Aerobic photooxidation in water by polyoxotungstates: The case of uracil

Article abstract of DOI:10.1002/ejoc.200500418

Uracil photooxygenation occurs in acidic water (pH = 1) at 25°C, under oxygen (1 atm), irradiating with γ > 300 nm in the presence of selected polyoxometalates (POM). A marked diversity of photocatalytic behavior is registered for different POMs in terms of oxidation rate and selectivity. H ₃PW₁₂O₄₀ (PW₁₂) appears to be the most reactive photocatalyst, by far superior to isostructural complexes, leading to a product distribution typical of OH^. dominated oxidations, while Na₄W₁₀O₃₂ (W₁₀) and Na ₁₂[WZn₃(H₂O)₂(ZnW₉O ₃₄)₂] (Zn₅W₁₉) exhibit a preferential reactivity towards uracil glycol. Kinetic studies and radical scavenger probes, performed on target intermediates and model diols, highlight a substantial difference in the mechanism of photocatalysis by the three complexes. Wiley-VCH Verlag GmbH & Co. KGaA, 2005.

Full text of DOI:10.1002/ejoc.200500418

FULL PAPER
Aerobic Photooxidation in Water by Polyoxotungstates: The Case of Uracil
Marcella Bonchio,*[a] Mauro Carraro,[a] Valeria Conte,[b] and Gianfranco Scorrano[a]
Keywords: Photooxidation / Polyoxometalates / Alcohols / Uracil / Dioxygen
Uracil photooxygenation occurs in acidic water (pH = 1) at
25 °C, under oxygen (1 atm), irradiating with λ Ͼ 300 nm in
the presence of selected polyoxometalates (POM). A marked
diversity of photocatalytic behavior is registered for different
POMs in terms of oxidation rate and selectivity. H3PW12O40
(PW12) appears to be the most reactive photocatalyst, by far
superior to isostructural complexes, leading to a product dis-
tribution typical of OH· dominated oxidations, while
Na4W10O32
(W10
)
and
Na12[WZn3(H2O)2(ZnW9O34)2]
(Zn5W19) exhibit a preferential reactivity towards uracil
glycol. Kinetic studies and radical scavenger probes, per-
formed on target intermediates and model diols, highlight a
substantial difference in the mechanism of photocatalysis by
the three complexes.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
POM based photocatalysis occurs via the formation of a
charge-transfer excited state (POM*) with strong oxidizing
properties (Scheme 1).[7–9,17] This primary photoreactant is
able to undergo multi-electron reduction,[7–9,18] while gener-
ating reactive radical intermediates in solution (substrate/
solvent activation).[19–22] In such catalytic schemes, di-
oxygen (i) intercepts the organic radicals giving rise to auto-
oxidation chains, (ii) provides for the re-oxidation of the
POM(red), and (iii) evolves to reduced species (dioxygen ac-
tivation).[23,24] The synergism of these multiple activation
pathways, whilst fostering highly efficient oxygenations in
terms of substrate conversion, poses a serious selectivity
challenge.[12]
Introduction
Dioxygen activation in water by photochemical methods
has a major appeal[1] vis-à-vis the environmental advantages
of low impact photooxygenations,[2,3] photoassisted degra-
dation of organic pollutants for wastewater treatment[4,5]
and its application within photodynamic therapy.[6] Poly-
oxometalates (POM) have been considered as the homogen-
eous analogs of semiconductor photocatalysts,[7–9] promot-
ing photooxygenation of various substrates, both in organic
solvents and in water.[7,8,10–13] This high versatility arises
from the extremely rich variety of known POM complexes
differentiated in terms of chemical composition, structure
and counterion.[14–16]
Especially in aqueous solution, the production of highly
reactive hydroxyl radicals OH· from the solvent activation
routine, might override more selective pathways, originating
within the substrate activation cycle, by interaction with the
POM photocatalyst.[10,25,26] The intervention of OH· as the
dominant oxidant during photocatalysis in water, is a mat-
ter of current debate, substantiated by ESR experiments,[27]
product distribution and kinetic studies.[22] The existence of
POM-substrate coordination equilibria preceding the oxi-
dation step has been evidenced by NMR experiments, kin-
etic and selectivity studies using radical scavenger
probes.[26,28] A POM-dependent selectivity in organic sol-
vents, discovered in photoassisted alkane dehydrogenation,
has been addressed by the elegant work of Hill and co-
workers.[29–31] These observations point to a potential tun-
ing of selectivity in aqueous photocatalysis by POMs, de-
spite the radical nature of the substrate activation step, thus
providing a valuable advantage over OH·-based oxidations.
In this work the activity and selectivity of several poly-
oxotungstates has been screened using uracil photooxi-
dation in water as a benchmark reaction. The photooxi-
dation of pyrimidine bases is relevant to the effect of ionis-
Scheme 1. Photocatalytic oxidation cycle by polyoxometalates in
water.
