Synthesis and application of a new pseudo C2-symmetric tertiary diamine for the enantioselective addition of MeLi to aromatic imines

Article abstract of DOI:10.1016/j.tetasy.2007.10.018

A new tertiary pseudo C₂-symmetric diamine derived from (1S,2S)-(+)-pseudoephedrine was synthesized and tested in the enantioselective addition of methyllithium on different aromatic imines. A comparative study with a similar C₂-symm

Full text of DOI:10.1016/j.tetasy.2007.10.018

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Available online at www.sciencedirect.com
Tetrahedron: Asymmetry 18 (2007) 2503–2506
Synthesis and application of a new pseudo C2-symmetric tertiary
diamine for the enantioselective addition of MeLi to aromatic imines
Quentin Perron and Alexandre Alexakis*
University of Geneva, Department of Organic Chemistry, 30 Quai Ernest Ansermet, CH-1211 Geneva, Switzerland
Received 11 September 2007; accepted 15 October 2007
Available online 5 November 2007
Abstract—A new tertiary pseudo C2-symmetric diamine derived from (1S,2S)-(+)-pseudoephedrine was synthesized and tested in the
enantioselective addition of methyllithium on different aromatic imines. A comparative study with a similar C2-symmetric ligand derived
from the cyclohexane diamine showed better reactivity and enantioselectivity up to 91%.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
were tested and two of them 3 and 4 appeared to be the
best (Scheme 1).
Many ligands in asymmetric catalysis are C2-symmetric to
reduce the number of possible transition states and conse-
quently the number of possible diastereoisomers.
After such promising results, it was reasonable to believe
that the cyclohexane core could be replaced by another chi-
ral backbone (more flexible), to optimize this enantioselec-
tive addition of methyllithium. Ligands 5 and 6 (Scheme 2)
were tested in the same reaction described in Scheme 1.2
In our previous article, we developed a new type of C2-sym-
metric tertiary diamine derived from a cyclohexane
diamine backbone for the addition of alkyllithium or aryl-
lithium to aromatic imines.1 We showed that the nitrogen
atom, complexed with a metal, such as lithium, could be-
come a stereogenic center in the reactive reagent due to a
chirality transfer from the chiral backbone. A large number
of ligands based on the cyclohexane diamine backbone
N
N
N
N
N
N
Ph
Ph
6
7
5
ee=27%
ee=60%
ee=29%
Scheme 2.
O
O
MeLi (3 eq)
3 or 4 (0.2 eq)
Their selectivity was almost the same3 but 5 is much more
difficult to synthesize compared to 6 prepared from the
commercial enantiomerically pure (1S,2S)-(+)-pseudo-
ephedrine. This represents an interesting feature given that
pseudoephedrine derivatives could be easily prepared enan-
tiomerically pure and with the same selectivity than the
analogous (1S-2S)-diphenyldiaminoethane. By analogy
with 3 and 4, ligand 7 was the first to be synthesized and
tested. It showed better reactivity but lower selectivity than
its analogue 4.4 In continuity of this work, we decided to
synthesize the analogue of 3 starting from (1S,2S)-(+)-
pseudoephedrine, which is commercially available. In addi-
tion, it was shown that a strong base, such as n-BuLi, could
N
H
N
toluene, -78°C, 15h
1a
(R)-2a
ee=67% ee with 3
ee=68% ee with 4
Ph
N
N
N
N
Ph
3
4
Scheme 1.
