Chiral bromine-lithium exchange catalyzed by diamines

Article abstract of DOI:10.1016/j.tetasy.2009.02.027

Different classes of prochiral polyhalide compounds have been tested in a chiral bromine-lithium exchange in the presence of different diamines with enantiomeric excesses of up to 63%.

Full text of DOI:10.1016/j.tetasy.2009.02.027

Tetrahedron: Asymmetry 20 (2009) 1004–1007
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Chiral bromine–lithium exchange catalyzed by diamines
*
Quentin Perron, Jezabel Praz, Alexandre Alexakis
University of Geneva, Department of Organic Chemistry, 30 Quai Ernest Ansermet, CH-1211 Geneva, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Different classes of prochiral polyhalide compounds have been tested in a chiral bromine–lithium
exchange in the presence of different diamines with enantiomeric excesses of up to 63%.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 4 February 2009
Accepted 19 February 2009
Available online 28 March 2009
1. Introduction
PG
Br
X
Br
The first halogen–lithium exchange was published by Marvel in
1927 where he noted the formation of toluene when n-BuLi was
mixed with o- or m-bromotoluene.1 In 1938 Wittig and Gilman
started to independently study this reaction and obtained the same
conclusions concerning orthometallation or halogen–lithium ex-
change with p-bromoanisole or o-bromoanisole, respectively, in
the presence of lithium base in ether.2 Since then, only new appli-
cations of this reaction or mechanistic studies have been reported.3
In our laboratory we studied organolithium compounds in pres-
ence of chiral diamines,4 and in order to prepare aromatic lithium
reagents starting from the corresponding aromatic halide we noted
that the exchange between an aromatic bromide and n-BuLi did
not occur at low temperature without the presence of a diamine
in toluene,4 in contrast to aromatic iodides, where a diamine is
not required. To the best of our knowledge, since Wittig and Gil-
man, no chiral version of this reaction has been reported. Hence
we started a program in applying our diamines in a chiral version
of this bromine–lithium exchange. However, during our investiga-
tions, Kagan et al. published a new concept where they present the
desymmetrization of prochiral aromatic or vinylic dihalide sub-
strates in presence of diamines.5 Herein, we report our results in
the state they were.
Br
Br
R'
Br
1
4
R' = Br, Ph or TMS
X = O or N
1 eq RLi
R'= Ph or TMS
1 eq RLi
R'= Br
2 eq RLi
Enantioselective
step
1 eq Diamine
1 eq of E+
2 eq Diamine
Diamine
1 eq
2 eq of E+
1 eq of E+
PG
E
Br
X
Br
Br
E
E
Br
Br
R'
E
5
2
3
Scheme 1.
The starting material 2,20,6,60-tetrabromobiphenyl 8 was pre-
pared in two steps, according to the procedure described by Ler-
oux,6c and was obtained with an overall yield of 47% thanks to
an Ullman coupling starting from the 1,3-dibromobenzene 6
(Scheme 2). In order to optimize this reaction and to prevent the
use of nitrobenzene, which is very difficult to separate from the
product, we decided to perform a one-pot procedure using benzo-
quinone or chloranil as a coupling agent.7
Thus, after the aromatic deprotonation of 6, and transmetalla-
tion to the Cu reagent, we added either benzoquinone or chloranil
to afford, after recrystallization in acetonitrile, the expected com-
pound in 42% and 47% isolated yields, respectively (Scheme 2). This
method gave similar overall yields but was more convenient on a
large scale.
2. Results and discussion
Two classes of substrates were studied. The first allows for an
axial chirality by the desymmetrization of prochiral aromatic ha-
lide compounds such as 1, by selectively exchanging only one of
two enantiotopic bromines (Scheme 1). Of particular interest was
1 (R = Br), because its halogen–lithium exchange has already been
studied by Leroux et al., but not in a chiral version.6 The second
class of substrates concerns more the ‘classical’ central chirality.
This could be obtained in the case of the protected diaryl alcohol
or amine 4 (Scheme 1).