[a] ITM-CNR sezione di Padova, Dipartimento di Scienze Chim-
iche, Università di Padova,
Via Marzolo 1, 35131 Padova, Italy
Fax: +39-0498275239
E-mail: marcella.bonchio@unipd.it
[b] Dipartimento di Scienze e Tecnologie Chimiche, Università di
Roma “Tor Vergata”,
Via della Ricerca Scientifica, 00133 Roma, Italy
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2005, 4897–4903
DOI: 10.1002/ejoc.200500418
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4897

M. Bonchio, M. Carraro, V. Conte, G. Scorrano
FULL PAPER
ing radiation on DNA and to the photocleavage mechanism pathway[39] or from heterocyclic peroxo-radicals originating
of nucleic acids.[32] In particular, oxidative damage to nuc- through dioxygen trapping.[40] The latter decay according
leobases is inflicted by OH· generated under different reac- to a parallel route responsible for the rearrangement to hyd-
tion conditions. Thus, product analysis of these target bi- antoine derivatives 5–7.[37]
omolecules can serve as a probe to identify dominant or
A first set of experiments was designed to assess the pho-
preferential reaction pathways.[33] Our study focuses on the tocatalytic activity of selected polyoxotungstates towards
product distribution deriving from uracil photocatalyzed uracil oxygenation. Photooxidation has generally been per-
oxidation by diverse systems including, for the sake of com- formed in acidic water (pH = 1) at 25 °C, under oxygen
parison, both homogeneous POMs and the heterogeneous (1 atm), irradiating with λ Ͼ 300 nm.
titanium dioxide (TiO2).[34] The selectivity trend observed
The wavelength cut-off prevents direct photolysis of the
will be substantiated by a detailed kinetic investigation per- uracil whilst leaving an absorbance tail for catalytic photo-
formed on the target substrate, on reaction intermediates, excitation (see Figure S1 in the Supporting Information).
namely uracil glycol and isodialuric acid, and on model Representative UV/Vis spectra are shown in Figure 1 along
alcohols.
with the POM structural types under examination.
Data in Table 1 are meant to evaluate the product distri-
bution along both the oxidative (products 2–4) and re-
arrangement (products 5–7) routes, which globally account
Result and Discussion
The generally accepted mechanism for uracil oxidation for 70–80% of substrate conversion. For the various sys-
by OH· involves the essential steps of Scheme 2, delineating tems under examination, pertinent results are collected at
a cascade of expected products under both anaerobic or similar substrate conversions including both a low 20–30%
aerobic conditions.[35–38] The heterocyclic system undergoes range, to allow a comparison with the less efficient anaero-
addition at the C(5)–C(6) double bond (k
= 5– bic experiments (entries 2, 5, 7 in Table 1), and the high
6.5×109 m–1 s–1)[32b] yielding hydroxylated pyrimidine radi- range at 95–100% conversion. Linear kinetics of substrate
cals. Such intermediates may evolve to uracil glycol 2 and to disappearance were generally obtained, indicating that, un-
its overoxidation products 3 and 4, either by an anaerobic der the conditions explored, uracil oxidation is not rate
Scheme 2. Expected products for uracil oxidation by OH· uracil glycol (cis + trans) (2), isodialuric acid (3), alloxan (4), 1-formyl-5-
hydroxyhydantoin (5), 5-hydroxyhydantoin (6), parabanic acid (7), isobarbituric acid (8), hydrated alloxan (9).
4898
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4897–4903

Aerobic Photooxidation in Water by Polyoxotungstates: The Case of Uracil
FULL PAPER
and selectivity. In particular PW12 appears the most reactive
photocatalyst, by far superior to isostructural complexes
(cf. entry 1, 8–10), leading to a product distribution typical
of OH· dominated oxidations as in the Fenton-like system
(entry 11).[43] Indeed, its reactivity is strongly inhibited
upon addition of KSCN, a well-known OH· scavenger (en-
try 3).[32b] Worthy of notice is the preferential reactivity of
W10 and Zn5W19 towards uracil glycol, along the oxidative
pathway 2–4 (entries 4, 6 and Figure S2 in supporting infor-
mation). This trend is highlighted by anaerobic experiments
where the hydantoine path is negligible (entries 5, 7). In
general, all POMs exhibit very different behavior with re-
spect to heterogeneous TiO2 which favors the rearrange-
ment of the pyrimidinic cycle (entry 12).
To further address the role of the POM photocatalyst in
controlling both the efficiency and selectivity of the process,
two sets of experiments have been performed. Firstly, the
kinetic dependence of the photooxidation rate on POM
concentration has been determined for PW12 (Figure 2,
Table S1 and Figure S4 in supporting information). Satura-
tion behavior is seen from a plot of the zero-order kobs val-
ues as a function of [PW12]. In particular a first-order de-
pendence is found for [PW12] Ͻ 5×10–4 molL–1 while the
rate levels off at higher POM concentrations reaching a lim-
Figure 1. UV/Vis absorbance of uracil (1) and selected POMs in
water (pH = 1, H2SO4).
determining while the slow step of the process lies within iting value. Moreover, under limiting-rate conditions, an ac-
the POM recycling phase (see Scheme 1 and further dis- cumulation of the reduced heteropolyblue complex is no-
cussion).[41] The corresponding zero-order rate constants ticed, thus indicating that only a steady-state concentration
have been reported in Table 1 which also includes: reactions of the photocatalyst is effectively participating in the turn-
performed under N2 and reference experiments by homo- over regime. Such a “stationary” state is probably due to
geneous Fe/H2O2 systems and heterogeneous TiO2.[42,8] As an equilibration of the rates within the intersecting turnover
expected, all anaerobic photooxidations exhibit a slower routines where the photoexcited POM* undergoes re-
rate and an abatement of rearrangement products (cf. en- duction (substrate/solvent activation) and re-oxidation (di-
tries 2, 5, 7 with 1, 4, 6 at low conversion).[35] A key obser- oxygen activation).[41] Therefore, the experimental condi-
vation is the marked diversity of photocatalytic behavior tions adopted (see footnotes in Table 1) feature the most
registered for the different POMs in terms of oxidation rate convenient loading of the catalyst.
