C. Grison, P. T. Anastas, ACS Sustainable Chem. Eng. 2017, 5,
[
17
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
3
214-3222.
[
12] a) M. Wang, J. Lu, J. Ma, Z. Zhang, F. Wang, Angew. Chem. Int. Ed. 2015, 54, 14061-14065;
Angew. Chem. 2015, 127, 14267-14271; b) M. Wang, X.-K. Gu, H.-Y. Su, J.-M. Lu, J.-P. Ma, M. Yu,
Z. Zhang, F. Wang, J. Catal. 2015, 330, 458-464; c) M. Wang, J. Lu, X. Zhang, L. Li, H. Li, N. Luo,
F. Wang, ACS Catal. 2016, 6, 6086-6090; d) H. Lu, M. Wang, H. Li, N. Luo, S. Xu, F. Wang, J. Catal.
2
017, 346, 170-179; e) H. Liu, M. Wang, H. Li, N. Luo, S. Xu, F. Wang, J. Catal. 2017, 346,
70-179; f) M. Wang, L. H. Li, J. M. Lu, H. J. Li, X. C. Zhang, H. F. Liu, N. C. Luo, F. Wang, Green
1
Chem. 2017, 19, 702.
13] C. Díaz-Urrutia, B. Sedai, K. C. Leckett, R. T. Baker, S. K. Hanson, ACS Sustainable Chem.
Eng. 2016, 4, 6244-6251.
[
[
[
14] Z. Zhou, M. Liu, C. Li, ACS Catal. 2017, 7, 3344-3348.
15] a) L. E. Aakel, F. Launay, A. Atlamsani, J.-M. Brégeault, Chem. Commun. 2001, 2218-2219; b)
L. E. Aakel, F. Launay, J.-M. Brégeault, A. Atlamsani, J. Mol. Catal. A 2004, 212, 171-182.
[
[
16] M. Vennat, P. Herson, J.-M. Brégeault, G. B. Shul’pin, Eur. J. Inorg. Chem. 2003, 908-917.
17] a) R. Wölfel, N. Taccardi, A. Bösmann, P. Wasserscheid, Green Chem. 2011, 13, 2759-2763; b) J.
Albert, R. Wölfel, A. Bösmann, P. Wasserscheid, Energy Environ. Sci. 2012, 5, 7956-7962.
[
[
18] J. Li, D.-J. Ding, L. Deng, Q.-X. Guo, Y. Fu, ChemSusChem 2012, 5, 1313-1318.
19] Z. Tang, W. Deng, Y. Wang, E. Zhu, X. Wan, Q. Zhang, Y. Wang, ChemSusChem 2014, 7,
1
557-1567.
[
[
20] J. Zhang, M. Sun, X. Liu, Y. Han, Catal. Today 2014, 233, 77-82.
21] a) W. Wang, M. Niu, Y. Hou, W. Wu, Z. Liu, Q. Liu, S. Ren, K. N. Marsh, Green Chem. 2014,
1
6, 2614-2618; b) M. Niu, Y. Hou, S. Ren, W. Wang, Q. Zheng, W. Wu, Green Chem. 2015, 17,
35-342; c) M. Niu, Y. Hou, S. Ren, W. Wu, K. N. Marsh, Green Chem. 2015, 17, 453-459.
3
18
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
[
[
[
22] T. Lu, M. Niu, Y. Hou, W. Wu, S. Ren, F. Yang, Green Chem. 2016, 18, 4725-4732.
23] H. Weiner, R. G. Finke, J. Am. Chem. Soc. 1999, 121, 9831-9842.
24] a) C.-X. Yin, R. G. Finke, J. Am. Chem. Soc. 2005, 127, 13988-13996; b) A. M. Morris, C. G.
Pierpont, R. G. Finke, J. Mol. Catal. A 2009, 309, 137-145.
25] a) S. K. Hanson, R. T. Baker, J. C. Gordon, B. L. Scott, D. L. Thorn, Inorg. Chem. 2010, 49,
611-5618; b) S. K. Hanson, R. Wu, L. A. Silks, Angew. Chem. Int. Ed. 2012, 51, 3410-3413; Angew.
[
5
Chem. 2012, 124, 3466-3469; c) B. Sedai, C. Díaz-Urrutia, R. T. Baker, R. Wu, L. A. Silks, S. K.