*
Corresponding author. Tel.: +41 22 379 65 22; fax: +41 22 379 32 15;
e-mail: alexandre.alexakis@chiorg.unige.ch
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.10.018

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2504
Q. Perron, A. Alexakis / Tetrahedron: Asymmetry 18 (2007) 2503–2506
deprotonate 4 and 7 and release styrene, even at ꢀ78 °C.5
This would not be the case with 3 and its analogues.
to 3. Nevertheless, it is interesting to note that for 1b the
reaction was performed at ꢀ65 °C due to low reactivity
at ꢀ78 °C with 3, but carrying out the reaction with 12
allowed us to stay at ꢀ78 °C and still obtain complete con-
version (entry 3 vs entry 4). Systematic inversion of config-
uration was obtained for all products, as expected with the
inversion of configuration at the chiral backbone of the
diamines. In terms of selectivity, in all cases, diamine 12
induced better enantioselectivity than 3. The bigger differ-
ence appeared for imine 1a (entry 1 vs entry 2). For imines
1d–f we observed an increase of reactivity but a decrease in
term of selectivity (entries 8, 10, and 12). In fact, in these
cases, the substrate itself behaves like a ligand in chelating
the MeLi. This is why the stronger the chelation, the better
the conversion is, but the less the selectivity is induced, due
to the achirality of the substrate. To avoid such behavior,
we changed the substrate 1d–f by their position isomers
1k–m where the chelation is not possible.
2. Results and discussion
Diamine 12 was synthesized in a four step sequence starting
from (1S,2S)-(+)-pseudoephedrine in 77% overall yield
(Scheme 3).
H
Pd/C, H2, EtOH,
N
O
N
N
rt, 72h, 92%
Ph
OH
O
K2CO3, Ether,
Reflux, 2h, 99%
OH
Ph
Ph
8
9
Et3N, MeNH2,
rt, 16h, 90%
Et3N, MsCl, Et2O
0°C, 30 min
N
N+
10
Ph
NH
Ph
11
Table 1. Enantioselective addition of MeLi to imines 1a–n promoted by
diamines 3 or 12
N
N
O
Entry
Product
Lig (equiv)
Conva (%)
eeb,c (%)
NaBH3CN, AcOH,
MeOH, rt, 24h, 94%
Ph
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
2a
2a
2bd
2b
2c
2c
2d
2d
2e
2e
2f
3g (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
3 (0.2)
12 (0.2)
12 (1)
— (98)
100 (72)
— (70)
100 (75)
— (70)
90 (81)
— (93)
80 (76)
— (94)
96 (83)
— (87)
100 (77)
— (78)
100 (97)
100 (90)
100 (74)
99 (88)
88 (85)
76 (53)
95 (90)
90 (93)
93 (92)
67 (R)
88 (S)
57 (R)
69 (S)
38 (R)
57 (S)
58 (R)
66 (S)
20 (R)
19 (S)
4 (R)
12
Scheme 3.
Oxazolidine 8 was the result of the condensation of 3,3-
dimethylbutyraldehyde with (1S,2S)-(+)-pseudoephed-
rine.6 Amino-alcohol 9 was then obtained after reduction
of oxazolidine 8 with palladium on charcoal (0.01 equiv)
under 1 atm of H2.7 Other reducing agents, such as NaBH4,
NaBH3CN, or LiAlH4 were tried but without any success.
Reacting 9 under basic conditions with methanesulfonyl
chloride gave an aziridinium intermediate salt 10.8 The lat-
ter was opened regiospecifically by a commercial methyl-
amine solution in ethanol, to afford 11 in quantitative
yield with an overall retention of the configuration.9 Final-
ly, ligand 12 was obtained in quantitative yield after reduc-
tive amination of 11 with sodium cyanoborohydride
activated by acetic acid in methanol.10 Diamine 12 was
then tested in the enantioselective addition of MeLi to var-
ious N-p-methoxyphenyl imines (Scheme 4). The same
reaction conditions as previously reported1 were applied
for the purpose of comparison with diamines 3.
2f
3 (S)
2ge
2ge
2g
2h
2i
42 (R)
45 (S)
79 (S)
76 (S)
82 (S)
91 (S)
81 (S)
75 (S)
62 (S)
88 (S)
12 (1)
12 (0.2)
12 (1)
12 (1)
12 (1)
2jf
2k
2l
2mf
2n
12 (1)
12 (1)
a Determined by 1H NMR spectroscopy of the crude. Yield of isolated
product in parenthesis.
b Determined by supercritical fluid chromatography.
c The absolute configuration of 2a was determined on the basis of previous
work. For other products, the configurations were assigned by consid-
eration of the stereochemical pathway and of the systematic inversion of
configuration occurring using 3 or 12.