Table 1. POM photocatalyzed oxidation of uracil in water.[a]
Entry
POM[b]
Atm.[c]
t[d]
[min]
Conv.[e]
(%)
k (×10–7)[f]
[ms–1]
Product distribution[g] (%)
2
3
4
5
6
1
PW12
O2
10
40
180
60
10
60
22
100
22
60
25
100
18
30
95
37
28
5
20.0
38
19
72
40
7
12
17
19
8
–
–
–
18
21
9
35
25
16
-
33
22
–
6
–
37
30
57
32
43
–
17
–
–
–
–
–
–
28
–
2
PW12
PW12/SCN–
W10
N2
O2
O2
0.8
3[h]
4
–
14.0
52
37
37
57
36
46
21
83
63
25
18
16
47
20
10
42
44
14
–
–
5
6
W10
Zn5W19
N2
O2
110
20
1.3
5.3
43[i]
–
150
240
240
180
180
30
–
7
Zn5W19
SiW12
ZnW12
H2W12
FeII/H2O2
TiO2
N2
O2
O2
O2
O2
O2
0.9
1.1
10
31
17
-
36
7
8[j]
9
10
11[k]
12[l]
9
25
50
–
–
10
–
9
8
60
[a] Uracil (5×10–3 m); POM (5×10–4 m); water (2 ml, pH = 1, H2SO4), 500 W Hg/Xe lamp; λ Ͼ 300 nm, T = 25 °C. [b] Catalyst legend:
H3PW12O40 (PW12); H4SiW12O40 (SiW12); H6ZnW12O40 (ZnW12); Na6H2W12O40 (H2W12) Na4W10O32 (W10) Na12[WZn3(H2O)2-
(ZnW9O34)2] (Zn5W19). [c] O2 or N2 (1 atm). [d] Reaction time. [e] substrate conversion. [f] Zero-order rate constant. [g] Photooxidation
products as in Scheme 2, product 7 observed only in traces. [h] [KSCN] = 1×10–3 m. [i] Including dehydrated form 8. [j] pH = 2.8. [k]
[Fe] = 1×10–3 m, [H2O2] = 3.3×10–2 m. [l] Heterogeneous photooxidation.
Eur. J. Org. Chem. 2005, 4897–4903
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4899

M. Bonchio, M. Carraro, V. Conte, G. Scorrano
FULL PAPER
to a substantial difference in the mechanism of photocataly-
sis by the three complexes. While the undifferentiated attack
on cis-2 and 3 suggests a free radical pathway dominated
by the highly reactive OH· discrimination between the two
substrates may involve a direct interaction of the substrate
with the POM-based oxidant. Such a working hypothesis is
supported by further experimental evidence involving the
photooxidation kinetics of model alcohols and the use of
radical scavenger probes. More precisely photooxidation of
cyclohexanol 13 and cis-cyclohexan-1,2-diol 14 has been
performed in aqueous acidic media, under analogous con-
ditions, but in the presence of a higher substrate/POM ratio
to favor association phenomena. Photooxidation of 13 and
Figure 2. Plot of zero order kobs vs. [PW12]. In all experiments:
uracil (5×10–3 m); POM (0.5–25×10–4 m); water (2 ml, pH = 1, 14 occurs yielding cyclohexanone 16 and 2-hydroxy-cyclo-
hexanone 17, respectively, as the main products.[45] In all
H2SO4), 500-W Hg/Xe lamp; λ Ͼ 300 nm, T = 25 °C.
cases, the kinetics of the substrate conversion has been ana-
lyzed in order to extract pseudo-first-order rate constants
(Table 3).
Our second aim was to probe the potential of POMs in
tuning the photooxidation selectivity in aqueous media. In
view of this, the kinetics of the two-step oxidation of uracil
cis-diol (cis-2) has been studied in more detail, thus zoom-
Table 3. Photooxidation of model alcohols by POMs.[a]
ing into one of the pathways of the general reaction Scheme Entry
POM
Substrate Additive
kobs ×10–4 [m–1 s–1]
(Scheme 2).
1
PW12
PW12
PW12
Zn5W19
Zn5W19
Zn5W19
W10
13
14
14
13
14
14
13
14
14
–
1.5 0.1
1.2 0.1
0.3 0.1
6.4 0.1
6.9 0.1
5.7 0.1
6.7 0.3
10.6 0.9
7.5 0.8
The reactions catalysed by PW12, W10 and Zn5W19 exhi-
bit uncomplicated kinetics, typical of first order consecutive
reactions (Scheme 2, oxidative pathway from 2 to products
3 and 4), as shown in one representative case (Figure 3). So,
pseudo-first order constants (k1 and k2) have been obtained
by a straightforward fitting of the experimental data
(Table 2).[44]
2
–
3[b]
4
SCN–
–
5
–
SCN–
–
6[b]
7
8
W10
W10
–
9[b]
SCN–
[a] Substrate (30×10–3 m), POM (3×10–4 m), in H2O (2 mL, pH
1.0), pO2 1 atm; T = 25 °C; λ Ͼ 300 nm, Substrate: cyclohexanol
13, cis-cyclohexan-1,2-diol 14. [b] 0.06 m KSCN was added as a
radical scavenger.
