Metal triflates-methanesulfonic acid as new catalytic systems: Application to the Fries rearrangement

Article abstract of DOI:10.1016/S0040-4039(03)01599-5

A surprising synergistic effect has been discovered between metal triflates (Mg, Ca, Sc, Cu, Zn, Y, Ln, Bi) and methanesulfonic acid, leading to active catalytic systems for the Fries rearrangement. In particular, the systems based on yttrium and copper(II) triflate proved to be very active and much cheaper than scandium triflate. The efficiency of these systems might result from the catalytic Lewis acid activation of Br?nsted acids.

Full text of DOI:10.1016/S0040-4039(03)01599-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 6379–6382
Metal triflates–methanesulfonic acid as new catalytic systems:
application to the Fries rearrangement
Omar Mouhtady,a Hafida Gaspard-Iloughmane,a,* Nicolas Roquesb and Christophe Le Rouxa,*
aHe´te´rochimie Fondamentale et Applique´e (UMR-CNRS 5069), Universite´ Paul-Sabatier, 118 route de Narbonne,
31062 Toulouse Cedex 04, France
bRhodia Organique Fine, Centre de Recherche de Lyon, 85 avenue des Fre`res Perret, 69192 Saint-Fons Cedex, France
Received 20 June 2003; revised 27 June 2003; accepted 27 June 2003
Abstract—A surprising synergistic effect has been discovered between metal triflates (Mg, Ca, Sc, Cu, Zn, Y, Ln, Bi) and
methanesulfonic acid, leading to active catalytic systems for the Fries rearrangement. In particular, the systems based on yttrium
and copper(II) triflate proved to be very active and much cheaper than scandium triflate. The efficiency of these systems might
result from the catalytic Lewis acid activation of Brønsted acids.
© 2003 Elsevier Ltd. All rights reserved.
Superacidic systems which may be formed by mixing
appropriate Lewis and Brønsted acids have been the
subject of much academic and industrial research.1
Trifluoromethanesulfonic acid (TfOH), which can be
considered as a super-acid,2 forms water-stable salts
with non-hydrolyzable metals.3 The use of the commer-
cially available metal triflates such as rare-earth metal
triflates (RET) as catalysts in organic synthesis has been
the subject of numerous publications.4 As triflic acid is
one of the most acidic monoprotic organic acids,1 one
might hypothesize that mixing catalytic amounts of
RET with any organic acid (such as carboxylic or
sulfonic acids) exhibiting a much higher pKa than
TfOH would not lead to the formation of TfOH but to
the complexation of the acid to the RET.5–7 Such
complexation (ideally between an inactive metal triflate
and an organic acid inactive towards a specific reaction)
which depends on the affinity between both partners
may lead via the enhancement of the Brønsted acidity
to active catalytic systems. Among all electrophilic aro-
matic substitutions we have studied the Fries rearrange-
ment (FR) (Eq. (1)).8
For this reaction, classical Lewis acids (such as AlCl3 or
TiCl4)8 must be used in excess (1.1–3 equiv.) and there-
fore lead to corrosion and environmental problems due
to the large amounts of acidic effluents and solid
wastes. More recently, a new generation of Lewis acids
such as metal triflates (scandium9 and hafnium10) have
been reported as catalysts. Our results using mixtures of
metal triflates and organic acids as catalytic systems are
summarized below.
Using the FR of naphthyl acetate (1a) as a model
reaction (Table 1), it emerges that most of components
are separately inactive (except for scandium and bis-
muth triflates, entries 4 and 12) or poorly active (see
entry 1 where methanesulfonic (MSA) was used alone)
and that powerful synergistic effects are observed when
metal triflates are mixed with MSA. Moreover, it
should be noted that MSA is a cheap sulfonic acid,
environmentally benign, commercially available in its
anhydrous form, and easy to handle.11 It should be
noted that the use of Group 2 triflates lead to active
catalytic systems and that the system Cu(OTf)2–MSA is
also active and cheaper than scandium triflate. As the
mixtures of scandium or yttrium triflates and MSA are
the most efficient (Table 1, entries 4 and 7), we decided
to use these catalytic systems in the rearrangement of
other phenolic esters (Table 2). From these results it
emerges that the combination of Y(OTf)3 or Sc(OTf)3
with MSA is very efficient towards the esters derived
from 1-naphthol and to a lesser extent towards those
derived from phenol. Conversely to the results previ-
(1)
* Corresponding authors. Fax: (33) 5 61 55 82 04; e-mail: leroux@
chimie.ups-tlse.fr
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01599-5

6380
O. Mouhtady et al. / Tetrahedron Letters 44 (2003) 6379–6382
13
Table 1. Screening of catalysts for the Fries rearrange-
ment of 1-naphthyl acetate (1a)a
Y(OMs)3 did not lead to active catalytic systems.
