Selective photooxidation of diols with silica bound W10 O324 -

Article abstract of DOI:10.1016/j.jcat.2007.11.008

The decatungstate anion W₁₀ O₃₂^{4 -} has been heterogenized on silica previously functionalized with different ammonium cations covalently bound to the surface of the solid support. These materials are investigated as photocatalysts for the oxygen-assisted oxidation of 1,3-butanediol and 1,4-pentanediol. Product distribution and adsorption experiments indicate that the polarity of the environment surrounding the photoactive anion plays a crucial role in controlling the adsorption of diols on the surface and, consequently, their reaction with the photoexcited decatungstate. Proper reaction conditions are found for obtaining more than 90% of 4-hydroxy-2-butanone from 1,3-butanediol and for stopping the oxidation of 1,4-pentanediol to 4-hydroxypentanal with good yield. The employed photocatalysts present a very good stability in repeated experiments.

Full text of DOI:10.1016/j.jcat.2007.11.008

Journal of Catalysis 253 (2008) 312–317
www.elsevier.com/locate/jcat
Research Note
4
3
−
Selective photooxidation of diols with silica bound W O
1
0
2
a,∗
a
b
A. Maldotti , A. Molinari , F. Bigi
a
Dipartimento di Chimica, Università di Ferrara, Via L. Borsari 46, I-44100 Ferrara, Italy
b
Dipartimento di Chimica Organica e Industriale, Università di Parma, Parco Area delle Scienze 17A, I-43100 Parma, Italy
Received 14 September 2007; revised 31 October 2007; accepted 14 November 2007
Abstract
The decatungstate anion W
4
−
O
10
has been heterogenized on silica previously functionalized with different ammonium cations covalently
3
2
bound to the surface of the solid support. These materials are investigated as photocatalysts for the oxygen-assisted oxidation of 1,3-butanediol
and 1,4-pentanediol. Product distribution and adsorption experiments indicate that the polarity of the environment surrounding the photoactive
anion plays a crucial role in controlling the adsorption of diols on the surface and, consequently, their reaction with the photoexcited decatungstate.
Proper reaction conditions are found for obtaining more than 90% of 4-hydroxy-2-butanone from 1,3-butanediol and for stopping the oxidation
of 1,4-pentanediol to 4-hydroxypentanal with good yield. The employed photocatalysts present a very good stability in repeated experiments.
©
2007 Elsevier Inc. All rights reserved.
Keywords: Heterogeneous photocatalysis; Polyoxotungstates; Oxidative catalysis; Alcohol oxidation
1
. Introduction
that the so obtained systems allow one to expand the reaction
solvent choice and present high efficiency and selectivity in the
catalytic oxidation of sulfides by hydrogen peroxide [10]. In
the present paper these materials are investigated as photocata-
lysts for the oxygen-assisted oxidation of one site of multifunc-
tional molecules such as diols. We were prompted to explore
Heterogenization of polyoxoanions for photocatalytic oxida-
tions by molecular oxygen has attracted much attention in the
last years. In particular, heterogenization of the decatungstate
W10O for photooxidation reactions has been previously per-
formed following different procedures, such as by impregnation
of a solid support [1–4], by embedding in polymeric mem-
branes [5–7] or inside the silica network [8,9]. Generally speak-
ing, the support makes the photocatalyst more easily handled
and recycled, it allows a free choice of the reaction medium
and, sometimes, it controls efficiency and selectivity of the
processes.
4
−
3
2
4−
this strategy because now it is well-established that W10O
32
is able to induce the photocatalytic oxidation of alcohols in
homogeneous solution as briefly summarized in the follow-
ing [1,7,11–15]. It is to outline that the preferential oxidation of
one of the alcoholic functionalities present in diols, with pro-
duction of hydroxyaldehydes or hydroxyketones as the main
products, is an important transformation in synthetic organic
chemistry [16].
