Synthesis of nitrile coordinated Lewis acids Al(OC(CF3) 2R)3 and their application in catalytic epoxide ring-opening reactions

Article abstract of DOI:10.1016/j.apcata.2010.06.029

Four nitrile ligated aluminum-based Lewis acids PhCNAl(OC(CF ₃)₂PhCH₃)₃ (1), CH ₃CNAl(OC(CF₃)₂PhCH₃)₃ (2), PhCNAl(OC(CF₃)₂Ph)₃ (3), CH ₃CNAl(OC(CF₃)₂Ph)₃ (4) were synthesized and characterized by nuclear magnetic resonance and infrared spectroscopy, as well as X-ray crystallography. The ring-opening reactions of epoxides with aromatic/aliphatic amines were efficiently catalyzed by compounds 1-4 as catalysts in a concentration of 1 mol% under solvent-free conditions at room temperature, affording 2-amino alcohols in high yields (up to 99%) within 4 h. Compound 1, being the best catalyst, yields 70% product within 10 min. Published by Elsevier B.V.

Full text of DOI:10.1016/j.apcata.2010.06.029

Applied Catalysis A: General 384 (2010) 171–176
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Synthesis of nitrile coordinated Lewis acids Al(OC(CF3)2R)3 and their application
in catalytic epoxide ring-opening reactions
Yang Lia, Yi Tana, Eberhardt Herdtweckb, Mirza Cokojab, Fritz E. Kühna,b,∗
a Molecular Catalysis, Catalysis Research Center, Technische Universität München, Lichtenbergstrasse 4, D-85747 Garching, Germany
b Chair of Inorganic Chemistry, Catalysis Research Center, Department of Chemistry, Technische Universität München, Lichtenbergstrasse 4, D-85747 Garching, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Four nitrile ligated aluminum-based Lewis acids PhCNAl(OC(CF3)2PhCH3)3 (1), CH3CNAl(OC(CF3)2
PhCH3)3 (2), PhCNAl(OC(CF3)2Ph)3 (3), CH3CNAl(OC(CF3)2Ph)3 (4) were synthesized and characterized
by nuclear magnetic resonance and infrared spectroscopy, as well as X-ray crystallography. The ring-
opening reactions of epoxides with aromatic/aliphatic amines were efficiently catalyzed by compounds
1–4 as catalysts in a concentration of 1 mol% under solvent-free conditions at room temperature, afford-
ing 2-amino alcohols in high yields (up to 99%) within 4 h. Compound 1, being the best catalyst, yields
70% product within 10 min.
Received 26 March 2010
Received in revised form 2 June 2010
Accepted 15 June 2010
Available online 22 June 2010
Keywords:
Homogeneous catalysis
Lewis acid
© Published by Elsevier B.V.
Ring-opening reaction
Epoxides
Aluminum
1. Introduction
recently as Lewis acid fragment in ‘frustrated Lewis Pair’-catalyzed
hydrogen activation [14]. Besides B(C6F5)3, other bulky molecular
Lewis acid promoted ring opening of epoxides by amines is an
important organic transformation both for academia and chemi-
cal/pharmaceutical industry as the resulting 2-amino alcohols are
versatile intermediates for the synthesis of a vast range of biolog-
ically active products [1], unnatural amino acids [2], and chiral
auxiliaries [3] for asymmetric synthesis. Various protocols have
been developed for activating epoxides to nucleophilic cleavage by
amines. These include the use of metal amides and triflamide [4],
metal alkoxides [5], zirconium sulfophenyl phosphonate [6], metal
and lanthanide halides [7], metal salts with weakly coordinating
anions [8], and others [9]. However, these traditional methodolo-
gies suffer from disadvantages such as long reaction times, elevated
temperatures, high pressure, moderate yields and so on. Therefore,
the development of more efficient procedures is in high demand.
The air stable, bulky Lewis acid tris(pentafluorophenyl) borane
B(C6F5)3 [10] has been extensively used as co-catalyst in the field
of cationic olefin polymerization [11]. Moreover, B(C6F5)3 can
be applied as catalyst for organic transformations such as ring-
opening reactions of epoxides [12] and aziridines [13], and very
Lewis acids are also reported, such as As(OTeF5)5 [15], B(OTeF5)3
[16], 1,2-((C6F5)2B)2C6F4 [17] and others. Very recently, Krossing
et al. communicated the synthesis and characterization of the non-
oxidizing superacid Al(OC(CF3)3)3 [18], which is relatively stable
and considered to be one of the strongest Lewis acids reported
to date. These extraordinary properties indicate that Al(OC(CF3)3)3
and similar compounds would have a great potential as Lewis acid
catalysts for various organic transformations, such as the above
mentioned catalytic ring-opening reactions of epoxides.
