Dimethylsulfoxide as a kinetic booster for the chemical generation of singlet oxygen in methanol

Article abstract of DOI:10.1016/j.tetlet.2010.10.022

The kinetic booster effect of dimethylsulfoxide on the chemical generation of singlet oxygen, ¹O₂, from the disproportionation of hydrogen peroxide catalyzed by molybdate ions in methanol has been evidenced by detection of the IR luminescence of ¹O₂ at 1270 nm and by ⁹⁵Mo NMR spectroscopy. DMSO interacts rapidly, through a direct oxygen transfer with the stable tetraperoxomolybdate Mo(O₂) ₄^2-, leading to DMSO₂ and to the unstable triperoxomolybdate Mo(0₂)₃^2-, which releases ¹O₂. The procedure was applied to accelerate the dark singlet oxygenation of β-citronellol and α-terpinene.

Full text of DOI:10.1016/j.tetlet.2010.10.022

Tetrahedron Letters 51 (2010) 6531–6534
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Dimethylsulfoxide as a kinetic booster for the chemical generation
of singlet oxygen in methanol
Marion Collinet-Fressancourt a,b, Nathalie Azaroual a,c, Jean-Marie Aubry a,b, Véronique Nardello-Rataj a,b,
⇑
a
Université Lille Nord de France, F-59000 Lille, France
Université Lille1 and ENSCL, EA 4478, Chimie Moléculaire et Formulation, F-59655 Villeneuve d’Ascq Cedex, France
b
c
CNRS UMR 8516, UDSL, Faculté de Pharmacie, BP 83, 59006 Lille, France
a r t i c l e i n f o
a b s t r a c t
The kinetic booster effect of dimethylsulfoxide on the chemical generation of singlet oxygen, 1O , from
Article history:
2
Received 6 August 2010
Revised 30 September 2010
Accepted 5 October 2010
Available online 29 October 2010
the disproportionation of hydrogen peroxide catalyzed by molybdate ions in methanol has been evi-
denced by detection of the IR luminescence of 1
95
O
2
at 1270 nm and by Mo NMR spectroscopy. DMSO
2ꢀ
interacts rapidly, through a direct oxygen transfer with the stable tetraperoxomolybdate MoðO
2
Þ
, lead-
4
2ꢀ
, which releases 1
-terpinene.
ing to DMSO
2
and to the unstable triperoxomolybdate MoðO
2
Þ
2
O . The procedure was
3
applied to accelerate the dark singlet oxygenation of b-citronellol and
a
Keywords:
Ó 2010 Elsevier Ltd. All rights reserved.
Singlet oxygen
Dimethylsulfoxide
Molybdate
Catalysis
Oxidation
1
. Introduction
Hence, in the present work, we reexamine the mechanism of
1O
generation from the H
O
/MoO /DMSO system in methanol
2ꢀ
2
2
2
4
Singlet oxygen, 1
O
(1
D
), can be chemically generated from
by monitoring the formation of O
1
through its IR chemilumines-
2
g
2
7
hydrogen peroxide disproportionation catalyzed by molybdate
ions in alkaline aqueous solutions. Four peroxomolybdates,
cence and by detecting the intermediate peroxomolybdates in-
1
95
volved in the process by Mo NMR. Then, we take advantage of
2
ꢀ
MoO4ꢀnðO
2
Þ
with n = 1, 2, 3, 4, are formed under these condi-
the accelerating effect of DMSO to speed up the ‘dark’ singlet oxy-
genation of a-terpinene and b-citronellol in MeOH.
n
2
ꢀ
tions but the triperoxomolybdate, MoOðO
2
Þ
, is the main precur-
3
1
2ꢀ
2
sor of O
2
whereas the tetraperoxomolybdate MoðO
2
Þ
is stable.
4
This reaction has been implemented in organic solvents, espe-
cially in dipolar protic solvents, such as methanol, for the ‘dark’
singlet oxygenation of hydrophobic substrates. In such solvents,
2. Experimental
3
2.1. NIR-chemiluminescence spectrometer
a continuous and slow addition of H
2 2
O is required for an optimal
2
ꢀ
reaction kinetics in order to favor the formation of MoOðO
2
Þ
and
All measurements of light emission attributable to the chemilu-
3
1
2
ꢀ
to avoid that of MoðO
2
Þ
. This procedure has been scaled up by
minescent transition of O
2
to the triplet ground state were carried
4
3
DSM company in 10 m reactors to produce rose oxide from b-cit-
out in the near infrared (NIR) at 1270 nm with a home-made
instrument fitted with a liquid nitrogen-cooled germanium photo-
diode detector (Model EO-817L, North Coast Scientific Co., Santa
Rosa, CA, USA) sensitive in the spectral region from 800 to
ronellol in ethylene glycol.4
On the other hand, the screening of various solvents had re-
vealed a puzzling behavior for DMSO as the chemiluminescence
1
2
7
signal of O
2
was intense (ꢁ285 mV) but much shorter (ꢁ2 min)
1800 nm with a detector of 0.25 cm and a sapphire window.
