Aluminium triflate: an efficient recyclable Lewis acid catalyst for the aminolysis of epoxides

Article abstract of DOI:10.1016/j.tetlet.2006.07.036

Epoxides are powerful starting materials for a range of useful materials and can be converted into, amongst others, amino alcohols. In this work, a range of epoxides was ring-opened using various alkyl- and arylamines under the action of aluminium triflat

Full text of DOI:10.1016/j.tetlet.2006.07.036

Tetrahedron Letters 47 (2006) 6557–6560
Aluminium triflate: an efficient recyclable Lewis acid catalyst
for the aminolysis of epoxides
D. Bradley G. Williams* and Michelle Lawton
Department of Chemistry, University of Johannesburg, PO Box 524, Auckland Park 2006, South Africa
Received 18 May 2006; revised 28 June 2006; accepted 7 July 2006
Available online 28 July 2006
Abstract—Epoxides are powerful starting materials for a range of useful materials and can be converted into, amongst others, amino
alcohols. In this work, a range of epoxides was ring-opened using various alkyl- and arylamines under the action of aluminium tri-
flate as a catalyst. This catalyst was found to be highly active, producing the desired amino alcohol products in high yields with low
catalyst loadings. Additionally, it was shown that the aluminium triflate catalyst is recyclable by simple extraction into water and
may be re-used several times.
Ó 2006 Elsevier Ltd. All rights reserved.
1,2-Amino alcohols are important molecules in both
organic synthesis and medicinal chemistry.1 They can
be prepared using a variety of routes, most commonly
through the ring opening of epoxides,2 often involving
heating of the epoxide in a protic solvent with an excess
of amine.3 Lewis acids, many of which suffer from deac-
tivation of the catalyst due to complex formation with
First, we examined the effects of different solvents on a
standard reaction. Using aniline as the nucleophile with
a variety of epoxides, the reactions were carried out in
DCM, ether or toluene (Table 1), while limiting the reac-
tion times to 5 h.10
The reactions carried out in toluene resulted in highest
yields. Both DCM and ether can act as Lewis bases
the amine, have also been used for this process.4–7
A
recent report describes the use of Al(OTf)3 present at
5 mol % loading in the preparation of ionic liquids,8 in
which this catalyst was used in the presence of 2-picolyl-
amine as nucleophile with a few epoxides.
Table 1. Solvent effect on aminolysis reactions
Entry Products
Yielda (%)
Ether
DCM
55
Toluene
82
Recently, we reported the use of Al(OTf)3 as a remark-
ably efficient Lewis acid catalyst for the ring opening
of epoxides in alcohols.9 We now report that this ver-
satile catalyst is also highly efficient (when present
from as little as 1.0 mol % loading) for the ring open-
ing of a range of epoxides with a variety of amines
(Scheme 1). Additionally, we investigated the influence
of steric and electronic effects on the outcome of this
reaction.
OH
H
N
1
56
64
OH
H
O
N
2
28
90
3
34 (8)b 36 (7)b 75 (18)b
N
H
OH
OH R1
H
N
Al(OTf)3
O
OH
N
+
R1
R2
R2
R
H
R
1.0 mol%
O
N
4
45
62
81
Scheme 1.
a 1.0% Al(OTf)3, 1.2 equiv aniline, temp = lower of reflux or 100 °C,
*
5 h.
Corresponding author. Tel.: +27 11 489 3431; fax: +27 11 489
2819; e-mail: dbgw@rau.ac.za
b Yields in parentheses refer to the other regioisomer.
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.07.036

