Tin(IV) bis(perfluoroalkanesulfonyl)amide complex as a highly selective Lewis acid catalyst for Baeyer-Villiger oxidation using hydrogen peroxide in a fluorous recyclable phase

Article abstract of DOI:10.1016/S0040-4039(03)01185-7

Sn[N(SO₂C₈F₁₇)₂]₄ catalyst was shown to give an excellent yield and selectivity in a fluorous biphasic catalytic system for Baeyer-Villiger oxidation of cyclic ketones by 35% aqueous hydrogen peroxide, a green, safe and cheap oxidant. Furthermore, the catalyst was completely recovered and reused in the fluorous immobilized phase without loss of activity.

Full text of DOI:10.1016/S0040-4039(03)01185-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 4977–4980
Tin(IV) bis(perfluoroalkanesulfonyl)amide complex as a highly
selective Lewis acid catalyst for Baeyer–Villiger oxidation using
hydrogen peroxide in a fluorous recyclable phase
Xiuhua Hao, Osamu Yamazaki, Akihiro Yoshida and Joji Nishikido*
The Noguchi Institute, 1-8-1 Kaga, Itabashi-ku, Tokyo 173-0003, Japan
Received 4 April 2003; revised 8 May 2003; accepted 9 May 2003
Abstract—Sn[N(SO2C8F17)2]4 catalyst was shown to give an excellent yield and selectivity in a fluorous biphasic catalytic system
for Baeyer–Villiger oxidation of cyclic ketones by 35% aqueous hydrogen peroxide, a green, safe and cheap oxidant. Furthermore,
the catalyst was completely recovered and reused in the fluorous immobilized phase without loss of activity. © 2003 Elsevier
Science Ltd. All rights reserved.
Since the technique of Fluorous Biphasic System (FBS),
as a phase-separation and catalyst immobilization tech-
nique, was first introduced by Horva´th and Ra´bai in
1994,1 it has been used for some organic syntheses.2 On
the other hand, the Baeyer–Villiger (BV) oxidation was
reported as early as 1899, and it still plays an important
role in practical processes.3 With the increase in envi-
ronmental concern, much research has focused on the
development of a catalytic BV oxidation process using
the green and cheap hydrogen peroxide,4 such as with
metal complexes (based on Re, Pt) and acid zeolitic
catalysts (like HZSM-5, USY) to facilitate the nucle-
ophilic attack of hydrogen peroxide on the electrophilic
carbonyl carbon. However, most of them showed either
a poor selectivity or a low turnover number (TON) for
the desired esters and lactones. Accordingly, some
researchers have begun to consider an alternative reac-
tion route consisting of an initial activation of carbonyl
groups to increase the electrophilicity of the carbonyl
carbon and a subsequent interaction with hydrogen
peroxide.5 Since Lewis acid catalysts can activate car-
bonyl groups, our interest was concentrated on metal
complexes with perfluorinated ligands, whose Lewis
acidity can be adjusted by the choice of different metals
(e.g. Sn, Yb) and nucleophilicity by the variation of the
fluorine load in different perfluorinated ligands (e.g.
-C8F17, -CF3). Here, we will report our newest results
on a successful approach to FBS, i.e. perfluoro(methyl-
cyclohexane)/1,2-dichloroethane and perfluoro(methyl-
cyclohexane)/1,4-dioxane systems for BV oxidation
using 35% H2O2 catalyzed by tin(IV) bis(perfluorooc-
tanesulfonyl)amide complex6 (Fig. 1).
Keywords: Baeyer–Villiger oxidation; biphasic catalysis; hydrogen
peroxide; perfluorinated ligand; tin.
* Corresponding author. Tel.: +81-3-5248-3596; fax:+81-3-5248-3597;
e-mail: nishikido.jb@om.asahi-kasei.co.jp
Figure
Sn[N(SO2C8F17)2]4.
