A practical and efficient synthesis of enantiomerically pure di-tert- butylethanediamine

Article abstract of DOI:10.1055/s-1999-3391

A diastereoselective synthesis of 1,2-diamino-1,2-di-tert-butylethane has been developed by addition of tert-butyl magnesium chloride to a chiral bis-imine derived from glyoxal and (S)-methylbenzylamine. Addition of the bis-imine to the Grignard reagent i

Full text of DOI:10.1055/s-1999-3391

228
SHORT PAPER
A Practical and Efficient Synthesis of Enantiomerically Pure Di-tert-butyl-
ethanediamine
Sylvain Roland,* Pierre Mangeney, Alex Alexakis
Laboratoire de Chimie des Organo-ElŽments, tr 44Ð45, 2 •me Žt., UniversitŽ P. et M. Curie, 4, Place Jussieu, 75252 Paris Cedex 05, France
Fax +33(1)44277567; E-mail: sroland@ccr.jussieu.fr
Received 3 August 1998
obtained 93.5% of the major diastereomer 5a, character-
ized by X-ray crystallography. Addition of phenylmagne-
Abstract: A diastereoselective synthesis of 1,2-diamino-1,2-di-tert-
butylethane has been developed by addition of tert-butyl magnesium
sium chloride and methylmagnesium bromide was also
studied by Simpkins and co-workers.7 These reagents
gave a mixture of diastereomers from which 5b and 5c
were respectively isolated in 47% and 35% yield.
chloride to a chiral bis-imine derived from glyoxal and (S)-methyl-
benzylamine. Addition of the bis-imine to the Grignard reagent in
hexane at 50¡C gave only one diastereomer detectable by 1H and 13
C
NMR. Hydrogenolysis of the phenylethyl groups led to the expected
free diamine 3 in good yields. The absolute configuration (R,R) of
the carbons bearing the tert-butyl groups has been conÞrmed by
X-ray spectroscopy.
Key words: chiral 1,2-diimine, diastereoselective addition, Grignard
reagent, hydrogenolysis, ligand precursor
In recent years, numerous applications have been devel-
oped in asymmetric synthesis using vicinal diamines as
the source of chirality. One of their major interest lies in
their use as precursors for the synthesis of a broad family
of bidentate ligands.1 Many reactions have also been de-
scribed using these diamines as chiral auxilaries and pro-
tecting groups of aldehydes.2
Scheme 1
Most of these applications generally use the framework of
diphenyldiaminoethane (1) or diaminocyclohexane (2)
whose preparations were fully described.3 However, in
several of our projects, we were in need of new 1,2 di-
amines. We were particularly interested in obtaining 1,2-
diamino-1,2-di-tert-butylethane (3) because of its in-
creased steric bulk and absence of benzylic protons.
We were attracted by the use of this procedure to develop
a diastereoselective synthesis of 1,2-diamino-1,2-di-tert-
butylethane (3) by addition of tert-butylmagnesium chlo-
ride to the chiral 1,2-bis-imine 4. To our knowledge, only
one method was described in the literature for the synthe-
sis of 3,8 by coupling of nitriles or N-(trimethylsilyl)im-
ines promoted by NbCl4(THF). In this procedure the
diamine was isolated as a 1.3:1 mixture of the (d,l:meso)
diastereoisomers and the (+)-diamine was obtained pure
in 18% yield by resolution with (Ð)-mandelic acid.9
Our initial attempts to study the reactivity of tert-butyl-
magnesium chloride towards the bis-imine showed that
the first tert-butyl group added easily at low temper-
ature (Ð70¡C) whereas a temperature above +45¡C was
necessary for the second addition. This high temperature
seemed at first sight to be incompatible with obtaining the
expected pure diamine 6. Indeed, dropwise addition at
Ð70¡C of tert-butylmagnesium chloride to the bis-imine,
followed by heating to +45¡C, generally led to a mixture
of compounds in which the major isomer 6 could be de-
tected by 1H NMR. Unfortunately, isolation of this latter
compound by chromatography on silica gel was not effi-
cient. Our best result was obtained by carrying out the ad-
dition in hexane where about 50% of this major
diastereomer was formed.10 We stipulated then that re-
verse addition of the bisimine to the Grignard reagent at a
temperature above 45¡C could increase the rate of the sec-
ond addition and avoid the formation of by-products
which probably arose from the degradation of the interme-
Syntheses of enantiomerically pure diamines have been
mainly developed through resolution of racemic di-
amines. Nevertheless, asymmetric synthesis of vicinal di-
amines is an attractive alternative solution, and few
methods have been thus far reported. The addition of
Grignard or zinc reagents to the carbonÐnitrogen double
bonds of the chiral 1,2-bis-imine precursor 44,6 derived
from glyoxal and (S)- [or (R)]-methylbenzylamine, fol-
lowed by removal of the phenylethyl group, has been
shown to be an attractive method for the preparation of
these compounds (Scheme 1). The first example was de-
scribed by Neumann et al.5 who showed that addition of
allylmagnesium chloride gave a 6:1 mixture of two dias-
tereomers. A few years later, Savoia and co-workers6 im-
proved this reaction by using allylzinc bromide and
Synthesis 1999, No. 2, 228Ð230 ISSN 0039-7881 © Thieme Stuttgart á New York

SHORT PAPER
Efficient Synthesis of Enantiomerically Pure Di-tert-butylethanediamine
229
diate imine. Indeed, addition of the chiral (S,S)-1,2-bis-
imine 4 to a suspension of tert-butylmagnesium chloride
in hexane at 50¡C, led very cleanly in 0.5 h to a single
diastereomer 6 with no other compound detectable by 1H
and 13C NMR.