Recently, Chakraborti and co-workers reported improved cat-
alytic synthesis of 2-amino alcohols under solvent-free conditions
at room temperature with cheap catalysts ranging from normal
metal salts LiBr [19], Zn(ClO4)2 [8c] and ZrCl4 [20], to solid acids
such as silica gel [21] and montmorillonite K10 [22]. The envi-
ronmentally benign procedures are marked with short reaction
times (several hours), high yields and good to excellent regio- and
diastereoselectivities.
In this work, we report the synthesis and characterization of
four nitrile ligated Lewis acids, namely PhCNAl(OC(CF3)2PhCH3)3
(1), CH3CNAl(OC(CF3)2PhCH3) (2), PhCNAl(OC(CF3)2Ph)3 (3) and
CH3CNAl(OC(CF3)2Ph) (4). Preliminary results of their cat-
alytic application in epoxide ring-opening reactions by amines
under solvent-free conditions at room temperature are also
presented.
∗
Corresponding author at: Chair of Inorganic Chemistry, Catalysis Research Cen-
ter, Department of Chemistry, Technische Universität München, Lichtenbergstrasse
4, D-85747 Garching, Germany.
E-mail address: fritz.kuehn@ch.tum.de (F.E. Kühn).
0926-860X/$ – see front matter © Published by Elsevier B.V.
doi:10.1016/j.apcata.2010.06.029

172
Y. Li et al. / Applied Catalysis A: General 384 (2010) 171–176
Scheme 1. Synthesis of complexes 1–4.
Table 1
27Al NMR resonances of complexes 1–4.
Complex
1
2
3
4
Signal (ppm)
58.60
44.20
56.35
42.63
2. Results and discussion
2.1. Synthesis and characterization
Complexes 1–4 were synthesized by reacting AlMe3 with
the polyfluorinated alcohols HOC(CF3)2R (R = Ph or PhCH3; see
Scheme 1). The complexes were isolated in yields of 70–90%
by removing the volatiles in vacuo. Compounds 1–4 can be fur-
ther purified by recrystallization from a mixture of corresponding
nitriles, dichloromethane and pentane at −35 ◦C. Nevertheless, in
most cases, the crude product is already pure enough to give sat-
isfactory characterization data. Attempts to use non-coordinating
solvents such as toluene or dichloromethane as reaction media only
led to mixtures of oily solids, which were not possible to identify
and presumably resulting from product decomposition, including
C–F activation and formation of aluminum fluorides [18].
Fig. 1. ORTEP drawing of MeCNAl(OC(CF3)2PhCH3)3 2 at 173 K with thermal ellip-
soids set at 50% probability. Hydrogen atoms were omitted for clarity. Selected
bond lengths [Å] and angles [◦]: Al1–O1 1.691(1), Al–O2 1.692(1), Al–O3 1.689(1),
Al1–N1 1.920(1), N1–C1 1.131(2); O1–Al1–O2 113.11(6), O1–Al1–O3 112.79(6),
O2–Al1–O3 111.59(5), O1–Al1–N1 104.51(6), O2–Al1–N1 104.85(5), O3–Al1–N1
109.39(6), Al1–N1–C1 169.3(1).
Complexes 1–4 were characterized by NMR (1H, 13C, 19F) mea-
surements. All spectra are in good agreement with expected and
reported values [18]. The 27Al NMR spectra of 1–4 (Table 1) exhibit
broad signals ranging from 43 to 57 ppm, which is quite similar
to the 27Al NMR shift of the parent compound PhFAl(OC(CF3)3)3
reported by Krossing et al. (38 ppm) [18].
Scheme 2. Ring-opening reaction of cyclohexene oxide with aniline catalyzed by
compounds 1–4.