The resulting reactivity trend again shows a superior re-
activity of Zn5W19 and W10 towards alcoholic functionali-
ties. Furthermore, addition of KSCN produces a 70% re-
duction of the rate in the PW12-photoassisted process (entry
3), while only a minor effect (up to 28%) results in the pres-
ence of W10 or Zn5W19 (entries 6 and 9). This probe sheds
light on the mechanistic gap separating the two classes of
photocatalysts and is consistent with the prevailing radical
nature of the PW12-initiated oxidation.[34] As far as the
other two systems are concerned, whilst the photooxidation
performance of Zn5W19 represents a novel observation,[46]
an extensive collection of literature studies has been pub-
lished dealing with W10 photocatalysis. In particular, it has
been shown that the competent photoreactant is an ex-
tremely reactive transient, generated from the initially ex-
cited LMCT state [W10O32]4–*.[47] This latter species, re-
ferred to as “wO”, exhibits an oxyradical-like character due
to the presence of an electron-deficient oxygen center, and
reacts with alcohols mainly through direct hydrogen-atom
abstraction (HA).[47] Moreover, several pieces of evidence
indicate that W10 can undergo strong pre-complexation and
Figure 3. Photooxidation kinetics of uracil cis-diol by W10. Experi-
mental data: cis-2 (᭿), 3 (♦),4 (᭡), and calculated curves. See foot-
note in Table 2.
Table 2. Kinetic constants for the two-step photooxidation of cis-2
by POMs.[a]
POM,
k1 ×10–4 [m–1 s–1]
k2 ×10–4 [m–1 s–1]
PW12
Zn5W19
W10
2.7 0.1
7.8 0.4
25.6 1.6
2.6 0.1
3.1 0.2
5.0 0.4
[a] Substrate (5×10–3 m); POM (5×10–4 m); water (2 ml, pH = 1,
H2SO4), 500 W Hg/Xe lamp; λ Ͼ 300 nm, T = 25 °C.
Table 2 highlights a direct POM influence on reaction that substrate pre-association is a general feature in photo-
rates. In particular, PW12 promotes the two oxidative steps catalyzed reactions by POMs.[8,47] All of these observations
with similar speed (k1 Ϸ k2), at variance with W10 and allow us to address the observed dependence of the photo-
Zn5W19 yielding, in both cases, k1 Ͼ k2. This result points oxidation outcome on the POM system. The composition,
4900
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4897–4903

Aerobic Photooxidation in Water by Polyoxotungstates: The Case of Uracil
FULL PAPER
ondary focussing lens to maximize the incident light and cut-off
filter allowing irradiation at λ Ͼ 300 nm. The lamp features a con-
tinuous emission spectrum from 250 nm to the near IR, overlapped
by Hg discontinuous emission in the UV field. For the scaled-up
reaction, a high-pressure immersion Hg lamp (Helios Italquartz,
250 W) was used. The software Scientist (Micromath) was em-
ployed to fit experimental kinetics providing values for rate con-
stants and errors.
shape, electronic structure and charge density of the POM
photocatalyst determine the unique character of the active
“wO” center, the nature of its interaction with hte substrate,
the stoichiometry and stability of substrate-POM com-
plexes before reaction and the rate of some fundamental
steps leading to substrate oxygenation.
Photooxidation Procedure: Photocatalytic experiments were carried
out employing a standard spectrophotometric quartz cell hosted in
a thermostatted holder (25 °C), placed at 8.5 cm distance from the
focusing lens, to collect all the focussed radiation. The reaction
solution (H2O, 2 mL, pH 1.0) containing uracil (5 mm,
1×10–2 mmol) and the catalyst (0.5 mm), was irradiated at λ Ͼ
300 nm, under magnetic stirring and oxygen (or nitrogen). The re-
actions were monitored over time at 217 nm by HPLC analysis.
Reaction aliquots (50 μL) were withdrawn at selected times and
diluted (450 μL) with a standard solution of 6-methyluracil in
water, as chromatographic standard (0.283 mm). Uracil and its oxi-
dation products were identified according to their retention times
and absorption spectra by comparison with authentic samples (ura-
cil 19.3 min, 6-methyl uracil 25.2 min). The stability of the POM
photocatalysts has been verified by FT-IR analysis of the recovered
complex after precipitation with tetrabutylammonium bromide.
Photooxidation of alcohols was performed following a similar pro-
cedure, using a reaction mixture (2 mL H2O, pH 1.0) containing
the substrate (30 mm, 6×10–2 mmol) and the catalyst (0.3 mm). The
reactions were monitored over time by quantitative GLC-analysis.
Reaction aliquots (50 μL) were extracted with a dichloromethane
solution (250 μL) containing biphenyl (0.9 mmolL–1) as internal
standard, and dried with anhydrous MgSO4.