Hanson, ACS Catal. 2011, 1, 794-804.
[
[
26] Y. Ma, Z. Du, J. Liu, F. Xia, J. Xu, Green Chem. 2015, 17, 4968-4973.
27] J. Mottweiler, M. Puche, C. Räuber, T. Schmidt, P. Concepción, A. Corma, C. Bolm,
ChemSusChem 2015, 8, 2106-2113.
[
[
[
[
[
[
28] Y.-Y. Jiang, L. Yan, H.-Z. Yu, Q. Zhang, Y. Fu, ACS Catal. 2016, 6, 4399-4410.
29] J. Liu, Z. Du, Y. Yang, T. Lu, F. Lu, J. Xu, ChemSusChem 2012, 5, 2151-2154.
30] J. Lan, J. Lin, Z. Chen, G. Yin, ACS Catal. 2015, 5, 2035-2041.
31] I. V. Kozhevnikov, Chem. Rev. 1998, 98, 171-198.
32] N. Mizuno, M. Misono, Chem. Rev. 1998, 98, 199-218.
33] N. Obara, S. Hirasawa, M. Tamura, Y. Nakagawa, K. Tomishige, ChemCatChem 2016, 8,
1
732-1738.
[
[
[
34] S. Van de Vyver, Y. Román-Leshkov, Catal. Sci. Technol. 2013, 3, 1465-1479.
35] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 2010, 39, 228-240.
36] a) D. C. Crans, A. S. Tracy, in ACS Symposium Series, Vol. 711, American Chemical Society,
1
998, pp. 2; D.C. Crans, J. J. Smee, E. Gaidamauskas, L. Yang, Chem. Rev. 2004, 104, 849-902.
[
37] N. Mizuno, M. Tateishi, M. Iwamoto, J. Catal. 1996, 163, 87-94.
19
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
[
38] a) R. Gopinath, B. K. Patel, Org. Lett. 2000, 2, 577-579; b) R. Gopinath, B. Barkakaty, B.
Talukdar, B. K. Patel, J. Org. Chem. 2003, 68, 2944-2947.
39] a) G. B. Shul’pin, M. C. Guerreiro, U. Schuchardt, Tetrahedron 1996, 52, 13051-13062; b) G.
Süss-Fink, L. Gonzalez, G. B. Shul’pin, Appl. Catal. A 2001, 217, 111-117.
[
[
40] M. Langpape, J. M. M. Millet, U. S. Ozkan, M. Boudeulle, J. Catal. 1999, 181, 80-90.
[
[
[
41] B. Delley, J. Chem. Phys. 1990, 92, 508-517.
42] B. Delley, J. Chem. Phys. 2000, 113, 7756-7764.
43] B. Delley, Phys. Rev. B 2002, 66, 155125.
20
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 1. Oxidation of 2-hydroxycyclohexanone with V O + carbon catalysts
2
5
Entry Catalysts
Conv. Selectivity [%]
[
%]
1.9
AA
69
80
63
72
74
59
77
76
35
63
48
66
GA
1
SA
<1
2
CDO COx
Others
1
2
3
4
5
6
7
8
9
1
1
1
None
V O
5
<1
<1
7
6
19
11
31
11
14
21
10
12
57
22
13
11
28
61
46
33
33
32
5
2
2
5
V O + FAC-10
4
<1
6
2
2
5
V O + Carboraffin
4
<1
6
2
5
V O + XC-72
4
2
<1
9
2
5
V O + graphene oxide
5
5
2
2
5
V O + carbon nanotube
4
1
<1
9
8
2
5
FAC-10
5.1
2
1
<1
<1
<1
<1
<1
XC-72
2.5
6.8
<1
2
<1
1
8
0
1
2
Carboraffin
Graphene oxide
Carbon nanotube
12
37
21
11
6.4
2
<1
<1
2
Reaction conditions: 2-Hydroxycyclohexanone 5 mmol, V O 0 or 9.1 mg (0.1 mmol-V), carbon 0 or 0.1 g,
2
5
water 10 g, O 0.3 MPa, 353 K, 1 h. AA = adipic acid, GA = glutaric acid, SA = succinic acid, CDO =
2
1
,2-cyclohexanedione. “Others” include both identified products and unidentified RID peaks.