The aluminum coordination geometry is a distorted tetrahedron
with average O–Al–O angles of 112.5◦ and average N–Al–O angles
of 106.3◦. The acetonitrile is loosely coordinated to the aluminum
atom as seen on the relatively long Al1–N1 bond of 1.920(1) Å,
which is comparable to the already reported analogous complex
[Al{OB(Mes)2}3(NCMe)] (1.594(2) Å) [24]. This observation again
shows the weaker Lewis acidity of Al(OC(CF3)2PhCH3)3 compared
to B(C6F5)3. The acetonitrile adduct of the latter shows a very
strong B–N bond (1.616(3) Å) [24]. Consistent to the IR measure-
ment, N1–C1 bond becomes slightly stronger (1.131(2) Å compared
to 1.141(2) Å of free acetonitrile [25]) due to the ␴-donation of
electron density from the lone pair of the nitrogen [26].
It is known that the stretching frequency of the C N bond of the
benzonitrile adduct is dependent on the Lewis acid strength [23].
The stronger a Lewis acid is, the higher the vibration frequency of
the C N bond. As seen in Table 2, compounds 1 and 3 show signif-
icantly higher C N stretching frequencies than free benzonitrile.
The nearly identical ꢀCN values of 1 and 3 (2286 and 2288 cm−1
)
suggest that substitution of a phenyl group by a methylphenyl
group has barely any influence on the Lewis acidity. On the other
hand, the ꢀCN value of benzonitrile coordinated B(C6F5)3 (5) is much
higher [24]. This clearly demonstrates that B(C6F5)3 is a stronger
Lewis acid than the other two. The same situation is also observed
for acetonitrile ligated complexes 2, 4 and 6 [25].
2.2. Studies of the catalytic activity of compounds 1–4
Single crystals of 2 suitable for an X-ray crystal structure anal-
ysis were obtained by slow diffusion of pentane into a mixture
of acetonitrile and dichloromethane solution of 2. The X-ray crys-
tal structure analysis of 2 confirms the formation of a 1:1 adduct
between acetonitrile and Al(OC(CF3)2PhCH3)3 (Fig. 1).
Complexes 1–4 and B(C6F5)3 were tested as catalysts for ring-
opening reaction of epoxides, with cyclohexene oxide and aniline
as test substrates (see Scheme 2). The results are summarized in
Table 3. Good to excellent yields were achieved when 1–4 were
Table 2
Selective FT-IR vibrational frequencies for 1–4.
Complex
1
3
2
4
5a
6b
PhCN
MeCN
ꢀCN (cm−1
)
2286
–
2288
–
2315
2341
2313
2344
2324
–
–
2367
2230
–
2253
2293
IR data from Refs. [24,25].
a
5 = PhCNB(C6F5)3.
6 = MeCNB(C6F5)3.
b

Y. Li et al. / Applied Catalysis A: General 384 (2010) 171–176
173
Table 3
Epoxide ring-opening reaction of cyclohexene oxide with aniline with various
catalysts.a.
No.
Catalyst
Yieldb (%)
1
2
3
4
5
6c
1
2
3
4
99
92
87
79
47
None
B(C6F5)3
None
a
5.0 mmol cyclohexene oxide, 5.0 mmol aniline, 1 mol% catalyst, 25 ◦C, 4 h.
Isolated yield.
24 h reaction time.
b
c
Fig. 2. Steric repulsion between bulky dibenzylamine and 1.
applied (79–99%), while only a moderate yield was obtained with
B(C6F5)3 as the catalyst (47%).
As can be seen from Fig. 2, assuming that the catalyst–epoxide
adduct is the key intermediate of the catalytic cycle [8c,20], then the
much bulkier dibenzylamine will encounter a larger steric repul-
sion than aniline, thus causing its very low activity.
Presumably, the higher yield in the case of carbazole versus
dibenzylamine is giving rise to the rigid structure of carbazole,
which is a flat molecule and thus provides a better access of the
amine to the aluminum center. Contrasting this, the rotation of
the two flexible benzyl rings of dibenzylamine creates extra steric
repulsion with the alkoxy moieties at the Al center. A moderate
yield (62%) was achieved with piperidine. Very poor yield (7%) was
obtained when allylamine was used as substrate, indicating the
system is incompatible with allylamines.
The system was also tested with various epoxides and aniline in
the presence of 1 (Table 5). Good yields of 2-amino alcohols were
obtained for cyclooctene oxide (86%) and cyclopentene oxide (73%).
In the case of 2-hexyloxirane, the yield was much lower (41%).