1. MNDO/FF
optimized
structure
of
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01185-7

4978
X. Hao et al. / Tetrahedron Letters 44 (2003) 4977–4980
BV oxidation of adamantanone in FBS was firstly
examined with various Lewis acid catalysts (Scheme 1),†
whose representative results were summarized in Table
1. In the absence of the catalyst (entry 11), both conver-
sion and selectivity were poorest. All of the metal
complexes (SnII, HfIV, ScIII, YbIII) with -N(SO2C8F17)2
ligand (entries 2, 5, 7 and 9) gave better yield and TON
than those with -OSO2CF3 ligand (entries 3, 6, 8 and
10), but no significant difference in selectivity. These
results can be attributed to the presence of the power-
fully electron-withdrawing -N(SO2C8F17)2 ligand bear-
ing the higher fluorine load on the metal center: the
electrophilicity of the carbonyl carbon would have been
increased making it be more easily attacked by the poor
nucleophile H2O2.7,8 This ligand effect was further
clearly proven by the Sn-based catalysts: yield and
TON followed the order of Sn[N(SO2C8F17)2]4>
Sn[N(SO2C8F17)2]2>Sn(OSO2CF3)2 (entries 1–3). As for
M[N(SO2C8F17)2]n (n=3, 4) complexes, yield, selectivity
and TON were in order of SnIV>HfIV>ScIII>YbIII
(entries 1, 5, 7 and 9), except that the selectivity of YbIII
was slightly higher than that of ScIII. With respect to
Sn[N(SO2C8F17)2]4 (entry 1), a conversion of 94% was
obtained which corresponded to a TON of 94 (for 2 h
at 25°C), while the selectivity towards the lactone was
99%. Such a result showed a great improvement in the
reaction conditions compared with the best reported
Sn-zeolite catalytic system5 (TON of about 116 for 6 h
at 56°C).
As our previously reported lanthanide complex cata-
lysts,9 Sn[N(SO2C8F17)2]4 was also found to be com-
pletely immobilized and could be reused in the fluorous
phase,10 i.e. completely remained in the fluorous phase
and recycled without depression of its catalytic activity
as shown in Figure 2. These results manifested that
even after the catalyst was recycled four times, the GC
yield still kept higher than 90%.
The possibility of recycling Sn[N(SO2C8F17)2]4 catalyst
in the BV oxidation of cyclobutanone was also investi-
gated (Scheme 2). It was found that there was almost
no reduction of the catalytic activity (yield: 93–95%)
and no other by-products detected except for the
desired lactone formation (selectivity: 96–100%).
Scheme 1. Baeyer–Villiger oxidation of 2-adamantanone.
Table 1. Baeyer–Villiger oxidation of 2-adamantanone in
fluorous biphasic system under the conditions in Scheme 1
In order to analyze and quantify the catalyst recovery,
the reaction rates (=activity) at different cycle times
were compared. In Figure 3, the representative results
obtained from cyclobutanone oxidation were shown. It
can be found that there was almost no difference for
Entry Catalyst
Yield (%)a Selectivity (%)b TONc
1
2
3
4
5
6
7
8
Sn[N(SO2C8F17)2]4 93 (91)d
Sn[N(SO2C8F17)2]2 48
99
83
87
79
92
91
69
66
73
83
26
94
58
42
52
88
43
77
47
41
23
–
Sn(OSO2CF3)2
SnCl4
37
41
Hf[N(SO2C8F17)2]4 82
Hf(OSO2CF3)4 41
Sc[N(SO2C8F17)2]3 53
Sc(OSO2CF3)3 31
Yb[N(SO2C8F17)2]3 31
9
10
11
Yb(OSO2CF3)3
None
19
2
a GC yields (internal standard: n-nonane).
b Selectivity: mmol of produced lactone/mmol of reacted ketone.
c Turnover number: mmol converted adamantanone/mmol catalyst.
d Value in parenthesis refers to the isolated yield.