O
MeLi (3 eq)
3 or 12 (0.2 or 1 eq)
O
Ar
N
Ar
N
toluene, -78°C,
H
15h or more
d Reaction carried out at ꢀ65 °C.
2a-n
1a-n
e Reaction carried out at ꢀ30 °C.
a, Ar = C6H5
h, Ar = 1-naphthyl
f Reaction time: 38 h.
b, Ar = p-Cl-C6H4
c, Ar = p-CF3-C6H4
d, Ar = 2-thienyl
e, Ar = 2-furyl
f, Ar = 2-pyridyl
g, Ar = 2-naphthyl
i, Ar = p-Me-C6H4
j, Ar = p-OMe-C6H4
k, Ar = 3-thienyl
l, Ar = 3-furyl
g Reactions with 3 see Ref. 1.
m, Ar = 4-pyridyl
n, Ar =
O
To further optimize the reaction conditions, we first
decided to increase the temperature to obtain complete
conversion, but a dramatic decrease of selectivity was
observed. As a result, we increased the catalyst amount
to 1 equiv in case of incomplete reaction after 15 h. For
1a–c and 1i, the conditions were already optimized and
O
Scheme 4.
As far as 12 is concerned, in almost all cases, very good
conversion occurred and isolated yields were very similar

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Q. Perron, A. Alexakis / Tetrahedron: Asymmetry 18 (2007) 2503–2506
Table 2. Tests of catalyst loading on imine 1i after 15 h
2505
Acknowledgments
Entry
Diamine 12 (mol %)
Conv (%)
ee (%)
We thank the Swiss National Science Foundation (No.
200020-113332) for financial support. Special thanks, also,
1
2
3
4
5
10
15
20
47
78
88
74
82
82
82
´
to Stephane Rosset (University of Geneva) for his help in
the separation of enantiomers on chiral SFC.
100
References
we obtain complete conversion with 20 mol % of 12, which
proved that this reaction, in term of conversion, is not very
sensitive to electronic effects of the substituent on the
aromatic part. However in terms of selectivity, substrates
bearing electronic donor substituents induced more
enantioselectivity contrary to those bearing an electron
acceptor (entry 17 vs entry 6). For the other substrates,
1 equiv of catalyst was used and the time was extended to
38 h in case of partial conversion. Under such conditions,
we managed to obtain almost complete conversion
with naphthyl or heteroaromatic derivatives, and good
selectivity up to 91% was observed (entries 15, 16, and
19–21). Catalyst loading (5, 10, and 15 mol %) has been
also tested on the substrate 1i under the same conditions
as previously described (Scheme 4).
1. (a) Cabello, N.; Kizirian, J. C.; Gille, S.; Alexakis, A.;
Bernardinelli, G.; Pinchard, L.; Caille, J. C. Eur. J. Org.
Chem. 2005, 4835; (b) Kizirian, J. C.; Cabello, N.; Pinchard,
L.; Caille, J. C.; Alexakis, A. Tetrahedron 2005, 61, 8939; (c)
Cabello, N.; Kizirian, J. C.; Alexakis, A. Tetrahedron Lett.
2004, 45, 4639; (d) Kizirian, J. C.; Caille, J. C.; Alexakis, A.
Tetrahedron Lett. 2003, 44, 8893.
2. Kizirian, J. C. Ph.D. Dissertation 3467, University of
Geneva, Geneva, 2003.
3. These diamines gave excellent results when associated with
PhLi instead of MeLi. See: Refs. 1a and c.
4. Gille, S.; Cabello, N.; Kizirian, J. C.; Alexakis, A. Tetra-
hedron: Asymmetry 2006, 17, 1045.
5. Gille, S. Ph.D. Dissertation 3765, University of Geneva,
Geneva, 2006.