Scheme 2
In these additions, it was postulated that the metal gener-
ates a rigid five-membered chelate with the two nitrogen
atoms, providing an activation of the imine double bonds
as well as an improvement of the stereocontrol. According
to the results obtained in additions of the allyl or phenyl
group,6,7 the more stable conformation of this complex
was stipulated to be the one where the CÐH benzylic bond
and the C=N bond are coplanar (Scheme 3). Following
this model, the stereoisomer 6 was effectively the expect-
ed one. Indeed, due to steric bulk, the first addition is di-
rected on the least hindered methyl side. The second
addition is then controlled both by the first t-butyl group
added and the chiral auxiliary.
Figure ORTEP diagram of compound 8
Deprotection of the chiral auxilary groups was performed
from the crude diamine 6, using ammonium formate11 and
palladium hydroxide in refluxing ethanol. This reaction
gave the free diamine 3 in 2 h (Scheme 5).
Scheme 5
Scheme 3
In conclusion, we developed an efficient synthesis of 1,2-
diamino-1,2-di-tert-butylethane from glyoxal and (S)-me-
thylbenzylamine. This diamine was obtained optically
pure in three steps with short reaction times, no interme-
diate purification and 75% overall yield. We believe that
the easy availability of diamine 3 will render it as attrac-
tive as 1 and 2 for testing a new carbon framework. Syn-
thetic applications using this diamine are under
investigation.
The diamine 6 was obtained as an oil and all our attempts
to get crystals for X-ray spectroscopy failed. Neverthe-
less, with the intention of carrying out the methylation of
6 by the Eischweiler-Clarke procedure, we obtained the
aminal 7 as a white solid in 77% isolated yield. The mono-
hydrochloride 8, obtained by treatment of 7 with hydro-
chloric acid, was then recrystallized twice in ethyl acetate
to give colourless crystals suitable for the analysis
(Scheme 4). The ORTEP diagram of 8 is shown in the Fig-
ure, and clearly confirms the expected (R,R) absolute con-
figuration of the carbons bearing the t-butyl groups.
Experiments involving organometallics were carried out under a
positive pressure of dry N2 or under Ar. (S)-(Ð)-a-methylbenzyl-
amine (98%) was purchased from Aldrich. tert-Butylmagnesium
chloride 2M in Et2O was purchased from Aldrich. Palladium hy-
droxide 20% on activated charcoal was purchased from Fluka. Hex-
ane was of analytical grade type and used without special drying or
distillation. NMR spectra were recorded on a Brucker ARX 400 or
AC 200 Q instrument, in CDCl3 as the solvent. Optical rotations
were measured on a Perkin-Elmer 141.
N,N'-Bis[(S)-1-phenylethyl]ethanediimine (4)
The previously reported procedures starting from 40% aq glyoxal4
or glyoxal trimer6 can be followed; However, we wish to present a
faster modified procedure:
Scheme 4
Synthesis 1999, No. 2, 228Ð230 ISSN 0039-7881 © Thieme Stuttgart á New York

230
S. Roland et al.
SHORT PAPER
A mixture of 40% aq glyoxal (1.13 mL, 7.8 mmol), (a)-(S)-methyl-
benzylamine (2.06 mL, 16 mmol), formic acid (50 mL) and anhyd
MgSO4 (4 g) in CH2Cl2 (15 mL), was stirred for 15 min. at 25¡C.
The suspension was filtered over Celite. The filtrate was concentrat-
ed and the residue dissolved in cyclohexane (15 mL), dried
(Na2SO4), filtered and concentrated to give 1.97 g (96 %) of 4 as an
orange oil. (store at Ð20¡C).