Conclusions
The results reported herein show a promising perspective
for POMs as photocatalysts in aerobic oxidations per-
formed in water. Because the generally accepted mechanism
involves the formation of radicals eliciting autoxidation cy-
cles, the ability to control POM-dependent intermediates
or relative rates of the oxygenation steps offers the unique
opportunity to tame the selectivity of these processes by an
appropriate choice of the polyoxometalate properties. In
this respect polyoxometalate synthesis can be controlled by
an extreme variety of parameters among which are number
and type of metal addenda, central heteroatom, and coun-
terion type: thus, providing a rich pool of complexes. Fu-
ture experiments will focus on catalyst design and mecha-
nism elucidation with the ultimate aim to devising new se-
lective systems for dioxygen activation in water or alterna-
tive media including fluorinated solvents.
Experimental Section
Synthesis and Analyses of Uracil Oxidation Products: Main pro-
ducts were isolated from a scaled-up photooxidation performed
with an immersion Hg lamp in a thermostatted photochemical re-
actor. Uracil (2.4 g, 21 mmol) in water (800 mL, pH 1, H2SO4) was
irradiated in the presence of H3PW12O40 (3.5 g, 1.2 mmol). Oxygen
(1 atm) was provided through a tank connected to the system and
irradiation was stopped after 90 h. The catalyst was precipitated
by adding CsSO4 and removed by filtration through celite. After
neutralization, the reaction mixture was dried and purified by
chromatography on a silica column. Elution was performed with
a mixture of ethyl acetate/2-propanol/water = 75:16:9 or CH3Cl/
CH3OH = 82:18, TLC monitoring and spot visualization by using
a UV lamp or Tollens’ reagent. The products were isolated from the
reaction mixture and characterized as detailed below. Experimental
data were compared with those of authentic samples, either com-
mercially available or synthesized according to literature pro-
cedures.
General Remarks: The following commercial reagents: tetrabu-
tylammonium bromide, 6-methyluracil (2,4-dihydroxy-6-methyl-
pyrimidine), uracil, isobarbituric acid (Aldrich); parabanic acid,
uric acid, 2,4-dihydroxypyrimidine, tetrahydrate alloxan, phos-
photungstic
acid
(H3PW12O40),
sodium
polytungstate
(Na6H2W12O40), tungstosilicic acid (H4SiW12O40), (Fluka) have
[48]
been used as received. Na4W10O32,
Na12[WZn3(H2O)2-
[50]
(ZnW9O34)2][49] and H6ZnW12O40 have been prepared following
a literature procedure. HPLC analyses were performed on a Spec-
trasystem P2000 instrument equipped with an UV SPD-6A detec-
tor, with a CR-3A integrator (Shimadzu) and using a reverse phase
C18 column (Aqua or Luna, Phenomenex) under gradient elution
(H2O/CH3CN
=
100:0×8 min, 80:20×20 min, flow rate
0.35 mLmin–1). GC analyses were performed with a Hewlett–Pack-
ard 5890 Series II instrument, equipped with flame ionisation de-
tector and capillary column HP 5 (30 m; I.D. 0.32 mm; 0.25 μm
film thickness, temperature program: at 70 °C×3 min, 15 °C/min,
13
at 250 °C×3 min). 1H and
C NMR spectra were collected at Uracil trans-Diol (trans-5,6-Dihydroxydihydropyrimidine-2,4-dione)
25 °C with a Bruker AC 250 instrument. MS (ESI) analyses were
done with a Finnigan MAT LCQ instrument. FT-IR spectra were
recorded with a Perkin–Elmer 1600 series instrument. Solutions of
uracil, its oxidation products (1 mm) in water (pH 1, H2SO4) and
(trans-2):[51] 1H NMR ([D6]DMSO): δ = 10.08 (s, 1 H, 3-H), 8.09
(s, 1 H, 1-H), 6.24 (s, broad, 1 H, 6-OH), 6.16 (s, broad, 1 H, 5-
OH), 4.52 (s, 1 H, 6-H), 3.67 (s, 1 H, 5-H) ppm. 13C NMR ([D6]
DMSO): δ = 171.0, 152.4, 75.9, 69.2 ppm. Ret. time (Luna) = 7.8Ј;
Ret. time (Aqua) = 8–8.5Ј; Rf (Tollens) = 0.8–0.9.
of the POM photocatalysts (0.1 mm, PW12 ε300nm
=
7.2×103 m–1 cm–1; W10 ε300nm = 8.1×103 m–1 cm–1; Zn5W19 ε300nm
= 1.5×104 m–1 cm–1; SiW12 ε300nm = 6.7×103 m–1 cm–1; H2W12
ε300nm = 6.6×103 m–1 cm–1) were analyzed by UV/Vis at 25 °C with
a Perkin–Elmer Lambda 45 spectrometer.
Uracil cis-Diol (cis-5,6-Dihydroxydihydropyrimidine-2,4-dione) (cis-
2):[52] A mixture containing uracil (567 mg, 5.06 mmol) and anhy-
drous MgSO4 (5 g), dispersed in acetone (30 mL), was sonicated
and cooled to 0 °C. A KMnO4 solution (400 mg, 2.53 mmol) in
acetone (30 mL) was added dropwise whilst stirring and left at 0 °C
for a ca. 5 h. The reaction mixture was then filtered through celite
to remove inorganic salts and quenched with an aqueous solution
of sodium metabisulfite to reduce the unreacted KMnO4. The
In photooxidation experiments, irradiation was performed with a
light source housing (Oriel Instruments), equipped with a 500 W
Hg-Xe arc lamp, 200–500 W power supply, F/1.5 UV grade fused
silica condenser, liquid (water) filter to absorb IR radiations, a sec-
Eur. J. Org. Chem. 2005, 4897–4903
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4901

M. Bonchio, M. Carraro, V. Conte, G. Scorrano
FULL PAPER
brown MnO2 precipitate was separated by centrifugation and the
solvent was removed under vacuum. The product was purified by
chromatography (column or preparative TLC), using ethyl acetate/
Supporting Information: UV/Vis spectra of uracil and of its oxi-
dation products (Figure S1); product evolution as a function of
time obtained for the W10 photocatalyzed oxidation (Figure S2);
zero-order kinetics as a function of PW12 concentration (Table S1
and Figure S4); fitting of the two-step oxidation kinetics of uracil
cis-diol by PW12, and Zn5W19 (Figures S4 and S5).