For alcohol 9 we performed a one-pot procedure using an
a-
bromo phenyl Grignard, prepared according to Knochel,8 yielding
67% of the expected compound. The alcohol was then protected
with different groups in order to examine the influence of the ste-
ric and coordination effects on the Br–Li exchange step (Scheme 3).
For the synthesis of the Boc-protected amine, the a-bromo phe-
nyl Grignard was reacted with imine 17 as the electrophile, to
provide the expected compound 18 in 60% overall yield.
* Corresponding author. Tel.: +41 22 379 65 22; fax: +41 22 379 32 15.
E-mail address: alexandre.alexakis@chiorg.unige.ch (A. Alexakis).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.02.027

Q. Perron et al. / Tetrahedron: Asymmetry 20 (2009) 1004–1007
1005
Table 1
I
Screening of organolithium reagents for the Br–Li exchange
1) nBuLi, Et2O, -78°C
Br 2) CuBr
1) LDA, THF/Hexane
-78°C, 2h
Br
Br
Br
Br
Br
Br
Br
Entry
Ligand
RLi
Time
Conva (%)
ee (%)
3) Nitrobenzene
56%
2) I2
84%
6
1
2
3
4
5
6
7
8
19
19
19
19
19
20
20
20
n-BuLi
MeLi
PhLi
s-BuLi
t-BuLi
n-BuLi
s-BuLi
t-BuLi
20 min
20 min
20 min
2 h
2 h
20 min
2 h
100
0
24
—
7
8
0
—
1) LDA, THF/Hexane
-78°C, 2h
32
53
100 (90)
32
100
63
50
50
44
55
2) CuCN.LiCl (THF), Benzoquinone
or Chloranil
42-47%
2 h
Scheme 2.
Conditions: see general procedure A (Ref. 10).
a
In parentheses, yield of the isolated product.
PG
Br
OH Br
proceed further, probably due to the insolubility of 8 at low
temperature.
Br
O
Br
Alcohol
protection
In addition to (ꢀ)-sparteine 19, we also tested our recently de-
scribed diamine 20.9 Under the same conditions as aforemen-
tioned, full conversion was observed with n-BuLi, and, after
quenching with DMF, a 90% isolated yield of adduct 8a was ob-
tained with 50% ee (Table 1, entry 6). No improvement was ob-
served with s-BuLi or t-BuLi (entries 7 and 8). An in situ quench
with DMF was also attempted, but the reaction of n-BuLi on DMF
was faster than the Br–Li exchange.
9
67%
10
PG= Me
84%
90%
11
12
13
14
TMS
Methylformate,
30min, -15°C
TBDPS 9%
MOM 60%
MEM 41%
iPrMgCl.LiCl, THF
-15°C, 2h
Br
Br
MgCl.LiCl
Catalytic versions of this reaction were tested and the results
are summarized in Table 2.
Br
6
Boc
Boc
Br HN
Br
N
Na2SO4,
SO2Ph
K2CO3, THF
reflux,16h
Table 2
30min, -15°C
Screening of catalyst loading
Amine Boc
NaSO2Ph
Formic acid
MeOH/H2O
RT, 48h
17
16
Entry
Ligand (equiv)
RLi (2 equiv)
Time
Conv (%)
ee (%)
Boc
Br HN
Br
1
2
3
4
5
6
7
8
20 (0.05)
20 (0.1)
20 (0.2)
20 (2)
19 (0.2)
19 (1)
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
2 h
2 h
1 h
20 min
30 min
30 min
20 min
20 min
27
20
28
100
10
80
16
21
22
50
8
18
24
26
Br
O
18 60%
15
Scheme 3.
19 (2)
19 (4)
100
100
Conditions: see general procedure A (Ref. 10).