One of the present authors has recently reported [10] that
decatungstate W10O4 can be heterogenized with silica pre-
−
Previous studies indicate that the photocatalytic oxidation
3
2
4
−
viously functionalized with different ammonium cations cova-
lently bound on the surface of the solid support. The polyoxoan-
ion is immobilized on the solid surface through an exchange
reaction by mixing the selected surface bound alkylammonium
salt and an aqueous solution of Na4W10O32. It has been found
of alcohols by W O occurs according to Scheme 1 [1,7,
10
3
2
11–15]. It is generally accepted that illumination of the de-
catungstate yields an oxygen to metal charge transfer excited
state (W10O4 *), which decays rapidly to a reactive transient
−
3
2
designed as wO. This intermediate reacts with aliphatic and
aromatic alcohols, through hydrogen atom abstraction mecha-
4
−
nism, leading to the reduced form of decatungstate (HW10O
)
*
32
Corresponding author. Fax: +39 0532240709.
E-mail addresses: andrea.maldotti@unife.it (A. Maldotti),
and to an organic radical [15]. The photocatalytic process
leads to the formation of a carbonylic compound as a conse-
alessandra.molinari@unife.it (A. Molinari), franca.bigi@unipr.it (F. Bigi).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.11.008

A. Maldotti et al. / Journal of Catalysis 253 (2008) 312–317
313
2
. Experimental
The decatungstates Na4W10O32 and (n-Bu4N)4W10O32
were prepared following literature procedures [17]. The photo-
catalysts Si-(alkyl)4NW, Si-(alkyl)3NHW and Si-alkylNH3W
were synthesized as previously described [10]. The prepara-
tion of anchored primary and tertiary ammonium salts was
performed by reacting silica (5 g) and the suitable (alkyl-
aminopropyl)trialkoxysilane (10 mmol) in toluene at reflux.
◦
After washing and drying under vacuum (60 C), the result-
ing materials in dichloromethane (5 g in 25 ml) were reacted
with trifluoromethane sulfonic acid (two equivalents respect
to the supported amino group) at room temperature, filtered
off, washed and dried. To prepare supported quaternary ammo-
nium salt, bromopropylated silica was first obtained condensing
4
−
Scheme 1. Photocatalytic behaviour of W
O
0
32
.
1
(
(
(
bromopropyl)trimethoxysilane (10 mmol) with silica silanols
5 g) by refluxing in toluene and then treated with triethylamine
50 mmol).The solid was filtered off, washed and dried. Im-
4−
mobilization of W10O32 was carried out by stirring a mixture
of the selected surface bound alkylammonium salt (4 mmol)
and an aqueous solution of sodium decatungstate (4 mmol) at
room temperature for 30 min. The white solids were filtered
off, carefully washed with distilled water, ethanol, diethyl ether
and with hot acetonitrile using a Soxhlet apparatus. The dried
materials were characterized as previously reported [17]. Pho-
tochemical excitations were performed with a Helios Italquartz
Q400 medium-pressure mercury lamp selecting wavelengths
with cut-off filters. Products analysis were carried out by using
a HP 6890 gas chromatograph, equipped with a flame ioniza-
tion detector and a DB-WAX capillary column. Quantitative
analyses was performed with calibration curves obtained with
Scheme 2. Oxidized products obtained upon irradiation of heterogeneous
authentic samples or prepared ones. For CO determination a
pH meter BasiC 20 CRISON equipped with a gas sensing probe
2
4
−
suspended in CH Cl solutions containing the diol in the presence
W
of O .