than in other solvents, strongly suggesting that DMSO interferes
with the chemical generation of O
1
1
3
or with O
2
itself. However,
2.2. Typical luminescence experiments
2
several papers report that sulfoxides do not react chemically with
1
to give the corresponding sulfone,5 while they interact with
O
2
A solution of Na
MeOH (10 mL) was placed in the thermostated brass cell holder
2
MoO
4
ꢂ2H
2
O (121.0 mg, 0.05 M, 0.5 mmol) in
6
peroxomolybdates at a very low reaction rate.
at 25 °C. When the background noise of the IR signal was stabi-
lized, H
intensity of the luminescence signal was recorded. After 10 min,
DMSO (360 L, 0.5 M, 5 mmol) was added in the solution.
2 2
O (50%, 600 lL, 1 M, 10 mmol) was introduced and the
⇑
Corresponding author. Tel.: +33 3 20 33 63 69; fax: +33 3 43 42 60.
E-mail address: veronique.rataj@univ-lille1.fr (V. Nardello-Rataj).
l
0
040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.10.022

6
532
M. Collinet-Fressancourt et al. / Tetrahedron Letters 51 (2010) 6531–6534
the signal is almost recovered since the product formed, that is,
ascaridole, does not interfere with the formation or the decompo-
sition of peroxomolybdates. On the contrary, when DMSO is added
(
curve C), the signal is dramatically different resulting in a 10-fold
1
higher intensity corresponding to a sudden burst of
2
O produc-
tion. This result strongly suggests that DMSO directly interacts
2
ꢀ
2ꢀ
3
with the red-brown MoðO
2
Þ
giving the orange MoOðO
2
Þ
,
4
1
which, in turn, rapidly decomposes into O
2
. It is noteworthy that
as evidenced by its life-
D
time in this solvent (s = 19 ls). Bunton and co-workers. reported
1
DMSO does not react chemically with O
2
8
the fading of a tetraperoxomolybdate solution by the addition of
thioanisole and explained it by a direct oxygen transfer from
2
ꢀ
MoðO
2
Þ
to the sulfide leading to its oxidation into the corre-
4
6
sponding sulfoxide. However, in contrast to our observation with
DMSO, they also reported that the oxidation of the sulfoxide itself
by the tetraperoxomolybdate into the corresponding sulfone is
very slow. DMSO appears then much more reactive toward
Figure 1. IR luminescence at 1270 nm of 1
containing Na MoO 0.05 M and H 1 M in MeOH alone (A) or after addition of
terpinene 0.25 M (B) or DMSO 0.25 M (C).
2
O generated at 25 °C from solutions
2
4
2
O
2
a-
2
ꢀ
MoðO
2
Þ
than the aryl sulfoxide. The main characteristics of the
4
2
.3. Oxidation experiments
chemiluminescence curves as a function of DMSO concentration
are reported in Table 1.
A solution of Na
106 L, 1.5 mmol), and the substrate (1.5 mmol) in MeOH
2.5 mL) was placed in a thermostated water bath at 25 °C, after
(50%, 180 L, 3 mmol) was added. The or-
ange-red reaction medium was stirred until it became almost col-
2
MoO
4
ꢂ2H
2
O (36.3 mg, 0.15 mmol), DMSO
1
(
(
l
3.2. Mechanism of O
2
production
which aqueous H
2
O
2
l
The disappearance of DMSO (0.25 and 0.5 M) was followed by
H NMR in methanol-d containing Na MoO (0.05 M) and H
(1 M) (see Supplementary data). The reaction is first order in each
reactant and leads to the total conversion of DMSO into DMSO
after 30 min. Mo NMR experiments were then carried out in
CH OD to determine the nature of the peroxomolybdates involved
1
4
2
4
2 2
O
1
orless while the disappearance of substrate was followed by
NMR analysis of the reaction mixture.
H
2
9
5
3
. Results and discussions
3
in the process.
4 2 4
Tetrabutylammonium salt, (NBu ) MoO , was used to increase
3
.1. Effect of DMSO on singlet oxygen production from H
2
O
2
/
2
ꢀ
the catalyst solubility in methanol and the medium was main-
tained at 0 °C to slow down the peroxomolybdate decomposition.