The chiral halogen–lithium exchange was first tested in the
presence of 2 equiv of n-BuLi and 2 equiv of (ꢀ)-sparteine 19 in tol-
uene at ꢀ80 °C, on substrate 8 (Scheme 4). We were surprised to
find that even with more than 2 equiv of n-BuLi, only two bromines
were substituted and each one on a different aryl portion. The last
observation was already noted by Leroux, but in THF.6 After
20 min, the reaction was over and DMF was added to trap the dior-
ganolithium species. An enantiomeric excess of 24% was deter-
mined by SFC analysis (Table 1, entry 1). In addition to toluene,
several other solvents were tested, but either the enantioselectivity
was lower (THF), or compound 8 was insoluble (Et2O). The Br–Li
exchange was also tested with other organolithium reagents. MeLi
and PhLi (entries 2 and 3) did not show any trace of the desired
lithium reagent, whereas s-BuLi and t-BuLi (entries 4 and 5) did
promote the exchange, albeit at a much slower rate. Interestingly,
the observed enantioselectivity was higher (63% and 50%, respec-
tively). However, even after prolonged times, the reaction did not
These results show a partial conversion when the reaction is
performed under catalytic conditions even with an extension of
the reaction time. This is due to the low solubility of the substrate
in toluene at ꢀ80 °C. It appears that the reaction proceeds at the
‘hot point’ during the addition of the substrate on the diamine
and n-BuLi. With 2 equiv of diamine the reaction is so fast that at
the end of the addition all starting material is consumed. In con-
trast, the rate of the Br–Li exchange is much slower with a catalytic
amount of ligand, and, at the end of the addition, the reaction is not
complete and the starting material freezes in the Schlenk, thus
stopping the reaction.
In addition to diamines 19 and 20, the screening of other
ligands has been performed and the results are summarized in
Table 3.
Due to the high cost of these ligands, only the catalytic version
was undertaken (except entries 1 and 8). Nevertheless, some inter-
esting trends could be observed. Inexpensive diamine 21, did not
afford any enantioselectivity, even in stoichiometric amount. In
all other cases the conversion and selectivity were low, certainly
due to a catalytic use of ligand. However, ligand 25 appeared to
be the most reactive one providing the expected compound in
85% conversion and 20% ee. Nevertheless, the only attempt with
stoichiometric amount of bis-oxazoline 27 did improve the conver-
sion, but the enantioselectivity remained poor.
1) (-)-sparteine 19 or
20 2 eq., RLi, toluene
O
N
N
-80°C
Br
Br
Br
Br
Br
Br
2) DMF
O
20
8
8a
Since many problems were related to the solubility of 2,20,6,60-
tetrabromobiphenyl 8, we decided to replace one bromine by a
Scheme 4.

1006
Q. Perron et al. / Tetrahedron: Asymmetry 20 (2009) 1004–1007
Table 3
Screening of other ligands
Br
Br
Br
Br
Entry
Ligand
Equiv
Conv (%)
ee (%)
PhLi, ZnCl2, Pd(PPh3)4
THF, -80°C-RT
1) nBuLi, THF, -80°C
2) TMSCl
8
1
2
89
0
21
N
N
Br
Br
TMS
Br
Br
Br
Ph
Br
29
28
42%
100%
2
3
0.2
0.2
29
39
2
O
N
N
O
Scheme 5.
22
Table 4
Enantioselective Br–Li exchange on substrates 28 and 29
N
15
23
N
Entry
Substrate
Ligand
Equiv
Time (min)
Conv (%)
ee (%)
1
2
3
4
28
28
28
29
20
20
19
20
0.2
2
2
20
20
20
5
33
56
97
48
22
28
3
0.2
18
Ph2
P
4
0.2
19
—
24
Conditions: see general procedure A (Ref. 10).
N
O
Table 5
O
O
Alcohol or ethers screening in chiral Br–Li exchange
5
6
7
8
0.2
0.2
0.2
2
85
20
4
N
N
Entry
Substrate
Ligand
RLi
Time
Conv (%)
ee (%)
25
1
2
3
4
5
6
7
8
9
10
11
12
13
9
TMEDA
19
19
19
19
19
20
21
19
20
19
20
19
n-BuLi
n-BuLi
s-BuLi
t-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
n-BuLi
1 h 30 min
45 min
3 h
3 h 30 min
2 h 30 min
2 h
2 h
2 h
1 h 30 min
2 h
2 h
4
73
53
16
15
nd
nd
nd
45
55
65
58
50
—
16
8
13
10
6
7
2
29
4
6
10
10
10
11
12
12
12
13
13
14
14
18
O
O
N
N
6
26
O
O
22
71
18
22
N
N
27
2 h
2 h
9
14
Conditions: see general procedure B (Ref. 10).