Based on the above observations and a previous investiga-
tion [8c], a mechanism is proposed (Scheme 3). The catalyst
RꢀCNAl(OC(CF3)2R)3 first loses the nitrile ligand to generate the
active species Al(OC(CF3)2R)3. The epoxide oxygen atom coor-
dinates to the aluminum atom immediately after the vacant
coordination site is generated, forming intermediate I. The Lewis
acid–base adduct I significantly weakens the C–O bonds and delo-
calizes the positive charge over the two adjacent carbon atoms,
either of which is then attacked by the nucleophilic nitrogen atom
of the amine in step c, leading to intermediate II.
The results show that the catalytic activities of the complexes
follow the inverse sequence of their Lewis acidity (catalytic reactiv-
ity: 1, 2 > 3, 4 > B(C6F5)3, Lewis acidity: B(C6F5)3 > 3, 4 > 1, 2), with
the strongest Lewis acid B(C6F5)3 being the least active catalyst. It
should also be noted that the benzonitrile ligated complexes show
higher catalytic activity than the acetonitrile ligated ones in both
pairs (1 > 2 and 3 > 4). As expected, no product was obtained after
24 h without a catalyst being present.
The general application of the most active catalyst 1 for the
ring opening of cyclohexene oxide was then studied with vari-
ous aromatic and aliphatic amines (Table 4). The electron deficient
perfluorinated aniline afforded only 52% yield, indicating that a
nucleophilic addition of amine to the epoxide is involved and
amines with higher pKa value are favored [8c]. With bulkier ani-
line derivatives such as carbazole, the yield was only slightly lower
(92%). However, no product was detected for dibenzylamine under
the applied reaction, probably due to steric hindrance.
Table 4
Epoxide ring-opening reaction of cyclohexene oxide with various amines catalyzed
by 1.a.
No.
1
Amine
Product
Yieldb (%)
99
An intramolecular proton transfer takes place via the transition-
state III as the oxyanionic site forms a hydrogen bond with one
2c
52
Table 5
Epoxide ring-opening reaction of various epoxides by aniline in the presence of 1.a.
No.
1
Epoxide
Product
Yieldb (%)
3c
92
99
4
5
None
2
3
86
73
41
62
7
6
4
a
5.0 mmol cyclohexene oxide, 5.0 mmol amine, 1 mol% catalyst, 25 ◦C, 4 h.
Isolated yield.
1 mL CH2Cl2 as solvent.
a
5.0 mmol epoxide, 5.0 mmol aniline, 1 mol% catalyst, 25 ◦C, 4 h.
Isolated yield.
b
c
b

174
Y. Li et al. / Applied Catalysis A: General 384 (2010) 171–176
base character allows the epoxide to coordinate to Al. The amine
coordination to Al i.e. the catalyst deactivation depends on the
nature of the amine: stronger ␴-donors such as allyl amine or
piperidine thus appear to be less suitable co-substrates for the ring
opening of epoxides.
Additionally, the time-dependent feature of the catalytic system
was also investigated (Fig. 3). The reaction of cyclohexene oxide
with aniline catalyzed by 1 was monitored by GC at different time
intervals. The high initial reaction rate leads to ca. 70% yield within
only 10 min (TOF = 420 h−1).
3. Conclusion
Acetonitrile and benzonitrile coordinated aluminum-based
Lewis acids 1–4 were successfully synthesized in high yields and
characterized by multinuclear NMR (1H, 13C, 19F, 27Al), FT-IR and
elemental analysis. Additionally, an exemplary structure of com-
plex 2 was obtained by X-ray single crystal diffraction. Complexes
1–4 are weaker Lewis acids than the well known B(C6F5)3 but show
higher catalytic activity towards ring-opening reaction of epox-
ides by amines. Yields up to 99% were obtained with 1 mol% of
compound 1 at room temperature under solvent-free conditions.
The preliminary results suggest that (a) less bulky amines and (b)
rather less strong ␴-donor amines strongly affect the product yield.
Detailed examinations concerning the applicability of the system
and reaction mechanism are currently being conducted in our lab-
oratories.
Scheme 3. Mechanism of epoxide ring-opening reaction by amine with
RꢀCNAl(OC(CF3)2R)3 as catalyst.
+
of the hydrogen atoms of the (R)ArNH2 moiety. This affords the
amino alcohol and releases the catalyst completing the catalytic
cycle. Therefore, the catalytic activity of the complex depends on:
(1) generation of the active species Al(OC(CF3)2R)3 to start the cat-
alytic cycle, (2) oxophility of the central aluminum atom to form a
strong coordination with the oxygen atom of the epoxide ring and
(3) proton transfer and dissociation of the catalyst from the final
product.