1
2-propanol/water = 75:16:9. H NMR ([D6]DMSO): δ = 10.05 (s,
1 H, 3-H), 8.13 (s, 1 H, 1-H), 6.09 (d, J = 4.1 Hz, 1 H, 6-OH), 5.48
(d, J = 6.2 Hz, 1 H, 5-OH), 4.62 (dd, J = 3.8, JЈ = 4.1 Hz, 1 H, 6-
H), 4.18 (dd, J = 6.2, JЈ = 3.8 Hz, 1 H, 5-H) ppm. 13C NMR ([D6]
DMSO): δ = 172.0, 153.1, 74.3, 69.0 ppm. Ret. time (Luna) = 7.9–
8.2Ј; Ret. time (Aqua) = 7.7–8.0Ј; Rf (Tollens) = 0.5.
Acknowledgments
Isodialuric Acid (6-Hydroxydihydropyrimidine-2,4,5-trione) (3):[53]
Br2 (550 mg, 3.50 mmol) in water (10 mL) was added whilst stirring
to a solution of isobarbituric acid (450 mg, 3.44 mmol) in water
(20 mL), maintained at 30 °C, until the solution remained orange.
The volume was reduced to 3.5 mL and an equivalent volume of
diluted H2SO4 was added. The mixture was concentrated to 2 mL
and then poured into an ice-water bath. The solid was filtered,
washed with cold water (5 mL), ethyl alcohol (10 mL) and ethyl
ether (10 mL). 1H NMR ([D6]DMSO): δ = 10.06 (s, 1 H, 3-H), 8.11
(d, J = 4.8 Hz, 1 H, 1-H), 6.75 (s, 1 H, 5-OH), 6.43 (s, 1 H, 5-OH),
5.89 (d, J = 4.0 Hz, 1 H, 6-OH), 4.34 (dd, J = 4.0, JЈ = 4.8 Hz, 1
H, 6-H) ppm, hydrated form. 13C NMR (D2O/rif. DMSO): δ =
132.7, 116.4, 50.9 ppm. Ret. time (Aqua, Luna) = 6.6–6.9Ј; Rf (Tol-
lens) = 0.7.
Alloxan (Pyrimidine-2,4,5,6-tetraone) (4 and 9): 1H NMR ([D6]-
DMSO): δ = 11.25 (s, 2H, 1-H and 3-H), 7.56 (s, broad, OH, tetra-
hydrate form) ppm. 13C NMR (D2O/ref. DMSO): δ = 170.7, 151.9,
86.2 ppm. MS (ESI+, CH3OH/H2O) m/z = 214 [M + 4H2O]+. Ret.
time (Luna) 9.0–11.1Ј; Ret. time (Aqua) = 8.7–8.8Ј; Rf (UV, Tollens)
= 1.1–1.2.
1-Formyl-5-hydroxyhydantoin (5):[35] Uracil cis-diol (150 mg,
1.03 mmol) was dissolved in water (65 mL) and reacted with so-
dium metaperiodate (0.1 m solution, 250 mL, 25 mmol). After a few
minutes at 30 °C, the mixture was carefully concentrated at 40 °C
and diluted with hot ethanol. The precipitated was removed from
the hot mixture and the solvent was removed under vacuum. The
product was purified by preparative TLC silica plate.1H NMR ([D6]
DMSO): δ = 11.71 (s, broad, 1 H, 3-H), 8.94 (s, 1 H, CHO), 7.53
(d, J = 6.8 Hz, 1 H, 5-OH), 5.47 (d, J = 6.8 Hz, 1 H, 5-H) ppm.
13C NMR ([D6]DMSO): δ = 171.3, 159.1, 154.4, 76.0 ppm. Ret.
time (Luna) = 28.2Ј; Ret. time (Aqua) = 12.1–12.4Ј; Rf (Tollens) =
1.6.
Financial support from the Italian National Research Council is
gratefully acknowledged.
[1] a) D. H. R. Barton, A. E. Martell, D. T. Sawyer, The Activation
of Dioxygen and Homogeneous Catalytic Oxidations Plenum
Press, New York, 1993; b) C. L. Hill, Nature 1999, 401, 436–;
c) A. Albini, M. Fagnoni, Chemisty Today 2004, 22, 36–38.
[2] a) E. L. Clennan, Tetrahedron 2000, 56, 9151–9179; b) A. Al-
bini, M. Fagnoni, Green Chem. 2004, 6, 1–6.
[3] M. Bonchio, T. Carofiglio, M. Carraro, R. Fornasier, U. Tonel-
lato, Org. Lett. 2002, 4, 4635–4637.
