Synthesis of heavily substituted 1,2-amino alcohols in enantiomerically pure form

Article abstract of DOI:10.1021/jo050960+

A simple and convenient methodology for the preparation of optically pure 2-amino-2-aryl-1,1-diphenylethanols is presented. Allylamine was found to produce the ring-opening of triaryloxiranes in a regioselective and a stereospecific fashion. Removal of the allyl protecting group provided the free 1,2-amino alcohols in enantiomerically pure form.

Full text of DOI:10.1021/jo050960+

Synthesis of Heavily Substituted 1,2-Amino
Alcohols in Enantiomerically Pure Form
Noem´ı Garc´ıa-Delgado,† Katamreddy Subba Reddy,†
Llu´ıs Sola`,‡ Antoni Riera,† Miquel A. Perica`s,*,†,‡ and
Xavier Verdaguer*,†
Unitat de Recreca en S´ıntesi Asime`trica (URSA-PCB),
Parc Cientı´fic de Barcelona and Departament de Qu´ımica
Orga`nica, Universitat de Barcelona, c/Josep Samitier,
1-5, E-08028 Barcelona, Spain, and Institute of Chemical
Research of Catalonia (ICIQ), E-43007 Tarragona, Spain
mapericas@iciq.es; xverdaguer@pcb.ub.es
Received May 13, 2005
FIGURE 1.
SCHEME 1
and diarylzinc to aldehydes.5 Our recent studies indicate
that the bulky triaryl structure in 4 is responsible for
this high catalytic activity.6 To extend the use of con-
gested 1,2-amino alcohols as a common building block
for other applications, an efficient synthesis of 2-amino-
1,1,2-triarylethanol structures is required.
A simple and convenient methodology for the preparation
of optically pure 2-amino-2-aryl-1,1-diphenylethanols is
presented. Allylamine was found to produce the ring-opening
of triaryloxiranes in a regioselective and a stereospecific
fashion. Removal of the allyl protecting group provided the
free 1,2-amino alcohols in enantiomerically pure form.
Traditionally, the synthesis of highly substituted 1,2-
amino alcohols is carried out from natural amino acids
by nucleophilic attack of the carboxylic moiety with a
Grignard or alkyllithium reagent7 (Scheme 1). For ex-
ample, the synthesis of 2-amino-1,1,2-triphenylethanol
(5) has been reported from phenylglycine and 10 equiv
of PhMgBr.8 However, this approach has several draw-
backs. It requires a large excess of Grignard reagent, and
more significantly, the use of an optically pure amino acid
as starting material limits the final structures that can
be acceded to commercially available amino acids. Given
that for many applications the nitrogen atom in the
amino alcohol moiety acts as the metal chelating element
it would be of great interest to modify at will the
electronic and steric properties of the amino group. An
alternative oxirane ring-opening approach should, in
principle, allow for an efficient introduction of diversity
in the 2 position, since different aryl groups can be
introduced at that center at the olefin construction stage
(Scheme 1). However, reliable procedures for the intro-
duction of primary amino groups via ring-opening of
sterically congested epoxides are not available. In this
context, here we report the development of a simple and
1,2-Amino alcohols are among the most valuable chiral
building blocks for asymmetric synthesis and catalysis.1
A large number of chiral auxiliaries and ligands hold 1,2-
amino alcohols as stereogenic fragments (Figure 1).
Oxazolidinone chiral auxiliaries (1), BOX ligands (2), and
Salen-type ligands (3) exemplify the importance of these
structures as building blocks.2 Furthermore, â-amino
alcohols are efficient catalysts for the ligand accelerated
addition of dialkyzinc species to carbonyls.3
Our group has developed families of modular chiral
amino alcohols for this and other transformations, start-
ing from synthetic enantiopure epoxides.4 Among these,
the bulky 2-piperidino-1,1,2-triphenylethanol (4), con-
taining a diphenylcarbynol moiety, is one of the most
active catalysts for the asymmetric addition of dialkyl
† Universitat de Barcelona.
‡ Institute of Chemical Research of Catalonia.
(1) For a review on catalytic applications of amino alcohols, see:
Fache, F.; Schulz, E.; Tommasino, L. M.; Lemaire, M. Chem. Rev. 2000,
100, 2159-2232.
(2) For references to selected structures, see: (a) Davies, S. G.;
Sanganee, H. J. Tetrahedron: Asymmetry 1995, 6, 671-674. (b) Evans,
D. A.; Scheidt, K. A.; Johnston, J. N.; Willis, M. C. J. Am. Chem. Soc.
2001, 123, 4480-4491. (c) Legros, J.; Bolm, C. Angew. Chem., Int. Ed.