Toluene can be replaced advantageously by nitroethane
(a similar yield was obtained under the same conditions
as entry 11, Table 2), an industrially acceptable solvent
(in contrast to nitromethane which is potentially explo-
sive in the presence of strong acids).14 Interestingly, the
FR of naphthyl benzoate has been carried out and gave
a 72% yield of 1-hydroxy-2-naphthyl phenyl ketone
(Table 2, entry 10).
Entry
M(OTf)n
Yield in 1b (%)b
Cat.: M(OTf)n
Cat:
M(OTf)n+MSAc
1
2
3
4
5
6
7
8
None
4d
46
67
Mg(OTf)2
Ca(OTf)2
Sc(OTf)3
Cu(OTf)2
Zn(OTf)2
Y(OTf)3
La(OTf)3
Nd(OTf)3
Dy(OTf)3
Yb(OTf)3
Bi(OTf)3
0
0
93 (67)f
74e (82)f
90 (34)f
84 (26)f
90 (52)f
90 (22)f
90
From a mechanistic point of view, we believe that the
observed synergistic effects might proceed: (i) via a pure
Brønsted catalysis (Scheme 1, path (a) since these rear-
rangements are known to be proton-catalyzed;8b–d (ii)
via a ligand exchange between MSA and the metal
triflate, generating metal sulfonate and TfOH which is a
known catalyst for the FR8d or (iii) via a Brønsted
Lewis acid catalysis (BLA)15 resulting in the double-
activation of the ester as previously reported in the case
of the ene-reaction6 (Scheme 1, path a+b). Further
experiments16 proved that the hypothesis of a ligand
exchange could be ruled out in the case of yttrium(III)
triflate and bismuth(III) triflate, this latter being previ-
ously known to be prone to give comparable ligand
exchanges.17 Moreover, intermolecular or intramolecu-
lar acylation transfer have been postulated for the
FR.18 In the case of 1a, when toluene was substituted
by anisole (same conditions as entry 7, Table 1) no
p-methoxyacetophenone was detected, suggesting that
the mechanism may involve an intramolecular transfer.
6
30
0
0
9
0
10
11
12
0
90
2g
90
80 (40)f
88 (70)f
a Reactions carried out at 100°C using toluene as the solvent (c=
0.867 M relative to 1a).
b Isolated yields of 1-hydroxy-2-naphthyl methyl ketone (1b) after 5 h
heating at 100°C (oil bath).
c 1a/M(OTf)n/MSA=1/0.05/0.1 molar ratio.
d In this case only MSA was used.
e The lower yield is mainly due to side reactions of 1b leading to not
analyzed highly colored adducts.
f In parentheses, yield after 1 h heating (oil bath).
g See Ref. 10.
ously reported,9 it should be noted that in our case, the
addition of acetic acid to Sc(OTf)3 did not increase the
yield of 1b. Furthermore, the combination of MSA
Our results suggest that the role of the metal could be
displaced from a Lewis acid behavior towards the reac-
tants to a Lewis acid behavior towards MSA, thus
12
with other salts of yttrium such as Y(NTf2)3 or
Table 2. Catalysis of Fries rearrangements by MX3–AH catalytic systems
Entry
Aromatic substratea
Catalytic system
Product, yieldb (%) and time
MX3
AH
1
2
3
4
5
6
7
8
1a
1a
1a
1a
1a
1a
2a
3a
4a
5a
6a
7a
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Y(OTf)3
Y(NTf2)3
Y(OMs)3
Y(OTf)3
Y(OTf)3
Y(OTf)3
Sc(OTf)3
Y(OTf)3
Y(OTf)3
None
CH3CO2H
MSA
1b, 67, 1 h
1b, 55, 1 h
1b, 82, 1 h
1b, 52, 1 h
1b, 20, 1 hc
1b, 3, 5 hd
2b, 82, 5 h
3b, 87, 5 h
4b, 80, 5 h
5b, 72, 16 he
6b, 60, 2 h 30f,g
7b, 60, 2 h 30
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
CH3CO2H
9
10
11
12
a Conditions: Aromatic substrate/MX3/AH=1/0.05/0.1 molar ratio except for entry 10 where this ratio is 1/0.1/0.2 and entries 11 and 12 where
this ratio is 1/0.2/0.4. Reactions carried out at 100°C using toluene as the solvent (c=0.867 M relative to the aromatic substrate) except for
entries; aromatic substrates: naphthyl acetate (1a), naphthyl hexanoate (2a), naphthyl isobutyrate (3a), naphthyl cyclohexylcarboxylate (4a),
naphthyl benzoate (5a), 3-methoxy-1-phenyl acetate (6a), 3-methyl-1-phenyl acetate (7a).