[4] B. Meunier, A. Sorokin, Acc. Chem. Res. 1997, 30, 470–476.
[5] O. Legrini, E. Oliveros, A. M. Braun, Chem. Rev. 1993, 93,
671–698.
[6] a) R. Bonnett, Chemical Aspects of Photodynamic Therapy
Gordon and Breach Science: Amsterdam, 2000; b) R. Bonnet,
G. Martinez, Tetrahedron 2001, 57, 9513–9547 and references
cited therein.
[7] C. L. Hill, C. M. Prosser-McCartha, in: Photosensitization and
Photocatalysis Using Inorganic and Organometallic Compounds
(Eds.: K. Kalyanasundaram, M. Grätzel), Kluwer Academic
Publishers: Dordrecht, The Netherlands, 1993, pp. 307–330.
[8] a) A. Hiskia, A. Mylonas, E. Papaconstantinou, Chem. Soc.,
Rev. 2001, 30, 62–69; b) E. Papaconstantinou, Chem. Soc., Rev.
1989, 18, 1–31.
[9] A. Maldotti, A. Molinari, R. Amadelli, Chem. Rev. 2002, 102,
3811–3836.
[10] M. A. Fox, R. Cardona, E. Gailard, J. Am. Chem. Soc. 1987,
109, 6347–6354.
[11] A. Hiskia, E. Papaconstantinou, Polyhedron 1988, 7, 477–481.
[12] M. Bonchio, M. Carraro, G. Scorrano, E. Fontananova, E.
Drioli, Adv. Synth. Catal. 2003, 345, 1119–1126.
[13] M. Bonchio, M. Carraro, G. Scorrano, A. Bagno, Adv. Synth.
Catal. 2004, 346, 648–654.
5-Hydroxyhydantoin (5-Hydroxyimidazolidine-2,4-dione) (6):[32c] Pa-
rabanic acid (100 mg, 0.88 mmol) was dissolved in water (10 mL)
and reacted with a solution of NaBH4 (16.6 mg) prepared in water
(100 μL) with NaOH (2 m, 20 μL) whilst stirring for 1 h. A few
droplets of dilute HCl were added to quench the hydride excess.
Gel chromatography on silica column yielded the expected product.
1H NMR ([D6]DMSO): δ = 10.57 (s, 1 H, 3-H), 8.28 (s, 1 H, 1-H),
6.70 (d, J = 8.5 Hz, 1 H, 5-OH), 5.07 (d, J = 8.5 Hz, 1 H, 5-
H) ppm. 1H NMR ([D6]DMSO/D2O = 3:1): δ = 5.10 (s, 1 H, 5-
H) ppm. 13C NMR ([D6]DMSO): δ = 174.5, 156.8, 77.1 ppm. MS
(ESI+, CH3OH/H2O) m/z = 97 [M+ – H2O]. Ret. time (Luna) 8.4–
9.6Ј; Ret. time (Aqua) = 8.8Ј; Rf (Tollens) = 1.2.
Parabanic Acid (Imidazolidine-trione) (7): 1H NMR ([D6]DMSO):
δ = 11.73 (s, 2 H, 1-H and 3-H) ppm. 13C NMR ([D6]DMSO): δ =
160.0; 154.9 ppm. Ret. time (Luna) = 27.7–28.0Ј; Ret. time (Aqua)
= 13.8Ј; Rf (UV) = 1.7.
Isobarbituric Acid (5-Hydroxy-1H-pyrimidine-2,4-di-one) (8): 1H
NMR ([D6]DMSO): δ = 11.11, (s, 1 H, 3-H), 10.13 (s, broad, 1 H,
1-H), 8.37 (s, broad, 1 H, 5-OH), 6.83 (d, J = 5.5 Hz, 1 H, 6-
H) ppm. Ret. time (Luna) 13.5Ј; Ret. time (Aqua) = 15Ј; Rf (UV,
Tollens) = 0.8–0.9.
[14] M. T. Pope, Heteropoly and Isopoly Oxometalates, Springer-
Verlag: New York, 1983.
[15] “Polyoxometalates” Special Issue: C. L. Hill, Chem. Rev. 1998,
98, 1–387.
[16] M. T. Pope, A. Muller Polyoxometalate Chemistry. From Top-
ology via Self-assembly to Applications Kluwer Academic Pub-
lishers: Dordrecht, The Netherlands, 2002.
[17] K. Nomiya, Y. Sugie, K. Amimoto, M. Miwa, Polyhedron 1987,
6, 519–524.
[18] D. Dimotikali, E. Papaconstantinou, Inorg. Chim. Acta 1984,
87, 177–180.
[19] C. L. Hill, Synlett 1995, 127–132.
[20] C. L. Hill, Z. Zheng, Chem. Commun. 1998, 2467–2468.
[21] R. C. Chambers, C. L. Hill, Inorg. Chem. 1989, 28, 2509–2511.
[22] A. Mylonas, A. Hiskia, E. Androulaki, D. Dimotikali, E. Pa-
paconstantinou, Phys. Chem. Chem. Phys. 1999, 1, 437–440.
[23] H. Hiskia, E. Papaconstantinou, Inorg. Chem. 1992, 31, 163–
167.
[24] R. Akid, J. Darwent, J. Chem. Soc., Dalton Trans. 1985, 395–
399.