6. 2-(2,2-Dimethyl-propyl)-3,4-dimethyl-5-phenyl-oxazolidine 8
was prepared according to the procedure described in:
Bergmann, E. D.; Zimkin, E.; Pinchas, S. Rec. Trav. Chim.
Pays-Bas. 1952, 71, 237; 3-3-dimethylbutyraldehyde (80 ll,
0.605 mmol) and (1S,2S)-(+)-pseudoephedrine (100 mg,
0.605 mmol) were mixed in ether. Heat was evolved and an
insoluble layer (mostly water) was formed. Potassium
carbonate (85 mg, 0.605 mmol) was added and the whole
mixture was refluxed for 2 h. After 2 h, TLC showed the
reaction was over, the mixture was then filtered and the
solvent evaporated under rotatory evaporator to yield 149 mg
(100%) of 8 as a colorless oil. 1H NMR (CDCl3, d = ppm,
J = Hz): d = 1.05 (s, 9H), 1.2 (d, 3H, J = 6), 1.6 (dd, 1H,
J = 7.8 and 14.4), 1.7 (d, 1H, J = 14.9), 2.3 (s and m, 4H), 4.2
(d, 1H, J = 7.8), 4.6 (d, 1H, J = 8,6), 7.2–7.5 (m, 5H).
7. 2-[(3,3-Dimethyl-butyl)-methyl-amino]-1-phenyl-propan-1-ol
9 was prepared according to the procedure described in:
Gil-Av, E. J. Am. Chem. Soc. 1952, 74, 1346: A solution of
As expected, the conversion decreased with less catalyst,
but the enantioselectivity was unaffected until 10 mol %
of catalyst. We also wanted to compare this new catalyst
with our previous one such as 3 or (ꢀ)-sparteine in the
addition of n-BuLi on imine 1a, in toluene at ꢀ78 °C
during 2 h with 2 equiv of ligand.
With 3, only 61% conversion occurred, whereas (ꢀ)-spart-
eine afforded 100%, and in both cases around 25% enantio-
meric excess was measured. With 12, the reaction was
quantitative and we managed to obtain the (S)-enantiomer
of 1a in 50% enantiomeric excess (Tables 1–3).
Table 3. n-BuLi addition to imine 1a
(2,2-dimethyl-propyl)-3,4-dimethyl-5-phenyl-oxazolidine
8
Entry
Diamine
Conv (%)
ee (%)
(1.49 g, 6 mmol) in 13.5 ml of ethanol was hydrogenated
with palladium on charcoal (0.01 equiv) at 35 °C under 1 atm
of H2 during 36 h. The mixture was filtered on Celite, then the
solvents were removed to yield 1.37 g (91%) of 9 as a yellow
oil; 1H NMR (CDCl3, d = ppm, J = Hz): d = 0.8 (d, 3H,
J = 6.6), 1 (s, 9H), 1.4–1.6 (m, 2H), 2.3 (s, 3H), 2.4 (td, 1H,
J = 5.76 and 10.7), 2.6 (td, 1H, J = 5.9 and 10.9), 2.7 (m, 1H),
4.25 (d, 1H, J = 9.7), 7.2–7.5 (m, 5H).
1
2
3
3
61
100
100
25
25
50
(ꢀ)-Sparteine
12
3. Conclusion
8. (a) Dieter, R. K.; Deo, N.; Lagu, B.; Dieter, J. W. J. Org.
Chem. 1992, 57, 1663–1671; (b) O’Brien, P.; Towers, T. D. J.
Org. Chem. 2002, 67, 304–307.
In conclusion, this new tertiary pseudo C2-symmetric
diamines 12 was easily synthesized in a four step sequence
starting from the commercial enantiomerically pure
(1S,2S)-(+)-pseudoephedrine. It was tested in the addition
of MeLi on different aromatic imines 1a–n, and with
n-BuLi on 1a. By comparison with the analogous
C2-symmetric cyclohexane diamines based 3, and even with
all the ligands previously described in our laboratory,
better conversions and enantioselectivities were obtained
in all cases with 12. This study confirmed the potential of
pseudo C2-symmetric diamines based on the pseudoephed-
rine core and describes a general procedure for the synthe-
sis of various other diamines.