b Products: 1-hydroxy-2-naphthyl pentyl ketone (2b), 1-hydroxy-2-naphthyl isopropyl ketone (3b), 1-hydroxy-2-naphthyl cyclohexyl ketone (4b),
1-hydroxy-2-naphthyl phenyl ketone (5b), 2-hydroxy-4-methoxyphenyl methyl ketone (6b), 2-hydroxy-4-methylphenyl methyl ketone (7b).
c Only traces of 1b were obtained under the same conditions when Y(NTf2)3 was used alone.
d No reaction occurred under the same conditions when Y(OMs)3 was used alone.
e This experiment has been carried out at 150°C using 4-chlorotoluene instead of toluene.
f This experiment has been carried out in nitroethane instead of toluene and gave a similar yield.
g Containing 4-hydroxy-2-methoxyphenyl methyl ketone (yield 7%).

O. Mouhtady et al. / Tetrahedron Letters 44 (2003) 6379–6382
6381
Scheme 1.
forming the active catalytic system. In this view, the
metal triflate could be described as a procatalyst for the
reaction considered, the other partner of the catalytic
system being voluntarily added (this work) or in situ
generated during the reaction. Effectively, Sc(OTf)3 has
been reported as a catalyst for the alkylation of aromat-
ics with alkyl methanesulfonates leading to alkyl aro-
matics and MSA.19 Based on our findings it seems
attractive to consider that the real catalyst might be
Sc(OTf)3-MSA and not Sc(OTf)3.
2. Olah, G. A.; Prahash, G. K. S.; Sommer, J. Science 1979,
206, 13–20.
3. Kobayashi, S.; Nagayama, S.; Busujima, T. J. Am. Chem.
Soc. 1998, 120, 8287–8288.
4. (a) Hertweck, C. J. Prakt. Chem. 2000, 342, 316–321; (b)
Kobayashi, S.; Sugiura, M.; Kitagawa, H.; Lam, W. W.
L. Chem Rev. 2002, 102, 2227–2302.
5. Such a hypothesis has been recently proposed for the
benzoic activation of lanthanum triflate-catalyzed allyla-
tions: (a) Aspinall, H. C.; Bissett, J. S.; Greeves, N.;
Levin, D. Tetrahedron Lett. 2002, 43, 319–321; (b)
Aspinall, H. C.; Bissett, J. S.; Greeves, N.; Levin, D.
Tetrahedron Lett. 2002, 43, 323–325.
Typical procedure for the rearrangement of naphthyl
acetate (1a). In a 50 mL Schlenk tube were introduced
under argon 70 mg (130 mmol) of yttrium triflate and
483 mg (2.6 mmol) of 1a. 3 mL of toluene was intro-
duced and finally 25 mg (260 mmol) of MSA was added
via a syringe. The suspension was magnetically stirred
and heated in a thermostated oil bath at 100°C for 5 h.
After this time, the red solution was cooled, 50 mL of
a saturated aqueous solution of sodium bicarbonate
were added and the organic products were extracted
with dichloromethane (2×50 mL). The combined
organic phases were dried over sodium sulfate and
concentrated under reduced pressure. The product was
purified by flash chromatography (silica gel, eluent:
pentane/dichloromethane=1/1) to give 435 mg (90%
yield) of 1-hydroxy-2-naphthyl methyl ketone (1b),
identified with an authentic sample.9
6. Another example is given in the case of ene-reactions
catalyzed by a mixture of Yb(fod)3 and AcOH: Ciufolini,
M. A.; Deaton, M. V.; Zhu, S.; Chen, M. Tetrahedron
1997, 53, 16299–16312.
7. The use of Lewis acid surfactant-Brønsted acid catalytic
systems have been reported in water: Manabe, K.;
Kobayashi, S. Tetrahedron Lett. 1999, 40, 3773–3776.