21
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 2. Oxidation of 2-hydroxycyclohexanone with V O under different pH values
2
5
Entry Catalysts
Initial Conv. Selectivity [%]
pH
3.7
5.0
5.8
3.0
1.8
1.7
[%]
7
AA
71
71
59
61
58
34
GA
5
SA
<1
<1
<1
<1
<1
<1
CDO COx
Others
19
1
2
3
4
5
6
V O
<1
2
5
3
5
3
4
1
2
5
V O + C
27
11
27
25
30
4
20
2
5
V O + NaOH
1
<1
2
35
2
5
V O + C + H SO
<1
2
33
2
5
2
4
V O + C + H SO
<1
6
37
2
5
2
4
a
H PV Mo O
5 10 40
3
55
2
Reaction conditions: 2-Hydroxycyclohexanone 4.3 mmol, V O 0 or 2 mg (22 mol-V), carbon (C; FAC-10)
2
5
0
or 0.1 g, water 10 g, O 0.3 MPa, 353 K, 1 h. AA = adipic acid, GA = glutaric acid, SA = succinic acid,
2
a
CDO = 1,2-cyclohexanedione. “Others” include both identified products and unidentified RID peaks.
H PV Mo O 11 mol.
5
2
10 40
22
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 3. Oxidation of 2-hydroxycyclohexanone with V O under different pH values
2
5
Entry Catalyst
Time
Dissolved
Dissolved
Conv. Selectivity / %
[
min]
a
V(V) [%]
V(IV) [%]
[%]
4
AA
56
72
57
76
-
CDO
<1
<1
<1
<1
-
Others
44
28
43
24
-
1
2
3
4
5
6
7
8
9
V O
10
60
10
60
0
34
<5
<5
<5
<5
<5
<5
<5
6
12
36
99
97
<3
60
56
94
82
2
5
a
V O
17
2
2
5
VOSO4
VOSO4
V O + C
7
-
2
5
V O + C
10
60
0
18
46
-
49
59
-
7
44
36
-
2
5
V O + C
5
2
5
VOSO + C
-
4
VOSO + C 60
27
64
1
35
4
Reaction conditions: 2-Hydroxycyclohexanone 0.86 mmol, catalyst 22 mol-V, carbon (C; FAC-10) 0 or 0.02
g, water 2 g, O 0.1 MPa, 353 K. Oxygen refilling was performed every 10 minutes. AA = adipic acid, CDO =
2
1
,2-cyclohexanedione. “Others” include both identified products and unidentified RID peaks. Dissolved
a
vanadium species were determined by UV-Vis spectra. The raw UV-Vis data are shown in Figure S1. Black
solid was also formed.
23
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 4. Oxidation of 2-hydroxycyclohexanone with V O and various additives
2
5
Entry Catalysts
Conv. Selectivity [%]
Dissolved Dissolved
a
a
[
%]
1.9
AA GA SA CDO COx Others V [%]
Mo [%]
1
2
3
4
5
6
7
8
9
1
None
V O
69
80
63
76
77
67
74
68
53
65
1
5
4
2
3
4
3
4
12
6
<1
2
5
<1
<1
9
6
19
11
31
13
17
26
18
25
20
17
-
-
28
61
2
48
54
-
-
2
5
V O + C
<1
<1
<1
<1
1
2
-
2
5
C
5.1
<1
2
-
V O + C + CsPMo 72
<1
1
48
37
45
42
-
3
2
5
V O + C + HPMo 68
2
32
3
2
5
V O + CsPMo
37
42
2
1
2
5
V O + HPMo
<1
<1
<1
3
1
43
n.d.
n.d.
2
5
C + CsPMo
CsPMo
9.6
8.2
15
12
<1
<1
0
-
Reaction conditions: 2-Hydroxycyclohexanone 5 mmol, V O 0 or 9.1 mg (0.1 mmol-V), carbon (C; FAC-10)
2
5
0
or 0.1 g, polyoxometalate 0 or 0.01 mmol, water 10 g, O 0.3 MPa, 353 K, 1 h. CsPMo = Cs H PMo O .
2 2.5 0.5 12 40
HPMo = H PMo O , AA = adipic acid, GA = glutaric acid, SA = succinic acid, CDO =
3
12 40
1
,2-cyclohexanedione. “Others” include identified products and unidentified RID peaks. n.d. = not determined.
a
Determined by ICP-OES; after reaction.