O
1
0
32
2
2
2
(
Crison 9666) was employed. When necessary, UV–vis spectra
were recorded with a KONTRON Uvikon 943 spectrophotome-
ter. Si-(alkyl)4NW, Si-(alkyl)3NHW and Si-alkylNH3W were
dispersed in 3 ml (7 g/l) of a CH2Cl2 or CH3CN solution
containing 1,3-butanediol or 1,4-pentanediol (0.01 M). The ob-
tained suspensions were stirred and irradiated with wavelengths
higher than 290 nm, at room temperature and under 760 Torr
of O2. On the basis of previous investigations, the amount of
7 g/l of supported decatungstate was the optimum in order to
ensure the maximum absorption of light [2,4] and was chosen
for all the photocatalysts. After irradiation, sample was cen-
trifuged, the products that remained eventually adsorbed on
the irradiated powder were extracted with CH2Cl2 (2 × 3 ml)
and analysed. Homogeneous reactions were carried out dissolv-
ing (n-Bu4N)4W10O32 in CH3CN solutions containing the diol
(0.01 M). The decatungstate concentration in the final solu-
tion (2 × 10 M) warranted the complete absorption of all
the incident photons. Three ml of this solution were irradiated
(20 min) with wavelengths higher than 290 nm, at room temper-
ature and under 760 Torr of O2. After irradiation, the organic
phase was analysed by gas chromatography. Each photocat-
alytic experiment was repeated three times in order to evaluate
the errors, which never exceeded ± 5%. No oxidation products
were obtained when blank experiments were run in the dark
quence of the reaction of this radical with O2, which also pro-
vides the reoxidation of the photoreduced decatungstate. In any
case, O2 undergoes reduction to reactive oxygenated species
[
1,7,15].
In the W10O4 -based materials employed in this work the
−
3
2
decatungstate is held on the support by its ionic bond with
tetraalkylammonium cations in Si-(alkyl)4NW, with trialkylam-
monium cations in Si-(alkyl)3NHW and with monoalkylam-
monium cations in Si-alkylNH3W (see insert in Scheme 2).
The structure of these three materials has been previously in-
vestigated through different physical–chemical techniques [10],
which evidenced that (i) the structure of the decatungstate is
still preserved on the silica surface; (ii) the loading of the
decatungstate is in the range 0.06–0.13 mmol/g; (iii) the de-
catungstate anion is bound to the solid support through ionic
interactions with four supported ammonium cations, since the
presence of sodium was not detected in the catalysts. Atten-
tion is here focused on the effects of the decatungstate counte-
rion in controlling the microenvironment where the photocat-
alytic process occurs and, consequently, its chemo- and regio-
selectivity. The diols investigated as oxidizable substrates are
−
4
1
,3-butanediol and 1,4-pentanediol.

314
A. Maldotti et al. / Journal of Catalysis 253 (2008) 312–317
or irradiating without the photocatalysts. Evaluation of the ad-
sorption of the two diols on the surface of photocatalysts was
carried out suspending each photocatalyst (7 g/l) in 3 ml of
the CH2Cl2 solution containing the diol (0.01 M) and keeping
it in the dark for 15 min under magnetic stirring. The amount
of diol adsorbed was established considering its decrease in
the solution. Analogous experiments were performed with so-
lutions containing mixtures of 1-pentanol/2-pentanol (0.005 M
each) and 1-butanol/2-butanol (0.005 M each). Evaluation of
the adsorption of carboxylic acids or carboxylates on the photo-
catalysts was carried out suspending each photocatalyst (7 g/l)
in 3 ml of a CH2Cl2 solution containing 4-oxopentanoic acid
or 3-hydroxybutiric acid (0.005 M) and keeping it in the dark
for 20 min under magnetic stirring. The amount of carboxylic
acid adsorbed was established measuring its decrease in the so-
lution by GC analysis; the possible release of the decatungstate
in the solution phase was evaluated by UV–vis analysis of the
solution.