8. (a) Fries, K.; Finck, G. Ber. Dtsch. Chem. Ges. 1908, 41,
4271–4284; (b) Gerecs, A. In Friedel–Crafts and Related
Reactions; Olah, G. A., Ed.; Wiley-Interscience: New
York, 1964; Vol. III, pp. 499–533; (c) Martin, R. Org.
Prep. Proc. Int. 1992, 24, 369–435; (d) Martin, R. In
Handbook of Hydroxybenzophenones; Kluwer Academic:
Dordrecht, 2000; (e) Effenberger, F.; Klenk, H.; Reiter,
P. L. Angew. Chem., Int. Ed. Engl. 1973, 12, 775–776; (f)
Bensari, A.; Zaveri, N. T. Synthesis 2003, 267–271.
9. Kobayashi, S.; Moriwaki, M.; Hachiya, I. Tetrahedron
Lett. 1996, 37, 4183-4186. Interestingly, it is reported in
this paper that the addition of a catalytic amount of
acetic acid leads to an improvement of the yield explained
by the authors by assuming that acetic acid works as an
acylating agent.
In conclusion, we have demonstrated that mixing metal
triflates with methanesulfonic acid leads to an
extremely synergistic effect, thus generating a new set of
active catalysts for the Fries rearrangement. Due to
their highly tunable potentialities (by reason of the
choice of the appropriate R%, S, M, L), we believe that
this new family of catalytic systems might enrich the
already wide field of catalytic applications of metal
triflates towards organic synthesis. The extension of this
work to other reactions is currently under investigation
in our laboratory.
10. Kobayashi, S.; Moriwaki, M.; Hachiya, I. Bull. Chem.
Soc. Jpn. 1997, 70, 267–273.
11. Commarieu, A.; Hoelderich, W.; Laffitte, J. A.; Dupont,
M. P. J. Mol. Catal. A 2002, 182-183, 137–141.
12. For the preparation of Y(NTf2)3, see: Mikami, K.;
Kotera, O.; Motoyama, Y.; Tanaka, M. Inorg. Chem.
1998, 10–11.
References
13. For the preparation of Y(OMs)3, see: Zinner, L. B. An.
Acad. Brasil. Cieˆnc. 1980, 52, 715–722.
1. (a) Olah, G. A.; Prahash, G. K. S.; Sommer, J.
Superacids; Wiley: New York, 1985; (b) Olah, G. A.;
Laali, K.; Farooq, O. J. Org. Chem. 1984, 49, 4591–4594;
(c) Engelbrecht, A.; Tschager, E. Z. Anorg. All. Chem.
1977, 433, 19–25.
14. Scott, N.; Deacon, G. B.; Appel, W. Mater. Sci. Forum
1999, 315-317, 484–489.
15. Ishihara, K.; Kurihara, H.; Matsumoto, M.; Yamamoto,
H. J. Am. Chem. Soc. 1998, 120, 6920–6930 and refer-
ences cited therein.

6382
O. Mouhtady et al. / Tetrahedron Letters 44 (2003) 6379–6382
16. The benzoylation of chlorobenzene (10 equiv.) by benzoic
anhydride (1 equiv.) has been conducted in the presence
of TfOH (0.3 equiv.) or a mixture of Y(OTf)3 (0.1 equiv.)
and MSA (0.3 equiv.). These experiments gave respec-
tively after 5 h at 110°C (a comparable temperature of
that used for the FR experiments) a yield of 27% and 1%
(with respect to Bz2O) in 4-chlorobenzophenone revealing
that no MeSO3−/TfO− exchange occurs. In the case of the
mixture of Bi(OTf)3 (0.1 equiv.) and MSA (0.3 equiv.) as
catalyst for the same reaction, only traces (<1%) of
4-chlorobenzophenone have been detected revealing that
no exchange occur.
17. Le Roux, C.; Dubac, J. Synlett 2002, 181–200 and refer-
ences cited therein.
18. (a) Martin, R.; Gavard, J. P.; Delfly, M.; Demerseman,
P.; Tromelin, A. Bull. Soc. Chim. Fr. 1986, 659–662; (b)
Ogata, Y.; Tabuchi, H. Tetrahedron 1964, 20, 1661–1666;
(c) Warshawsky, A.; Kalir, R.; Patchornik, A. J. Am.
Chem. Soc. 1978, 100, 4544–4550.
19. (a) Kotsuki, H.; Ohishi, T.; Inoue, M.; Kojima, T. Syn-
thesis 1999, 4, 603–606; (b) Kotsuki, H.; Ohishi, T.;
Inoue, M. Synlett 1998, 255–256.