6558
D. B. G. Williams, M. Lawton / Tetrahedron Letters 47 (2006) 6557–6560
Table 3. N-Methylaniline reactions with epoxidesa
lowering the Lewis acidity of the Al(OTf)3, which could
account for the lower yields in these solvents.
Entry
Products
Yieldb (%)
1 mol %
Al(OTf)3
2 mol %
Al(OTf)3
These initial experiments also demonstrated that higher
concentrations of Al(OTf)3 were required for these reac-
tions than for the alcoholysis reactions.9 Presumably,
this was due to the nitrogen of the amine and the oxygen
of the epoxide competing for complexation to the
Al(OTf)3, since both oxygen and nitrogen are hard
Lewis bases (see below).
OH
N
1
2
3
4
87
89
OH
O
N
50
93
A variety of aryl- and alkylamine nucleophiles were then
used in reactions with the same four epoxides, in order
to determine the role that steric and electronic effects
would play in these reactions. In general, the alkyl-
amines (Table 2) are harder bases than the aromatic
amines and therefore more efficiently compete with the
epoxides for the catalyst. Longer reaction times and/or
higher concentrations of Al(OTf)3 were required for
diethylamine and isopropylamine than those for the aro-
matic amines (Tables 1 and 3) (for isopropylamine,
diminished reactivity probably arose due to a combina-
tion of steric effects and catalyst deactivation). This is
in line with the ‘hard–soft acid–base’ theory proposed
by Pearson.11 According to this theory, ethers are con-
sidered hard bases, as are alkylamines. In contrast, aryl-
amines are considered borderline cases. This being the
case, the harder amines will tend to retard the rate of
the reaction, in line with our observations, and concomi-
tantly require slightly elevated levels of catalyst to
observe similar rates and conversions.
OH
74 (8)c
80 (10)c
N
OH
O
N
65
91
a 1.2 equiv amine, 100 °C, 5 h.
b Isolated yield (%).
c Yields in parentheses refer to the other regioisomer.
The reactions involving the glycidyl ethers with the
alkylamines afforded much higher yields than those with
the alkyl epoxides. This is presumably because the gly-
cidyl ethers can form a chelate structure with the
aluminium triflate, which is more stable than the simple
Al–O complexes formed by regular oxiranes, and are
thus able to compete more effectively with the amines
for the metal centre.
Table 2. Alkylamine reactions with epoxidesa
Entry Products
Yieldb (%)
The aniline reactions (Table 1) that were initially per-
formed showed the best results overall for the aromatic
amines, in terms of yield and selectivity at 1.0 mol % cat-
alyst loading. N-Methylaniline, which is sterically more
hindered than aniline, performed equally well in the
presence of 2.0 mol % of Al(OTf)3 (Table 3). Albeit less
efficient, even deactivated arylamines can be used in
these reactions (Table 4). For example, 2-chloroaniline
showed a decrease in the yield when using low catalyst
loadings, but afforded good yields of product in the
presence of 2 mol % of the catalyst (Table 4, entries
1–4).
1 mol % 2 mol % 10 mol %
Al(OTf)3 Al(OTf)3 Al(OTf)3
O
N
1
48
45
75b
80b
—
—
OH
OH
2
O
N
OH
H
3
14
—
—
—
—
43c
N
The use of deactivated nucleophiles in the form of para-
nitroaniline (Table 4, entries 5–7) or diphenylamine (Ta-
ble 4, entries 8–9) also allowed acceptable to good yields
of products to be obtained when making use of
10 mol % of the catalyst (Table 4, entries 5–7). In the
presence of this nucleophile, the chelate effect, together
with the deactivated nucleophile, proved deleterious to
the reaction. The latter set of reactions were carried
out under solventless conditions (Table 4, entries 5–7)
and afforded essentially identical results to those in
toluene.12
HO
H
8 (3)c,d
33 (26)c,d
31 (31)c,d
30 (12)c,d
38 (30)c,d
35 (34)c,d
N
4
OH
H
5
O
N
O
NH
6
OH
a 1.2 equiv amine, 100 °C, 5 h.
Several reactions also showed that the catalyst is suit-
able for recovery and re-use without a reduction in effi-
cacy (Table 5). After the first reaction, the catalyst was
extracted into water, which was subsequently removed
b Isolated yield (%).
c 24 h reaction time.
d Yields in parentheses refer to the other regioisomer.

D. B. G. Williams, M. Lawton / Tetrahedron Letters 47 (2006) 6557–6560
Table 4. Deactivated arylamine reactions with epoxides
6559
Entry
Product
Yielda (%)
Yieldb (%)
5 mol % Al(OTf)3 solventless
2 mol % Al(OTf)3
80
5 mol % Al(OTf)3
—
OH
Cl
H
N
1
—
OH
Cl
H
O
N
2
3
4
5
6
7
88
63
94
—
—
—
—
—
—
52
54
32
—
—
—
58
57
16
OH
Cl
H
N
OH
Cl
H
N
O
OH
H
N
NO
2
OH
H
N
NO2
OH
H
N
O
NO
2
OH
8
9
—
—
57b
57c
O
N
OH
41b
—
O
N
a 1.2 equiv amine, 100 °C, 5 h, isolated yield.
b 1.2 equiv amine, 100 °C, 24 h, isolated yield.
c 10 mol % Al(OTf)3, 1.2 equiv amine, 100 °C, 24 h, isolated yield.
Table 5. Yields (%) of catalyst recycling reactions
The slightly enhanced reactivity of the catalyst (see first
recycle) may be ascribable to the removal of Al-bound
water, affording a more active form of the catalyst. Dif-
ferential scanning calorimetry (DSC) of the commercial
Al(OTf)3 used in this study indicated an endotherm at
170 °C and another at 260 °C. We believe these endo-
therms to arise due to Al-bound water. The first endo-
therm is allocated to loosely bound water (probably
via hydrogen bonding in an outer ligand sphere) and
the second to tightly bound water (as a Lewis acid–
Lewis base pair); neither endotherm was observed for
pre-dried Al(OTf)3 (dried under vacuum at 120 °C for
48 h). In proof thereof, the dried Al(OTf)3 was exposed
to the atmosphere for 1 min at ambient temperature,
and application to DSC to the sample afforded a scan
showing the higher temperature endotherm, which we
ascribe to tightly bound water. Longer exposure
(5 min) showed a DSC scan demonstrating both
endotherms.
Product
Yielda(%)
Initial
1st recycle 2nd recycle
reaction
OH
H
N
82
90
90
92
—
88
OH
H
O
N
a Isolated yield (%).
under vacuum at elevated temperature. This cycle was
repeated for the second recycling reaction. It is clear
from Table 5 that the activity and selectivity are retained
during three cycles (two recycling steps).