Table 1
Photocatalytic properties of heterogenized decatungstate in the oxidation of
,3-butanediol and of 1,4-pentanediol in CH Cl
a
1
2
2
Diol
Photocatalyst
nab
µmoles of
aldehyde)
nbb
(µmoles of
ketone)
na/nb
(
1
,3-butanediol (n = 1)
Si-(alkyl) NW
0.1
1.6
1.1
1.2
0.06
0.18
0.63
4
Si-(alkyl) NHW 0.2
3
Si-alkylNH W
0.76
3
1
,4-pentanediol (n = 2)
Si-(alkyl) NW
3.0
1.0
0.4
0.2
3.0
4.0
7.5
4
Si-(alkyl) NHW 1.6
3
Si-alkylNH W
1.5
3
a
In a typical experiment the employed photocatalyst (7 g/l) was suspended in
a CH Cl solution (3 ml) containing 1,3-butanediol or 1,4-pentanediol (0.01 M)
2
2
and irradiated (20 min, λ > 290 nm) at 298± 1 K and 760 Torr of O2. Reported
values are the mean of three repeated experiments and are ± 5%.
b
Amount of carbonylic compounds as µmoles obtained in 3 ml of solution.
3
. Results and discussion
oxidation through Scheme 1 [1,7,11–15] is retained also after
its heterogenization on silica. The nature of the photoproducts
is consistent with the oxidation of one alcoholic functional-
ity of the diol, although two-site attack has been suggested as
important in the oxidation of diols in conventional electrolytic
cells [19]. It is noteworthy that, for both the investigated diols,
partially oxidized product can be accumulated in good yields
with no formation of ketoacids and no mineralization of the
substrate to carbon dioxide.
As far as the stability of the photocatalysts is concerned,
UV–vis spectra demonstrated that the decatungstate anion was
not released in detectable amounts into the solution during the
photocatalytic experiment, thus confirming that we were in the
presence of real heterogeneous catalytic processes. We have
also experimental evidence that the loss of activity of the three
photocatalysts in terms of detected end products was less than
Dichloromethane suspensions of Si-(alkyl)4NW or Si-(al-
kyl)3NHW or Si-alkylNH3W (7 g/l) containing 1,3-butanediol
were irradiated (λ > 290 nm) at room temperature and under
7
60 Torr of O2. Twenty minutes irradiation caused the conver-
sion of the diol to 3-hydroxybutanale (1a) and 4-hydroxy-2-bu-
tanone (1b) (see Scheme 2), which derive from the oxidation of
primary or secondary alcoholic functionality of the diol. Specif-
ically, these two products represented more than 90% of the
oxidized 1,3-butanediol and gave reason of about 90% of the
overall gas chromatographic area of the obtained photoprod-
ucts. Analogous photocatalytic experiments indicated that 1,4-
pentanediol (0.01 M) could be converted to 4-hydroxypentanal
(2a) and 5-hydroxy-2-pentanone (2b). Other two chromato-
graphic peaks have been attributed to the cyclic emiacetalic
diastereoisomers (2c) derived from the closure of 2a. Minor
amounts of unidentified compounds likely due to further pho-
tooxidation processes were also formed. 2a, 2b and 2c repre-
sent about 80% of the overall gas chromatographic areas of
the detected products. We could estimate that the mass balance
between the detected products and diol disappeared was about
1
5% after five repeated cycles. This slow deactivation may be
4−
tentatively ascribed to the degradation of W O to form small
10
32
undetectable fragments. Control experiments allowed us to rule
out the possibility that small amounts of photogenerated car-
boxylic acids and carboxylates may produce ion exchange in
some extent. In fact, (i) GC analysis indicates that the con-
centration of 4-oxopentanoic acid or 3-hydroxybutiric acid in
9
0% also for this diol.
Carbon dioxide, deriving from complete oxidative degra-
CH Cl solutions (0.005 M) did not decrease after contact with
2
2
dation of the substrate, must be counted among the possible
reaction products in photocatalysis with polyoxoanions [18].