24
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 5. Oxidation of dihydroxyacetone with V O -carbon-polyoxometalate catalysts
2
5
Entry Catalysts
Time [h] Conv. C-based yield [%]
[
%]
Glycolic acid Formic
acid
CO2
1
2
3
4
5
6
7
8
V O
1
1
1
1
4
4
4
4
46
13
45
24
53
32
43
33
43
14
26
19
35
38
40
36
41
1
4
2
5
V O + C
81
2
5
V O + CsPMo
55
6
2
5
V O + C + CsPMo
>99
88
6
2
5
5
V O
2
8
V O + C
>99
91
13
9
2
5
V O + CsPMo
2
5
V O + C + CsPMo
>99
13
2
5
Reaction conditions: Dihydroxyacetone 5 mmol, V O 9.1 mg (0.1 mmol-V), carbon (C; FAC-10) 0 or 0.1 g,
2
5
CsPMo (Cs H PMo O ) 0 or 0.01 mmol, water 10 g, O 0.3 MPa, 353 K.
2
.5 0.5
12 40
2
25
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 6. Oxidation of sugars with V O -carbon-polyoxometalate catalysts
2
5
Entry Substrate Catalysts
Conditions Conv. C-based yield [%]
[
%]
Fructose
Glycolic Formic CO2
acid
8
acid
11
20
18
21
<1
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
Fructose V O5
A
A
A
50
79
62
81
2
-
3
9
2
V O + C
-
10
9
2
5
V O + CsPMo
-
5
2
5
V O + C + CsPMo A
-
9
10
<1
<1
<1
<1
<1
<1
1
2
5
Glucose V O5
A
A
A
<1
<1
9
<1
<1
<1
<1
<1
<1
1
2
V O + C
8
2
5
V O + CsPMo
39
13
25
<1
49
98
>99
>99
82
6
2
5
V O + C + CsPMo A
3
3
2
5
CsPMo
C
A
A
B
B
B
25
<1
<1
2
<1
<1
13
43
46
42
18
0
1
2
3
4
5
Glucose V O5
2
V O + C
2
26
48
57
2
2
5
V O + CsPMo
<1
<1
8
<1
<1
4
2
5
V O + C + CsPMo B
2
5
a
H PV Mo O
5 10 40
B
2
Reaction conditions: (A) substrate 5 mmol, V O 0 or 9.1 mg (0.1 mmol-V), carbon (C; FAC-10) 0 or 0.1 g,
2
5
CsPMo (Cs H PMo O ) 0 or 0.01 mmol, water 10 g, O 0.3 MPa, 353 K, 4 h. (B) O 0.6 MPa, 24 h; other
2
.5 0.5
12 40
2
2
a
conditions were the same as (A). H PV Mo O , 0.1 mmol-V.
5
2
10 40
26
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Table 7. Oxygen pressure dependence of 2-hydroxycyclohexanone oxidation with
V O -carbon-polyoxometalate catalysts
2
5
Entry
O2
Conv.
Selectivity [%]
[
MPa] [%)]
AA
GA
SA
CDO COx Others
1
2
3
4
0.1
0.3
0.7
1.2
20
52
67
74
59
84
88
85
1
<1
1
<1
<1
<1
<1
11
<1
<1
<1
<1
<1
<1
<1
29
14
9
1
13
Reaction conditions: 2-Hydroxycyclohexanone 4.3 mmol, V O 1.8 mg (0.02 mmol-V), carbon (FAC-10) 0.1
2
5
g, H PV Mo O 1 mol, water 10 g, O 0.3 MPa, 353 K, 0.5 h. AA = adipic acid, GA = glutaric acid, SA =
5
2
10 40
2
succinic acid, CDO = 1,2-cyclohexanedione. “Others” include identified products and unidentified RID peaks.
27
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Scheme 1. Vanadium-catalyzed oxidation of ketol and carbohydrate.
28
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700566
Scheme 2. Adipic acid yield of different systems. Conditions: 2-hydroxycyclohexanone 4.3 mmol,
V O 0.1 mmol-V, carbon (FAC-10) 0 or 0.1 g, CsPMo (Cs H PMo O ) 0 or 0.01 mmol, water
2
5
2.5 0.5
12 40
1
0 g, O 0.3 MPa, 353 K.
2
29
This article is protected by copyright. All rights reserved.