2004, 43, 4225-4228.
(3) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis; John
Wiley & Sons: New York, 1994. (b) Chen, Y. K.; Costa, A. M.; Walsh,
P. J. J. Am. Chem. Soc. 2001, 123, 5378-5379.
(5) (a) Sola`, L.; Reddy, K. S.; Vidal-Ferran, A.; Moyano, A.; Perica`s,
M. A.; Riera, A.; Alvarez-Larena, A.; Piniella, J. F. J. Org. Chem. 1998,
63, 7078-7082. (b) Fontes, M.; Verdaguer, X.; Sola`, L.; Perica`s, M. A.;
Riera, A. J. Org. Chem. 2004, 69, 2532-2543.
(6) Caldentey, F. X.; Jimeno, C.; Perica`s, M. A. Institute of Chemical
Research of Catalonia (ICIQ), 2004.
(7) Amino alcohols can be also synthesized from mandelic acid.
However, from the point of view of diversity, this approach has the
same limitation as the amino acid route. See: Braun, M.; Fleischer,
R.; Mai, B.; Schneider, M.-A.; Lachenicht, S. Adv. Synth. Catal. 2004,
346, 474-482.
(8) Bach, J.; Berenguer, R.; Garcia, J.; Loscertales, T.; Vilarrasa, J.
J. Org. Chem. 1996, 61, 9021-9025.
(4) (a) Vidal-Ferran, A.; Moyano, A.; Perica`s, M. A.; Riera, A. J. Org.
Chem. 1997, 62, 4970-4982. (b) Vidal-Ferran, A.; Moyano, A.; Perica`s,
M. A.; Riera, A. Tetrahedron Lett. 1997, 38, 8773-8776. (c) Reddy, K.
S.; Sola`, L.; Moyano, A.; Perica`s, M. A.; Riera, A. J. Org. Chem. 1999,
64, 3969-3974. (d) Reddy, K. S.; Sola`, L.; Moyano, A.; Perica`s, M. A.;
Riera, A. Synthesis 2000, 2000, 165-176. (e) Jimeno, C.; Pasto, M.;
Riera, A.; Perica`s, M. A. J. Org. Chem. 2003, 68, 3130-3138. (f) Pasto´,
M.; Riera, A.; Perica`s, M. A. Eur. J. Org. Chem. 2002, 2337-2341. (g)
Garc´ıa-Delgado, N.; Fontes, M.; Perica`s, M. A.; Riera, A.; Verdaguer,
X. Tetrahedron: Asymmetry 2004, 15, 2085-2090.
10.1021/jo050960+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/04/2005
7426
J. Org. Chem. 2005, 70, 7426-7428

SCHEME 2
SCHEME 3
SCHEME 4
TABLE 1. Ring-Opening of (S)-Triphenylethylene Oxide
with Various Ammonia Equivalents
entry
R1
Bn
PMB
allyl
allyl
R2
T (°C) time (h) yield (%)a product
1
H
140
140
120
120
24
24
24
72
70
72
97
47
7
8
2
H
3b
4
H
9
allyl
10
a Yields refer to compounds purified by flash chromatography.
b Allylamine has a boiling point of 53-54 °C. Heating the reaction
to 120 °C produces an overpressure of approx 1.5 bar. The reaction
was regularly run in a glass pressure tube.
general synthesis of 2-amino-2-aryl-1,1-diphenylethanol
structures from the corresponding optically pure oxiranes
(Scheme 1).
substituted carbon. Among these reagents, allylamine
afforded amine 9 in an excellent 97% yield (Table 1, entry
3). Secondary amines such as diallylamine gave, in
general, lower yields even when reaction times were
extended (Table 1, entry 4).
Removal of the amino-protecting group from 7, 8, and
9 was next examined. While benzylic-protecting groups
for 7 and 8 yielded complex reaction mixtures under
hydrogenolytic (H2, Pd/C) and oxidative conditions (DDQ,
CAN), for 9, double bond isomerization and imine hy-
drolysis in water provided 5 in 91% yield (Scheme 3).