† Typical procedure: To a mixture of perfluoro(methylcyclohexane) (3
mL) and 1,2-dichloroethane (3 mL) were added Sn[N(SO2C8F17)2]4
(81 mg, 0.02 mmol), 2-adamantanone (300 mg, 2.00 mmol) and 35%
H2O2 (194 mg, 2.00 mmol). The reaction mixture was stirred
continuously at 25°C for 2 h. Once the stirring was stopped, the
reaction mixture separated into two liquid phases within 10 s. An
upper 1,2-dichloroethane phase afforded pure lactone (302 mg, 91%
yield) by silica gel chromatography, while the lower fluorous phase
was reused in subsequent reactions. The amounts of H2O2 both in
the upper organic phase and lower fluorous phase were titrated with
Na2S2O3 after their respective sampling. With respect to the system
of 1,4-dioxane/perfluoro(methylcyclohexane), H2O2 can be dis-
solved in the upper 1,4-dioxane phase completely. In the system of
1,2-dichloroethane/perfluoro(methylcyclohexane), H2O2 was par-
tially (<10% ) dissolved in the upper 1,2-dichloroethane phase and
the other stayed above this phase.
Figure 2. Recycles of Sn[N(SO2C8F17)2]4 catalyst in Baeyer–
Villiger oxidation of 2-adamantanone.
Scheme 2. Baeyer–Villiger oxidation of cyclobutanone.

X. Hao et al. / Tetrahedron Letters 44 (2003) 4977–4980
4979
the formation rate of g-butyrolactone at cycles 1 and 4
within 1 h reaction time, which confirmed that there
was not only no loss of the catalyst but also no
depression of catalytic activity during the repetition.
Accordingly, it can be deduced 99% of the catalyst
should still remain in the fluorous phase even after the
fourth cycle.11 In fact, the product yield kept a constant
level in a specific period (e.g. about 1 h for cycle 1 in
Fig. 3), which is a criteria for us to determine the
reaction time.
relatively high selectivity to the desired lactone (>91%)
for all of these BV reactions (the upper one in case of
entries 2, 3, 5 and 6), which can be viewed as superior
to most of the reported catalytic systems with 35%
H2O2 where a mixture of lactone, epoxide and epoxy-
lactone was produced.4 Furthermore, in our catalytic
system, no significant H2O2 decomposition was
detected. It may be noticed the conversions for
cyclopentanone, cyclohexanone, menthone and cam-
phor only reached a moderate level (entries 2, 3, 5 and
6), a little lower than the reported ones.12 This can be
contributed to the differences of aqueous hydrogen
peroxide concentration and the ratio of substrate/H2O2,
i.e. 50% higher and excess H2O2 were used in their
works, whereas 35% and equimolar H2O2 in ours. As
an ideal catalytic synthesis, it not only takes account of
the high catalytic activity, but also accounts for the
atom economy, safety and easy handing.11 Since combi-
nation of higher concentration H2O2 (>50%) with
ketones was potentially explosive, it is very significant
to develop a practical system with 35% H2O2 for the
industrial operation. On the other hand, the present
turnover frequency (TOF) of the system Sn[N-
(SO2C8F17)2]4/H2O2/cyclohexanone (upper one of entry
3 in Table 2) amounted to 36 h−1, which was compara-
ble to the reported TOF of 23 h−1 with the catalyst
(Sn-MCM-41) for the same reaction.13 In order to
increase the conversion of cyclopentanone, cyclohex-
anone, menthone and camphor, we have conducted
these reactions with an excess amount of catalyst and/
or a prolonged reaction time. As shown in Table 2 (the
lower one in case of entries 2, 3, 5 and 6), the conver-
sion was improved, but the selectivity was not signifi-
In Table 2, a series of BV oxidation of ketone sub-
strates were examined to evaluate the above catalytic
system. It was found that Sn[N(SO2C8F17)2]4 can give a
Figure 3. Lactonization of cyclobutanone catalyzed by
Sn[N(SO2C8F17)2]4 at different cycles under the conditions in
Scheme 2.