The results of Tables 2–4 can be rationalized with the help of
the mechanism proposed in Scheme 3. The higher catalytic activ-
ity of 1 and 3 can be explained by benzonitrile being a weaker and
more steric hindered Lewis base than acetonitrile [27]. Therefore,
in case of 1 and 3, the reaction equilibrium shifts slightly to non-
coordinated Al(OC(CF3)2R)3 (step a, Scheme 3), leading to a higher
concentration of the active species. The relative low catalytic effi-
ciency of the strongest Lewis acid B(C6F5)3 is most likely a result of
the fact that the borane forms stronger donor-acceptor bonds with
e.g. the amine, which consequently deactivates the catalyst. The
coordination of the epoxide to B(C6F5)3 is obviously not favored. In
the case of compounds 1–4, instead, their less pronounced Lewis
4. Experimental
All preparations and manipulations were carried out under
argon using standard Schlenk techniques. All solvents used were
dried by standard procedures. 1H, 13C and 19F NMR measurements
were performed on a JEOL 400 MHz spectrometer. 27Al NMR mea-
surements were performed on a Bruker AVANCE-DPX-400 MHz
spectrometer. FT-IR spectra were recorded on a Perkin Elmer FT-IR
spectrometer using Nujol. Elemental analyses were carried out at
the Mikroanalytisches Labor of TU München.
4.1. General procedure for the synthesis of 1 and 3
To a benzonitrile solution (3 mL) of HOC(CF3)2R, (R = Ph, PhCH3)
(3 mmol), a heptane solution of Al(CH3)3 (1.1 mmol, 1.1 mL) was
added dropwise at 0 ◦C within 30 min. The reaction solution was
stirred for 1 h at 0 ◦C and another 4 h at room temperature. The
product was obtained by removing the volatiles under high vac-
uum.
PhCNAl(OC(CF3)2PhCH3)3 1: HOC(CF3)2PhCH3 (3 mmol,
0.62 mL), 1 M solution of AlMe3 in n-heptane (1.1 mmol, 1.1 mL);
yield: (82%). Elemental Anal. Calc. (%) for C37H26AlF18NO3 (901.56):
C 49.29; H 2.91; F 37.93; N 1.55. Found: C 48.10; H 3.04; F 37.45; N
1.45; 1H NMR (400 MHz, CD2Cl2, 25 ◦C, ı (ppm)): 7.88 (d, 6H), 7.81
(t, 6H), 7.64 (t, 4H), 7.58 (d, 1H), 2.34 (s, 9H); 13C NMR (101 MHz,
CD2Cl2, 25 ◦C, ı (ppm)): 139.61, 137.23, 134.03, 130.21, 129.44,
128.87, 125.03, 122.15, 119.28, 118.85, 105.25, 29.74, 20.77; 19F
NMR (376.5 MHz, CD2Cl2, 25 ◦C, ı (ppm)): −75.33 (s); 27Al NMR
(104.3 MHz, PhCN, 25 ◦C, ı (ppm)): 58.6. Selected IR (Nujol, cm−1):
ꢀ (CN) 2725, 2392, 2286, 2234, 2000, 1973, 1920, 1903, 1811, 1765.
PhCNAl(OC(CF3)2Ph)3 3: HOC(CF3)2Ph (3 mmol, 0.505 mL), 1 M
solution of AlMe3 in n-heptane (1.1 mmol, 1.1 mL); yield: (90%).
Elemental Anal. Calc. (%) for C34H20AlF18NO3 (859.48): C 47.51; H
2.35; F 39.79; N 1.63. Found: C 46.30; H 2.33; F 39.76; N 1.66; 1H
NMR (400 MHz, CD2Cl2, 25 ◦C, ı (ppm)): 7.71 (d, 3H), 7.67 (d, 6H),
7.47 (d, 6H), 7.37–7.31 (m, 5H); 13C NMR (101 MHz, CD2Cl2, 25 ◦C,
ı (ppm)): 139.49, 136.76, 133.73, 129.81, 129.43, 125.03, 122.15,
Fig. 3. Time-dependent yield of 2-(phenylamino)cyclohexanol in the presence of
1 mol% 1 at 25 ◦C.