TMEDA = N,N,N0,N0-tetramethylethylenediamine.
Conditions: see general procedure A (Ref. 10).
organolithium could be introduced in the reaction. The chiral ex-
change has been tested in our classical conditions, and the lithiated
intermediate was quenched by methanol (Scheme 6).
more lipophilic group such as phenyl or trimethylsilyl. To prepare
such substrates we performed a monolithiation of 8 in presence of
1 equiv of n-BuLi in THF at ꢀ80 °C.6 Trapping with TMSCl allowed
us to obtain 28 in 100% yield. Alternatively, after transmetallation
to zinc, a Negishi coupling afforded 29 in 42% yield (Scheme 5).
These new substrates were tested in the chiral Br–Li exchange
under our standard conditions. The results are summarized in Ta-
ble 4.
PG
PG
1) (-)-sparteine 19 or
20 2eq, RLi, toluene
-80°C
Br
X
Br
X
Br
2) MeOH
10 Me
=
X= O, PG
X= N, PG
TMS
11
12
13
14
TBDPS
MOM
MEM
=18Boc
Unfortunately, at ꢀ80 °C, these new substrates did not appear
more soluble in toluene than 2,20,6,60-tetrabromobiphenyl 8, hence
both the conversion and the enantioselectivity were not better.
This means that during the enantiodetermining step the discrimi-
nation between a bromine atom and a phenyl or a TMS is not better
than that between a bromine atom and a lithium, which is the case
during the second bromine–lithium exchange with 8.
In addition to atropoisomeric substrates, we then turned our
attention to the protected alcohols and amine 4 (Table 5). In
contrast to 2,20,6,60-tetrabromobiphenyl 8, the use of an excess
of base led to a double Br–Li exchange, hence only 1 equiv of
Scheme 6.
For the free alcohol 9, only 4% conversion was observed with
TMEDA, hence the chiral version was not attempted. In the case
of silyl ethers 11 and 12, a Brook rearrangement was observed;
these reactions are not clean due to the presence of monolithiated,
bislithiated, Brook species and starting materials, even with a large
protecting group such as 12. In the other cases the conversions
were moderate, again due to the bislithiated compound or the
remaining starting material. It is interesting to note, again, that

Q. Perron et al. / Tetrahedron: Asymmetry 20 (2009) 1004–1007
1007
9303; (g) Schlosser, M. Angew. Chem., Int. Ed. 2005, 44, 376–393; (h) Clayden, J.
In Organolithium: Selectivity for Synthesis; Baldwin, J. E., Williams, R. M., Eds.;
Pergamon: Oxford, 2002.
the Br–Li exchange proceeds more slowly with s-BuLi and t-BuLi
(entries 3 and 4). The best enantiomeric excess was obtained with
the MOM protecting group in the presence of (ꢀ)-sparteine 19 (en-
try 7). Our ligand 20 appeared, in these cases, to be less efficient
than 19. The Boc-protected amine 18 was also desymmetrized
with 14% enantiomeric excess. Other types of protecting groups
and diamines need to be tested to improve this result.
4. (a) Alexakis, A.; Vrancken, E.; Mangeney, P. Synlett 1998, 1165; (b) Cabello, N.;
Kizirian, J.-C.; Alexakis, A. Tetrahedron Lett. 2004, 45, 4639–4642; (c) Cabello,
N.; Kizirian, J.-C.; Gille, S.; Alexakis, A.; Bernardinelli, G.; Pinchard, L.; Caille, J.-
C. Eur. J. Org. Chem. 2005, 4835–4842.
5. Fan, C.-A.; Ferber, B.; Kagan, H. B.; Lafon, O.; Lesot, P. Tetrahedron: Asymmetry
2008, 19, 2666–2677.