13C NMR: d = 18.0, 23.5, 28.7, 29.9, 35.5, 35.6, 59.0, 63.7, 69.0,
69.9, 72.2, 127.6, 128.6, 129.1, 130.8, 132.7, 146.4.
C27H41N2Cl (429.087): calcd. C 75.58, H 9.63, N 6.53; found C
75.52, H 9.58, N 6.65.
(R,R)-1,2-Diamino-1,2-di-tert-butylethane (3)
To a solution of diamine 6 (2 g, 5.3 mmol) in EtOH (50 mL) were
added (0.5 g,10% mol) of Pd(OH)2ÐC 20% wt. and 2 g (31.6 mmol,
6 eq.) of anhyd ammonium formate. The mixture was refluxed 2 h
with vigourous stirring under inert atmosphere, filtered and concen-
trated. The residue was dissolved in Et2O (25 mL), stirred 15 min
on K2CO3, filtered and concentrated to give 943 mg of crude 3 as an
orange liquid. Distillation under reduced pressure led to 760 mg
(83%) of pure 3 as a colourless liquid.
C18H20N2 (M = 264.37).
1H NMR: d = 1.6 (d, 6H, J = 6.7 Hz), 4.53 (q, 2H, J = 6.7 Hz), 7.24Ð
7.38 (m, 10H), 8.07 (s, 2H).
N,N'-Bis[(S)-1-phenylethyl]-(R,R)-1,2-diamino-1,2-di-tert-bu-
tylethane (6)
A solution of 100 mL (0.2 mol) of t-butylmagnesium chloride (2M
in Et2O) in hexane (1 L) was heated to 50¡C and stirred for 15 min.
At this temperature was added dropwise a solution of diimine 4
(18.5 g, 0.07 mol) in hexane (300 mL). The mixture was stirred
0.5 h at 50¡C then cooled to 20¡C, quenched with sat. NH4Cl (300
mL), diluted with Et2O (300 mL) and stirred for 30 min. The aque-
ous layer was extracted with Et2O (2 × 200 mL). The combined or-
ganic layers were dried (K2CO3), filtered over silica gel (Et2O) and
concentrated to give 25.2 g (95%) of crude 6 as an orange oil.
bp = 240¡C. [a]D25 = Ð15 (c = 0.145, CH2Cl2). [a]lit = +12 for the
(S,S) diamine.9
1H NMR: d = 0.88 (s, 18H), 2.69 (s, 2H), 3.8 (s, 4H).
13C NMR: d = 27.0, 35.8, 57.6.
References
[a]D25 = +31.6 (c = 0.364, CHCl3).
(1) Tomioka, K. Synthesis 1990, 541.
Corey, E. J.; Sarshar, S.; Bordner, J. J. Am. Chem. Soc. 1992,
114, 7938.
Corey, E. J.; Kim, S. S. J. Am. Chem. Soc. 1990, 112, 4976.
Bennani, Y. L.; Hanessian, S. Chem. Rev. 1997, 97, 3161.
Jacobsen, E. J. Catalytic Asymmetric Synthesis, Ed.: Ojima,
VCH, 1993, p 159/179.
1H NMR: d = 0.83 (s, 18H), 1.3 (d, 6H, J = 6.4 Hz), 2.4 (s, 2H), 3.75
(q, 2H, J = 6.4 Hz), 7.15Ð7.45 (m, 10H).
13C NMR: d = 23.6, 35.9, 57.1, 62.2, 126.5, 126.9, 128.2, 147.9.
C26H40N2 (380.615): calcd. C 82.05, H 10.59, N 7.36; found C
80.99, H 10.01, N 7.19.
Katsuki, T. J. Mol. Cat. 1996, 113, 87.
(1R,2R)-N,N'-Bis[(S)-1-phenylethyl]-1,2-di-tert-butylimidazoli-
dine (aminal) (7)
Mukaiyama, T. Aldrichimica Acta 1996, 29, 59.
Mukaiyama, T. and Yamada, T. Bull. Chem. Soc. Jpn. 1995,
68, 17 and 1455.
The crude diamine 6 (380 mg, 1 mmol) was dissolved in formic acid
(380 µL, 10 mmol) and 400 µL (5 mmol) of 37% aq formaldehyde.
The solution was stirred under reflux for 5 h. After cooling, H2O
(1 mL) and Et2O (2 mL) were added and the mixture was neutra-
lised under stirring by slow addition of solid K2CO3 until the gas
evolution stopped. The aqueous layer was separated and extracted
with Et2O (3 × 3 mL). The combined organic layers were dried on
K2CO3, filtered and concentrated. Purification by chromatography
on silica gel (c-Hex : EtOAc, 98:2) led to 300 mg (77%) of the ex-
pected aminal as a colourless oil that crystallizes after a few hours.