Most rewardingly, chiral HPLC analysis of the amino
alcohol product confirmed that no racemization had taken
place either in the initial epoxide ring-opening step with
allylamine or in the deprotection of the allyl group with
Pd/C in the presence of methansulfonic acid in refluxing
water.11
To determine the scope of the present sequence for the
synthesis of 2-amino-2-aryl-1,1-diphenylethanols, several
triarylethylene oxides were prepared from the corre-
sponding olefins by Jacobsen epoxidation (Scheme 4).12
Metalation of diphenylmethane with n-BuLi and reaction
with the appropriate aromatic aldehyde provided an
intermediate secondary alcohol that, upon dehydration
(TsOH in toluene at reflux), afforded the corresponding
alkene precursors in good overall yields. Alternatively,
the synthesis of 1-(3,5-dimethylphenyl)-2,2-diphenylethene
was performed under Wittig olefination conditions. Asym-
metric salen-manganese-catalyzed epoxidation of the
resulting olefins provided the oxirane substrates in
moderate to excellent yields (44-91%) and enantiomeric
We have reported several examples of stereospecific
ring-opening of triarylethylene oxides (Scheme 2). When-
ever a cyclic, secondary amine is involved as the nucleo-
phile in this process, the reaction occurs regioselectively,
the less substituted carbon of the epoxide being exclusive-
ly5a or preferentially4d attacked. The reaction failed,
however, for ammonia and other standard ammonia
equivalents. The use of sodium azide under thermal
conditions, Crotti conditions,9 or microwave activation led
to no detectable levels of reaction on triphenylethylene
oxide. Much in the same way, aqueous ammonia under
thermal or microwave activation (pressure tube in both
cases) was also ineffective.10 In contrast, reaction with
diisopropoxytitanium diazide provided stereospecifically
an azide ring-opening product with opposite regiochem-
istry (Scheme 2).4c
To overcome this limitation, and in search for a large-
scale, safe process leading to 5, a series of benzylic and
allylic ammonia equivalents were studied for the ring-
opening of 6 (Table 1). Initial experiments under the
original Crotti conditions (LiClO4, 70 °C, acetonitrile)
failed to yield any of the ring-opening products. In
contrast, good yields of the corresponding amino alcohols
were obtained when the reaction was run at higher
temperatures (120-140 °C) using the nucleophilic amine
as a solvent (Table 1). Primary benzylic and allylic
amines gave better results and provided in good yield the
reaction products resulting from ring-opening at the less
(9) (a) Chini, M.; Crotti, P.; Macchia, F. J. Org. Chem. 1991, 56,
5939-5942. (b) Chini, M.; Crotti, P.; Macchia, F. Tetrahedron Lett.
1990, 32, 4661-4664.
(10) For examples of epoxide ring-opening in aqueous ammonia,
see: Pasto´, M.; Rodriguez, B.; Riera, A.; Perica`s, M. A. Tetrahedron
Lett. 2003, 44, 8369-8372.
(11) (a) Liu, Q.; Marchington, A. P.; Boden, N.; Rayner, C. M. J.
Chem. Soc., Perkin Trans. 1 1997, 511-525. (b) Picq, D.; Anker, D.;
Rousset, C.; Laurent, A. Tetrahedron Lett. 1983, 24, 5619-5622.
(12) Brandes, B. D.; Jacobsen, E. N. J. Org. Chem. 1994, 59, 4378-
4380.
J. Org. Chem, Vol. 70, No. 18, 2005 7427

TABLE 2. Synthesis of
2,2,3-triphenyloxirane 5 (0.500 g, 1.84 mmol), LiClO4 (0.391 g,
3.67 mmol), and allylamine (1.4 mL, 18.40 mmol). The suspen-
sion was then stirred at 120 °C overnight. The resulting solution
solidified when cooled to room temperature. The crude was then
disolved in dichloromethane, and the organic layer was washed
successively with brine. The organics were dried (MgSO4) and
removed together with the excess amine (bp 53 °C) under
vaccum. The pure product was obtained as a white solid in 97%
yield (0.590 g, 1.8 mmol); 99% ee. mp ) 110-112 °C; [R]D +168.9
(c 0.945, CHCl3); 1H NMR (400 MHz, CDCl3) δ 7.70 (d, J ) 7
Hz, 2H), 7.40 (t, J ) 8 Hz, 2H), 7.28 (m, 1H), 7.15-6.95 (m,-
10H), 5.80 (m, 1H), 5.00 (m, 2H), 4.70 (s, 1H), 4.30 (broad s,
1H), 3.10 (dd, J ) 14 Hz, J′ ) 5 Hz, 1H), 2.98 (dd, J ) 14 Hz, J′
) 7 Hz, 1H), 1.60 (broad s, 1H) ppm; 13C NMR (100 MHz, CDCl3)
δ 145.8 (C), 143.5 (C), 137.5 (C), 136.2 (CH), 129.5 (CH), 128.5
(CH), 127.4 (CH), 127.3 (CH), 127.2 (CH), 127.2 (CH), 126.6 (CH),
126.2 (CH), 126.1 (CH), 116.4 (CH2), 79.9 (C), 68.1 (CH), 49.4
(CH2) ppm; IR (KBr) νmax 3400, 3100, 2850, 1489 cm-1; MS (CI-
2-Amino-2-aryl-1,1-diphenylethanolsa
a Conditions A: Allylamine (10 equiv), LiClO4 (2 equiv), 120 °C
in a pressure tube, 24 h. Conditions B: 5 mol % Pd/C, MsOH (2
equiv), H2O, reflux. b Yields refer to isolated crystalline compounds
of >95%. Purity as determined by elemental analysis.