While this is an advantage in the abatement of pollutants, it is a
drawback in the present case, where the aim is the accumulation
of valuable reaction intermediates. Some specific experiments
carried out in a closed reactor in order to collect the CO2 even-
tually formed enable us to conclude that carbon dioxide was not
formed in detectable amounts during the photocatalytic oxida-
tion of both 1,3-butanediol and 1,4-pentanediol. We have also
verified that no oxidation products were obtained when blank
experiments were run in the absence of light or irradiating the
dispersing medium without the photocatalysts.
the solid photocatalysts and (ii) UV–vis spectra gave evidence
that the decatungstate was not released in detectable amounts
into the solution phase.
Table 1 compares the photocatalytic activities of Si-(al-
kyl) NW, Si-(alkyl) NHW and Si-alkylNH W dispersed in
4
3
3
CH2Cl2 in the oxidation of 1,3-butanediol and 1,4-pentanediol.
Columns 3 and 4 report the µmoles of aldehyde (na) and ke-
tone (nb) obtained for both the investigated diols. In the case
of 1,4-pentanediol, 2a is the sum of 4-hydroxypentanal and of
its cis-trans emiacetalic isomers (2c, see Scheme 2) which are
formed by cyclization of the aldehyde and remain in equilib-
rium with it. In the last column the concentration ratio between
aldehyde and ketone enable us to evaluate how much the nature
On the basis of the described results, we can state that the
well known ability of photoexcited W10O4 to induce alcohol
−
3
2

A. Maldotti et al. / Journal of Catalysis 253 (2008) 312–317
315
Table 2
Adsorption % of diols and monoalcohols on photocatalysts
a
Photocatalyst
Adsorption (%)
+
+
Si-(alkyl) NW
10
12
19.8
8
9
22.4
4
10
13
4
9
8
<1
2.8
8.5
<1
2.1
6.5
4
Si-(alkyl) NHW
3
Si-alkylNH W
3
a
In a typical experiment the photocatalyst (7 g/l) was suspended under magnetic stirring for 15 min in a CH Cl solution containing the diols or mixtures of their
2
2
monofunctional models. Columns 2 and 3: 1,4-pentanediol or 1,3-butanediol, respectively (0.01 M). Column 4: a mixture of 1-pentanol and 2-pentanol (0.005 M
each). Column 5: a mixture of 1-butanol and 2-butanol (0.005 M each).
of the counterion bound on the solid support affects the relative
reactivity of the two alcoholic functional groups.
tanediol or 1,4-pentanediol and evaluating the amount of diol
adsorbed after 15 min. It is seen in columns 2 and 3 of Table 2
that the adsorption of both 1,3-butanediol and 1,4-pentanediol
occurs in the order Si-alkylNH3W > Si-(alkyl)3NHW > Si-
(alkyl)4NW, indicating that the enhanced polarity of the sur-
face due to the presence of hydrogen atoms instead of alkyl
chains favours the adsorption of the diols. Analogous experi-
ments (columns 4 and 5 of Table 2) were carried out suspend-
ing the photocatalysts in CH2Cl2 solutions containing mixtures
of monofunctional alcohols simulating the two different hy-
droxy functional groups present in the diols: 1-pentanol and
2-pentanol for 1,4-pentanediol (column 4), 1-butanol and 2-bu-
tanol for 1,3-butanediol (column 5). The results obtained con-
firm that adsorption phenomena play a key role in controlling
Previous studies regarding kinetic and mechanistic aspects
of photocatalysis by W10O4 have demonstrated that the pri-
−
3
2
mary reaction between wO and alcoholic substrates involves
hydrogen-atom abstraction from the C–H bond α to the OH
group (Scheme 1) [11–15]. From this point of view, both the
primary and the secondary alcoholic functions of 1,3-butane-
diol and of 1,4-pentanediol can undergo hydrogen-atom ab-
straction by wO from both the α C–H bonds. Since this reaction
leads to carbon centered radicals, the preferential oxidation of
the secondary carbon to give a ketone would be predicted on
the basis of pure thermodynamic considerations [15]. On the
other hand, it is seen in Table 1 that the ratio between aldehyde
and ketone depends on the nature of the counterion. Specifi-
cally, this ratio increases markedly as the alkyl chains of the
cation are substituted by hydrogen atoms. Interestingly enough,
we note that optimal conditions have been found in CH Cl for
4−
the regioselectivity of the W O -assisted photooxidation of
10
32
1,3-butanediol or 1,4-pentanediol. In fact, Table 2 shows that
primary alcohols are adsorbed in a greater extent than are sec-
ondary substrates and, therefore, they can interact easily with
2
2
4
−
obtaining more than 90% of 4-hydroxy-2-butanone (1b) from
,3-butanediol (with Si-(alkyl)4NW) and about 90% of 4-hydr-
oxypentanal (2a) and its cyclic emiacetalic isomers (2c) from
the photoexcited W O . The adsorption of both 1-butanol
10
3
2
1
and 2-butanol become negligible on the less polar photocata-
lyst Si-(alkyl) NW.