Table 2. Baeyer–Villiger oxidation of ketones

4980
X. Hao et al. / Tetrahedron Letters 44 (2003) 4977–4980
cantly satisfactory. In addition, the possibility of the
present catalytic system for BV oxidation of aromatic
ketone (e.g. acetophenone) was also investigated, but
obtained 20% yield. How to achieve a higher yield for
BV oxidations of aromatic ketone left us further work.
in-Print No. 91: Fluorous Chemistry’; Pergamon Press:
Oxford, 2002; Vol. 58, No. 20.
3. Renz, M.; Meunier, B. Eur. J. Org. Chem. 1999, 64,
737–750.
4. (a) Frisone, M. D. T.; Pinna, F.; Strukul, G.
Organometallics 1993, 12, 148–156; (b) Herrmann, W. A.;
Fischer, R. W.; Correia, J. D. G. J. Mol. Catal. 1994, 94,
213–223; (c) Strukul, G. Angew. Chem., Int. Ed. Engl.
1998, 37, 1198–1209; (d) Gavagnin, R.; Cataldo, M.;
Pinna, F.; Strukul, G. Organometallics 1998, 17, 661–667;
(e) Fischer, J.; Ho¨lderich, W. F. Appl. Catal. A 1999, 180,
435–443; (f) Palazzi, C.; Pinna, F.; Strukul, G. J. Mol.
Catal. A 2000, 151, 245–252.
Additionally, it was also confirmed that such a high
selectivity could not be achieved when the catalyst was
used in an organic monophase, i.e. either 1,2-
dichloroethane as a solvent for adamantanone BV oxi-
dation (selectivity: 81%) or 1,4-dioxane for
cyclobutanone BV oxidation (selectivity: 65%), which
were considerably lower than those of FBS (selectivity:
99% of entry 1 in Table 1; 99% of cycle 1 in Scheme 2).
Another exceptional advantage in FBS is that the cata-
lyst can be recovered and reused as described above,
which is an impossible task in the organic monophase
system. Moreover, it is worth pointing out that such
results benefit greatly from both the efficient Lewis
acidity14 of Sn[N(SO2C8F17)2]4 and the unique solution
property of FBS.
5. Corma, A.; Nemeth, L. T.; Renz, M.; Valencia, S. Nature
2001, 412, 423–425.
6. A mixture of 1,2-dichloroethane (15 mL) and acetonitrile
(5 mL) was added by HN(SO2C8F17)2 (6.00 mmol) and
tin(IV) acetate (1.50 mmol), which was stirred continu-
ously for 15 h at 50°C. The reaction mixture was then
cooled to room temperature and perfluorodecane (20 mL)
was added. After the perfluorodecane layer was evapo-
rated and dried at 80°C/0.01 mmHg for 8 h, tin(IV)
bis(perfluorooctanesulfonyl) amide complex was there-
fore obtained as a hygroscopic white powder in 96% yield
(5.87 g). Anal. calcd for Sn[N(SO2C8F17)2]4: C, 19.03; Sn,
2.94. Found: C, 19.10; Sn, 2.91. 19F NMR: l −126.1,
−121.2, −114.2, −81.4 ppm.
7. Nishikido, J.; Yamamoto, F.; Nakajima, H.; Mikami, Y.;
Matsumoto, Y.; Mikami, K. Synlett 1999, 10, 1990–1992.
8. Koppel, I. A.; Taft, R. W.; Anvia, F.; Zhu, S.-Z.; Hu,
L.-Q.; Sung, K.-S.; DesMarteau, D. D.; Yagupolskii, L.
M.; Yagupolskii, Y. L.; Ignat’ev, N. V.; Kondratenko, N.
V.; Volkonskii, A. Y.; Vlasov, V. M.; Notario, R.; Maria,
P.-C. J. Am. Chem. Soc. 1994, 116, 3047–3057.