Y. Li et al. / Applied Catalysis A: General 384 (2010) 171–176
175
119.28, 118.85, 105.25, 29.74, 20.77; 19F NMR (376.5 MHz, CD2Cl2,
25 ◦C, ı (ppm)): −75.55 (s); 27Al NMR (104.3 MHz, PhCN, 25 ◦C, ı
(ppm)): 56.3. Selected IR (Nujol, cm−1): ꢀ (CN) 2728, 2287, 2239,
2008, 1973, 1903, 1811, 1765.
dichloromethane (6 mL). The reaction began with the addition of
cyclohexene oxide (5 mmol, 0.505 mL). The course of the reaction
was monitored by quantitative GC analysis. Samples were taken
(7.2 ␮L) in regular intervals (10, 30, 50 and 90 min), diluted with
1 mL acetonitrile, and treated with one drop of water to quench
the reaction. The resulting solution was injected into a GC column.
The product was identified by comparing its retention time with
that of authorized sample. The yield at each sampling time was
calculated from calibration curve recorded prior to the reaction
course.
4.2. General procedure for the synthesis of 2 and 4
To an acetonitrile (1 mL)/dichloromethane (2 mL) solution of
HOC(CF3)2R, R = Ph, PhCH3 (3 mmol), a heptane solution of Al(CH3)3
(1.1 mmol, 1.1 mL) was added dropwise at 0 ◦C within 30 min. The
reaction solution was kept stirring for 1 h at 0 ◦C and another 24 h
at room temperature. The product was obtained by removing the
volatiles in vacuo (1 × 10−3 mbar).
Acknowledgements
MeCNAl(OC(CF3)2PhCH3)3 2: HOC(CF3)2PhCH3 (3 mmol,
0.62 mL), 1 M solution of AlMe3 in n-heptane (1.1 mmol, 1.1 mL);
yield: (80%). Elemental Anal. Calc. (%) for C32H24AlF18NO3 (839.49):
C 45.78; H 2.88; F 40.74; N 1.67. Found: C 44.41; H 2.71; F 40.38;
N 1.62; 1H NMR (400 MHz, CD2Cl2, 25 ◦C, ı (ppm)): 7.56 (d, 6H),
7.17 (d, 6H), 2.49 (s, 3H), 2.35 (s, 9H); 13C NMR (101 MHz, CD2Cl2,
25 ◦C, ı (ppm)): 139.64, 130.11, 128.87, 125.03, 122.15, 119.28,
126.61, 124.97, 122.09, 20.78, 1.96; 19F NMR (376.5 MHz, CD2Cl2,
25 ◦C, ı (ppm)): −75.33 (s); 27Al NMR (104.3 MHz, MeCN, 25 ◦C, ı
(ppm)): 44.2. Selected IR (Nujol, cm−1): ꢀ (CN) 2726, 2315, 1919,
1813, 1653, 1615.
MeCNAl(OC(CF3)2Ph)3 4: HOC(CF3)2Ph (3 mmol, 0.505 mL), 1 M
solution of AlMe3 in n-heptane (1.1 mmol, 1.1 mL); yield: (87%). Ele-
mental Anal. Calc. (%) for C29H18AlF18NO3 (797.41): C 43.68; H 2.28;
F 42.89; N 1.76. Found: C 43.41; H 2.51; F 42.39; N 1.62; 1H NMR
(400 MHz, CD2Cl2, 25 ◦C, ı (ppm)): 7.72–7.48 (m, 12H,) 7.25 (s, 3H),
2.49 (s, 3H); 13C NMR (101 MHz, CD2Cl2, 25 ◦C, ı (ppm)): 139.53,
129.97, 128.77, 125.03, 122.15, 126.23, 124.76, 121.46, 29.85, 20.64,
2.97; 19F NMR (376.5 MHz, CD2Cl2, 25 ◦C, ı (ppm)): −75,32 (s); 27Al
NMR (104.3 MHz, MeCN, 25 ◦C, ı (ppm)): 42.6. Selected IR (Nujol,
cm−1): ꢀ(CN) 2725, 2392, 2292, 2241, 2007, 1931, 1915, 1821, 1765.
Y.L. is grateful to the University Bavaria e.V. for a Ph.D. grant.
Dr. Zhenggang Lan is acknowledged for useful discussions. The
International Graduate School of Science and Engineering (IGSSE)
is acknowledged for financial support.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apcata.2010.06.029.