NMR spectra reported in Figure 2 clearly show that the addition
MoO4
1
The influence of the addition of an efficient trap of O
2
,
a-ter-
2
ꢀ
of DMSO to the red-brown solution of MoðO
2
Þ
(spectrum F)
triperoxomolybdate
(spectrum E). As a reference, curve D shows a mixture
pinene, and of DMSO on the IR-luminescence signal arising from
4
2
ꢀ
gives,
almost
instantaneously,
the
the H
At high H
stable MoðO
2 2
O /MoO4 system in MeOH was first studied (Fig. 1).
2ꢀ
O
2
/Mo ratios, MoO42 is mainly converted into the
ꢀ
MoOðO Þ
3
2
2
2
ꢀ
2
ꢀ
2ꢀ
of tri- and tetra-peroxomolybdates obtained by mixing MoO
2
Þ
4
, beside a minor amount of MoOðO
2
Þ
3
which is
4
with a lower concentration of H
2 2
O . These findings led us to pro-
responsible for the weak luminescence (ꢁ18 mV) (curve A). As
pose the mechanism described in Scheme 1 accounting for the
the reaction proceeds, H
2
ꢀ
O
2
is slowly consumed and the equilib-
1
2ꢀ
boosting of O2 from the H O /MoO
4
2
2ꢀ
system in MeOH in the
rium between MoðO
2
Þ
and MoOðO
2
Þ
is shifted toward the
2
2
4
3
presence of DMSO.
less peroxidized intermediate. Accordingly, the dark red solution
2
ꢀ
fades into orange while the 1
The quenching of MoðO Þ
by DMSO probably occurs through
O
2
signal progressively increases in
2
4
a direct oxygen transfer from one peroxo group to the sulfoxide as
such a way that a maximum of intensity (ꢁ55 mV) is reached at
5 h. When -terpinene is added to the medium (curve B), the sig-
nal sharply drops reflecting the chemical trapping of
6
reported by Bunton in the case of thioanisole. Kinetically, the
ꢁ
a
1
addition of DMSO leads to an increase of the decomposition rate
of H O by a factor of 3–7 depending on the added concentration.
2 2
O
2
by the
substrate as expected. Once all
a-terpinene has been consumed,
Table 1
Effect of DMSO on the chemiluminescence signal of 1
O
generated from the H
/MoO4 system in methanol. Conditions: [Na
2ꢀ
= 0.05 M, [H
= 1 M, 25 °C
2
2
O
2
2
MoO
4
]
0
2
O
2
]
0
f
1O
yieldg (%)
V 1O i
ꢀ1
Entry
[DMSO] (M)
Curve
A
D
t
end (min)
Area (mV min)
2
2
(mM min )
1
2
3
4
5
6
0
0
0.25
0.5
0.5
336
337
117
59
60
61
7320
2550
5290
3800
3830
3730
70h
—
51
36
37
36
1.0
1.0
2.2
3.0
3.0
2.9
d
B
a
C
a
e
b
e
e
2 x 0.25c
a
b
c
d
e
f
DMSO added 10 min after H
DMSO added before H
DMSO added at 10 and 20 min after H
Addition of -terpinene 0.25 M.
Curves presented in Supplementary data.
Corresponding to a residual signal intensity equal to 5 mV.
2 2
O .
2 2
O .
2 2
O .
a
g
h
i
1
[
O
2
]
cumulative/0.5ꢃ[H
2 2 o
O ] calculated from the underlying area of the luminescence curve.
According to Ref. 3.
1
[
2 cumulative end
O ] /Dt .

M. Collinet-Fressancourt et al. / Tetrahedron Letters 51 (2010) 6531–6534
6533
Table 2
Effect of DMSO on the peroxidation of b-citronellol 0.20 M and
the H
M, conversion P95%, T = 25 °C
a
-terpinene 0.25 M by
2
ꢀ
] = 1a or 2b
2 2 2 4 2 2
O /MoO4 system in MeOH. Conditions: [Na MoO ] = 0.05 M, [H O
ꢀ
Entry
Substrate
1a
(DMSO) (M)
D
t
95% (min)
TOF (h 1
)