4
1
,4-pentanediol (with Si-alkylNH W).
The results of Table 1 show that the selectivity of the pri-
3
The observed effect of the counterion on the regioselectiv-
mary alcohol oxidation is higher for 1,4-pentanediol than for
1,3-butanediol, in line with the fact that this regioselectivity is
expected to be higher when the two competitive sites are more
distant each other. In order to evaluate the possibility that the re-
gioselective oxidation of 1,4-pentanediol to 4-hydroxypentanal
can be also controlled by the nature of the solvent, other pho-
ity of the photoprocess suggests that a first important parameter
affecting the reactivity of the investigated photocatalysts may
be the polarity of the solid surface and, consequently, its in-
teraction with the diol: preferential adsorption effects can be
significant in controlling the selectivity, considering that the re-
gion of the molecule which is more closely associated with the
surface and near to the photoexcited decatungstate is expected
to undergo more rapid oxidation. In particular, the enhanced
polarity of the surface due to the presence of hydrogen atoms
instead of alkyl chains should favour the preferential adsorp-
tion of the primary OH group of the more hydrophilic head of
the diol molecule with respect to the secondary OH group of
its more hydrophobic tail, thus fostering the interaction of the
primary alcoholic group with the photoexcited decatungstate
and, consequently, its oxidation. This statement is supported
by a previous investigation by Fox et al. on the oxidation of
tocatalytic experiments were performed using CH CN as dis-
3
persing medium. We have experimental evidences that, when
the photocatalytic experiments were carried out in this sol-
vent the concentration ratios between aldehyde and ketone were
always lower (in the range 0.6–0.7) than those obtained in
CH Cl independently of the photocatalyst employed (see Ta-
2
2
ble 3 as supplementary material). A plausible explanation is
that the more polar CH CN reduces the diol adsorption and,
3
consequently, the above described surface effects. Accordingly,
adsorption experiments confirm that the monoalcohols models
of 1,4-pentanediol were only slightly adsorbed on all the em-
1
,4-pentanediol by photoexcited TiO where they attributed the
ployed photocatalysts when dissolved in CH CN (<3%).
2
3
observed regioselective oxidation of the primary OH group to a
site selective adsorption on the polar surface of the active pho-
tocatalyst [20].
The activity of the investigated heterogeneous photocat-
4−
alysts can not be compared with that of W O
1
0
in ho-
32
mogeneous solution from a quantitative point of view since
Some experiments were carried out suspending each hetero-
(n-Bu N) W O , which is the salt usually employed for
4
4
10 32
geneous photocatalyst in a CH Cl solution containing 1,3-bu-
2 2
obtaining homogeneous solutions, is not soluble in CH Cl .