6. (a) Leroux, F.; Mettler, H. Synlett 2006, 5, 0766–0770; (b) Leroux, F.; Nicod, N.;
Bonnafoux, L.; Quissac, B.; Colobert, B. Lett. Org. Chem. 2006, 3, 165–169; (c)
Leroux, F. R.; Simon, R.; Nicod, N. Lett. Org. Chem. 2006, 3, 948–954; (d) Leroux,
F. R.; Mettler, H. Adv. Synth. Catal. 2007, 349, 323–336.
3. Conclusion
7. (a) Lipshutz, B. H.; Siegmann, K.; Garcia, E.; Kayser, F. J. Am. Chem. Soc. 1993,
115, 9276–9282; (b) Del Amo, V.; Dubbaka, S. R.; Krasovskiy, A.; Knochel, P.
Angew. Chem., Int. Ed. 2006, 45, 7838–7842; (c) Miyake, Y.; Wu, M.; Rahman, M.
J.; Kuwatani, Y.; Iyoda, M. J. Org. Chem. 2006, 71, 6110–6117.
The enantioselective Br–Li exchange is a promising new way to
introduce chirality. Although the enantioselectivity is still moder-
ate (63% at best) a better choice of substrates and ligands should
soon improve these results.
8. Krasovskiy, A.; Knochel, P. Angew. Chem., Int. Ed. 2004, 43, 3333–3336.
9. Perron, Q.; Alexakis, A. Tetrahedron: Asymmetry 2008, 19, 1871–1874.
10. General procedure: (A) To a clean and dry Schlenk tube were added diamine
(0.2 mmol) and 1.5 ml of toluene. Then the whole was cooled to ꢀ80 °C and n-
BuLi (0.2 mmol) was added dropwise to the mixture. The substrate (0.1 mmol)
diluted in toluene (0.5 ml) at rt was added dropwise to the previous mixture.
At the end of the reaction, DMF was added (0.4 mmol) and the mixture was
allowed to warm until rt. At rt, HCl 1 M was added and the organic phase was
separated. The aqueous one was extracted three times with dichloromethane.
The organic phases were mixed and dried over sodium sulfate and
concentrated on a rotatory evaporator. The diamine can be recovered after a
basic workup of the aqueous phase. The product was purified on column
chromatography on silica gel, and was eluted with ethyl acetate and
cyclohexane 1/9.
Acknowledgment
We thank the Swiss National Science Foundation (No. 200020-
113332) for financial support.
References
1. Marvel, C. S.; Hager, F. D.; Coffman, D. D. J. Am. Chem. Soc. 1927, 49, 2323–2328.
2. (a) Wittig, G.; Pockels, U.; Dröge, H. Ber. Dtsch. Chem. Ges. 1938, 71, 1903–1912;
(b) Gilman, H.; Langham, W.; Jacoby, A. L. J. Am. Chem. Soc. 1939, 61, 106–109;
(c) Gilman, H.; Moore, F. W. J. Am. Chem. Soc. 1940, 62, 1843–1846; (d)
Langham, W.; Brewster, R. Q.; Gilman, H. J. Am. Chem. Soc. 1941, 63, 545–548.
3. (a) Reuben, G. J.; Gilman, H. Chem. Rev. 1954, 54, 853–890; (b) Bailey, W. F.;
Patricia, J. J. J. Organomet. Chem. 1988, 352, 1–46; (c) Beak, P.; Musick, T. J.;
Chen, C.-W. J. Am. Chem. Soc. 1988, 110, 3538–3542; (d) Beak, P.; Allen, D. J. J.
Am. Chem. Soc. 1992, 114, 3420–3425; (e) Eisch, J. J. Organometallics 2002, 21,
5439–5463; (f) Nájera, C.; Sansano, J. M.; Yus, M. Tetrahedron 2003, 59, 9255–
(B) To a clean and dry Schlenk tube were added diamine (0.1 mmol), substrate
(0.1 mmol), and 2 ml of toluene. Then, the solution was cooled to ꢀ80 °C and n-
BuLi (0.1 mmol) was added dropwise to the mixture. At the end of the reaction,
drops of methanol were added. At rt the product was extracted with ether in
the presence of HCl 1 M. The organic phases were mixed and dried over sodium
sulfate and concentrated in
a rotatory evaporator. The diamine can be
recovered after a basic workup of the aqueous phase.