9. Mikami, K.; Mikami, Y.; Matsumoto, Y.; Nishikido, J.;
Yamamoto, F.; Nakajima, H. Tetrahedron Lett. 2001, 42,
289–292.
10. The catalyst recovered from the fluorous phase without
apparent weight loss, i.e. 80 mg modified catalyst was
recovered after the original 81 mg was recycled four
times. The recovered catalyst was characterized by ele-
mental analysis. Anal. calcd for Sn[N(SO2C8F17)2]4: C,
19.03; Sn, 2.94. Found: C, 19.50; Sn, 2.90. For further
supporting evidence, see Figure 3.
11. Gladysz, J. A. Pure Appl. Chem. 2001, 73, 1319–1324.
12. (a) Neimann, K.; Neumann, R. Org. Lett. 2000, 2, 2861–
2863; (b) ten Brink, G.-J.; Vis, J.-M.; Arends, I. W. C. E.;
Sheldon, R. A. J. Org. Chem. 2001, 66, 2429–2433; (c)
Berkessel, A.; Andreae, M. R. M. Tetrahedron Lett. 2001,
42, 2293–2295.
In conclusion, Sn[N(SO2C8F17)2]4 has proved to be a
favorable catalyst for the BV oxidation of cyclic
ketones in FBS not only due to the improved selectivity
but also due to the environmental benignity. We believe
that the following results endow the catalytic approach
with great potential for synthetic application: (1) 35%
aqueous hydrogen peroxide, a safe and economic oxi-
dant for practical processes, was shown to provide
satisfactory oxidizing ability; (2) the catalyst, com-
pletely immobilized in the fluorous phase, can be recov-
ered and reused; (3) high yields and purities of lactone
can be obtained under very mild conditions.
Acknowledgements
This work was supported by New Energy and Indus-
trial Technology Development Organization (NEDO)
and Japan Chemical Innovation Institute (JCII)
through the R&D program for Process Utilizing Multi-
Phase Catalytic Systems.
References
1. Horva´th, I. T.; Ra´bai, J. Science 1994, 266, 72–75.
2. (a) Cornils, B. Angew. Chem., Int. Ed. Engl. 1997, 36,
2057–2059; (b) Curran, D. P. Angew. Chem., Int. Ed.
Engl. 1998, 37, 1174–1196; (c) Horva´th, I. T. Acc. Chem.
Res. 1998, 31, 641–650; (d) Barthel-Rosa, L. P.; Gladysz,
J. A. Coord. Chem. Rev. 1999, 192, 587–605; (e) Barrett,
A. G. M.; Braddock, D. C.; Catterick, D.; Chadwick, D.;
Henschke, J. P.; McKinnell, R. M. Synlett 2000, 11,
847–849; (f) Xiang, J.; Toyoshima, S.; Orita, A.; Otera, J.
Angew. Chem., Int. Ed. 2001, 40, 3670–3672; (g) Colonna,
S.; Gaggero, N.; Montanari, F.; Pozzi, G.; Quici, S. Eur.
J. Org. Chem. 2001, 66, 181–186; (h) Gladysz, J. A.;
Curran, D. P. Tetrahedron 2002, 58, 3823–3825; (i)
Ghosez, L.; Wasserman, H. H. Tetrahedron ‘Symposium-
13. Corma, A.; Navarro, M. T.; Nemeth, L.; Renz, M. Chem.
Commun. 2001, 2190–2191.
14. HN(SO2C8F17)2 (4 mol%) and Sn[N(SO2C8F17)2]4 (1
mol%) were compared with adamantaone BV oxidation.
The former only obtained 71% yield and 73% selectivity
for the lactone product, but the latter gave 93% yield and
99% selectivity. Additionally, HN(SO2C8F17)2 was found
to be difficult for efficient separation and recycle due to
its weight loss (>40%) from the lower fluorous phase and
distribution in the upper organic phase (confirmed by 19F
NMR) after the reaction. These results verified that the
efficient and recyclable catalysis requires the unique com-
plexation of SnIV with HN(SO2C8F17)2.