2
2

316
A. Maldotti et al. / Journal of Catalysis 253 (2008) 312–317
(A)
(B)
Fig. 1. µmoles of products vs time obtained irradiating (λ > 290 nm, 298 K, 760 Torr of O ) (A) Si-alkylNH W (7 g/l) in CH Cl solutions containing 1,4-pen-
2
3
2
2
tanediol (0.01 M). (B) Si-(alkyl) NW (7 g/l) in CH Cl solutions containing 1,3-butanediol (0.01 M).
4
2
2
Moreover, the microenvironment variations due to the size
and the nature of the counterions that occur on the surface
are not reproducible in the homogeneous medium. However,
a semiquantitative comparison of the photocatalytic properties
of Si-(alkyl) NW, Si-(alkyl) NHW and Si-alkylNH W with
was not formed in detectable amounts neither for 1,3-butane-
diol or 1,4-pentanediol. These findings are particularly relevant
since it is quite difficult to stop the oxidation of an alcohol at the
stage of aldehydes, which are known to be important building
blocks in applied organic synthesis.
4
3
3
that of (n-Bu4N)4W10O32 merits to be done for evaluating
the potential future applications of silica bound decatungstate
in photocatalysis. In these experiments we were forced to
employ CH CN in order to dissolve (n-Bu N) W O . Ir-
During prolonged irradiations, both Si-(alkyl)4NW and Si-
alkylNH3W became light blue coloured, indicating that their re-
duced forms were accumulated in some amounts on silica sup-
port. In order to evaluate the amount of reduced decatungstate
formed during irradiation, the powdered photocatalysts were
analysed by titration with KMnO4 at the end of the photocat-
alytic experiments. On the basis of this analysis, we concluded
that the amount of reduced decatungstate formed during irra-
diation was about 2.5% of the total decatungstate, a negligible
quantity that did not affect significantly the photocatalytic effi-
ciency of decatungstate.
3
4
4
10 32
radiation (20 min, λ > 290 nm) of CH3CN solutions of
−
4
(
1
n-Bu4N)4W10O32 (2 × 10 M) containing 1,3-butanediol or
,4-pentanediol (0.01 M) led to the formation of the same prod-
ucts obtained in heterogeneous conditions with about the same
total overall yield. This is a good result in view of the employ-
4
−
ment of the silica bound W O in heterogeneous catalysis,
10
3
2
where, usually, a significant loss of activity in comparison with
the homogeneous phase is observed. Lack of surface effects
and restriction in choice of the solvent have a negative effect
on the selectivity. In fact, we found that the concentration ratios
between aldehyde and ketone in the homogeneous phase exper-
iments were 0.55 for 1,3-butanediol and 1.2 for 1,4-pentane-
diol.
4. Conclusions
4
−
Heterogenization of W O on silica gel by chemical bond
1
0
3
2
gives a robust photocatalyst for the selective oxidation of diols
by O upon near UV light and under mild pressure and temper-
2
Some experiments have been carried out in order to investi-
ature conditions. The polarity of the environment surrounding
4−
gate the photocatalytic activity of silica bound W O under
the photoactive anion plays a crucial role in controlling the ad-
sorption of diols on the surface and, consequently, their reaction
with the photoexcited decatungstate. We also conclude that site
selective adsorption, a basic requirement to obtain high regios-
electivity, can be fostered through a correct choice of the dis-
persing medium. The described reactions are a new appealing
synthetic route for the selective oxidation of diols since proper
reaction conditions have been found for obtaining more than
90% of 4-hydroxy-2-butanone from 1,3-butanediol. The oxi-
dation of 1,4-pentanediol yields 4-hydroxypentanal with good
selectivity. This is a particularly relevant result since it is quite
difficult to stop the oxidation of an alcohol at the aldehyde,
which are known to be important building blocks in organic
synthesis.