[25] C. L. Hill, D. A. Bouchard, J. Am. Chem. Soc. 1985, 107, 5148–
5157.
[26] M. Misono, H. Einaga, Bull. Chem. Soc. Jpn. 1997, 70, 1551–
1557.
4902
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4897–4903

Aerobic Photooxidation in Water by Polyoxotungstates: The Case of Uracil
FULL PAPER
[27] T. Yamase, R. Sasaki, T. Ikawa, J. Chem. Soc., Dalton Trans. [42] V. Conte, F. Di Furia, S. Moro, Gazz. Chim. Ital. 1995, 125,
1981, 628–634.
563–568.
[28] R. R. Ozer, J. L. Ferry, J. Phys. Chem. B 2000, 104, 9444–9448.
[29] C. L. Hill, R. F. Renneke, L. A. Combs, New J. Chem. 1989,
13, 701–706.
[43] An analogous reactivity trend, which is hardly predicted simply
on the basis of the POM spectral properties (Figure 1), has
been previously observed in the photocatalyzed aerobic oxi-
dation of phenol and p-cresol in water, see A. Mylonas, E. Pa-
pacostantinou, Polyhedron 1996, 15, 3211–3217.
[44] Fittings for PW12 and Zn5W19 kinetics are reported in the sup-
porting information (Figures S4 and S5).
[30] R. F. Renneke, M. Pasquali, C. L. Hill, J. Am. Chem. Soc. 1990,
112, 6585–6594.
[31] L. A. Combs-Walker, C. L. Hill, J. Am. Chem. Soc. 1992, 114,
938–946.
[32] a) M. I. Al-Sheikhly, A. Hissung, H. P. Schuchmann, M. N.
Schuchmann, C. Von Sonntag, J. Chem. Soc., Perkin Trans. 2
1984, 601–608; b) C. L. Greenstock, J. W. Hunt, Trans. Faraday
Soc. 1969, 3279–3287; c) R. Ducolomb, J. Cadet, R. Teoule,
Bull. Soc. Chim. Fr. 1973, 3, 1167–1174.
[33] C. J. Burrows, J. G. Muller, Chem. Rev. 1998, 98, 1109–1151.
[34] S. Kim, H. Park, W. Choi, J. Phys. Chem. B 2004, 108, 6402–
6411 and references cited therein.
[45] Detected products account for 20–50% of substrate conver-
sion.
[46] Under the conditions adopted, Zn5W19 undergoes a fast struc-
ture rearrangement. Preliminary 183W NMR spectroscopic ex-
periments, performed at higher POM concentrations, show the
evolution of the original 10 signal spin system to a spectrum
dominated by one resonance at –102.50 ppm, whose assign-
ment is unknown so far.
[35] T.-J. Yu, R. G. Sutherland, R. E. Verrall, Can. J. Chem. 1980,
[47] a) C. Tanielian, R. Seghrouchni, C. Schweitzer, J. Phys. Chem.
A 2003, 107, 1102–1111 and references cited therein; b) C. Tan-
ielian, Coord. Chem. Rev. 1998, 178–180, 1165–1181; c) I. Tex-
ier, J. A. Delaire, C. Giannotti, Phys. Chem. Chem. Phys. 2000,
2, 1205–1212.
58, 1909–1915.
[36] T. Wada, H. Ide, S.-I. Nishimoto, T. Kagiya, Chem. Lett. 1982,
1041–1044.
[37] M. N. Schuchmann, C. Von Sonntag, J. Chem. Soc., Perkin
Trans. 2 1983, 1525–1531.
[38] R. L. Willson, Int. J. Radiat. Biol. 1970, 17, 349–358.
[39] A disproportionation mechanism has been suggested to ac-
count for the formation of 2 under conditions of deareated
sonolysis[35] or radiolysis.[36] Due to the photocatalytic proper-
ties of POMs, the production of OH· radicals can also be en-
abled in deoxygenated reactions.[34]
[48] D. C. Duncan, T. L. Netzel, C. L. Hill, Inorg. Chem. 1995, 34,
4640–4646.
[49] C. M. Tourné, G. F. Tourné, F. Zonnevijlle, J. Chem. Soc., Dal-
ton Trans. 1991, 143–155.
[50] D. H. Brown, J. A. Mair, J. Chem. Soc. 1958, 2597–2599.
[51] B. S. Hahn, S. Y. Wang, Bioch. Biophys. Res. Commun. 1977,
77, 947–951.
[40] A Russel type bimolecular decay of the peroxo-radicals is ex-
pected to yield 2 and 3 in equal amounts (not shown in
Scheme 2). The two-step overoxidation of 2 to 3 and 4 or paral-
lel decomposition routes are responsible for different product
distributions generally observed (Table 1).
[41] K. Nomiya, T. Miyazaki, K. Maeda, M. Miwa, Inorg. Chim.
Acta 1987, 127, 65–69.
[52] a) H. Hayatsu, S. Iida, Tetrahedron Lett. 1969, 10, 1031–1034;
b) M. H. Benn, B. Chatamra, A. S. Jones, J. Chem. Soc. 1960,
1014–1017.
[53] R. Behrend, O. Roosen, Justus Liebigs Ann. Chem. 1889, 251,
242–247.
Received: June 9, 2005
Published Online: September 21, 2005
Eur. J. Org. Chem. 2005, 4897–4903
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4903