9. N2-(3,3-Dimethyl-butyl)-N1,N2-dimethyl-1-phenyl-propane-
1,2-diamine 11 was prepared according to the procedure
described in Ref. 8b: At 0 °C, MsCl (511 ll, 6.6 mmol) was
added dropwise to a stirred solution of 2-[(3,3-dimethyl-
butyl)-methyl-amino]-1-phenyl-propan-1-ol 9, (1.37 g, 5.49
mmol) and dry Et3N (1.3 ml, 9.3 mmol) in 27 ml of dry Et2O.
A white solid in suspension appeared. After 30 min, dry Et3N
(1.5 ml, 11 mmol) was added and the mixture was allowed to
warm to room temperature. An 8 M solution of methylamine
in ethanol (11.6 ml, 93.3 mmol) was then added and the
mixture was stirred vigorously for 48 h at room temperature.
Ethanol was evaporated, water and ether were added to the

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2506
Q. Perron, A. Alexakis / Tetrahedron: Asymmetry 18 (2007) 2503–2506
mixture and the organic phase was separated. The aqueous
phase was extracted with Et2O three times and the combine
organic phases were washed with a aqueous solution of 5%
NaHCO3, water, and finally dried over sodium sulfate and
concentrated under vacuo to give 1.3 g of crude 11 (80%) as a
for 24 h, methanol was then evaporated and the residue
diluted in ether. The organic layer was washed with a 10%
NaOH aqueous solution. The product was purified by
bulb-to-bulb distillation (155 °C, 0.4 mmHg), to yield
860 mg (94%) of (1S,2S)-N1,N2-bis(3,3-dimethylbuthyl)-
N1,N2-dimethyl-1-phenylpropane-1,2-diamine,12, as a white
1
yellow oil; H NMR (CDCl3, d = ppm, J = Hz): d = 0.6 (d,
20
3H, J = 6.6), 0.9 (s, 9H), 1.35 (td, 1H, J = 4.8 and 12.6), 1.49
(td, 1H, J = 5 and 11.6), 2.25 (s, 3H), 2.3 (td, 1H, J = 5.1 and
11.6), 2.5 (td, 1H, J = 5.3 and 12.9), 2.75 (m, 1H), 3.2 (d, 1H,
J = 9.8), 7.2–7.5 (m, 5H).
solid. mp = 35 °C, ½aꢁD ¼ þ36:2 (c 0.97, CHCl3); 1H NMR
(CDCl3, d = ppm, J = Hz): d = 0.6 (d, 3H, J = 6.32), 0.8 (s,
9H), 0.9 (s, 9H), 1.3–1.5 (m, 4H), 2.05 (m, 1H), 2.1 (s, 3H),
2.25 (s, 3H), 2.3 (m, 1H), 2.4–2.5 (m, 2H), 3.3 (m, 1H), 3.6 (d,
1H, J = 10.4), 7.1–7.3 (m, 5H); 13C NMR (CDCl3, d = ppm,
J = Hz): d = 10.4, 29.9, 36.8, 38.2, 41.4, 42.1, 50.4, 56.1, 69.7,
127, 127.9, 129.7; IR (neat): 2953, 1465, 1362, 703; MS (m/z)
347 (M+1), 331, 277, 275, 232, 204, 156, 142, 118, 81, 43, 29;
HRMS calcd for C23H43N2 347.3420, found 347.3404.
10. To a solution of N2-(3,3-dimethyl-butyl)-N1,N2-dimethyl-1-
phenyl-propane-1,2-diamine 11 (800 mg, 2.64 mmol) in 375 ll
of methanol were added 3-3-dimethylbutyraldehyde (524 ll,
3.97 mmol), sodium cyanoborohydride (330 mg, 5.29 mmol)
and acetic acid (76 ll, 1.32 mmol). The mixture was stirred