10
32
prolonged irradiation. Si-alkylNH3W in CH2Cl2 has been cho-
sen for the photooxidation of 1,4-pentanediol, since the data
of Table 1 indicate that this photocatalyst presents the high-
est selectivity toward 4-hydroxypentanal. Fig. 1A shows that
the aldehyde 2a, including also its cyclic isomers 2c, increased
quite linearly during the photocatalytic experiment and repre-
sented the main product also after three hours irradiation, when
about 50% of the initial diol underwent oxidation. Table 1 indi-
cates that the most selective photocatalyst for the oxidation of
1
,3-butanediol is Si-(alkyl) NW. Fig. 1B shows that its main
4
product during the entire photocatalytic experiment was the
ketone 1b, whose concentration increased quite linearly vs irra-
diation time. It is noteworthy that after 225 min irradiation CO2

A. Maldotti et al. / Journal of Catalysis 253 (2008) 312–317
317
Acknowledgments
[6] M. Carraro, M. Gardan, G. Scorrano, E. Orioli, E. Fontananova, M. Bon-
chio, Chem. Commun. (2006) 4533.
[
7] M. Bonchio, M. Carraro, M. Gardan, G. Scorrano, E. Drioli, E. Fonta-
nanova, Top. Catal. 40 (2006) 133.
The authors are very grateful for the financial support of
MIUR (PRIN 2006) and INCA Consortium.
[8] Y. Guo, C. Hu, X. Wang, Y. Wang, Y. Zou, H. Ding, S. Feng, Chem.
Mater. 13 (2001) 4058.
Supplementary material
[9] Y. Guo, C. Hu, J. Mol. Catal. A Chem. 262 (2007) 136.
10] F. Bigi, A. Corradini, C. Quarantelli, G. Sartori, J. Catal. 270 (2007) 222.
11] C.L. Hill (Ed.), Chem. Rev. 98 (1998) 1, A special issue of Chemical Re-
views is devoted to polyoxometalates.
[
[
Supplementary material associated with this article can be
found, in the online version, at DOI: 10.1016/j.jcat.2007.11.008.
[12] C. Tanielian, Coord. Chem. Rev. 178–180 (1998) 1165.
13] D.C. Duncan, M.A. Fox, J. Phys. Chem. A 10 (1998) 4559.
[
References
[14] I.N. Lykakis, M. Orfanopoulos, Tetrahedron Lett. 46 (2005) 7835.
[
[
[
[
15] C. Tanielian, I.N. Lykakis, R. Seghrouchni, F. Cougnon, M. Orfanopoulos,
J. Mol. Catal. A Chem. 262 (2007) 170.
16] R.A. Sheldon, J.K. Kochi, Metal-Catalyzed Oxidation of Organic Com-
pounds, Academic Press, New York, 1981.
17] A. Chemseddine, C. Sanchez, J. Livage, J.P. Launay, M. Fournier, Inorg.
Chem. 23 (1984) 2609.
18] A. Hiskia, A. Mylonas, E. Papaconstatinou, Chem. Soc. Rev. 30 (2001)
[
[
1] A. Maldotti, A. Molinari, R. Amadelli, Chem. Rev. 102 (2002) 3811.
2] A. Maldotti, A. Molinari, G. Varani, M. Lenarda, L. Storaro, F. Bigi,
R. Maggi, A. Mazzacani, G. Sartori, J. Catal. 209 (2002) 210.
3] A. Maldotti, R. Amadelli, I. Vitali, L. Borgatti, A. Molinari, J. Mol. Catal.
A Chem. 204 (2003) 703.
[
[
[
4] A. Molinari, R. Amadelli, A. Mazzacani, G. Sartori, A. Maldotti, Lang-
muir 18 (2002) 5400.
62.
[
[
19] G. Horanyi, K. Torkos, J. Electroanal. Chem. 125 (1981) 105.
20] M.A. Fox, H. Ogawa, P. Pichat, J. Org. Chem. 54 (1989) 3847.
5] M. Bonchio, M. Carraro, G. Scorrano, A. Bagno, Adv. Synth. Catal. 346
(2003) 1119.