Trost, B. M.; Van Vrancken, D. L. Chem. Rev. 1996, 96, 395.
(2) Alexakis, A.; Mangeney, P. Advanced Asymmetric Synthesis;
Stephenson, G. R., Ed.; Chapman & Hall, 1996; p 93.
Barett, A. G. M.; Doubleday, W. W.; Tustin, G. J.; White, A.
J. P.; Williams, D. J. J. Chem. Soc. Chem. Comm. 1994, 2739.
(3) For the diamine 1, see:
Pini, D.; Iuliano, A.; Rosini, C.; Salvadori, P. Synthesis 1990,
1023.
Lohray, B. B.; Ahuja, J. R., J. Chem. Soc. Chem. Comm. 1991, 95.
Oi, R.; Sharpless, K. B.; Tetrahedron Lett. 1991, 32, 999.
Corey, E. J.; Imwinkelried, R., Pikul, S., Xiang, Y. B. J. Am.
Chem. Soc. 1989, 111, 4486.
Shimizu, M.; Kamei, M.; Fujisawa, T. Tetrahedron Lett.
1995, 36, 8607.
For the diamine 2, see:
Wieland, A.; Schlichtung, O.; Langsdorf, W. V. Z. Phys.
[a]D25 = + 68.6 (c = 0.37, CHCl3). mp = 78¡C.
1H NMR: d = 0.92 (s, 18H), 1.35 (d, 6H, J = 7 Hz), 3.08 (s, 2H),
3.59 (s, 2H), 4.12 (q, 2H, J = 7 Hz), 7.15Ð7.45 (m, 10H).
13C NMR: d = 23.3, 29.1, 36.1, 61.5, 68.2, 75.3, 126.6, 127.9, 128.3,
146.3.
C27H40N2 (392.626): calcd. C 82.60, H 10.27, N 7.13; found C
82.47, H 10.17, N 7.03.
Chem. 1926, 161, 74.
Swift, G.; Swern, D. J. Org. Chem. 1967, 32, 511.
Whitney, T. A. J. Org. Chem. 1980, 45, 4214.
(4) Tom Dieck, H.; Dietrich, J. Chem. Ber. 1984, 117, 694.
(5) Neumann, W. L.; Rogic, M. M.; Dunn, T. J. Tetrahedron Lett.
1991, 32, 5865.
(6) Alvaro, J.; Grepioni, F.; Savoia, D. J. Org. Chem. 1997, 62,
4180.
(7) Bambridge, K.; Begley, M. J.; Simpkins, N. S. Tetrahedron
Lett. 1994, 35, 3391.
Aminal Monohydrochloride 8
To a solution of 281 mg (0.72 mmol) of aminal 7 in Et2O (2 mL)
were added 6 M HCl (3 mL). A white precipitate appeared and the
mixture was stirred for 15 min at 20¡C. The aqueous layer was ex-
tracted with CH2Cl2 (3 × 3 mL). The combined organic layers were
dried (MgSO4) filtered and concentrated to give 301 mg (97%) of
the expected chlorohydride as a white crystalline solid. The crude
was recrystallized twice in EtOAc to give colourless crystals suit-
able for X-ray crystallography.
(8) Roskamp, E. J.; Pedersen, S. F. J. Am. Chem. Soc. 1987, 109,
3152.
[a]D25 = +103.2 (c = 0.155, CH2Cl2), mp = 226¡C.
(9) Rasmussen, K. G.; Thomsen, D. S.; J¿rgensen, K. A. J. Chem.
Soc. Perkin Trans 1 1995, 2009.
(10) The yield was determined by formation and isolation of the
aminal 7 from the crude mixture.
(11) Anwer, M. K.; Spatda, A. F. Synthesis 1980, 929.
1H NMR: d = 0.46 (s, 9H). 1.26 (d, 3H, J = 6.5 Hz), 1.45 (s, 9H),
1.65 (s, 1H), 2.0 (d, 3H, J = 7.2 Hz), 3.24 (d, 1H, J = 3.5 Hz), 3.5
(dd, 1H, J = 6.9, 3.5 Hz), 3.85 (dd, 1H, J = 9.5, 8.3 Hz), 4.53 (dd,
1H, J = 9.5, 5 Hz), 4.75 (q, 1H, J = 6.5 Hz), 4.86 (m, 1H), 7.18Ð7.64
(m, 10H).
Synthesis 1999, No. 2, 228–230 ISSN 0039-7881 © Thieme Stuttgart · New York