1
2
3
4
—
0.25
—
0.25 ꢃ 2
450
165
850
220
0.63
1.73
0.29
1.05
2b
a
One addition of H
Two additions of H
2
O
2
1 M.
1 M.
b
2
O
2
2
ꢀ
Actually, when MoðO
2
Þ
0.02 M in MeOH is kept at
4
2
5.0 ± 0.1 °C, it slowly decomposes according to a first order pro-
ꢀ
4 ꢀ1
cess (t1/2 = 77.01 min, k = 1.5 ꢃ 10
s ) whereas its disappear-
ance is about 100 times faster in the presence of 0.25 M DMSO
Figure 2. 95Mo NMR spectra at 0 °C of solutions in CH
OD of (NBu
without DMSO (E) or with
3
4 2 4
) MoO 0.1 M
ꢀ
2 ꢀ1
(
t
O
1/2 = 0.97 min, k = 1.2 ꢃ 10
s
). As a consequence, the yield of
decreases since the oxidation of DMSO into DMSO by
production
and 1 M H
2
O
2
(D), (NBu
4
2
)
2
ꢀ
MoO
4
0.2 M and 3 M H
2
O
2
1
2ꢀ
1
.5 M DMSO (F). MoðO
2
Þ4 and MoOðO
2
Þ3 signals are assigned from Ref. 2.
2
2
2
ꢀ
1
MoðO
rate increases by a factor of ꢁ2.2–3 (Table 1).
It is noteworthy that addition of DMSO in batches (entry 4 vs
entry 6) and before or after H (entry 4 vs entry 5) has no influ-
yield as well as on the reaction time. However, the
profile of the luminescence curves obtained with 1 M of H and
.5 M of DMSO added before (entry 5) or after H (entry 4) is dif-
2
Þ
2 2 2
consumes a part of H O whereas the O
4
2 2
O
1
ence on the O
2
2 2
O
0
2 2
O
ferent (see Figs. 6e and 6f in Supplementary data) although they
have the same underlying area. This indicates that the instanta-
1
neous rates of
2
O formation are different in the two cases and
mainly result from different compositions of the reaction mixtures.
2
ꢀ
Actually, when DMSO is added 10 min after H
2
O
2
, the MoOðO
2
Þ
/
3
2
ꢀ
MoðO
2
Þ
equilibrium is reached and DMSO reacts with the max-
4
2
ꢀ
1
imum concentration of MoðO
2
Þ
leading to a high initial O
2
pro-
4
duction rate and to a high signal intensity (I ꢁ 300 mV). On the
1
other hand, when DMSO is added first, a lower initial O
2
produc-
is continuously
before reaching its maximal con-
centration. When all DMSO has been consumed, after about
2
ꢀ
Scheme 1. Mechanism of 1
O
generation from H
O
2
/MoO4 in the presence of
2ꢀ
tion rate (I ꢁ 180 mV) is obtained, as MoðO Þ
2
2
2
4
2
ꢀ
DMSO.
being converted into MoOðO
Þ
2
3
1
3
0 min, only the remaining H O is responsible for the O forma-
2 2 2
This acceleration is probably due to the shunt of the slow hydroly-
tion. This confirms that the presence of DMSO does not interfere
with the kinetics and equilibrium of the peroxomolybdate forma-
2ꢀ
2ꢀ
sis of MoðO
2
Þ
into MoOðO
2
Þ
by the rapid quenching of
4
3
2
ꢀ
2ꢀ
MoðO
2
Þ
by DMSO leading to the formation of MoOðO
2
Þ
that
2ꢀ
4
3
tion and that the reaction between DMSO and MoðO Þ
is much
2
4
1
2ꢀ
rapidly releases
O
2
. The dramatic effect of DMSO on MoðO
2
Þ
2ꢀ
2ꢀ
4
faster than the time required to attain the MoOðO Þ /MoðO2Þ4
2
3
is clearly shown by monitoring the fading of the red-brown tetra-
peroxo species by visible spectroscopy at 600 nm (Fig. 3).
equilibrium.
Figure 4. Peroxidation of b-citronellol 0.20 M with two additions of H
2
O
2
1 M
Figure 3. Evolution of the absorbance at 600 nm of Na
alone (gry) or with DMSO 0.25 M (black) at 25 °C.
2
Mo(O
2
)
4
0.02 M in MeOH
(gray) and in the presence of DMSO 0.25 M (black). Conditions: methanol,
[Na MoO ] = 0.05 M, T = 25 °C.
2
4

6
534
M. Collinet-Fressancourt et al. / Tetrahedron Letters 51 (2010) 6531–6534
the rate of 1
O production when DMSO is added to the reaction
2
medium. It actually shortens the time required for complete oxida-
tion of the substrate. The process is simple and kinetically im-
proved compared to peroxidations conducted in MeOH without
DMSO for which a careful controlled addition of H
2 2
O is required.