References
[1] (a) K.B. Lindsay, S.G. Pyne, Tetrahedron 60 (2004) 4173;
(b) E. Ruediger, A. Martel, N. Meanwell, C. Solomon, B. Turmel, Tetrahedron
Lett. 45 (2004) 739.
[2] (a) P. O’Brien, Angew. Chem. Int. Ed. 38 (1999) 326;
(b) G.G. Li, H.T. Chang, K.B. Sharpless, Angew. Chem. Int. Ed. 35 (1996) 451.
[3] D.J. Ager, I. Prakash, D.R. Schaad, Chem. Rev. 96 (1996) 835.
[4] (a) L.E. Overman, L.A. Flippin, Tetrahedron Lett. 22 (1981) 196;
(b) M.C. Carre, J.P. Houmounou, P. Caubere, Tetrahedron Lett. 26 (1985) 3107;
(c) Y. Yamamoto, N. Asao, M. Meguro, N. Tsukade, H. Nemoto, N. Adayari, J.G.
Wilson, H. Nakamura, J. Chem. Soc., Chem. Commun. (1993) 1201;
(d) J. Cossy, V. Bellosta, C. Hamoir, J.-R. Desmurs, Tetrahedron Lett. 43 (2002)
7083.
[5] (a) S. Sagava, H. Abe, Y. Hase, T. Inaba, J. Org. Chem. 64 (1999) 4962;
(b) S. Rampalli, S.S. Chaudhari, K.G. Akamanchi, Synthesis (2000) 78.
[6] M. Curini, F. Epifano, M.C. Marcotullio, O. Rosati, Eur. J. Org. Chem. (2001) 4149.
[7] (a) M. Chini, P. Crotti, F. Macchia, Tetrahedron Lett. 31 (1990) 4661;
(b) J. Iqbal, A. Pandey, Tetrahedron Lett. 31 (1990) 575;
4.3. Single crystal X-ray structure determination
Compound 2: colorless plate, C32H24AlF18NO3, Mr = 839.50;
orthorhombic, space group Pbca (Nr. 61), a = 17.4940(3),
(c) C. Bonini, G. Righi, Synthesis (1994) 225;
(d) B.L. Chng, A. Ganesan, Bioorg. Med. Chem. Lett. 7 (1997) 1511;
(e) X.L. Fu, S.H. Wu, Synth. Commun. 27 (1997) 1677;
b = 13.7218(2),
ꢁ(MoK␣) = 0.71073 Å,
T = 173(1) K,
c = 29.6752(5) Å,
V = 7123.5(2) Å3,
,
Z = 8,
ꢂ = 0.186 mm−1
ꢃcalc. = 1.566 g cm−3
,
(f) S. Chandrasekhar, T. Ramachandar, S.J. Prakash, Synthesis (2000) 1817;
(g) L.R. Reddy, M.A. Reddy, N. Bhanumathi, K.R. Rao, Synthesis (2001) 831.
[8] (a) J. Auge, F. Leroy, Tetrahedron Lett. 37 (1996) 7715;
(b) D. Brandley, G. Williams, M. Lawton, Org. Biomol. Chem. 3 (2005) 3269;
(c) B. Shivani, A.K. Pujala, Chakraborti, J. Org. Chem. 72 (2007) 3713;
(d) M. Fujiwara, M. Imada, A. Baba, H. Matsuda, Tetrahedron Lett. 30 (1989)
739;
F(0 0 0) = 3376,
0.10 mm × 0.33 mm × 0.36 mm,
ꢄmax
:
25.34◦, R1 = 0.0323 {5741 observed data [Io > 2ꢅ(Io)]},
wR2 = 0.0908 (all 6518 data), GOF = 1.069, 500 parameters,
ꢆꢃmax/min = 0.25/−0.21 e Å−3
. Crystallographic data (excluding
structure factors) for the structure of complex 2 have been
deposited with the Cambridge Crystallographic Data Centre
as supplementary publication no. CCDC-776087. Copies of the
data can be obtained free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223 336 033;
e-mail: deposit@ccdc.cam.ac.uk). For more details see Supporting
Information.
(e) G. Sekar, V.K. Singh, J. Org. Chem. 64 (1999) 287.
[9] (a) G.H. Posner, D.Z. Rogers, J. Am. Chem. Soc. 99 (1977) 8208;
(b) M. Onaka, M. Kawai, Y. Izumi, Chem. Lett. (1985) 779;
(c) J. Yamada, M. Yumoto, Y. Yamamoto, Tetrahedron Lett. 30 (1989) 4255.