NH3) m/z: 329 [(M)+, 100%]; HRMS (CI-CH4): calcd for C23H23
-
NO (M + H+) 330.1858, found 330.1863; HPLC Daicel CHIRAL-
CEL-ODH. Hexane/i-PrOH 95:5, 0.5 mL/min, λ ) 254 nm, tR
(R) ) 8.7 min, tR (S) ) 10.1 min.
excess between 70 and 91% ee. In most cases, a single
crystallization sufficed to afford enantiomerically pure
(99% ee) products.
(R)-2-Amino-1,1,2-triphenylethanol (5). (R)-2-(Allylamino)-
1,1,2-triphenylethanol (9) (114 mg, 0.35 mmol), methanesulfonic
acid (45 µL, 0.69 mmol), 10% Pd/C (47 mg), and water (5 mL)
were added to a two-necked round-bottom flask. The mixture
was heated under reflux overnight. A slow flow of argon passed
through the solution to aid removal of propionaldehyde. The
solution was then basified with NaOH to pH 10, and a white
solid appeared. To remove the catalyst the suspension was
filtered through Celite, and the residue was washed with
dichloromethane. The organic layer was washed with brine,
dried (MgSO4), and concentrated under vacuum. The residue
was purified by column chromatography on silica gel (hexane/
ethyl acetate) to afford 5 (91 mg, 0.31 mmol) as white solid in
With these oxirane substrates in hand, the ring-
opening with allylamine and subsequent deprotection
sequence was assayed (Table 2). Reaction with allylamine
was monitored by TLC for the disappearance of the
starting material (approximately 24 h) and afforded the
corresponding secondary allylamines as crystalline solids
in high yields (70-99%). Allyl deprotection was effected
in refluxing water, the propanal produced being continu-
ously removed from the reaction medium with the aid of
a small stream of inert gas. The resulting 1,2-amino
alcohols were obtained as crystalline solids in high optical
purity.13
In summary, we demonstrated here that allylamine is
an effective synthetic equivalent of ammonia for the
transformation of epoxides into 1,2-amino alcohols. For
encumbered oxirane rings in which azide fails to provide
ring-opening products, allylamine produces a smooth
stereospecific reaction. The present methodology is a
simple and convenient technique for the preparation of
bulky 2-amino-2-aryl-1,1-diphenylethanols and is the
only method currently available when the corresponding
arylglycines are not accessible in enantiomerically pure
form.
1
91% yield and 99% ee. H NMR (400 MHz, CDCl3) δ 7.75 (d, J
) 8 Hz, 2H), 7.40 (t, J ) 7 Hz, 2H), 7.30-7.00 (m, 11H), 5.00 (s,
1H), 1.62 (broad s, 2H); HPLC Daicel CHIRALCEL-ODH.
Hexane/i-PrOH 80:20, 0.5 mL/min, λ ) 254 nm, tR (R) ) 19.7
min, tR (S) ) 21.1 min.
Acknowledgment. We thank the DGI-MCYT (Grant
BQU2002-02459), the DURSI (Grant 2001SGR50), and
the “Fundacio´n Ramon Areces” for financial support.
D.C. thanks the MECyD for a fellowship. M.A.P. and
X.V. thank the DURSI for the “Distincio´ per a la
Promocio´ de la Recerca Universitaria” awards. N.G.D.
thanks the DURSI and the Universitat de Barcelona
for a fellowship.
Supporting Information Available: General experimen-
tal information, complete experimental details, and charac-
terization data for compounds 11a-d, 12a-d, and 13a-d.
Experimental Section
(R)-2-(Allylamino)-1,1,2-triphenylethanol (9). A glass pres-
sure tube equipped with a stirring bar was charged with (S)-
1
Copy of H and 13C NMR spectra for compounds 9, 11d, 12d,
13b, and 13d. This material is available free of charge via the
Internet at http://pubs.acs.org.
(13) By analogy with 5, the resulting 1,2-amino alcohols we assumed
to preserve the optical purity of the starting epoxides.
JO050960+
7428 J. Org. Chem., Vol. 70, No. 18, 2005