6560
D. B. G. Williams, M. Lawton / Tetrahedron Letters 47 (2006) 6557–6560
4. (a) Sekar, G.; Singh, V. K. J. Org. Chem. 1999, 64, 287–
289; (b) Ollevier, T.; Lavie-Compin, G. Tetrahedron Lett.
2004, 45, 49–52.
5. Auge, J.; Leroy, F. Tetrahedron Lett. 1996, 64, 7715–7716.
6. Fugiwara, M.; Imada, M.; Bab, A.; Matsuda, H. Tetra-
hedron Lett. 1989, 30, 739–742.
7. Sagawa, S.; Abe, H.; Hase, Y.; Inaba, T. J. Org. Chem.
1999, 64, 4962–4965.
8. Fringuelli, F.; Pizzo, F.; Toroolli, S.; Vaccaro, L. J. Org.
Chem. 2004, 69, 7745–7747.
In summary, aluminium triflate is an effective, recyclable
Lewis acid catalyst for the aminolysis of epoxides, with
possible application in ‘green chemistry’ scenarios by
virtue of its activity under solventless conditions and
its ability to be recycled. The amine nucleophiles, being
hard bases, sometimes effectively compete with the oxy-
gen atom of the oxirane ring for complexation to the
aluminium triflate, especially in the cases where alkyl-
amines were employed. The putative stable chelate
structures, which presumably formed as intermediates
from the glycidyl ethers, were generally able to compete
more effectively for the aluminium than the simple alkyl
epoxides leading to improved reactions. However, when
present together with a strongly deactivated nucleophile,
the chelate structures proved deleterious.
9. Williams, D. B. G.; Lawton, M. Org. Biomol. Chem. 2005,
3, 3269–3272.
10. General experimental procedure for reactions in a solvent:
2.5 mL of toluene was added to the relevant amount of
Al(OTf)3 (as shown in the Tables) under an inert atmo-
sphere. Epoxide (3.5 mmol) and 1.2 equiv of the amine
were added and the mixture was heated at 100 °C for the
time indicated. Aqueous extraction and chromatography
over silica or vacuum distillation allowed the pure products
to be isolated. All compounds provided satisfactory 1H
NMR, 13C NMR, IR, MS and HRMS data consistent
with the structures shown. Regioisomers were identified
making use of HSQC and HMBC 2D NMR spectra.
11. Pearson, R. G.; Songstad, J. J. Am. Chem. Soc. 1967, 89,
1827–1836.
We are currently investigating the stereochemical out-
come of these reactions to determine whether or not
the stereochemistry present in the epoxide is retained
in the product.
References and notes
12. General reaction conditions for solventless reactions:
3.5 mmol of the epoxide and 1.2 equiv of the amine were
added to the relevant amount of Al(OTf)3 (as shown in the
Tables) under an inert atmosphere. The mixture was
heated at 100 °C for 24 h. Aqueous extraction and
chromatography over silica allowed the pure products to
be isolated.
1. Ager, D. J.; Prakash, I.; Schaad, D. R. Chem. Rev. 1996,
96, 835–875.
´
ˇ
2. Cepanec, I.; Litvic, M.; Mikuldas, H.; Bartolincic, A.;
ˇ ´
´
Vinkovic, V. Tetrahedron 2003, 59, 2435–2439.
3. Chini, M.; Crotti, P.; Macchia, F. J. Org. Chem. 1991, 56,
5939–5942.