[10] G. Erker, Dalton Trans. (2005) 1883.
[11] (a) M. Bochmann, J. Chem. Soc., Dalton Trans. (1996) 255;
(b) E.Y.-X. Chen, T.J. Marks, Chem. Rev. 100 (2000) 1391;
(c) M.C. Baird, Chem. Rev. 100 (2000) 1471.
[12] S. Chandrasekhar, C.R. Reddy, B.N. Babu, G. Chandrashekar, Tetrahedron Lett.
43 (2002) 3801.
4.4. General procedure for epoxides ring opening by amines
[13] I.D.G. Watson, A.K. Yudin, J. Org. Chem. 68 (2003) 5160.
[14] (a) G.C. Welch, R.R.S. Juan, J.D. Masuda, D.W. Stephan, Science 314 (2006) 1124;
(b) P.A. Chase, D.W. Stephan, Angew. Chem. Int. Ed. 47 (2008) 7433;
(c) D. Holschumacher, T. Bannenberg, C.G. Hrib, P.G. Jones, M. Tamm, Angew.
Chem. Int. Ed. 47 (2008) 7428;
The epoxide (5 mmol) was added to a mixture of amine (5 mmol)
and catalyst (1 mol%). The mixture was stirred for 4 h at room tem-
perature. Then resulting mixture was diluted with Et2O (15 mL),
washed with water (5 mL), dried over Na2SO4, and concentrated in
vacuo to afford the respective 2-amino alcohol.
(d) T.A. Rokob, A. Hamza, A. Stirling, T. Soos, I. Papai, Angew. Chem. Int. Ed. 47
(2008) 2435;
(e) V. Sumerin, F. Schulz, M. Nieger, M. Leskela, T. Repo, B. Rieger, Angew. Chem.
Int. Ed. 47 (2008) 6001.
[15] M.J. Collins, G.J. Schrobilgen, Inorg. Chem. 24 (1985) 2608.
[16] F. Sladky, H. Kropshof, O. Leitzke, J. Chem. Soc., Chem. Commun. (1973) 134.
[17] V.C. Williams, W.E. Piers, W. Clegg, M.R.J. Elsegood, S. Collins, T.B. Marder, J.
Am. Chem. Soc. 121 (1999) 3244.
[18] L.O. Müller, D. Himmel, J. Stauffer, G. Steinfeld, J. Slattery, G. Santiso-Quinones,
V. Brecht, I. Krossing, Angew. Chem. Int. Ed. 47 (2008) 7659.
[19] A.K. Chakraborti, S. Rudrawar, A. Kondaskar, Eur. J. Inorg. Chem. (2004) 3597.
4.5. Kinetic measurements
Aniline (5 mmol, 0.456 mL), catalyst
dard 4-methylbenzophenone (2.5 mmol, 490 mg) were added to
the reaction vessel under standard conditions and diluted with
1 and internal stan-

176
Y. Li et al. / Applied Catalysis A: General 384 (2010) 171–176
[20] A.K. Chakraborti, A. Kondaskar, Tetrahedron Lett. 44 (2003) 8315.
[21] A.K. Chakraborti, S. Rudrawar, A. Kondaskar, Org. Biomol. Chem. 2 (2004) 1277.
[22] A.K. Chakraborti, A. Kondaskar, S. Rudrawar, Tetrahedron 60 (2004) 9085.
[23] K. Starowieyski, S. Pasynkiewicz, M. Boleslawski, J. Organomet. Chem. 10 (1967)
393.
[25] H. Jacobsen, H. Berke, S. Doring, G. Kehr, G. Erker, R. Frohlich, O. Meyer,
Organometallics 18 (1999) 1724.
[26] R.T. Henriques, E. Herdtweck, F.E. Kühn, A.D. Lopes, J. Mink, C.C. Romão, J. Chem.
Soc., Dalton Trans. (1998) 1293.
[27] (a) M. Berthelot, M. Helbert, C. Laurence, J.-Y.L. Questel, J. Phys. Org. Chem. 6
(1993) 302;
[24] C.H. Lee, S.J. Lee, J.W. Park, K.H. Kim, B.Y. Lee, J.S. Oh, J. Mol. Catal. A: Chem. 132
(1998) 231.
(b) J.-Y.L. Questel, M. Berthelot, C. Laurence, J. Phys. Org. Chem. 13 (2000) 347.