It is thus an interesting alternative to the more intricate micro-
1
0
emulsions for the dark singlet oxygenation of low molecular
weight hydrophobic substrates.
Acknowledgments
We are grateful to Université Lille1 for financial support.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.10.022.
Figure 5. Peroxidation of
2 2
a-terpinene 0.25 M with H O 1 M (gray) and in the
2 4
presence of DMSO 0.25 M (black). Conditions: methanol, [Na MoO ] = 0.05 M,
T = 25 °C.
References and notes
3
H
.3. Oxidation of a-terpinene and b-citronellol with the system
1.
2.
3.
(a) Aubry, J.-M.; Cazin, B. Inorg. Chem. 1988, 27, 2013–2014; (b) Aubry, J. M. J.
Am. Chem. Soc. 1985, 107, 5844–5849.
Nardello, V.; Marko, J.; Vermeersch, G.; Aubry, J. M. Inorg. Chem. 1995, 34,
2
ꢀ
2
2
O /DMSO/MoO4
4950–4957.
This procedure was then applied to the peroxidation of
a
-ter-
Nardello, V.; Bogaert, S.; Alsters, P. L.; Aubry, J.-M. Tetrahedron Lett. 2002, 43,
8731–8734.
1
pinene 1 and b-citronellol 2. Compound 1 reacts with O
2
accord-
ing to a [4 + 2] cycloaddition leading to ascaridole whereas 2
4. Alsters, P. L.; Jary, W.; Nardello-Rataj, V.; Aubry, J.-M. Org. Process Res. Dev.
010, 14, 259–262.
2
provides a mixture of two hydroperoxides according to the ene
5.
(a) Zhou, W.; Clennan, E. L. J. Am. Chem. Soc. 1999, 121, 2915–2916; (b) Clennan,
E. L.; Yang, K. Heteroat. Chem. 1993, 4, 197–201; (c) Akasaka, T.; Ando, W. J.
Chem. Soc., Chem. Commun. 1983, 1203–1204; (d) Baciocchi, E.; Chiappe, C.; Del,
G. T.; Fasciani, C.; Lanzalunga, O.; Lapi, A.; Melai, B. Org. Lett. 2009, 11, 1413–
9
reaction. Results are reported in Table 2.
Addition of DMSO clearly boosts the rates of peroxidation of
substrates in MeOH. At 25 °C, the time required to oxidize more
than 95% of b-citronellol is more than three-times shorter when
1
416; (e) Baciocchi, E.; Giacco, T. D.; Elisei, F.; Gerini, M. F.; Guerra, M.; Lapi, A.;
Liberali, P. J. Am. Chem. Soc. 2003, 125, 16444–16454.
6. (a) Wagner, G. W.; Procell, L. R.; Yang, Y.-C.; Bunton, C. A. Langmuir 2001, 17,
809–4811; (b) Bunton, C. A.; Gillitt, N. D. J. Phys. Org. Chem. 2002, 15, 29–35;
c) Chiarini, M.; Cerichelli, G.; Foroudian, H. J.; Gillitt, N. D.; Yunes, S. F.; Bunton,
0
.25 M DMSO is added (Fig. 4). In the case of
of 0.25 M DMSO accelerates the reaction rate by a factor 2.7
Fig. 5). In all cases, the reactions lead to the expected oxidation
products.
a-terpinene, addition
4
(
(
C. A. Langmuir 2004, 20, 5201–5208; (d) Chiarini, M.; Gillitt, N. D.; Bunton, C. A.
Langmuir 2002, 18, 3836–3842; (e) Yunes, S. J.; Gillitt, N. D.; Bunton, C. A.
Colloids Surf., A 2006, 281, 1–7.
7
8
9
.
.
.
Pierlot, C.; Barbillat, J.; Nardello-Rataj, V.; Mathieu, D.; Sergent, M.; Marko, J.;
Aubry, J.-M. Photochem. Photobiol. Sci. 2009, 8, 1024–1031.
Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1995, 24, 663–
4
. Conclusion
677.
The intrinsic reactivity of peroxomolybdates, which may be a
Aubry, J.-M.; Bouttemy, S. J. Am. Chem. Soc. 1997, 119, 5286–5294.
drawback because it can lead to side reactions, such as epoxida-
tion, is proved herein to be beneficial as it significantly accelerates
10. Nardello, V.; Caron, L.; Aubry, J.-M.; Bouttemy, S.; Wirth, T.; Saha-Möller
Chantu, R.; Adam, W. J. Am. Chem. Soc. 2004, 126, 10692–10700.